ICCA Volume 4

I nt er nat i onal Conf er ence on Comput er Appl i cat i ons 2012
Volume 4

I nt er nat i onal Conf er ence on Comput er Appl i cat i ons 2012
Volume 4
In association with
Association of Scientists, Developers and Faculties (ASDF), India
Association of Computer Machinery(ACM)
Science & Engineering Research Support society (SERSC), Korea

Human Computer Interaction, Information Communication Systems

27-31 January 2012
Pondicherry, India

Editor-in-Chief
K. Kokula Krishna Hari

Editors:
E Saikishore, T R Srinivasan, D Loganathan,
K Bomannaraja and R Ponnusamy

Published by

Association of Scientists, Developers and Faculties
Address: 27, 3
rd
main road, Kumaran Nagar Extn., Lawspet, Pondicherry-65008
Email: admin@asdf.org.in || www.asdf.org.in

International Conference on Computer Applications (ICCA 2012)
VOLUME 4

Editor-in-Chief: K. Kokula Krishna Hari
Editors: E Saikishore, T R Srinivasan, D Loganathan, K Bomannaraja and R Ponnusamy

Copyright 2012 ICCA 2012 Organizers. All rights Reserved

This book, or parts thereof, may not be reproduced in any form or by any means, electronic or mechanical, including
photocopying, recording or any information storage and retrieval system now known or to be invented, without written
permission from the ICCA 2012 Organizers or the Publisher.

Disclaimer:
No responsibility is assumed by the ICCA 2012 Organizers/Publisher for any injury and/ or damage to persons or
property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products
or ideas contained in the material herein. Contents, used in the papers and how it is submitted and approved by the
contributors after changes in the formatting. Whilst every attempt made to ensure that all aspects of the paper are uniform
in style, the ICCA 2012 Organizers, Publisher or the Editor(s) will not be responsible whatsoever for the accuracy,
correctness or representation of any statements or documents presented in the papers.

ISBN-13: 978-81-920575-7-6
ISBN-10: 81-920575-7-6

PREFACE

This proceeding is a part of International Conference on Computer Applications 2012 which was
held in Pondicherry, India from 27-Dec-2012 and 31-Dec-2012. This conference was hosted by
Techno Forum Research and Development Centre, Pondicherry in association with Association of
Computer Machinery(ACM), Association of Scientists, Developers and Faculties (ASDF), India,
British Computer Society (BCS), UK and Science and Engineering Supporting Society (Society),
Korea.
The world is changing. From shopping malls to transport terminals, aircraft to passenger ships, the
infrastructure of society has to cope with ever more intense and complex flows of people. Today,
more than ever, safety, efficiency and comfort are issues that must be addressed by all designers. The
World Trade Centre disaster brought into tragic focus the need for well-designed evacuation systems.
The new regulatory framework in the marine industry, acknowledges not only the importance of
ensuring that the built environment is safe, but also the central role that evacuation simulation can
play in achieving this.
An additional need is to design spaces for efficiency ensuring that maximum throughput can be
achieved during normal operations and comfort ensuring that the resulting flows offer little
opportunity for needless queuing or excessive congestion. These complex demands challenge
traditional prescriptive design guides and regulations. Designers and regulators are consequently
turning to performance-based analysis and regulations facilitated by the new generation of people
movement models.
When a greater changes are achieved these past years, still more is to be achieved which still seems
to be blue sky of 1970s. But for all the challenges, capabilities continue to advance at phenomenal
speed. Even three years ago it may have been considered a challenge to perform a network design
involving the evacuation of 45,000 people from a 120 story building, but with todays sophisticated
modelling tools and high-end PCs, this is now possible. Todays challenges are much more ambitious
and involve simulating the movement and behaviour of over one million people in city-sized
geometries. The management of these network is also easy and more specifically all the 45,000
people can be monitored by a single person sitting in his cabin. This has been the evidence of the
development these days.
As such, the conference represents a unique opportunity for experts and beginners to gain insight into
the rapidly.
Also I would like to thank all the co-operators for bringing out these proceedings for you which
majorly includes my mom Mrs. K. Lakshmi and my dad Mr. J. Kunasekaran. Apart from them my
biggest worthy gang of friends including Dr. S. Prithiv Rajan, Chairman of this conference, Dr. R. S.
Sudhakar, Patron of this Conference, Dr. A. Manikandan and Dr. S. Avinash, Convener of this
conference, Dr. E. Sai Kishore, Organizing Secretary of this Conference and the entire team which
worked along with me for the rapid success of the conference for past 1 year from the date of
initiating this Conference. Also I need to appreciate Prof. T. R. Srinivasan and his team of Vidyaa
Vikas College of Engineering and Technology for helping to make the publication job easy.
Finally, I thank my family, friends, students and colleagues for their constant encouragement and
support for making this type of conference.
-- K. Kokula Krishna Hari
Editor-in-Chief

Organizing Committee

Chief Patron
Kokula Krishna Hari K, Founder & President, Techno Forum Group, Pondicherry, India

Patron
Sudhakar R S, Chief Executive Officer(CEO), Techno Forum Group, Pondicherry, India

Chairman
Prithiv Rajan S, Chairman & Advisor, Techno Forum Group, Pondicherry, India

Convener
Manikandan A, Chief Human Resources Officer(CHRO), Techno Forum Group, India

Organizing Secretary
Sai Kishore E Chief Information Officer, Techno Forum Group, India.
Operations Chair
G S Tomar Director, MIR Labs, Gwalior, India

International Chair
Maaruf Ali Executive Director, (ASDF) - Europe, Europe

Hospitality
Muthualagan R Alagappa College of Technology, Chennai

Industry Liaison Chair
Manikandan S Executive Secretary, Techno Forum Group, India

Technical Panels Chair
Debnath Bhattacharyya, Executive Director, (ASDF) - West Bengal, India

Technical Chair
Samir Kumar Bandyopadhyay Former Registrar, University of Calcutta, India
Ponnusamy R President, Artificial Intelligence Association of India, India
Srinivasan T R,Vice-Principal, Vidyaa Vikas College of Engineering and Technology

Workshops Panel Chair
Loganathan D Department of Computer Science and Engineering, Pondicherry
Engineering College, India

MIS Co-Ordinator
Harish G Trustee, Techno Forum Research and Development Centre, Pondicherry

Academic Chair
Bommanna Raja K, Principal, Excel College of Engineering for Women, India
Tai-Hoon Kim Professor & Chairman, Dept. of Multimedia, Hanmam University, Korea

TECHNICAL REVIEWERS
Adethya Sudarsanan Cognizant Technology Solutions, India
Ainuddin University of Malaya, Malaysia
Ajay Chakravarthy University of Southampton, UK
Alessandro Rizzi University of Milan, Italy
Al-Sakib Khan Pathan International Islamic University, Malaysia
Angelina Geetha B S Abdur Rahman University, Chennai
Aramudhan M PKIET, Karaikal, India
Arivazhagan S Mepco Schlenk Engineering College, India
Arokiasamy A Anjalai Ammal Mahalingam Engineering College, India
Arul Lawrence Selvakumar A Adhiparasakthi Engineering College, India
Arulmurugan V Pondicherry University, India
Aruna Deoskar Institute of Industrial & Computer Management and Research, Pune
Ashish Chaurasia Gyan Ganga Institute of Technology & Sciences, Jabalpur, India
Ashish Rastogi Guru Ghasidas University, India
Ashutosh Kumar Dubey Trinity Institute of Technology & Research, India
Avadhani P S Andhra University, India
Bhavana Gupta All Saints College of Technology, India
Bing Shi University of Southampton, UK
C Arun R. M. K. College of Engineering and Technology, India
Chandrasekaran M Government College of Engineering, Salem, India
Chandrasekaran S Rajalakshmi Engineering College, Chennai, India
Chaudhari A L University of Pune, India
Ching-Hsien Hsu Chung Hua University, Taiwan
Chitra Krishnamoorthy St Josephs College of Engineering and Technology, India
Christian Esteve Rothenberg CPqD (Telecom Research Center), Brazil
Chun-Chieh Huang Minghsin University of Science and Technology, Taiwan
Darshan M Golla Andhra University, India
Elvinia Riccobene University of Milan, Italy
Fazidah Othman University of Malaya, Malaysia
Fulvio Frati University of Milan, Italy
G Jeyakumar Amrita School of Engineering, India
Geetharamani R Rajalakshmi Engineering College, Chennai, India
Gemikonakli O Middlesex University, UK
Ghassemlooy Z Northumbria University, UK
Gregorio Martinez Perez University of Murcia, Spain
Hamid Abdulla University of Malaya, Malaysia
Hanumantha Reddy T Rao Bahadur Y Mahabaleswarappa Engineerng College, Bellary
Hari Mohan Pandey NMIMS University, India
Helge Langseth Norwegian University of Science and Technology, Norway
Ion Tutanescu University of Pitesti, Romania
Jaime Lloret Universidad Politecnica de Valencia, Spain
Jeya Mala D Thiagarajar College of Engineering, India
Jinjun Chen University of Technology Sydney, Australia
Joel Rodrigues University of Beira Interior, Portugal
John Sanjeev Kumar A Thiagarajar College of Engineering, India
Joseph M Mother Terasa College of Engineering & Technology, India
K Gopalan Professor, Purdue University Calumet, US
K N Rao Andhra University, India
Kachwala T NMIMS University, India
Kannan Balasubramanian Mepco Schlenk Engineering College, India
Kannan N Jayaram College of Engineering and Technology, Trichy, India
Kasturi Dewi Varathan University of Malaya, Malaysia
Kathirvel A Karpaga Vinayaga College of Engineering & Technology, India
Kavita Singh University of Delhi, India
Kiran Kumari Patil Reva Institute of Technology and Management, Bangalore, India
Krishnamachar Sreenivasan IIT-KG, India
Kumar D Periyar Maniammai University, Thanjavur, India
Lajos Hanzo Chair of Telecommunications, University of Southampton, UK
Longbing Cao University of Technology, Sydney
Lugmayr Artur Texas State University, United States
M HariHaraSudhan Pondicherry University, India
Maheswaran R Mepco Schlenk Engineering College, India
Malmurugan N Kalaignar Karunanidhi Institute of Technology, India
Manju Lata Agarwal University of Delhi, India
Mazliza Othman University of Malaya, Malaysia
Mohammad M Banat Jordan University of Science and Technology
Moni S NIC - GoI, India
Mnica Aguilar Igartua Universitat Politcnica de Catalunya, Spain
Mukesh D. Patil Indian Institute of Technology, Mumbai, India
Murthy B K Department of Information and Technology - GoI, India
Nagarajan S K Annamalai University, India
Nilanjan Chattopadhyay S P Jain Institute of Management & Research, Mumbai, India
Niloy Ganguly IIT-KG, India
Nornazlita Hussin University of Malaya, Malaysia
Panchanatham N Annamalai University, India
Parvatha Varthini B St Josephs College of Engineering, India
Parveen Begam MAM College of Engineering and Technology, Trichy
Pascal Hitzler Wright State University, Dayton, US
Pijush Kanti Bhattacharjee Assam University, Assam, India
Ponnammal Natarajan Rajalakshmi Engineering College, Chennai, India
Poorna Balakrishnan Easwari Engineering College, India
Poornachandra S RMD Engineering College, India
Pradip Kumar Bala IIT, Roorkee
Prasanna N TMG College, India
Prem Shankar Goel Chairman - RAE, DRDO-GoI, India
Priyesh Kanungo Patel Group of Institutions, India
Radha S SSN College of Engineering, Chennai, India
Radhakrishnan V Mookamibigai College of Engineering, India
Raja K Narasu's Sarathy Institute of Technology, India
Ram Shanmugam Texas State University, United States
Ramkumar J VLB Janakiammal college of Arts & Science, India
Rao D H Jain College of Engineering, India
Ravichandran C G R V S College of Engineering and Technology, India
Ravikant Swami Arni University, India
Raviraja S University of Malaya, Malaysia
Rishad A Shafik University of Southampton, UK
Rudra P Pradhan IIT-KGP, India
Sahaaya Arul Mary S A Jayaram College of Engineering & Technology, India
Sanjay Chaudhary DA-IICT, India
Sanjay K Jain University of Delhi, India
Satheesh Kumar KG Asian School of Business, Trivandrum, India
Saurabh Dutta Dr B C Roy Engineering College, Durgapur, India
Senthamarai Kannan S Thiagarajar College of Engineering, India
Senthil Arasu B National Institute of Technology - Trichy, India
Senthil Kumar A V Hindustan College, Coimbatore, India
Shanmugam A Bannari Amman Institute of Technology, Erode, India
Sharon Pande NMIMS University, India
Sheila Anand Rajalakshmi Engineering College, Chennai, India
Shenbagaraj R Mepco Schlenk Engineering College, India
Shilpa Bhalerao FCA Acropolis Institute of Technology and Research
Singaravel G K. S. R. College of Engineering, India
Sivabalan A SMK Fomra Institute of Technology, India
Sivakumar D Anna University, Chennai
Sivakumar V J National Institute of Technology - Trichy, India
Sivasubramanian A St Josephs College of Engineering and Technology, India
Sreenivasa Reddy E Acharya Nagarjuna University, India
Sri Devi Ravana University of Malaya, Malaysia
Srinivasan A MNM Jain Engineering College, Chennai
Srinivasan K S Easwari Engineering College, Chennai, India
Stefanos Gritzalis University of the Aegean, Greece
Stelvio Cimato University of Milan, Italy
Subramanian K IGNOU, India
Suresh G R DMI College of Engineering, Chennai, India
Tulika Pandey Department of Information and Technology - GoI, India
Vasudha Bhatnagar University of Delhi, India
Venkataramani Y Saranathan College of Engineering, India
Verma R S Joint Director, Department of Information and Technology - GoI, India
Vijayalakshmi K Mepco Schlenk Engineering College, India
Vijayalakshmi S Vellore Institute of Technology, India
Ville Luotonen Hermia Limited, Spain
Vimala Balakrishnan University of Malaya, Malaysia
Vishnuprasad Nagadevara Indian Institute of Management - Bangalore, India
Wang Wei University of Nottingham, Malaysia
Yulei Wu Chinese Academy of Sciences, China

Part I
Proceedings of the Second International Conference on
Computer Applications 2012
ICCA 12
Volume 4
Proc. of the Second International Conference on Computer Applications 2012 [ICCA 2012] 1
www.asdf.org.in 2012 :: Techno Forum Research and Development Centre, Pondicherry, India www.icca.org.in
Proc. of the Intl. Conf. on Computer Applications
Volume 1. Copyright 2012 Techno Forum Group, India.
ISBN: 978-81-920575-7-6 :: doi: 10. 74201/ISBN_0768
ACM #: dber.imera.10. 74201
ABSTRACT- Human Computer interaction in the field
of input and output techniques has developed a lot of new
techniques over the last few years. With the recently released
full multi-touch tablets and notebooks the way how people
interact with the computer is coming to a new dimension. As
humans are used to handle things with their hands the
technology of multi-touch displays or touchpad's brought
much more convenience for use in daily life. But for sure the
usage of human speech recognition will also play an
important part in the future of human computer interaction.
This paper introduces techniques and devices using the
humans hand gestures for the use with multi-touch tablets
and video recognition and techniques for voice interaction.
Thereby the gesture and speech recognition take an
important role as these are the main communication methods
between humans and how they could disrupt the keyboard or
mouse as we know it today.

Keywords-Microsoft surface,Displax tm,GUI etc

I. INTRODUCTION
As mentioned before, much work in the sector of
human computer interaction, in the field of input and
output techniques, has been made since the last years.
Now since the release of the multi-touch tablets and
notebooks, some of these developed multi-touch
techniques are coming into practical usage. Thus it will
surely take not much time that sophisticated techniques
will enhance these techniques for human gesture or voice
detection. These two new methods will for sure play an
important role of how the HCI in future will change and
how people can interact more easily with their computer in
daily life.
Hewett, et al defined that "Human computer
interaction is a discipline concerned with the design,
evaluation and implementation of interactive computing
systems for human use and with the study of major
phenomena surrounding them. So since the invention of
the Human Computer Interface in the 1970s at Xerox Park,
we are used to have a mouse and a keyboard to interact
with the computer and to have the screen as a simple
output device. With upcoming new technologies these
devices are more and more converting with each other or
sophisticated methods are re placing them. Therefore this
paper mainly deals with these new developments, how
they should be implemented in future and how they could
influence and change the daily computer interaction.
For example with the techniques used for multi-touch
devices, described in section 2.1, the screen recently
becomes a new input and output tool in one device. In
doing so it is seen that there is no need for extra input
devices. Thus even this fact is a completely new way, as
we are used to having more than just one device. Included
in section 2.2 is another awarding method concerning
human gesture interaction in the field of detecting gestures
via video devices? In section 2.3 then another future
technique is pointed out which is concerned with the
human speech detection as an input method. Section 2.4
then deals with a method of using the combination of
video and speech detection. After these sections about the
different types of recent human computer interaction work
the section 3 then deals with the opportunities of these
new techniques and how they can be used in future
especially how life can be changed. It is then also pointed
out in which field these new developments can be adopted.

II. RECENT DEVELOPMENTS
Human gestures and the human speech are the most
intuitive motions which humans use to communicate with
each other. Al though after the invention of the mouse and
the keyboard, no further devices which
could replace this two objects as a computer input
method have been developed. Relating to the fact that it
are more human like methods of a lot of research of how
this fact can be used for communication between
computers and human beings has been done. As there are
some different ways of how gestures should be used as
input this section is divided into multi-touch, video,
speech and mutli-modal interaction Sections. Nowadays
there are already many tablets with touch screens available
and with the new Apple iPad a new full multi-touch
product has been released. But there is also a trend
noticeable that these methods can also be used on bigger
screens like the approach of Miller or Microsoft's Surface,
only to mention these two. Thus the trend is going more
and more not only in the direction of merging input and
output devices but rather using every surface as an input
and output facility. In this relation the video is becoming
of greater interest as it uses the full range of human motion
gestures and it is usable on any ground. In the end the
input via speech also takes a part in the HCI as it is the
human's easiest way to communicate, but for sure
something completely different, compared with the other
two types of input and output methods, as it is more an
algorithm than a device. The combination of different
Future Human Computer Interaction with special focus on input and
output techniques
P.A.Sathishkumar
Department: Computer Science and Engineering
Organization: Chandy college of Engineering Tuticorin,India

input methods, called multi-modal interaction, is then
described in the last Section.

B. Multi-Touch Devices
As mentioned before this section is dealing with the
technique of the recently released multi-touch devices and
with some new enhanced approaches. This method is now
becoming common in the tablet pc's like for example the
new Apple iPad in the sector of notebooks and the HP
Touch Smart in the desktop sector. Thereby the screen
becomes an input and output device. But multi-touch is
also used today in many normal touch pads which are
offering four-finger navigation. With this invention of a
new human computer interaction many more work in this
sector has been done and should sooner or later also come
into practical use. Nowadays the usage of touch screens
and multi-touch pads seems to be really common and that
this is going to be the future of human computer
interaction, but there is for sure more enhancements which
can be seen in many approaches. In the field of multi-
touch products the trend to bigger touch pads in terms of
multi-touch screens can be seen. Therefore the technique
of a single-touch touchpad as it is known from former
notebooks is enhanced and more fingers offering natural
human hand gestures can be used. Thus the user can use
up to 10 fingers to fully control things with both hands as
with the 10/GUI system which R. Clayton Miller [2]
introduced in 2009. With another upcoming tool even the
usage of any surface for such a touch screen can be used in
future, like for the example DisplaxTMMultitouch
Technologyfrom DISPLAXTMInteractiveSystems. From
these examples it can be clearly seen that new high
potential techniques are pushing into the market and are
going to replace the Apple's iPad and Microsoft's Surface.
In the next following sections these new tools and also the
related devices which are already in the market are
described in detail.

A. iPad
The recently introduced iPad from Apple is one of the
many implementations of full multi-touch displays
whereas it is a completely new way how people can
interact with their computer. Thus the possibility to use the
screen not only as a single-touch display is taking the
Human Computer Interaction to the next level. With the
introduced iPad it is possible to use all the finger
movements which are also possible with the build-in multi
touchpad's in the Apple MacBooks that can be found on
the Apple homepage. In doing so the user is able to use up
to four fingers at the same time to navigate through the
interface. For example two fingers can be used to zoom
and four fingers to browse through windows. With using
the screen as a big touchpad the techniques of the normal
touchpad have been enhanced. Although this technique is
just the beginning of the new multi-touch display
revolution which will be for sure expanded by increasing
the display size.
Apples Homepage
http://www.apple.com

B. Microsoft Surface

Microsoft's called their example of a multi-touch
screen just simply "Surface". With this tool they generated
a large touch screen tabletop computer. Therefore they are
using infrared cameras for recognition of objects that are
used on this screen. An object cans thereby the human
fingers or even other tagged items which can be placed on
the screen. With this opportunity the device supports
recognition of human's natural hand gestures as well as
interaction with real objects and shape recognition. With
this opportunity no extra devices are required for a usage
of this tool and interaction can be made directly with the
hands. With the large 30 inch display even more people
then only one person at the same time can interact with the
system and with each other at the same time. The
recognition of the objects placed on the table top pc then
provides more information and interaction. So it is perhaps
possible to browse through different information menus
about the placed item and obtain more digital information.
On the other hand the size of the display and the therefore
needed infrared cameras underneath is leading to an
increase of size. Thus the tool is mainly designed for
stationary use, like for example as a normal table with
which it is then possible to interact with.
C. 10/GUI
The 10/GUI system which Miller invented is an
enhanced touchpad for desktop computer purpose which
can recognize 10 fingers. With this opportunity human
beings can interact with the computer with both hands and
use it as a tracking and maybe also as a keyboard device.
Therefore Miller designed this new touch surface tool
especially for use in the desktop field.
To have the best ergonomically position he argues that
it is better to have a full multi-touch pad in front of a
screen as the keyboard and mouse replacement then
having the whole screen as an input device, like it is
known from other touch-screens. This novel system is a
whole different method of computer interaction and is
extended with a special graphical user interface for a full
use of all 10 fingers. The most remarkable thing about this
touchpad except the recognition of more fingers is also the
pressure detection of every finger, which is directly stated
on the screen. Therefore every finger can be used as a
pointing device like the mouse. With this feature it can be
also used in future as a keyboard in the daily used position
of the hands, without the need of selecting Letters with
only one finger. But for sure this first development of the
10/GUI is mainly concerned with having the 10 fingers act
instead of the mouse," as many activities today need only a
mouse and windowed information display on the top of
that..."
The other innovation of this system is its special
designed user interface for the usage of these 10 fingers.
Thereby Miller is proposing a way to solve the problem of
multiple open windows. His solution therefore is a linear
arrangement of the active windows where it is possible to
browse throu
right side of t
Figure 1: Di
Touch sur
C
Another
comes from t
They have in
into an intera
goal they ar
transparent
DisplaxTMM
paper they ar
50 inch big
work directly
This can be
Microsoft Su
stationary pla
of 16 fingers
user can wor
weight of jus
beside the fac
on the back o
other damage
Figure 1
surface wher
possible.
Great effo
video input
SixthSense th
Mistry, et
such devices
interaction w
screens or -
recognize hum
of additional
approach of F
algorithm for
their paper w
as a pointing
[5] have imp
development
only an input
have an impa
gh them with
the touch-pane
splax's thin transp
rface [4]
C. DisplaxTMM
forward-look
the Future La
nvented a new
active multito
re using in
paper tha
Multitouch co
re then able to
touchscreen.
y on a big sc
seen as the
urface tool w
ace. Additiona
at the same
rk on the sc
st 300g it is
ct that it is w
of the surface
e.
illustrates a
e even the u
D. V
orts have bee
and outpu
hat Pranav
t al. invented
thus lies in th
with the com
-pads. There
man gestures
handheld poi
Feng, et. al [6]
r real-time na
with which it i
device like t
proved this in
of a wearable
t but also an
act on the futu
h 2 buttons on
el.
parent multi-
Multitouch Te
king multi-t
abs of the Dis
w method to t
ouch surface.
their approa
at is att
ntroller. Wit
o turn any sur
With this p
creen by just
e main adva
which is tied
ally this interf
time so that m
creen simultan
also a very
ell durable as
to protect it f
detail of the
sage on trans
Video Devices
en made also
t devices l
in 2009. The
he potential of
mputer then w
eby these te
like the hand
inting devices
], they have a
atural hand ge
is possible to
the mouse. Pr
nput pointing
e gesture inte
output tool w
ure human com
n the outer lef
echnology
touch techno
splaxTMComp
urn "...any su
" To achieve
ach an very
tached to
th this ultra
rface into an
ossibility you
using your h
antage agains
to a more a
face allows a u
more than jus
neously. With
transportable
s the film is p
from scratche
thin touch-s
sparent surfac
in the sectio
ike for exa
e main purpo
f having even
with normal t
chniques ten
ds without the
s. According t
lready provid
esture detectio
use only the
ranav Mistry,
g method with
erface which i
which will for
mputer interac
ft and

ology
mpany.
urface
e this
thin
the
thin
up to
u can
hands.
st the
a less
usage
st one
h the
e tool
placed
es and
screen
ces is
on of
ample
ose of
more
touch
nd to
need
to the
ded an
on in
hand
et. al
h the
is not
r sure
ction.
With
inter
F
P
clear
comp
I
inter
HCI
hand
outp
this
direc
the m
a ce
comm
the
Ther
for s
T
proje
outp
they
other
it rec
addit
direc
T
hand
E
gestu
gestu
Carn
anoth
as an
hin this sectio
raction are now
Figure 2: Compon
Pranav Mistry
rly the begin
puter interacti
n their appro
rface, shown
to the next l
ds without any
ut of the devi
device. With
ctly projected
main advantag
ertain place.
munication w
users input a
refore the inp
sorting or editi
Thereby the i
ected over the
ut and input
have also inc
r things then j
cognizes that
tional media
ctly on the new
The above m
dling with hum
Especially it
ures and let
ures over a
negie Mellon U
her highly top
n input surface
on these new
w explained.
nents of the Sixth

2.2.1 Six

y, et al. [5] d
nning of ne
ion.
oach they are
in Figure 2,
evel where co
y handheld to
ice for this int
h a simple sm
to any surfac
ge against oth
So they ar
with the comp
and a simple
ut can range
ing images to
input can also
e small beame
at the same t
cluded in the v
just on the hum
the user is rea
to this topic
wspaper.

III. SKIN

mentioned v
man motion ge
is dealing w
tting people
certain outp
University and
pical approach
e, called Skinp
w methods of
h Sense device

xth Sense

developed wit
ew technique
e using a we
where they a
ommunication
ools. Just as e
terface which
mall projector
ce you like, w
her devices wh
re using for
puter a simpl
e projector fo
from simple
more specific
o be a keybo
er and so it is
time. With th
video recognit
uman hand ges
ading a newsp
c by projecti

NPUT

video recogn
estures.
with recogniz
interact wit
put device. R
d Microsoft h
h by using th
put.
f handling the

h Sixth Sense
es for human
arable gesture
are taking the
n is done with
asy is also the
is included in
the output i
which is clearly
hich are tied to
r their whole
e webcam fo
or the output
hand gesture
c tasks.
oard which i
s fulfilling the
his opportunity
tion to react to
stures. Perhap
paper and add
ing the media
nition tool i
zing the hand
th their hand
Researchers a
have developed
e humans arm
e
e
n
e
e
h
e
n
s
y
o
e
or
t.
s
s
e
y
o
s
ds
a
s
d
d
at
d
m
V
A
a
t
i
S
a
w
N
c
i
n
e
c
e
w
i
e
a
4
t
s
e
p
e
d
i
s
Proc. of th
Volume 1. Co
ISBN: 978-8
Abstract T
artificial brain
the human b
interface (BCI
Special propri
a new technol
with a group o
Keywords-A
Nanotechnolog
rain hack
cure me
controlle
A b
called a dir
interface, is a
numan or an
external devic
commands fr
example, to r
would allow
information in
Neur
electro chemic
are many
4microns(,004
their length va
Quan
semiconducto
electronic pro
point of culmi
.Thei
electromagnet
devices.
.
II.
T
he brain a
including neu
send and rece
B
he Intl. Conf. o
opyright 201
81-920575-7-6
ACM #: dber
D
This paper co
n cell and its f
rain, and a l
I) which carr
iety of nanotec
logy that whic
of such artificia
Artifical brain
gy.
I. I
king is just li
ntally disable
ed cars, house
brain-computer
rect neural i
a direct comm
nimal brain (a
ce. In one-way
rom the brai
restore vision
brain and
n both directio
rons and brai
cal signals to b
type of n
4mm) to 100m
aries from one
ntum dots,
or matrix w
oprieties, pres
ination of sem
ir proprites re
tic cell, rende
BRAIN CELL
and spinal co
urons and glia
eive electro-ch
on Computer A
12 Techno For
6 :: doi: 10. 74
r.imera.10. 742
M
Department of
Sree V
onsist of bas
function while
little bit abou
ries informatio
chnology opens
ch control the
al cells.
cell; Brain-Co
INTRODUCTION

ike hypnotism
e person, for m
etc
r interface (
interface or
munication pa
and brain cel
y BCIs, comp
in or send s
n) but not bot
external devi
ons.
in cells that
brain from ner
neuron they
microns(.1mm
e inch to sever
coherent i
hich truly z
sent the utmo
mi conductor ph
esembles thos
ering possible
L AND ITS FU
ord are made
al cells. Neur
hemical signal
Applications
rum Group, In
4208/ISBN_07
208
Mind Contr
f Computer Sc
Vidyanikethan
sic functionin
it is injected
ut brain comp
on to our sys
s a new gate w
brain of a hu
onmputer interf
N
m which is use
military and m
(BCI), somet
a brain-mac
athway betwe
ll culture) an
puter either ac
signals to it
th, two-way B
ices to exch
send and rec
rvous system,
vary in
m) in diameter
ral feet.
inclusions in
zero dimensi
ost challenge
hysics
se of atoms i
fascinating n
UNCTIONING
up of many c
rons are cells
ls to and from

ndia.
768
roller (Br
A.Sunil Sa
cience and Sys
Engineering C
Tirupati, Indi

ng of
in to
puter
stem.
way to
uman
rface:
ed to
mind
times
chine
een a
d an
ccept
(for
BCIs
hange
ceive
they
size
r and
n a
ional
and
in an
novel
cells,
that
m the
br
ne
pr
nu
mi
Th
fee
.
to
co
(si
co
ax
a
rec
Ge
W
"c
tak
ax
ne
br
pe
tra
by
20
rain Hack
ai
stem and Syste
College [Auto
ia.
rain and nerv
eurons in the b
rovide support
umerous than n
There is muc
icrons (.004 m
heir length va
et
Neuron
and from th
onsists of a ce
ignal receiver
onduct the ner
xon terminals
synapse (the
ceiving cell).
erman scien
Waldeyer-Hartz
chromosome")
The ax
ke informatio
xons are know
ervous system
ring informatio
Myelin
eriodic break
ansmission spe
y Schwann's c
0-30% protein
king)
ems Engineeri
onomous]
vous system.
brain. There ar
t functions for
neurons.
h type of neur
mm) to 100 m
aries from a f
Neuron (B
ns are nerve ce
he brain at u
ell body (or so
rs) and a proj
rve signal. At
transmit the e
gap between
The word "
ntist Heinrich
z in 1891
).
xon, a long ex
on away from
wn as nerves
m), as nerve t
on to the cell b
n coats and in
s called nod
eed along the
ells, and cons
.
ing,
There are ab
re many more
r the neurons,
rons. They var
microns (.1 m
fraction of an
Brain cell)
ells that transm
up to 200 mp
oma) with bran
ection called
the other end
electro-chemic
n the axon te
"neuron" was
h Wilhelm
(he also co
xtension of a
m the cell bo
or, within th
tracts or pathw
body.
nsulates the a
des of Ranv
axon. Myelin
ists of 70-80%
bout 100 billi
e glial cells; th
and are far mo
ry in size from
mm) in diamet
n inch to seve
mit nerve sign
ph. The neur
nching dendri
an axon, wh
of the axon, t
cal signal acro
erminal and t
s coined by t
Gottfried v
oined the te
a nerve cell, a
ody. Bundles
he CNS (cent
ways. Dendri
axon (except
vier), increasi
is manufactur
% lipids (fat) a
ion
hey
ore
m 4
ter.
eral

nals
ron
ites
ich
the
oss
the
the
von
erm
and
of
tral
ites
for
ing
red
and
n
D
s
n

c
(
c

The
nucleus (with
Dendrites bran
A ty
synapses (that
neurons, musc
III.D
Ther
carry electro-c
(the number
cell body) and
a.Sen
mess
ears,
proce
neuro
epith
b.Mo
signa
Thes
from
of a
neuro
c.Inte
from
have
dend
cord;
neuro
root g
a.LIFE SP
Unlike
after damag
Fortunately
brain.
b.GLIAL C
Glial cells m
cells are ne
various gli
important fu
neurons, m
physical an
cell body (
h DNA and
nch from the c
ypical neuron
t is, it commu
cle cells, gland
DIFFERENT T
e are differen
chemical nerv
of processes,
d are found in
nsory neuron
sages from th
etc.) to the
esses. Sensory
ons. (Exampl
helium cells.)
otoneurons o
als from the C
e neurons ha
m the cell body
all neurons.
ons, pyramida
erneurons or
m all the neura
two axons
drite). One axo
; one with eit
ons have two
ganglia cells.)
PAN OF NEUR
e most other
ge (except ne
y, there are ab
CELLS
make up 90 p
erve cells that
ial (meaning
functions, inclu
manufacturing
nd nutritional
(soma) conta
d typical nu
cell body and r
n has about
unicates with 1
ds, etc.).
TYPES OF NE
nt types of n
ve signals but
or axons, em
different parts
ns or bipola
e body's sens
CNS. These n
y neuron accou
les are retina
or Multipolar
CNS to the mu
ave many pro
y. Motoneuron
(Examples a
al neurons, Pur
Pseudopolar
l wiring withi
(instead of
on communica
ther the skin
processes. (Ex
)
RONS
cells, neuron
eurons from t
bout 100 billi
percent of the
don't carry ne
"glue") cell
uding: digestio
myelin for n
support for n
ains the neu
clear organe
receive messa
1,000 to 10
1,000-10,000 o
EURONS
neurons. They
differ in struc
manating from
s of the body.
ar neurons c
se receptors (e
neurons have
unt for 0.9% o
al cells, olfac
r neurons c
uscles and gla
ocesses origina
ns account for
are spinal m
rkinje cells.)
re (Spelling)
in the CNS. T
an axon an
ates with the sp
or muscle. T
xamples are d
ns cannot reg
the hippocamp
on neurons in
brain's cells. G
erve impulses.
ls perform m
on of parts of
neurons, provi
neurons, and m
uron's
lles).
ages.
0,000
other
y all
cture
m the

carry
eyes,
two
of all
ctory
carry
ands.
ating
r 9%
motor
cells
These
nd a
pinal
These
dorsal
grow
pus).
n the
Glial
. The
many
dead
iding
more.
ce
fo
a c
to
ce
br
of
a.N
sy
wo
pr
us
co
I
of
si
fa
po
It
so
Types of gli
Cells, Microg
Neuroglia (m
ell. These cells
III. ROL
After injuncti
r a right move
cell which is i
the brain. By
ells in to the hu
rain can do di
f brain cells.
Nanotechnolo
Nanotechno
ystems at the
ork and conce
In its origi
rojected ability
sing technique
omplete, high p
I want to bu
f each other
multaneous
ar as I can s
ossibility of
is not an at
omething, in
al cells inclu
glia, Oligoden
meaning nerv
s guide neuron
LE OF ARTIFIC
BR
ion of artifici
ement to enga
in damages sta
y this way we
uman brain be
ifferent operat
IV. QUA
ogy
logy is the
molecular sca
epts that are m
nal sense, 'n
y to construc
es and tools be
performance p
uild a billion
r, which are
sly. . . The pr
ee, do not sp
f maneuverin
ttempt to vio
n principle,
de Schwann's
droglia, and A
ve glue) is th
ns during fetal
CAL BRAIN CEL
RAIN

al cell in to th
age with the br
ate. Then it di
e attach severa
ecause various
tions so we n
ANTUM DOT

engineering
ale. This cov
more advanced.
nanotechnology
ct items from
eing develope
products.
n tiny factor
manufactur
rinciples of
peak agains
ng things at
olate any law
that can be
s Cells, Satell
Astroglia.
he type of bra
development
LL IN HUMAN
he brain it wa
rain by replaci
irectly connect
al artificial br
s parts of hum
eed few numb
g of function
ers both curr
.
y' refers to t
the bottom u
ed today to ma
ries, models
ring
f physics, as
st the
tom by atom
ws; it is
done; but in
lite
ain
.
aits
ing
ted
ain
man
ber
nal
ent
the
up,
ake

m.
n
p
b
p

u
a
q

d
c
l
a
Q
c
s
p
q
I
q
p
c
e
t
practice, it h
big
Richa
physics

a.Creation
If we
using top dow
and cut it do
quantum dot a
Quan
dimensions (z
characters. W
like cube, sp
artificial atom
b.Quantum
Quantum dot
currently the
sulphide, lead
phosphide. M
quantum dots
In order to
quantum dot
polymer.
Quant
We d
concentric qua
excellent rota
technique. V
has not bee
ard Feynman
n of Quantum
e want to crea
wn fabrication
wn until it be
as small as (5-
ntum dot is
zero) and nano
We can form th
phear and py
ms.
m Dot Materia
s can be ma
most common
d sulphide, c
Many of the
will see them
avoid toxic
s, they are
tum dots with its l
demonstrate th
antum double
ational symm
arying the g
n done beca
n, Nobel Pr

nano dot

ate quantum d
process that w
ecame an ato
-50) nm in size
a confineme
o in size can
his quantum i
yramids; this
als
ade from a ra
nly used mate
cadmium sele
e promising
m used within
materials le
also coating
layers appears as
he self-assem
e rings with hi
etry using th
growth proces
ause we are
rize winner i
dot its possibl
which take a
om. We make
e.
ent in all sp
give some sp
in different sh
is also calle
ange of mater
erials include
enide and ind
applications
n the human b
eaching from
g in a prote
a rings
mbled formatio
igh uniformity
he droplet epi
ss conditions
too
in
le by
solid
e this
pecial
pecial
hapes
ed as
rials,
zinc
dium
for
body.
m the
ctive
on of
y and
itaxy
can
co
fro
qu
tra
the
the
se
A.
eit
ph
wi
oc
ne
sa
wi
on
Ad
rea
tec
the
re
no
su
of
rec
sh
ne
the
ontrol each rin
om an individ
uantized level
ajectories.
We can cre
ey are as follo

1) Heterojun
2) Photovol
3) Devicer c
a) Electron
b) An activ
c) Adjacen
d) Electrod
e) Hole co
These
ey can rearran
lf if any dama
V.
Reading the
a.Neuroimagi
Neuroimagin
ther directly o
harmacology o
ithin medicine
b.Direct Neur
This is the m
ccuring in the
euron in the b
ay that this me
ith respect to
ngoing
dvances in na
ality.
c.Electroencep
This procedu
chnique disco
e brain. Due
solution this m
owadays. Ho
usceptible to n
f user training
cent research
hown that this
etworks to shi
e computer.
c.Magnetoenc
ng's size. Photo
dual quantum
s that are spe

eate different l
ows
nction
ltaic
compreses
n conductive la
ve photo voltic
nt electro cond
de layer
nductive layer
quantum dot
nge their struc
ages occur.
BRAIN CO
brain Neuro
ing definition
ng includes th
or indirectly im
of the brain. It
e and neurosci
ral Contact
most accurate m
e brain as it
rain, e.g. by u
ethod is highl
o our current
anotechnology
ephalography
ure is the firs
overed. It mea
to its ease of
method is the m
owever, in
noise and thus
in order to b
at the Fraunh
problem can
ft the learning
cephalography
oluminescence
m ring complex
ecified by the
layers in a sin
ayer
c layer
ductive layer
r

are self-assem
cture and they
MPUTER INGE

oimaging in B

he use of vario
mage the struct
t is a relatively
ience. [WIKI

method of reco
has direct c
using nanorob
ly invasive an
t technology.
y this method

(EEG)
st non-invasiv
asures the elec
f use, cost an
most widely u
practice EEG
s require a sig
be operable in
hofer Society [
be overcome
g overhead fro

y (MEG)
e spectra emitt
x show pecul
e carriers' orbi
ngle quantum d
mbled that whi
y can repair th
RFACE
BCIs
ous techniques
ture, function,
y new discipli
I_NI]
ording potenti
contact to eve
bots. Needless
nd impracticab
However, w
d might becom
ve neuroimagi
ctrical activity
nd high tempo
used one in BC
Gs are high
gnificant amou
a BCI. Lucki
[WIKI_BCI] h
by using neu
om the human
ted
liar
ital
dot
ich
em
s to
, or
ine
ials
ery
s to
ble
with
me
ing
y of
oral
CIs
hly
unt
ily,
has
ural
n to
t
t
t
b
r
f
c
K
(
a
R
t
b
t
s
m
e
a
B

Though sim
technology th
technology. In
the brain this
by it. The mai
requirements
filled with s
cooling heliu
[NA_BCI] B
Kollotzek / H
c.Functiona
This techni
(blood flow
activity in th
Resonance Im
as Magneti
that there is a
brains haemo
tool. In contr
structure this
method requi
equipment and
D.Direct I
Pic.
In the fie
approaches to
Many of th
But for some
milar to the E
h MEG is a m
nstead of mea
technology re
in drawbacks o
in equipment
uper-conducti
um tanks sur
Brain Comput
ske
al Magnetic Re
ique measure
and blood o
he brain by
maging formerl
ic Resonance
a correlation b
odynamics ma
rast to the M
method studi
ires MRI tech
d thus is quite
Interfaces via E
Intro
.6 : Test perso
eld of Body
an interface d
hese are in fac
people who e
EEG in that it
much newer a
asuring the ele
ecords magnet
of this technol
. Using MEG
ing magnets
rrounded by
ter Interfaces

esonance Imag
es the haemod
oxygenation)
the use of
ly known
Tomography
between neura
akes the fMR
MRI which st
ies the brains
hnology it ne
costly.

EEG
oduction
on attached to
y-Computer-In
design have ev
ct haptic or lin
experienced a
t is a non-eva
and more accu
ectrical activi
tic fields prod
logy are its hig
G requires a r
and giant su
shielded wa
s - 6 - Beh
ging (fMRI)
dynamic resp
related to ne
f MRI (Mag
y MRT). The
al activity and
RI a neuroima
tudies the br
s function. As
eeds very sp

an EEG
nterfaces mul
volved.
nguistic interf
cerebral apop
asive
urate
ty in
duced
gh
room
uper-
alls.
hm /
ponse
eural
gnetic
fact
d the
aging
rains
s this
pecial
ltiple
faces.
plexy
an
the
di
br
In
ex
Fo
di
fu
fro
ce
of
thi

nd are locked
e operation o
fferent techni
rain via an ele
n the following
xamples are pr
a.Advantage
This is mostl
or example: if
fficult to foun
urther. By usin
om a terrorist

d.Disadvant
May be it ca
ells is less than
f his control an
Hear I refer s
is paper and m
[1] http://mec
uantumdo
[2] http://ww
792&id=R
wiki&prin
ot%20wik
[3] http://ww
m-Kollotz
[4] http://ww
[5] http://boo
&oi=fnd&
MNzM&s
FSUHdG
20dot%20
d-in, trapped
f those device
ique the dir
ectro-encephal
g four differe
resented and c
VI. ACKNO
es
ly used to chan
f police men
nd what is in h
ng this techno
to a police me
tages
an turn a huma
n artificial bra
nd do whateve
REFER
several books
my guide this i
chse.illinois.edu/m
ots.20100425.4bd
ww.google.co.in/pa
Rj7nAQAAEBAJ
ntsec=abstract#v=
ki&f=false
ww.ics.uci.edu/~ab
zek-Hueske_BCIs
ww.azonano.com/a
oks.google.co.in/b
&pg=PR9&dq=qu
sig=-
0IsFIWt3EhInq7A
0definition&f=fal
in their own
es is impossib
rect commun
logram (EEG)
ent techniques
ompared subs
OWLEDGMENT
nge the minds
caught a terr
his mind and w
ology we chan
en
an in to a robo
ain cell then th
er the input is g
RENCES
and mostly w
is the only refe

media/uploads/cou
d4c1c5be8fe7.734
atents?hl=en&lr=
J&oi=fnd&dq=qu
=onepage&q=qua
behm/class_report
s.pdf
article.aspx?Artic
books?hl=en&lr=&
uantum+dot+defin
AYOyUk#v=one
lse
body so to s
ble. Therefore
ication with t
) is applied he
and applicati
sequently.
T
set of a person
rorist, it is ve
what he is doi
nge his mind
ort, if no of br
hat person is o
given
websites to cre
erence
urse_websites/lee
485281.pdf
=&vid=USPAT79
uantum+nano+dot
antum%20nano%2
ts/stuttgart/bci/Be
leID=1814
&id=2dcNOpIRfJ
nition&ots=47eBn
page&q=quantum
ay,
e, a
the
ere.
ion
ery
ing
set
ain
out
eate
e_q
968
t+
20d
eh
JkC
n6
m%
ISBN: 978-81-920575-7-6 :: doi: 10. 74215/ISBN_0768
Goggle, NOT Google
A realistic approach towards the real world interaction

Abhishek Joshi
Shri G.S Institute of Technology & Science, Indore
SGSITS
Indore, India
Alok Panchal
Institute of Engineering and Technology, Indore
IET- DAVV
Indore, India

AbstractResearchers have found several means of putting the
idea of REAL WORLD INTERACTION into practical. But
most of them are purely an artistic approach with no real and
"practical" use. Projection based technologies consume higher
power and it is clear that using such technologies to interact
with the real world objects is NOT the way to a brighter world.
To overcome all these barriers we started researching on
optical theory, Bluetooth, infrared emitters and photo-diodes
(receivers). Our research work does not require any
projection. It is based on pure circuitry and sensor based
tracking instead of image processing. Hence we get precise
values with least power consumption. Also no background is
needed for interaction with the system. We came through some
fascinating results during their practical implementation. With
further improvements, on addition of infrared emitters,
collector (photo-diode) strips; the system now is able to track
the finger movements and work accordingly. Where fingers
work like any other infrared reflector; the reflected infrared
lights are sensed by the system.
Keywords- Real world Interaction, touchscreen (key words)
I. INTRODUCTION
Mobile phones, laptops, digital cameras are part of our
day to day life. These technologies can be considered as one
of the most promising invention of the twentieth century.
They play a major role in personal and professional life.
Communication with these devices has improved
significantly. For example touch-screens have taken place of
keypad based navigation; hence providing better interaction
with the system (not limited to cell phones). Amplified
interaction is one of the primary concerns of these devices.
Combination of different functionalities, services and
applications such as camera and document viewer in a cell
phone, one-click upload feature of digital cameras and other
such features attract users from every group.
The question persists and indeed grows whether the
computer will make it easier or harder for human beings to
know who they really are, to identify their real problems, to
respond more fully to beauty, to place adequate value on life,
and to make their world safer than it now is[1].
Most users find these devices (gadgets in this context)
difficult to operate, carry, and use effectively. Interaction
with the system is though enhanced but there are only few
applications which enhance the real time decision making
capability, virtualization and performance of the user. Our
focus is primarily on combination of several technologies
and applications to achieve goal of real world interaction
being as practical as possible.
IV.CURRENT SCENERIO & NEED
Researchers have known for years that pointing, clicking
and dragging are not ideal forms of interaction for many
tasks. Its hard to draw a flower or sign name using a mouse.
Similarly, keypad based navigation in cell phone has already
been replaced by touch-screens. Even after the introduction
of touch screen, cell phone size has gradually increased to
provide better interaction and display but they require great
care due to fragile nature. Users apparently make sure that
their fingers are clean before touching them, consciously
consider every move they make whilst holding said precious
gadget, swath their gadgets in all manner of protective case
and screen solutions; how well these products justify the
basic principle of good design.
Good design is as little as possible. Less, but better
because it concentrates on the essential aspects, and the
products are not burdened with non-essentials [2].
Today industrys prime focus is how well human
interacts with the machine and not how well machine
proves its usefulness in real world or how well the
machine survives in the real world conditions.
Digital technologies will continue to proliferate, enabling
ever more ways of changing how we live. But will such
developments improve the quality of life, empower us, and
make us feel safer, happier and more connected? Or will
living with technology make it more tiresome, frustrating,
angst- ridden, and security-driven?
V.EXISTING RESEARCHES AND THEIR IMPERFECTIONS
A.The Sixth Sense
The Sixth Sense prototype is comprised of a pocket
projector, a mirror and a camera. The hardware components
are coupled in a pendant like mobile wearable device. Both
the projector and the camera are connected to the mobile
computing device in the users pocket. The projector projects
visual information enabling surfaces, walls and physical
objects around us to be used as interfaces; while the camera
r
o
s
b
(
u
a
g
a
h
t
b
c
s
s
t
t
T
g
r
p
i
a
p
i
g
d
r
o
s

recognizes an
objects using
software prog
by the camera
(visual trackin
using simple
and arrangem
gestures that a
application int
The Sixth
hence consum
that the cell p
beam reporte
capability. [4]

An object
secure than th
started from se
to human body
The prime
that user is alr
There are a nu
goggles, mob
research devi
powerful enou
input, proce
accordingly. I
people use go
including GOG
glasses as co
digital camera
LCD mon
removal it ma
of using transp
seemed to be p
Following

Figu
nd tracks use
g computer-v
gram processes
a and tracks the
ng fiducials)
computer-vis
ments of thes
act as interact
terfaces.[3]
h Sense work
mes high amou
phones with pr
edly have po

II.THE
closer to hum
he object prese
earching the e
y; so that user
e focus of our
ready habited
umber of devi
bile phones, P
ice should b
ugh to carry
ss them pr
It took no lon
oggles. We b
GGLE into it.
mbination of
a or any other P
nitors use a
kes the LCD a
parent TFT in
practical here.
images show
ure 1. Inner view
Figure 2. Glos
er's hand ges
vision based
s the video st
e locations of
at the tip of
ion technique
e fiducials a
tion instruction
ks on the pro
unt of energy.
rojection capa
oor battery li
RESEARCH
man body is
ent at a distan
existing objects
can operate th
research was
of; so as to al
ces, such as:
PDA, mp3 p
be portable
out tasks suc
recisely and
ng to find that
broadened our
The basic ide
devices such
PDA device.
thin layer of
almost transpa
nstead of regul

the research in
w of TFT; opened
sy sheet and origi
tures and ph
d techniques.
tream data cap
the colored m
f the users f
es. The move
are interpreted
ns for the proj
ojection techn
It should be
ability like Sam
ife and proj

more persona
nce. So our res
s that are very
he device easil
to develop a d
llow quick lea
digital wrist w
players. The
and compact
ch as reading
generate o
t a large numb
r research wo
ea was to use g
h as mobile p
f glossy shee
arent; hence th
ar glasses in g
n action.

d in the lab.

inal tft
hysical
The
ptured
markers
fingers
ements
d into
ojected
nique;
noted
msung
ection
al and
search
y close
ly.
device
arning.
watch,
aimed
t; yet
user's
output
ber of
ork by
goggle
phone,
et; on
he idea
goggle
regu
bett
key
inte
emi
that
wit
usin
are
F

On attachme
ular display.
Alternatively,
ter transparenc

Figu
To interact w
yboard, but ou
eract with the
itters and cam
t the user doe
th the Operatin
ng mouse or k
The steps inv
as follows:

1. Th
di
2. Th
lig
3. W
ro
ca
an
4. Co
da
co
5. M
co
zo
po

Figure 3. TFT an
ent with PC, t
, OLED scree
cy [5]
ure 4. Philips' tr
with Operating
ur research cr
Operating sys
mera (web-cam
es not require
ng system; use
keyboard.
volved in iden
he System con
irectly on the b
he Infrared em
ghts in front of
When a stylus,
od, deflects th
amera senses
nd sends to the
ontroller conv
ata and sen
omputer.
Mouse contro
oordinates as a
ooms, clicks
osition.
nd object visiblity
the LCD was
ens can be us
ransparent OLED
system user re
eated a differ
stem we used s
m). The mott
any type of d
er can use his f
ntification of f
nsists of a fra
bezel of goggle
mitters emit in
f the goggle su
such as a fin
e beams of in
the deflected
e controller.
verts this data
ds the coor
oller driver
an input and m
or hovers

y
s still capable
sed which giv

[6]
equires mouse
rent scenario.
series of Infrar
to behind this
device to inter
fingers instead
finger(s) positi
ame that mou
e.
nvisible infrar
urface.
nger or a plas
nfrared light, t
d infrared lig
into meaning
rdinates to t
takes the
moves the curs
to the desir
of
ves
e or
To
red
s is
ract
d of
ion
unts
red
stic
the
ghts
gful
the
ese
sor,
red
p
s
i
F
The worki
perform the ac
sensed by the
identified at ru

Figure 7. TFT
Figure 5. Workin
Figure 6. 3
III.CONC
ng model can
ctions accordin
camera are in
untime for pro
T displaying conte
ng of emitter and
D model of gogg

CEPT INTO ACT
track the locat
ngly. The refle
raw form and
oper interaction
ent of webpage an

d camera

gle.
ION
tion of fingers
ected infrared
d blobs are
n with the syst

nd visible outer w
s and
rays
tem.
world
suc
Wi-
mil
con
enr
how
and
mu
com
[1]
[2]
[3]
[4]
[5]
[6]

Figur
There is n num
ch as mobility
-Fi connectiv
lestone in the h

This research
nsiders new k
rich life. Howe
w accurate the
d their actions
ch creatures o
mplex in their n

Norman Cousin
Dieter Rams: T
P. Mistry, P. M
the Proceeding
Japan. 2009.
http://mgitecete
projector-phone
http://electronic
http://www.eng
lighting-for-pos
re 8. Detected fi
VI.POSSI
mber of possib
y can be incre
vity. The rese
history of Real
VII.CONC
h looks at t
kinds of techn
ever, the succe
e system is in
at a given m
of habit, they c
needs and des
REFER

ns The Poet and
en principles for
Maes. SixthSense
gs of SIGGRAP
ech.wordpress.com
e/
cs.howstuffworks
gadget.com/2009/0
st-modern-identity
finger position (bl
IBILITIES
bilities with th
eased by addi
earch can pr
l World Intera
CLUSION
the need of
ologies that c
ess of research
nferring a per
moment. While
can also be hig
sires.
RENCES
d the Computer, 19
good design.
A Wearable G
PH Asia 2009,
m/2011/08/13/sam
.com/oled4.htm
03/06/philips-tran
y-mask/

lobs)
his research wo
ing Bluetooth
rove itself as
action.
innovation a
could be used
h will depend
rsons intentio
e people are ve
ghly random a
966
Gestural Interface
Sketch. Yokoha
msung-beam-
nsparent-oled-
ork
or
s a
and
d to
on
ons
ery
and
e. In
ama,
ISBN: 978-81-920575-7-6 :: doi: 10. 74222/ISBN_0768
Ontology Based Question Answering for Agriculture Domain
Arijit De
Tata Consultancy Services
Yantra Park, 2 Pokhran Road
Thane (W), Mumbai, India, 400601,

AbstractWith the advent of the semantic web, ontologies as
knowledge representation structures have gained immense
popularity. Ontologies have been used very effectively in
representing domain knowledge for domain-specific
Information Retrieval (IR) and Question Answering that
require Information Extraction and Natural Language
Processing techniques for query parsing and understanding.
In this paper we propose an automated question answering
system for the agriculture domain, that uses graph algorithms
to extract answers to user questions stored inside a agro-
ontology ontology. We evaluate our system using standard
questions asked by farmers to measure accuracy in terms of
precision and recall. We compare the performance of our
system with the widely popular AquaLog system when the
latter is configured for the agriculture domain. Our
experiments demonstrate the superior performance of our
system.
Semantic Search Systems, Knowledge Based Systems,
Ontology, Information Retrieval.
I. INTRODUCTION
The semantic web was largely an effort to organize the
hypertext content of the World Wide Web (WWW) into a
large organized repository of domain knowledge that could
be mined by automated intelligent agents to provide more
meaningful information for information seekers. In order to
transform the multi-media and text repository contained
within the WWW into a knowledge repository, scientists
and knowledge engineers came borrowed the concept of
ontology out of philosophy. Ontology which in philosophy
was the science of being was defined as (Chandrasekharan
[1]) a formal conceptualization of a domain or subject
matter knowledge. The semantic web demonstrated how
ontologies could be used effectively in representing domain
knowledge. The use of Ontologies has gone far beyond the
semantic web. Ontologies have become a popular choice as
domain knowledge representation structures for Information
Retrieval (IR), Question Answering (QA), Information
Extraction (IE) and a host of other Information Systems (IS)
applications.
In this paper we propose an ontology based question
answering system for the agriculture domain. We start off
discussing our proposed system as a conceptual domain
independent automated question answering system that uses
ontologies as the primary knowledge base for question
answering. Our question answering system treats the
domain ontology as combinatorial digraph that would be
traversed using traditional graph algorithms. We then
discuss the specifics of a agriculture domain specific
question answering system that answers questions put
forward by farmers and agro-advisors leveraging an
ontology as the sole knowledge base. We experiment with
standard queries asked in this domain to determine the
accuracy of the automated agro-question answering system
built. We also feed the same queries to the widely popular
AquaLog [2] question answering system which we
reconfigure for the agriculture domain. We compare and
contrast the performance of our system with that of
AquaLog.
Our paper is organized as follows. In the next section,
we describe previous work in this area. This includes a
discussion on key components of ontologies, ontology
applications in IR, IE and QA. This includes a detailed
description of AquaLog and our reconfiguration of
AquaLog, as this is the benchmark system for our
experiments. We also discuss standard agro-ontologies in
this section. In Section 3, we discuss our proposed
theoretical ontology-based question answering system.
Subsequently Section 4 we discuss, the design of our agro-
ontology and our ontology based automated question
answering system for the agriculture domain. Section 5 we
discusses our experiments and we conclude our paper with
a discussion on future work and how an ontology-based
question answering system can be a sub-system in a broader
automated question answering platform.
II. PREVIOUS WORK
As mentioned earlier, ontologies are structures for
representing domain knowledge. A first step in ontology
building involves identifying domain specific concepts and
entities. These are modeled into broader classes, individuals
that belong to these classes and class and individual
attributes or properties. The relationship between these
classes and individuals are established and represented as
object properties. Relationship between classes and
individuals are connected through a graph like structure
where each node represents a class or an individual and a
object property is an edge connecting two nodes that
defines a relationship between them. Ontologies thus
consist of (1) individuals entities that have specific
properties. (2) classes that are groups of individuals with
common properties that characterize them, (3) axioms and
rules which define necessary and sufficient conditions for
membership in specific classes, (4) object properties that
help establish relationship between different classes or
individuals and are represented as edges in the ontology
graph, (5) data properties that every entity (class or
individual) possesses, (6) annotation Properties that hold
additional information about a particular class that can not
be modeled otherwise. Classes and individuals form nodes
in the ontology graph and object properties for the edges.
Ontologies require formal description languages for
their construction. A popular formal language for
representing an ontology is Resource Description
Framework (RDF) and its extension Web Ontology
Language (OWL). RDF described by Klyne [14], is an
extension of XML and is used to create a conceptual model
for representing the knowledge structure of a domain.
Bechhofer [12] provides latest specifications of Web
Ontology Language (OWL). OWL extended RDF to add
more expressivity for knowledge representation on the
semantic web. OWL not only presents information for
human consumption, but can be also be used to process
information content. Bechhofer [12] describe key
components of OWL. OWL ontology components include
classes, properties, defining instances and its operations as
discussed earlier. Protg [13] is a tool developed at
Stanford University to create ontologies using OWL.
Lopez [2] built AquaLog an ontology based Question
Answering (QA). Lee [3] used ontologies for news
summarization and Wimalasuriya [4, 7], Wu [6, 8]
published the workings of Ontology Based Information
Extraction Systems (OBIE). Wu [6] created an OBIE
system, Kylin, which extracts data from Wikipedia
Infoboxes and organizes them in the form of an ontology.
Wu [8] improves the performance (in terms of recall) of
Kylin further in case of spare classes. Wimalasuriya [9]
explore the benefits of using multiple ontologies when
different ontologies are applied to different components of a
information extraction system and when multiple ontologies
are applied to each information extraction system
component. Wimalasuriya [4] builds on this component
based approach to creating OBIE systems that can be
reused. The works of Wimalasuriya [4, 7] demonstrates the
benefits of using multiple ontologies and ontology reuse.
a.Agro Ontologies
Since the objective of this paper is to build a ontology
based question answering system for the agriculture
domain, and would require building an agro-ontology, it is
prudent to review previous efforts at building agro-
ontologies. Ontologies in the agriculture domain have been
developed primarily for sharing domain knowledge in the
field of agriculture. Two critical works in creating agro-
ontologies are the AGROVOC project [9] and the
Agropedia Indica project [10], developed by IIT-Kanpur.
AGROVOC [9] was a United Nations (UN) Food and
Agriculture Organization (FAO) initiative to build the
worlds most exhaustive multilingual agricultural
vocabulary built on a standardized data description model.
AGROVOC is arranged as set of terms and their inter-
relationships that represent 40,000 agricultural concepts
world-wide covering not only agriculture but other related
topics that include forestry, fisheries, ranching, land use,
rural livelihoods and development, food security.
AGROVOC vocabulary is in over 20 most popular
languages world wide. AGROVOC is also an integrated
linked open data SOA platform for interoperability,
reusability and cooperation with other organizations that
use similar open standards. VocBench is a multilingual
vocabulary editor and workflow management tool available
on the AGROVOC webpage that allows for parallel and
distributed editing of the multi-lingual vocabulary. A global
community of librarians, terminologists, information
managers and software developers use VocBench for
maintaining and updating AGROVOC. Over the years,
AGROVAC ontologies have been downloaded by users in a
dozen countries to index and search over large document
and web collections.
Agropedia Indica [10] was a project of the Indian
Institute of Technology, Kanpur to build a platform for
digital knowledge management for agriculture in India.
Agropedia Indica has created a comprehensive, seamlessly
integrated model of digital information content for the
agricultural domain in India.
Agropedia Indica is a Web 2.0 application where
knowledge models are used to store and organize content
created by various knowledge actors through the Alfresco
Content Management System (CMS) that allows the
knowledge actor to upload not only text but also multimedia
content. Agropedia also provides users and collaborators
with an extensive tag library, called Agrotags provided in
English, Hindi and French. This creates a common medium
for knowledge transfer between knowledge actors.
Agropedia knowledge models are mostly centered on
specific crops and cover a broad range of topics like
irrigation plant disease, etc related to the crop. Agropedia
knowledge models provide a rich vocabulary of crop related
terms. This allows content to be stored, created, exchanged,
shared and used by knowledge actors. It also enables
semantic search of content so created.
a.AquaLog.
AquaLog is a Ontology Based Question Answering
system developed by Lopez [2]. It serves as a benchmark
system based on which we compare the performance of our
proposed ontology based QA system. Hence we go into
relatively more elaborate description of AquaLog.
Question Answering (QA) systems provide answers to
questions asked in natural language (NL). Although similar
to keyword-based Information Retrieval (IR) systems in
functionality, the latter return a set of documents that might
contain information relevant to a query. On the other hand
QA systems return a short answer to a question. Ontologies
often contain an organized vocabulary that is useful in
understanding user questions asked in NL which is often
critical to digging up the right answer. Ontologies can also
be used to represent knowledge as a knowledge base from
which suitable short answers can be retrieved. Let us
explore how AquaLog answers questions in natural
language.
Lopez [2] developed a ontology based question-
answering system, AquaLog that answers questions asked
in natural language using a domain-specific ontology that
can be plugged onto the system. Through AquaLog, Lopez
[2] demonstrated how an ontology based question
answering system can be separated from the domain
ontology that it uses to answer questions. The ontology
plugs-in to the AquaLog system AquaLog, includes a
learning component, which ensures that the performance of
the system improves over time, in response to a particular
domain. Key system components of AquaLog are (1)
Linguistic Query Classifier, (2) Relational Similarity
Services (RSS), and (3) Inference engine.
The Linguistic Component uses the General
Architecture for Text Engineering (GATE) [5] API and
other resources to map the Natural Language (NL) input
query to a query triplet. For example the query Who are
the researchers in the semantic web research area? is
translated to <person/organization, researchers, semantic
web research area>.
Relational Similarity Services (RSS) uses WordNet, a
domain dependent lexicon, and similarity matching to
translate the query triplet from the Linguistic Component to
an ontology compatible triplet. So the linguistic triplet
<person/organization, researchers, semantic web research
area> gets converted to the ontology triplet <researcher,
has-research-interest, semantic-web-area>. The domain
dependent lexicon is automatically updated through the use
of a Learning Mechanism.
One key feature of RSS within AquaLog is that it is
interactive in resolving ambiguity. When the system is
unsure of how to resolve ambiguities between linguistic
terms or relations in an effort to interpret a question it will
ask the user to resolve it manually. There is an in-built
learning mechanism that automatically updates the domain
dependent lexicon based on user feedback. Class Similarity
Services is a sub-component within RSS which maps
linguistic terms to ontology classes. Proper names are
mapped to individuals in the ontology by means of the use
of standard string distance metrics algorithms. If this
mapping fails a partial solution is implemented for
affirmative/negative type of questions, where we make
sense of questions in which only one of two instances is
recognized.
Since users can ask questions using a large vocabulary
of terms and interpretation and answering them is limited to
the specific domain ontology used, discrepancies often
occur between the natural language questions and the set of
terms in the ontology. Resources such as WordNet, that
provide synonyms and related words, can help making
sense of unknown terms, which could be detected in the
knowledge base. However, in many cases, the RSS
component finds it difficult to create genuine onto-triple
from the query triple formed by the Linguistic Component.
When this happens the learning mechanism built into the
RSS learns the new terms employed by the user and
disambiguates them in order to produce an adequate
mapping of the classes of the ontology.
Inference Engine forms the final component of the
AquaLog system. It makes sense of the ontology triple and
digs the answer out of the knowledge base.
III. PROPOSED QUESTION ANSWERING SYSTEM
a.Motivation
While AguaLog is widely popular ontology based
Question Answering system, it is hard to configure. The
configuration of AquaLog goes far beyond plugging in a
OWL ontology to the system. The knowledge base needs to
be created in so that the Inference Engine can dig out
information. It requires significant configuration at the
linguistic component level, significant changes at the RSS
level etc. Also a deep understanding of AquaLog system
source code is required to make AquaLog work properly.
Our motivation for creating a Question Answering system
was to create a system that would require users to plug in a
ontology which had to meet one simple specification. The
ontology would be a self containing knowledge base from
which information would be retrieved. There would be no
requirement for advanced lexicons etc. Any dictionary of
terms would be created from text content contained within
the ontology.
b.Graphical Nature of Ontologies
Our proposed Question Answering system is designed
on the premise that all information is contained within the
ontology. There would be only one simple specification for
creating the ontology. The ontology would need to be
created using OWL markup with each domain entity
modeled as an ontology class or individual that would
contain all relevant information stored as annotation
comments defined within it. The user asks questions
seeking information related to one or more domain entities
or sub-entities that are modeled using classes, sub-classes or
individuals respectively and the information which the user
requires is stored within the class or individual modeling
the entity either as a data property or as an annotation. The
task of the question answering system is therefore two fold.
The first is to identify which domain entity or entities is the
query related to. The second task is to retrieve the necessary
granules of information for that entity.
Let us consider the ontology as a graph. Let us say we
have an ontology O = <C, P>. Let C be a set of classes and
individuals of specific classes, P is a set of object properties
that create relationships between classes, individuals and
classes-individuals. So we can define a n object (classes and
individuals) system as C = {c
i
, i, 1 i n}. Each class c
i

also has a set of child classes or individuals C
h
associated
with it, such that |C
h
|0. Each class c
i
has an annotated text
A
i
attached to it. We can define P = {P
ij
, i, j, c
i
, c
j
}.
c.Preparing Lookup Table
The first key task in question answering is to identify
which domain entity or entities the user is querying about.
This is done by creating lookup tables associated with each
domain entity or class. Algorithm 1 defines the procedure to
create a lookup table for a specific class. Notice that the
procedure is recursive. The annotation associated with the
class contains textual information. The text content is
tokenized into words and n-grams. Word tokens are
screened for stop words and the filtered contents are added
to the lookup table associated with the class. Each child of
the class or individual is then processed the same way and
keywords associated are returned as a lookup table. The
contents of the lookup table of each child are then merged
with the lookup table of the parent class or individual.
Algorithm 1 can be repeated for each class/individual in the
set C defined.

b.Algorithm 1: CreateLookupTable
Input:
OWL Class/Individual c
n largest n-gram token to be created

Output:
L lookup table associated with c
L consists of n-gram tokens.

Algorithm:
A

annotated text in c.
C
h
set of all child classes and individual.
Define k
i
is a set of all i-gram tokens created by
tokenization
k
1
tokenize(A, 1)
Remove stop words from k
1
.
add contents of k
1
to L
for i = 2 to n
begin
k
i
tokenize(A, n)
add contents of k
i
to L
end loop
for all children c
h
C
h

begin
L
h
CreateLookupTable(c
h
, n)
LL L
h

end loop
return L
c.Question Answering
Algorithm 2 specified below describes the actual
process of question answering in response to a query. Each
user query is treated as a sequence of words. This sequence
of words is tokenized into words, and n-grams. The size of
each n-gram token varies from 2 to n where n is the number
of words in the query. Thus the largest n gram is the whole
query. N-gram tokens are matched to lookup tables. If a N-
gram token matches a lookup table, all N-grams and words
associated with it are thrown out of the set. An association
is created with that particular n-gram and the lookup table.
At the end of this matching, we obtain a set of lookup
tables. Lookup tables are associated with classes. From this
we can ascertain what classes or domain entities might be
associated with the query. For the set of classes associated
with query, if any class is a super-class another class the
super-class is eliminated and the sub-class is retained in the
set. Final answer to the question is obtained by merging the
annotated text associated with the classes associated with
the query.

d.Algorithm 2: Question Answering
Input:
n-class ontology system
Set of Lookup tables L = { l
i
,1in }
Set of Lookup tables C = { c
i
,1in }
T = L X C = {t
ik
, L
i
L,C
k
C}
Query Q, sequence of words{w
1
,.,w
m
}.

Output:
A
n
Answer String to query Q

Algorithm:
m number of words in the query
Q
m
set of all tokens 1 through m-grams
for j = 1 to m
begin
Q
j
set of j-gram tokens
Q
j
= Tokenize-Into-M-Grams(Q)
Q
m
= Q
m
Q
j
.
end loop
Define L
Q
A set of Lookup tables
Define C
Q
A set of associated Classes
for j = m to 1
begin
Q
j
set of j-gram tokens
get Q
j
from Q
m
.
for each token k in Q
j
begin
match k to tokens in lookup tables in L
if (k match tokens in any lookup table l
k
in L)
add L
Q
= L
Q
l
k
,
add C
Q
= C
Q
c
k
.
end loop
for each pair (c
i
, c
j
) in C
Q

begin
if ( c
i
is a sub-class of c
j
)
Remove c
j
from C
Q
.
end loop
for each c
i
in C
Q

begin
get A
i
c
i
// Annotated text A
i

A
n
= A
n
A
i

End loop

IV. QUESTION ANSWERING IN AGRICULTURE
In the previous section we have described a theoretical
Question Answering system that uses ontologies to answer
user questions. In this section we build on this theoretical
foundation to implement our Question Answering system
for the agriculture domain. The first step was to manually
b
n
f
i
r
t
t
f
c
c
T
c
p
i
c
c
o
T
c
e
c
P
W
a
s
h
P
d
s
c
w
b
h
o
T
i
p
p
t
a
W
o
w
c
r
r
N
D
r
A
F
N
build domain
next section w
for the agricul
implementatio
related queries

a.Building
At the hea
the agro ontol
that we build
focus of the
cotton etc. T
cultivation wi
The domain b
cultivation.
Our propo
practices for g
identify the cr
created. The c
class in the o
object propert
The next sta
cultivation an
each cultivat
cultivation w
Preparation, (2
Weed, Pest an
and sub-stage
stages would
hierarchical fa
Preparation w
different appr
sub classes to
class of land p
with each clas
become indivi
have some un
of a crop. Ind
This might ma
individual hie
properties suc
properties. Ide
through object
and various cu
Using this
Wheat, from
ontology creat
was the crop
cultivation w
relationship to
represented b
Nitrates, Sulp
DAP were ide
represented in
AgronomicPra
FieldPreparati
NormalTillage
ontology for
we describe th
lture domain a
on of the QA
s.
g Ontologies
art of the auto
logy. Like Agr
is crop centri
ontology is a
The ontology
ith wheat bein
being modele
osed ontology w
growing the c
rop for which
crop becomes
ontology. All o
ties to this cla
age was ident
nd defining cla
tion activity
would involve
2) Seeding, (3
nd Disease Con
would have a
be connecte
fashion. For e
would involv
roaches to Till
a Tillering cla
preparation. Id
ss of either cr
iduals. A very
nique propertie
dividuals woul
ake it prudent
erarchically u
ch as height, w
entifying and
t properties tha
ultivation stage
simple proces
best practice
ted is shown in
Wheat. From
we identified
o the wheat c
by a rectangl
phurPhosphate
entified as Indi
n Figure 1 u
actice clas
ion. Various
e and ZeroMin
the agricultu
he process of
and then proce
system that a
omated agro-a
ropedia Indica
ic, which mea
a specific cro
we build i
ng at the centr
d is therefore
was built base
crop wheat. T
the ontology
the central en
other classes
ass either dire
tifying variou
asses and sub
. Typically
e (1) Agron
) Fertilization
ntrol, (6) Harv
a class associa
d to their pa
example Agro
ve Tillering.
lering which
ass which wou
dentifying indi
rop or cultivat
y specific varia
es and might b
ld have specif
to model this
under the cro
weight etc mig
establishing
at link up betw
e class.
ss we build an
es obtained i
n Figure 1. Th
m the various
the stage c
class. In Figu
le. Specific f
, Phosphates,
ividuals as opp
using oval ob
ss was
types of t
nimumTillage
ure domain. In
building onto
eed to describ
answers agricu
advisory syste
a [10] the onto
ans that the ce
op such as w
is one for w
re of the ontol
e limited to w
ed on accepted
The first step
model needs t
ntity modeled
as related thro
ectly of indire
us stages of
b classes base
stages in
nomics and L
n, (4) Irrigation
vesting. Each s
ated with it. T
arent stages in
onomics and L
There would
would come i
uld in turn be a
ividuals assoc
tion activity w
ant of a crop m
become individ
fic data prope
crop variant a
op. Specific
ght identify as
class relations
ween the crop
n agro-ontolog
in Gill [11].
he central key
s stages of w
classes and
ure 1 each cla
fertilizers suc
Urea, Potash
posed to class
bjects. Within
the sub-
tillering such
were identifie
n the
ology
e the
ulture
em is
ology
entral
wheat,
wheat
logy.
wheat
d best
is to
to be
as a
ough
ectly.
crop
ed on
crop
Land
n, (5)
stage
These
n an
Land
d be
in as
a sub
ciated
would
might
duals
rties.
as an
data
data
ships
class
gy for
The
class
wheat
their
ass is
h as
h and
ses in
n the
class
h as
ed as
ke
be
us
us
do
OW
hig
sm
cla
ob
inf
an
try
sto
en
ce
cro
cla
ca
co
cu
the
co
an
Al
as
sta
us
Al
an
on
ag
cro
ey subclasses
e observed look
Figur
Our propose
sing the ontolo
sing OWL [
ocument [11]
WL Annotatio
ghly structure
mall granules o
e.Question A
The agro-on
asses that we
bject propertie
formation asso
nswering user
y to retrieve
ored as an ann
ntity. Looking
enter of the on
op Wheat as
asses modelin
alled these cul
onsists of a
ultivation stagi
e central class
ontains some
nnotation prope
Our Quest
lgorithm 1 (
sociated with
artup, before a
ser query com
lgorithm 2 (Se
nd retrieve the
V.
In the previo
ntology based
griculture dom
op centric ont
of FieldPrepa
king at Figure
re 1: Proposed W
ed best practi
ogy design too
[12]. Textual
was stored as
on property.
d text docume
of information
Answering
ntology that we
e connected b
s. Each class w
ociated with t
questions to
the most app
notation to th
at the ontolo
ntological gra
an entity. The
ng various sta
ltivation stagi
host of sub
ing classes w
s modeling W
information
erties.
tion Answer
(See Section
h each class (
any queries co
mes into the
ee Section 4)
appropriate an
. EXPERIMEN
ous section we
d question
main. We also
tology for the
aration. Simila
e 1.
Wheat Agro-On
ces ontology
ol Protg [13]
l content of
s the level of
Thus the ont
ent, with each
stored.
e created cons
by means of
would contain
the entity the
the agro-advi
propriate piece
he class model
ogy created in
aph is the cla
ere is next lay
ages in wheat
ing classes. T
classes conn
which are in tu
Wheat as an en
stored by m
ring system
4) to create
or entity mod
ome into the
system, the
to dive into th
nswer.
NTS AND RESU
e described th
answering s
o discussed th
Wheat crop.
ar subclasses c
ntology
was construct
] and represent
f best practic
each class as
tology created
class containi
ists of a series
hierarchies a
n granules of t
class models.
sory system,
e of informati
ling the relev
Figure 1, at t
ass modeling t
yer of connect
cultivation. W
The second lay
nected with t
urn connected
ntity. Each cl
means of OW
would app
e lookup tab
deled) at syst
system. When
system emplo
he annotated t
ULTS
e creation of o
system for t
he creation of
Our ontology
can
ted
ted
ces
an
d a
ing
s of
and
ext
In
we
ion
ant
the
the
ted
We
yer
the
d to
ass
WL
ply
bles
em
n a
oys
ext
our
the
f a
y is
created from best practices for growing wheat. Earlier we
had discussed the workings of a widely popular ontology
based question answering system AquaLog. We had also
discusses the Agropedia Indica project and the ontology
based vocabulary that it creates. Agropedia Indica has an
ontology for the crop Wheat which is crop centric. We have
four cases of experiments. These are outlined in the next
section.
a.Experimental Cases
There are four cases of experiments as outlined in Table
1. In the first case of experiments we use the a OWL-
reconstructed version of the Agropedia Wheat Ontology
and the AquaLog system configured for the agriculture
domain. In the second case we use the reconstructed Agro-
Wheat Ontology with our proposed Graph Based QA
system. In the third case we use our Wheat Ontology with
the with the reconfigured AquaLog. In the fourth case we
use our Wheat Ontology with our proposed Graph Based
QA system.
TABLE I. EXPERIMENTAL CASES
Cases Ontology QA System
Case 1
Reconstructed
Agropedia
Wheat
Ontology
Reconfigured
AquaLog
Case 2
Reconstructed
Agropedia
Wheat
Ontology
Proposed Graph Based
QA System
Case 3 Wheat Ontology
Reconfigured
AquaLog
Case 4 Wheat Ontology
Proposed Graph Based
QA System
a.Reconfigured AquaLog
In Section 2 we discussed the working of AquaLog. To
reconfigure AquaLog for the agriculture domain we need to
plug in the ontology. In our case we plug in the
Reconstructed Wheat Ontology from Agropedia or the
Wheat Ontology we create. The former is reconstructed in
OWL with every class or node in the system having related
text information regarding best practices. To plug-in the
ontology we modify service_properties.xml file. To create
linguistic tuples for the agriculture domain
query_properties.xml file needs to be modified. The lexicon
for the learning mechanism and the knowledge base also
required modification.
b.Question Set
For our experiments we obtain a set of standard
questions that are asked by users about wheat. The question
set contained 500 standard questions on the crop wheat.
Questions ranged from Disease Management, Irrigation,
Land Preparation, Fertilizers and Nutrient Management,
Pest Management, Harvesting, Seeds, Seeding and Seed
Variety, Weed Management. Each question was provided
with a correct answer. We passed these questions into the
system and received an answer from the system. Table 2
shows the category of questions and assigns a label to each
category.

c.Performance Metric
We measured our system performance on the basis of
standard Information Retrieval (IR) measures, precision and
recall as defined in Raghavan [15]. Precision is the fraction
of retrieved answers that were correct, while recall
represented the fraction of times the correct answer was
present and retrieved accurately.
f.Results
The precision and recall results of our experiments with
all four cases are tabulated in Table 3 and 4 respectively.
Clearly the results for Case 4 are the best in terms of both
average precision and recall. It can be observed that recall
values are generally higher than average precision in all
cases. This means that when the answer is present in the
knowledge base the system is able to pick up the answer
with reasonable accuracy. Correct answers were provided
with less accuracy in all cases. Clearly results for case 2 and
case 4 are the best. This demonstrated the performance of
our QA system over the AquaLog system. Analyzing the
results breaking down the category of questions, we observe
that the results are poorest for answering questions on
Harvesting (H). Recall and precision are both lower. This is
primarily because there was less knowledge in the
knowledge base in the topic of harvesting for wheat.
Overall our proposed ontology (Precision = 52.90%, Recall
= 56.88%) did slightly better than the Agropedia ontology
when used in the either QA system (Precision = 50.88%,
Recall = 52.81%).
TABLE II. QUESTION CATEGORY
Question Category Symbol
Disease Management DM
Irrigation IR
Land Preparation LP
Fertilizers Nutrient Management FNM
Pest Management PM
Harvesting H
Seeds, Seeding and Seed Variety SSV
Weed Management WM
TABLE III. PRECISION (%) OF QUESTION ANSWERING FOR CASES
Precision Case 1 Case 2 Case 3 Case 4
DM 47.63 64.26 50.12 68.35
IR 45.42 60.76 47.84 65.23
LP 46.93 62.94 49.25 67.16
FNM 42.78 59.56 44.97 61.32
PM 41.20 54.83 43.83 58.75
H 33.78 42.96 31.87 42.12
SSV 44.52 61.38 44.26 61.73
WM 46.43 58.63 47.23 62.35

TABLE IV. RECALL (%) OF QUESTION ANSWERING FOR CASES
Recall Case 1 Case 2 Case 3 Case 4
DM 52.73 67.80 54.35 74.23
IR 50.21 62.75 53.86 68.70
LP 50.45 65.60 56.68 71.82
FNM 45.92 57.80 44.66 63.29
PM 43.48 56.06 45.01 61.38
H 32.86 41.64 34.66 45.59
SSV 46.50 60.97 49.53 66.75
WM 47.69 62.52 51.19 68.45
TABLE V. SUMMARY OF RESULTS
Precision (%) Recall(%)
Case 1 43.59 46.23
Case 2 58.17 59.39
Case 3 44.92 48.74
Case 4 60.88 65.03

VI. CONCLUSION
In this paper we have discussed a theoretical system that
used domain ontologies to do domain based question
answering. We have built on this discussion to create an
agro-ontology and an agro-advisory system that answers
questions on cultivation of the crop wheat in the Indian
context. Our ontology is build using best practices for
wheat cultivation in the Indian context. Our system
performs with a high level of precision and recall for most
categories of questions with regards to wheat cultivation.
Our ontology is built manually and encapsulated best
practices knowledge. We compare the performance of the
system to that of Lopezs AquaLog system when the latter
is configured to answer questions for the agriculture
domain.

REFERENCES

[1] B. Chandrasekharan, J. R. Josephson, V. R. Benjamins. What Are
Ontologies and Why Do We Need Them?. In IEEE Journal of
Intelligent Systems, IEEE Press, 1999.
[2] V. Lopez, M. Pasin, and E. Motta. AquaLog: An Ontology-Portable
Question Answering System for the Semantic Web, In 2nd
European Semantic Web Conference, Heraklion, Crete, 2005.
[3] C. S. Lee, Z. W. Jian, L. K. Huang. A Fuzzy Ontology and Its
Application to News Summarization. In IEEE Transactions on
Systems, Man, and Cybernetics, 35(5), 2005.
[4] D. C. Wimalasuriya, D. Dou. Components for Information
Extraction: Ontology-Based Information Extractors and Generic
Platforms. In 20th ACM Conference on Information and Knowledge
Management, Toronto, Ontario, Canada, 2010.
[5] H. Cunningham, D. Maynard, K. Bontcheva, V. Tablan. GATE: A
Framework and Graphical Development Environment for Robust
NLP Tools and Applications. In Proceedings of the 40th
Anniversary Meeting of the Association for Computational
Linguistics, Philadelphia, 2002.
[6] F. Wu, R. Hoffmann and D. S. Weld. Information extraction from
Wikipedia: moving down the long tail. In Proceedings of 14th ACM
SIGKDD International Conference on Knowledge Discovery and
Data Mining, Las Vegas, NV, 2008.
[7] D. C. Wimalasuriya and D. Dou. Using multiple ontologies in
information extraction. In Proceedings of the 18th International
Conference on Information and Knowledge Management, 2009.
[8] F. Wu and D. Weld. Autonomously semantifying Wikipedia. In
Proceedings of the Sixteenth ACM Conference on Information and
Knowledge Management, Lisboa, Portugal, 2007.
[9] A. C. Liang, B. Lauser, M. Sini, J. Keizer, S. Katz. From
AGROVOC to the Agricultural Ontology Service / Concept Server,
An OWL model for managing ontologies in the agricultural domain.
FAO Press, 2004.
[10] M. Sini, V. Yadav. Building Knowledge Models for Agropedia
Indica - Requirements, Guidelines, Suggestions. IIT Kanpur Press,
2009.
[11] M. S. Gill, H. S. Bajwa. Package of Practices for Crops of Punjab,
Punjab Agriculture University Press, 2010.
[12] Bechhofer, S., van Harmelen, F., Hendler, J., Horrocks. I.,
McGuinness, D. L., Patel-Schneider,P. F., Stein, L. A.,2009. OWL
Web Ontology Language Reference. W3C Online Publication.
[13] J. H. Gennari, M. A. Musen, R. W. Fergerson, W. E. Grosso, M.
Crubzy, H. Eriksson, N. F. Noy, S. W. Tu. The evolution of
Protg: an environment for knowledge-based systems
development. In Journal International Journal of Human-Computer
Studies, v.58, 1, Academic Press, 2003.
[14] G. Klyne, J. J. Carroll. Resource Description Framework (RDF):
Concepts and Abstract Syntax. W3C Online Publication, 2004.
[15] V. V. Raghavan, G. S. Jung, P. Bollmann. A critical investigation of
recall and precision as measures of retrieval system performance. In
ACM Transactions on Information Systems, v7, pp.205-229, 1989.

ISBN: 978-81-920575-7-6 :: doi: 10. 74229/ISBN_0768
Can Human Computer Interaction (HCI) bridge the Digital Divide?
A Case of Akash in India

Shilpa Tripathi
Assistant Professor, Dept of Information Technology,
S.S. Dempo College of Commerce and Economics,
Altinho, Goa.

Abstract - The Human Computer Interaction has for long
been regarded as that unique area in computer
applications which blends information technology with
its human interface, thereby providing an individual the
ease of access, usability and output. HCI therefore not
only focuses on the technical dimension, but also
underscores the importance of the human element which
is the key to the success of any technology and its
application. While most of the current literature on HCI
focuses on the issues on challenges that emanate from
advanced users, very little work as yet is evident with
regard to the potential and challenges for HCI for the
population which is on the IT margins. The present
paper tries to look at the issues and challenges relating to
the HCI from the prism of Digital Divide and tries to
explore whether innovative HCI also has the potential to
reduce the user gap between those few (urban) in India
who have access to the latest technology and its
application and those who are in the periphery (rural)
and uses the case study of the Akash Tablet PC recently
introduced in India as the 35 $ answer to Indias vast
problem of digital divide. It then defines and explores the
manifestations of the digital divide in a country like
India. Then it blends the two by giving an account of the
origin of the Akash Program its potential and
limitations and suggests some ways of addressing the
potential concerns.
Keywords -HCI, Akash,digital divide
I. INTERPRETING HCI
Human Computer Interaction is one of the
most interesting aspect of Informational Technology
application studies that has two connected yet distinct
dimension of the Human- computer interface. First, it
has a direct bearing on how we as human beings relate
and adapt to the ever complex array of technological
advancement in the field of IT applications. It enables
us to become more innovative, experimenting and
thinking. But there is a second equally important
dimension- that of the technology also adapting and
reinventing itself to suit and meet human requirements
and expectations. The state of research and work in the
domain of HCI today therefore is an interesting
intermixes of these two converging trends.
This has had a significant bearing on research
relating to HCI as well and a need has been asserted to
move beyond mere technological dimensions of HCI
to the human and physical dimension as well .As
rightly pointed out in one of the interesting studies
relating to HCI research,
Research in Human-Computer Interaction (HCI)
has been spectacularly successful, and has
fundamentally changed computing. Just one example
is the ubiquitous graphical interface used by Microsoft
Windows 95, which is based on the Macintosh, which
is based on work at Xerox PARC, which in turn is
based on early research at the Stanford Research
Laboratory (now SRI) and at the Massachusetts
Institute of Technology. Another example is that
virtually all software written today employs user
interface toolkits and interface builders, concepts
which were developed first at universities. Even the
spectacular growth of the World-Wide Web is a direct
result of HCI research: applying hypertext technology
to browsers allows one to traverse a link across the
world with a click of the mouse. Interface
improvements more than anything else has triggered
this explosive growth. Furthermore, the research that
will lead to the user interfaces for the computers of
tomorrow is happening at universities and a few
corporate research labs..there is also a need to
discuss HC more from psychological, ergonomic, and
human factors as well.

II. AREAS OF HCI RESEARCH AND STUDY

The domain of HCI research has expanded and
unfolded in multiple ways. While the initial years of
work in the area focused more on the basic user
techniques and applications such as the Direct
manipulation of graphic objects, mouse, windows etc,
over the years, it expanded to include drawing
programs, spreadsheets, computer aided design and
video games etc. The more recent areas have included
videos, multi-media, web-based applications,
portability all of which taken together has had the
cumulative impact of making Human computer
interaction not only more interactive but also
diversified in terms of its content, applicability and
reach.
The HCI research and innovation, while being
initiated and led by the corporate and research
combine of the west, has had a very important bearing
on its implications for the developing world as well.
Developing countries, characterized by nascent state
of their technological and research institutions,
deficient governance and delivery mechanisms and
under-equipped human capability
1
to make best
possible use of the technology t address the basic
development concerns- are often seen at the receiving
end of IT spectrum. Nevertheless, for the developing
countries forming the middle bracket which are also
seen as emerging economies and potential user as well
as source of IT manpower, using HCI to address the
issue of digital divide becomes an interesting area of
study and research. The point shall be addressed
further after identifying some of the critical
characteristics of the digital divide.
Digital Divide: Meaning and Manifestations
The term digital divide carries multiple
interpretations depending from the location and usage
the use of Information and Communication
Technologies (ICT) is being referred to. At a very
basic level, it refers to the gap between individuals,
households, businesses and geographic areas at
different socio-economic levels with regard both to
their opportunities to access ICT and to their use of the
Internet for a wide variety of activities. It includes the
imbalance both in physical access to technology and
the resources and skills needed to effectively
participate as a digital citizen. Knowledge divide
reflects the access of various social groupings to
information and knowledge, typically gender, income,
race, and by location. The term global digital divide
refers to differences in access between countries in
regards to the internet and its means of information
flow.
2

A country like India finds itself at a rather
paradoxical situation with regard to the Digital divide.
While on one hand the country is home to one of the
most skilled mass of human resource when it comes to
IT expertise, it also is marked by abysmally low levels
of I.T literate population as compared to its total
population and the situation gets complicated further
as the overall literacy levels are also much less I
comparison to what a potential I.T. Superpower must
have. Besides irrespective of the ever growing
significance of IT exports in Indias GDP, the basic
infrastructure for IT expansion across the rural
hinterland remains at abysmally deficient levels. The
methods and means to overcome this paradox
therefore have to be focus on a number of factors
bearing on human, social, and technological capacity
enhancement. There have indeed been some
meaningful yet limited attempts by both the state and
private sector aimed at bridging the digital divide by
addressing some of these areas relating to both e-
literacy, governance and infrastructure .These are
summarized as follows:
e.CARD Project
The Computer Aided Administration of
Registration Department (CARD) project initiated by
the government of Andhra Pradesh illustrates the
effective use of IT to improve citizengovernment
interface. Under this project, land registration offices
through out Andhra Pradesh are now provided with
computerized counters. Citizens can now complete
registration formalities without much hassle.
f. Sourkaryan and ESeva
One project of the government of Andhra Pradesh
has been quite popular among the people. Sourkaryan,
which is now operational in the port city of
Visakhapatnam, provides the facility for a citizen to
pay property taxes online and also view details of
plans and projects of the government and local bodies.
Similarly the ESeva Kendras in the Hydrabad state
city is an innovative experiment towards eliminating
personal contact between citizen and the bureaucracy.
Here a citizen can pay sales taxes, insurance
premiums, property taxes, land taxes, etc.
Additionally, the government of Andhra Pradesh has
formulated a prolonged strategy to further the prospect
of egovernance in the state. In a major attempt to
bring remote rural areas into the information
technology fold, the first Cyber Grameen, a rural
broadband venture, was started. This project has been
launched by a nongovernmental organisation, Swarn
Bharat Trust, basically to set up IT convergence hubs
in rural areas of the country. By harnessing the power
of rural internet broadband the Cyber Grameen
seeks to provide a range of applications and services to
stimulate the rural economies. The services provided
include telephony, telemedicine, distance learning, e
mail, digital entertainment, and delivery of
government services and information.
.The Bhoomi Project of Karnataka state covers 6.7
million farmers and holds millions of records of land
ownership. The project has earned the goodwill of
many people and also international funding agencies.
This project has reduced the delays involved in
interacting with the bureaucratic hierarchy of the state
revenue department. Bhoomi centres are located all
over the state. Any land record can be reviewed
through a touch screen at these kiosks; the project can
also be used as a databank for various projects of
public and private sector organizations.
The project has won the 2002 Commonwealth
Association of Public Administration and
Management award for creating self content
governance and opening up new frontiers. Both the
UNDP and the World Bank have lauded Bhoomi for
bold vision and implementation. With the success of
the Bhoomi project other states of India, viz. Tamil
Naidu, Maharashtra and Madhya Pradesh have started
evolving models based on Bhoomi in their respective
states.
b. The Gyandoot Project
Gyandoot, which literally means Knowledge
Messenger, is the first ever project in India for a rural
information network in the Dhar district of Madya
Pradesh which has the highest percentage of tribes and
dense forest. Every village has a computer centre or
soochnalayas at prominent market places or major
roads. People can easily log in and complain or request
information on crops, forest fields, water resources,
etc. of the district. Twentyone village Panchayats in
the District have been connected with computers or
information centres; several private sector information
centres called Soochnalays have also been opened.
One such popular centre is in Manwar Agriculture
Mandi, where the latest crop prices are made
available to the farmers. The land records of a few
tehsils of district Dhar are also available on these
computers. Also, Internet connections have been
provided to get global information by linking to the
World Wide Web.
The government of Madhya Pradesh is attempting
to make Gyandoot Project a great success by
extending it to other districts. The state is in the
process of starting 7,800 IT kiosks with the help of the
private sector. To train common people to be computer
literate, 7,500 Jan Shiksha public instruction centres
have also been identified, and policy is being
formulated to bring IT to the common peoples need
and benefit. Efforts are also being made by the
government to involve public libraries in this project.
In fact, public libraries can play a vital role in making
the programme successful by acting as information
centres (soochnalayas) for the Gyandoot Project. For
this to occur, a Public Libraries and Information
Centers Act needs to be passed. Also, strong will and
commitment among the professionals and policy
makers are required .
a.FRIENDS Project
The Fast, Reliable, Instant Efficient Network for
Disbursement of Services (FRIENDS) Project has
been launched by the state of Kerala in the southern
part of the country with a view toward mitigating the
hardship of citizens paying taxes by eliminating
middlemen, delays and long queues. Essentially,
FRIENDS is a centralised collection counter which
accepts almost all types of tax and utility payments.
This project has expanded to serve 13 million people
in 12 districts of Kerala. The basic philosophy of
FRIENDS is to treat citizens as valued customers.
Similarly in Tamil Naidu state a private outfit N
Louge offering low cost telesolutions has worked
wonders in Madurai districts by using the local loop
technology and making available fiber optic lines
running across the district, by helping private
entrepreneurs run services including that of e
governance.
b.Lokamitra/Smart Project
Himachal Pradesh (HP), the hill state of the
country, has initiated the Lokamitra project with grants
from NABARD to provide the general public,
especially those living in distant rural areas, easy
access to government information and facilities of e
governance to their door steps. Lokamitra
Soochnalaya Kendras (information centres) have
been set up in 25 panchayat areas run by unemployed
youth.
These Kendras provide current information
relating to the district and government information.
The government of HP has also developed IT Vision
2010 in collaboration with NASSCOM (National
Association of Software Companies) to convert the
hill state into an IT destination and also make Simple
MoralAccessibleResponsive and Transparent
(SMART) Government.
3

While the above certainly are important initiatives
taken by state and central governments, there still
needs to be done a lot more by way of overcoming
barriers to digital coverage and expansion and most
importantly making the same relevant for Indias
present and future developmental needs. In one of the
most significant studies on the issue, the following
have been identified as key barriers to bridging the
digital divide in India.
Infrastructural barriers
Despite the incredible growth of the Internet since
the early 1990s, India still lacks a robust
telecommunication infrastructure with sufficient
reliable bandwidth for Internet connection. Due to
higher costs the necessary upgrading of hardware and
software is difficult; hence, despite the rapid spread of
the Internet the gap is growing wider as the
technological standard grows even higher. Faster
networks, higher level machines, more complex
software and more capable professionals are required,
but in many nations including India the funding is not
available to support these developments.
Libraries and information centres, with their
commitment to freedom of access to information and
promotion of lifelong learning in India, are yet to
have a robust infrastructure. Public libraries which can
provide access to the Internet do not have computers
and Internet access. Although cybercafes have been
increasing, poor people cannot afford to have access
due to high cost.
Literacy and skill barriers
Education and information literacy will play an
important role in keeping society from fragmenting
into information haves and havenots. In the
perspective of the digital divide, IT literacy is very
important to allow access to digital information. In a
country like India where roughly 50 percent of people
do not have reading and writing skills for functioning
in everyday life, IT literacy is out of the question.
Generally, online content and information have been
designed for an audience that reads at an average or
advanced literacy level and those who have
discretionary money to spend.
Education in information literacy will play an
important role in keeping the society from fragmenting
into a population of information haves and havenots.
The lack of skill in using computer and
communication technology also prevents people from
accessing digital information.
g.Economic barriers
Poor access to computer and communication
technology also causes a digital divide. In India the
ability to purchase or rent the tool for access to digital
information is less among the masses. The lower
income group does not have discretionary money to
spend on cybercafes or to get Internet connectivity on
their own to access digital information.
h.Content barriers
The Internet allows ideas and information to be
shared freely from citizen to citizen globally. In many
ways the strength of the Internet is a function of the
number of people and organisations creating quality
content. Since no entity controls the Internet, anyone
with Internet access has the potential to contribute
information. Therefore, to solve the digital divide,
steps should be taken by the government to ensure that
all citizens are able to receive diverse content relevant
to their lives as well as to produce their own content
for their communities and for the Internet at large.
i.Language barriers
India is a country having a multicultural and
multilingual population. Today a large percentage of
information content on the Internet is in English,
which is a barrier for the people whose primary
language is not English.
4

j.Akash as an innovator in bridging digital
divide?
While a number of the above mentioned concerns
and challenges remain, The Government of India has
come out with a path-breaking 35$ Tablet PC named
as Akash (sky) epitomizing Indias first bold foray in
making computer access , literacy and usage across
Indias hinterland. The original prototype named
Sakshat (Sanskrit: "Embodiment") was
unveiled in 2009, purported as a "$10 laptop".A new
prototype, costing $35 after government subsidy, was
unveiled in 2010. The device was formally launched
on 5 October 2011 as Aakash, made by the UK-based
company DataWind. Aakash will be assembled at
DataWind's new production centre in the southern city
of Hyderabad. Initially, the device will be sold to the
Government of India at $50. It is expected to be sold
for $60 in the retail stores.
5
The device has been
developed as part of the National Mission on
Education through Information and Communication
Technology that aims to link 25,000 colleges and 400
universities on the subcontinent in an e-learning
program via an existing Sakshat portal.

As per the Government of Indias guidelines, the
following specifications have to be adhered to in the
manufacturing and delivery of such devices:
"Support for video web conferencing facility;
multimedia content viewer for example .pdf, .docx,
.ods, .adp, .doc, .xls, .txt, .jpeg, .gif, .bmp, .odt, .zip,
AVI, AC3, etc.; searchable PDF reader; unzip tool to
unzip zip files; possibility to install suitable firmware
upgrades; computing abilities such as Open Office,
SciLab, cups (for printing support); media player able
to play streamed and stored media files; Internet
browsing, JavaScript, PDF plug-in Java; wireless
communication for audio/video I/O; cloud computing
option; remote device management ability; rendering
YouTube and other online video services (open source
Flash players, e.g., gnash or swfdec)."
Other preferable
Playback: AVCHD
Multimedia I/O interfaces: DTV, IPTv, DTH
Internet browsing: Flash player (Adobe)
Specifications
Other preferable
HDMI port
Hardware
7-inch 800x480 resistive touch screen
Rugged casing with a rubberized feel
Wi-Fi enabled (802.11 a/b/g WiFi)
Mini and full USB
miniSD card slot
Subscriber Identity Module (SIM card) slot
Video out
Headphone jack
256MB of RAM
2 GB of storage memory. 32 GB Expandable
microSD memory card.
2 Watts of power consumption with solar
charging option
366 MHz + HD video co-processor
Audio out: 3.5mm jack / Audio in: 3.5mm
jack
Display and Resolution: 7 display with
800x480 pixel resolution
Software
Android 2.2 operating system
Document Rendering
Supported Document formats: DOC, DOCX,
PPT, PPTX, XLS, XLSX, ODT, ODP
Educational software developed at Indian
Institute of Technology
Web browsing, video conferencing and word
processing software
PDF viewer, Text editor
Multimedia and Image Display
Image viewer supported formats: PNG, JPG,
BMP and GIF
Supported audio formats: MP3, AAC, AC3,
WAV, WMA
Supported video formats: MPEG2, MPEG4,
AVI, FLV
Communication and Internet
Web browser Standards Compliance:
xHTML 1.1 compliant, JavaScript 1.8 compliant
Separate application for online YouTube
video
Safety and other standards compliance
CE certification / RoHS certification
150000+ applications
6

While Akash certainly marks a path-breaking
development by way of conceptualizing and
implementing e-literacy, given the record of Indias
past forays in I.T literacy and existing bottlenecks, one
needs to do a careful assessment of the pros and cons
before giving the latest introduction a thumbs up! The
reasons for avoiding premature accolades are as
follows:
a.Wi-fi (dis)ability- The tablet PC platform to
become user friendly and universally applicable needs
an uninterrupted wire free environment through which
the basic applications as well as user interface-
multimedia applications can make a meaningful
impact. Wi-fi connectivity as it exist in India applies
merely to institutional and public spaces where there is
already a parallel mode of connectivity in some form
or the other available. I such circumstances, Akash
may at best be a devise which works better in a more
conventional cable based working environment and
therefore may have a limited reach.
b.Supporting infrastructure- While the net based
wi-fi connectivity is an important requirement, equally
important is the basic infrastructural requirement by
way of regular and uninterrupted power supply and
backup which would be essential for the tablet to
remain operational and effective. Again while such
kind of power backup facilities may be more readily
available in aforementioned urban public spaces, their
ability to sustain in environments where power
shortages and power absences are more a norm than
exception may simply make the whole idea of mobile
computing redundant. India needs a sound rural power
infrastructure as well to make Akash like experiments
feasible across the much needed pockets in the
country.
c.Literacy-basic and electronic- As mentioned
earlier, for Akash to become even a functionally viable
platform if not a successfully running one, we need a
twin prong intervention at enhancing both the basic as
well as the electronic literacy at the primary and the
secondary level. Given the relative performance of
Indias literacy missions, this may itself appear to be a
tall order, and may end up creating a situation when
the tablet may not have as many takers and
beneficiaries as it was intended to be.
d.The Economic argument- While the above
indeed are crucial factors in shaping the success or
otherwise of Akash, the role of economic factors
cannot be undermined. First in a country with annual
per capita income of Rs54,527 (about $1,100), very
few may have the privilege to spend $200 on a fancy
mobile with internet facilities.
7
Secondly, there needs
to be a significant catchment area for the economies of
scale argument to work: implying as the demand for
the users of the tablet increases, costs may
automatically come down therefore making the
possibility of entry of big telecom players more
realistic. Given the present state of our telecom
companies and the government skepticism relating to
the spectrum imbroglio, it is still time before a leading
operator would like to make a big entry in the sector.
Left alone, it would be difficult for the government to
carry the Akash forward on its own. It may at best
create a policy, but the real thrust will have to come
from basic infrastructural and social interventions that
increase the catchment area for Akash in future.
Conclusion
What do the above arguments imply? Is Akash too
unrealistic at the moment to be seen as a panacea for
Indias digital divide? Will it lead to greater deepening
of the digital divide than reducing it?
The authors would like to contend the opposite.
With a basic infrastructure in place, Akash is indeed a
good point to begin with. Indians over the years have
proved themselves capable of adapting and confirming
to the new technology showing a very unique blend of
the two dimensional HCI that was spoken of at the
beginning of the paper. For example, mobile phone,
which was considered a luxury ten years back, with its
limited coverage and applicability, is today an
important means of communication and work in the
hands of millions of low income Indians today. The
Akash with an appropriate mix of state policy and
private participation indeed can add one more
dimension to the perennial adaptability that the Indian
has shown himself to be capable of. Just to give an
example, a user of mobile in the rural hinterland, who
takes orders from a local agent for handicrafts to be
shown to an interested foreign tourist, may directly be
able to graphically display his product profile to the
user located abroad, reducing intermediary and
bringing greater accuracy and specialization in his
work thereby enhancing his operating margins by
volumes. However the really challenge shall remain in
creating the right kind of conditions for such a rural
artisan to acquire basic e-literacy and an operating
environment to sustain his activity at optimum costs.
Once done, Akash may become one key instrument,
but not the only instrument of bridging the economic
and social digital divide in India
1
Human Capability is described as those faculities related to
basic education, health, standard of living which enables
Human beings to make use of the opportunities being
opened by state and economic reform. For further details see
Amartya Senand Jean Dreze : India Economic Development
and Social Opportunity, Oxford University Press, New
Delhi, 1996

2
Wikepedia http://en.wikepedia.org/Digital_divide

3
Adapted from Neena Singh, Bridging the Digital Divide in
India: Some Challenges and Opportunities
http://www.worlib.org/vol17no1/singh_v17n1.shtml

4
Neena Singh, Bridging the Digital Divide in India: Some
Challenges and Opportunities
http://www.worlib.org/vol17no1/singh_v17n1.shtml
5
Wikepedia, http://en.wikipedia.org/Akash_tablet

6
Source : Wikipedia

7
James Fontanella-Khan, http://blogs.ft.com/beyond-
brics/2011/10/06/indias-digital-revolution-needs-more-than-
aakash/#ixzz1aEb3oB5n

A
p
m
u
o
s

i
p
B
v
t
t
G
c
h
w
p
e
d
s
c
h
r
s
s
r
v
s
Proc. of th
Volume 1. Co
ISBN: 978-8
GRID C
Abstract To
prefers to enj
multiple down
up or slows do
of the entire p
serious proble
So we
in this paper w
proposed Grid
By using our G
very fast depe
the Grid. We
this purpose.
Globus Archit
currently used
have proposed
we consider as
practically im
experience the
downloading m

Grid comp
simple yet l
computer ou
heterogeneous
resources. A
storage, analy
spectrometer,
resource. The
virtual sense
sharing and co

he Intl. Conf.
opyright 20
81-920575-7-6
ACM #: dber
COMPUT
oday we are in
joy fast acces
nloading, there
own the perform
process from t
ems that need
e have taken th
we are providin
d Model that c
Grid we can eas
ending on the
have used the
The Grid form
tecture, which
d worldwide fo
d an algorithm
s a blueprint fo
mplemented, o
e streak of lig
multiple files.
I. I
puting is all a
large and po
ut of a lar
s systems sh
'resource' m
ysis/visualizati
telescope an
e Grid aims to
-- allowing
ollaboration.
on Computer
012 Techno Fo
6 :: doi: 10. 74
r.imera.10. 74
TING - HI
n the Internet w
ss to the Inte
is a chance tha
mance that lea
the beginning.
the attention
his problem fo
ng a layout for
can access the
sily download
number of sy
e concept of G
mulated by us
h is the only
or developing
m for laying ou
or further imp
our Grid prov
ghtening over
INTRODUCT
about to create
owerful self
rge collectio
aring various
may be compu
ion tools, scie
nd so on) or
o bring all the
large or sma
Applications
orum Group, In
4180 /ISBN_0
4180
IGH LEV
F
R D
SSN
Kalavak
world and ever
ernet. But du
at the system h
ads to the resta
This is one o
of the researc
or our research
r implementing
Internet very
any number of
ystems employe
Grid Computin
uses the stan
Grid Architec
the Grid. An
ur Grid Model
plementation. W
vides the use
the Internet w
TION
e the illusion
managing vi
on of conne
s combination
uting power,
entific hardwa
any other rem
se together --
all scale reso

ndia.
768
VEL COM
FACILIT
Dinesh, C.Aj
N College of E
kkam,Chennai
ryone
ue to
hangs
arting
of the
chers.
h and
g our
fast.
f files
ed in
g for
ndard
cture
d we
l that
When
er to
while
of a
irtual
ected
ns of
data
are (a
mote
in a
ource
he
un
av
eq
wa
wh
ind
an
thr
wh
vie
ind
(p
ne
re
ha
va
the
tec
or
vir
pla
on
en
tra
MPUTING
TY
jit Nataraj
Engineering,
i,India-603 11
The standa
eterogeneous s
niformity for
vailability of
qually large ev
In a nutshel
asted CPU cy
hich can comp

II.
Grid compu
dividuals. Th
nalogy to powe
rough wall s
here or how t
ew of grid com
dividual user
rocessors, sto
eeded with l
sources are lo
ardware, opera
Grid compu
arious technol
e context o
chnologies th
rganizational a
rtualization
atforms, and o
nly achievable
nvironment th
ansparently a
G TO ACC
10
ardization of
systems create
r sharing r
higher bandw
volutionary ste
ll the grid com
ycles in the ne
pete with supe
GENESIS
uting can mean
he grand visio
er grids where
ockets with n
the electricity
mputing, comp
rs gain acce
orage, data,
little or no
ocated or what
ating system.
uting can be s
logies and sol
of distribute
hat are evol
application an
virtualizat
organizations.
e through the
hat provides
ccess resourc
CESS INT
f communica
ed the Interne
resources, al
width, are dri
ep in grid com
mputing is all
twork as one
ercomputer spe
OF GRID CO
n different th
on is often p
e users get acc
no care or c
is actually g
puting becom
ess to comp
applications,
knowledge o
t the underlyi
seen as a pat
lutions. Its ke
d computing
lving in sup
nd resource sh
tion across
. This kind of
e use of open
the ability
ces across a
TERNET
ations betwe
t explosion. T
long with t
iving a possib
mputing.
about using t
virtual machi
eed.
OMPUTING
hings to differ
presented as
cess to electric
onsideration
enerated. In t
mes pervasive a
puting resourc
and so on)
of where tho
ing technologi
th of integrati
ey values are
g infrastructu
pport of cro
haring in a wo
technologi
f virtualization
n standards. A
to share a
distributed a

een
The
the
bly
the
ne,
ent
an
city
for
this
and
ces
as
ose
ies,
ing
on
ure
oss-
ord,
ies,
n is
An
and
and
h
t
t
s
c
b
i
l
i
w
n
r
r
n
g
c
s
p
heterogeneous
technology t
technologies
security, acco
computing can
be able to id
informing man

The Grid a
layer is defin
independently
well-organize

a.Fabric:
Fabric is th
normal comp
represents log
represents lo
networking an

b. Connect
Connectiv
giant node of
connections a
such as Int
protocols (DN

s environme
to virtualize
and standard
ounting, syst
n come into p
entify importa
nagement of th
III. GRID
architecture h
ed in their, o
y. On integrat
d grid model.
he lowest laye
puter architec
gic gates, the f
cal computin
nd computatio
tivity:
vity layer con
f fabric. Con
and is implem
ernet protoco
NS), etc.
ent not on
certain reso
ds in the are
tems managem
play here too. V
ant trends thr
hose that requ
D ORANIZAT
has a layered a
wn behaviora
tion it works
er in grid arch
cture where
fabric is an ab
ng resources
nal resources.
nnects several
nnectivity laye
mented using n
ol (TCP/IP)
ly requires
ources, but
as of schedu
ment. Autono
Various tools
roughout the
uire attention.
TION:
approach and
al manner to w
as one comp
hitecture. Unlik
the lowest l
bstract layer w
such as stor

l fabrics into
er provides se
network proto
and applica
the
also
uling,
omic
may
grid,
each
work
mplete
ke in
layer
which
rage,
one
ecure
ocols
ation
co
pr
loa
ne
an
am
dir
se
arc
de
we
be

a.Resource:
Resource la
onnectivity lay
rotocols used
ad and usage
egotiate the p
nd operations.

b. Collective
Collective la
mong several
rectory servi
rvices, and sch

c. Applicatio
Application
chitecture. T
esired actions.
e are working

The three p
elow:

ayer deals w
yers. Resourc
to obtain info
e policies, an
policies for ha
e:
ayer consist o
different res
ices, account
heduling servi
on:
layer is the hi
his layer cal
. Application
with to solve
IV. GR
rimary types
with manage
ce layer can
ormation abou
nd managemen
andling resou
of the protocol
sources. This
ting paymen
ices to name a
ighest layer in
lls other lay
layer is simp
our problems
RIDS & TYPE
of grids and
ement of ma
be informati
ut configuratio
nt protocol th
urce requirem
ls of interactio
s layer includ
t, collaborati
a few.
n grid computi
yers to perfo
ply the progr
.
ES:
are summariz

any
ion
on,
hat
ent
ons
des
ion
ing
orm
am
zed
a.Computational grid
A computational grid is focused on setting aside
resources specifically for computing power. In this type of
grid, most of the machines are high-performance servers.
b.Scavenging grid
A scavenging grid is most commonly used with large
numbers of desktop machines. Machines are scavenged for
available CPU cycles and other resources. Owners of the
desktop machines are usually given control over when their
resources are available to participate in the grid.
c.Data grid
A data grid is responsible for housing and providing
access to data across multiple organizations. Users are not
concerned with where this data is located as long as they
have access to the data.

.GRID MODEL PROPOSED:
We are using the Scavenging Grid for our
implementation as large numbers of desktop machines are
used in our Grid and later planning to extend it by using
both Scavenging and data Grid. Figure2 gives an idea
about the Grid that we have proposed.

e.PROBLEMS DUE TO MULTIPLE
DOWNLOADING:
While accessing Internet most of us might have faced
the burden of multiple downloading and in particular with
downloading huge files i.e., there can be a total abrupt
system failure while a heavy task is assigned to the system.
The system may hang up and may be rebooted while some
percentage of downloading might have been completed.
This rebooting of the system leads to download of the file
once again from the beginning, which is one of the major
problems everyone is facing today.
Let us consider N numbers of files of different
sizes (in order of several MBs) are being downloaded on a
single system (a PC). This will take approximately some
minutes or even some hours to download it by using an
Internet connection of normal speed with a single CPU.
This is one of the tedious tasks for the user to download
multiple files at the same time. Our Grid plays a major role
here.

d.EMPLOYING THE GLOBUS ARCHITECTURE
IN PROPOSED GRID:
While planning to implement a Grid project, we must
address issues like security, managing and brokering the
workload, and managing data and resources information.
Most Grid applications contain a tight integration of all
these components.
The Globus Project provides open source software
tools that make it easier to build computational Grids and
Grid-based applications. These tools are collectively called
the Globus Toolkit. Globus Toolkit is the open source Grid
technology for computing and data Grids. On the server
side, Globus Toolkit 2.2 provides interfaces in C. On the
client side, it provides interfaces in C, Java language, and
other languages. On the client side, the Java interfaces are
provided by the Java Commodity Grid (CoG) Kit. Globus
runs on Linux, AIX, HP-UX, Solaris, and also on windows
operating systems.
The Globus architecture represents a multiple-layer
model. The local services layer contains the operating
system services and network services like TCP/IP. In
addition, there are the services provided by cluster
scheduling software (like IBM Load Leveler) -- job-
submission, query of queues, and so forth. The cluster
scheduling software allows a better use of the existing
cluster resources. The higher layers of the Globus model
enable the integration of multiple or heterogeneous
clusters.
The Globus Toolkit (GT) was developed by Global
Alliance, a division of Global Grid Forum. Global Alliance
comprises of R&D research groups based at several
universities such as the University of Chicago, the
University of Edinburgh and the University of Southern
California. GT is the de facto standard for grid computing
[2] and it is comprised of 3 main services:

The core services:
Basic infrastructures to enable grid computing such as:
resource management for naming and locating
computational resources on remote systems, security and
system level services, and monitor status.

Security services:
Security is implemented using the standard GSI (Grid
Security Infrastructure) and CAS
(Community Authorization Service). GSI offers
services such as basic certification, PKI, and many other
security libraries.

Data/Resource Management
Protocols to ensure rapid and secure data transfer
among resource nodes. There are 4 protocols:
GridFTP, Reliable File Transfer (RFT), Replica
Location Service (RLS), and Extensible
Input/output (XIO). GRAM (Globus Resource
Allocation Managers) is the Data Management for GT.
TeraGrid [5], TIGER, and Taiwan UniGrid [6] is examples
of grid projects that use GT.

f.Grid Security Infrastructure (GSI):
GSI provides elements for secure authentication and
communication in a grid. The infrastructure is based on the
SSL protocol (Secure Socket Layer), public key
encryption, and x.509 certificates. For a single sign-on,
Globus add some extensions on GSI. It is based on the
Generic Security Service API, which a standard API is
promoted by the Internet Engineering Task Force (IETF).

These are the main functions implemented by GSI:
Single/mutual authentication
Confidential communication
Authorization
Delegation

e.Grid Resource Allocation Manager (GRAM):
GRAM is the module that provides the remote
execution and status management of the execution. When a
job is submitted by a client, the request is sent to the
remote host and handled by the gatekeeper daemon located
in the remote host. Then the gatekeeper creates a job
manager to start and monitor the job. When the job is
finished, the job manager sends the status information back
to the Client and terminates.
Global Access to Secondary Storage (GASS)
GRAM uses GASS for providing the mechanism to
transfer the output file from servers to clients. Some APIs
are provided under the GSI protocol to furnish 138
Introduction to Grid Computing with Globus

Secure transfers. This mechanism is used by the
globusrun command, gatekeeper, and job manager.

f.Monitoring and Discovery Service (MDS):
MDS provides access to static and dynamic
information of resources. Basically, it contains the
following components:
Grid Resource Information Service
(GRIS)
Grid Index Information Service (GIIS)
Information Provider
MDS client

h.Grid Resource Information Service (GRIS)
GRIS is the repository of local resource information
derived from information providers. GRIS is able to
register its information with a GIIS, but GRIS itself does
not receive registration requests. The local information
maintained by GRIS is updated when requested, and
cached for a period of time known as the time-to-live
(TTL). If no request for the information is received by
GRIS, the information will time out and be deleted. If a
later request for the information is received, GRIS will call
the relevant information provider(s) to retrieve the latest
information.

i.Grid Index Information Service (GIIS):
GIIS is the repository that contains indexes of resource
information registered by the GRIS and other GIISs. It can
be seen as a grid wide information server. GIIS has a
hierarchical mechanism, like DNS, and each GIIS has its
own name. This means client users can specify the name of
a GIIS node to search for
information.

j.Security
Security is a much more important factor in planning
and maintaining a grid than in conventional distributed
computing, where data sharing comprises the bulk of the
activity. In a grid, the member machines are configured to
execute programs rather than just move data. This makes
an unsecured grid potentially fertile ground for viruses and
Trojan horse programs. For this reason, it is important to
understand exactly which components of the grid must
be rigorously secured to deter any kind of attack.
Furthermore, it is important to understand the issues
involved in authenticating users and properly executing the
responsibilities of a Certificate Authority.

h.Certificate authority
The primary responsibilities of a Certificate Authority
are:
Positively identify entities requesting
certificates
Issuing, removing, and archiving
certificates
Protecting the Certificate Authority
server
Maintaining a namespace of unique
names for certificate owners
Serve signed certificates to those
needing to authenticate entities
Logging activity

Briefly, a Certificate Authority is based on the public
key encryption system. In this system, keys are generated
in pairs, a public key and a private key. Either one can be
used to encrypt some data such that the other is needed to
decrypt it. The private key is guarded by the owner and
never revealed to anyone. The public one is given to
anyone needing it. A Certificate Authority is used to hold
These public keys and to guarantee who they belong to.
When a user uses his private key to encrypt something, the
receiver uses the corresponding public key to decrypt it.
The receiver knows that only that users public key can
decrypt the message correctly. However, anyone could
intercept this message and decrypt it because anyone can
get the originators public key. If the originator instead
doubly encrypts the message with his private key and the
intended recipients public key, a secure communication
link is formed. The receiver uses his private key to decrypt
the message and then uses the senders public key for the
second decryption. Now the recipient knows that if the
message decrypts properly, then only the sender could
have sent it and furthermore, the sender knows that only
the intended receiver can decrypt it. The beauty of all of
this is that nobody had to securely carry an encryption key
from the sender to the receiver,

k.Schedulers
Most grid systems include some sort of job scheduling
software. This software locates a machine on which to run
a grid job that has been submitted by a user. In the simplest
cases, it may just blindly assign jobs in a round-robin
fashion to the next machine matching the resource
requirements. However, there are advantages to using a
more advanced scheduler. Some schedulers implement a
job priority system. This is sometimes done by using
several job queues, each with a different priority. As grid
machines become available to execute jobs, the jobs are
taken from the highest priority queues first. Policies of
various kinds are also implemented using schedulers.
Policies can include various kinds of constraints on jobs,
users, and resources. For example, there may be a policy
that restricts grid jobs from executing at certain times of
the day.
More advanced schedulers will monitor the progress of
scheduled jobs managing the overall work-flow. If the jobs
are lost due to system or network outages, a good
scheduler will automatically resubmit the job elsewhere.
However, if a job appears to be in an infinite loop and
reaches a maximum timeout, then such jobs should not be
rescheduled. Typically, jobs have different kinds of
completion codes, some of which are suitable for re-
submission and some of which are not.

V. Monitoring progress and
recovery
A grid system, in conjunction with its job scheduler,
often provides some degree of recovery for sub jobs that
fail. A job may fail due to a:
Programming error: The job stops part
way with some program fault.
Hardware or power failure: The machine
or devices being used stop working in some way.
Communications interruption: A
communication path to the machine has failed or
is overloaded with other data traffic.
Excessive slowness: The job might be in
an infinite loop or normal job progress may be
limited by another process running at a higher
priority or some other form of contention.
It is not always possible to automatically determine if
the reason for a jobs failure
is due to a problem with the design of the application or
if it is due to failures of
various kinds in the grid system infrastructure.
Schedulers are often designed to
categorize job failures in some way and automatically
resubmit jobs so that they
are likely to succeed, running elsewhere on the grid.

i.Communications
A grid system may include software to help jobs
communicate with each other. The open standard Message
Passing Interface (MPI) and any of several variations is
often included as part of the grid system for just this kind
of communication

j.ACCESSING THE INTRANET GRID:
When any user wants to access our proposed Intranet
Grid in order to download multiple files over the Internet,
then he should follow certain procedures that we consider
necessary for the security of our Grid. The main
Requirements for Processing in Grid Environment are:
Security: Single sign-on, authentication,
authorization, and secure data transfer.
Resource Management: Remote job submission
and management.
Data Management: Secure and robust data
movement.
Information Services: Directory services of
available resources and their status.
Fault Detection: Checking the
intranet.
Portability: C bindings (header files) needed to
build and compile programs
.
VI. Existing Algorithm for Globus Architecture
Create security proxy via GSI
services
Access a MDS-GIIS server
Search for required machine(s)
Rank the machine list based on a scheduling
policy
Prepare the data
Transfer the data to the target machine by
using GASS services
Prepare a RSL document
Submit the program using GRAM services
React to status changes from GRAM
Get results via GASS

Here, we have got the resources available in the
Network which is automatically done by have the Globus
Toolkit in the server. When we want to download a file this
information has to be matched with the client module and
then the downloading has to be carried out in the clients.
For this we have added some modules to the Grid
Architecture.
Added module:
Get the Information about files to be
downloaded.
Match the files with appropriate Machines.
Store files in common database.
Retrieval of data from database is done after
proper authentication.

20. PPROPOSED ALGORITHM FOR
INTRANET PROPOSED GRID:
Steps to perform multiple downloading in the Grid, The
host details are got from the server of the LAN in order to
identify the various hosts.

The host information is got whenever needed on the
priority queue basis.
//module for downloading files
Start lookup // look for file size and
resource information
Declare nres, nfile // no of resources
available and no of files
Input nres, nfiles
Input size // the file size
Initialize P1 res info // store the
resource information in priority queue P1 with highest
system configuration as priority
Initialize P2 file size // store the file
information in the priority queue P2 with maximum file
size as priority
If condition (nfiles == nres) // check
whether the no of resources is equal to no of files
Initialize counter
For (counter =1; counter <= nres;
counter++) // initialize the loop to assign the files.
Assign the 1
st
file of P2 to the 1
st
node in P1. //
first node will be node with highest configuration and
first file will be the file with maximum size.
Start processing // files directed to the
appropriate system for accessing their wasted CPU
cycles.
Loop
Else:
Start timer
Delay 1 min
Collect incoming files // the files that the user
clicked to download in this duration.
Assign the files P2
Goto step 8
Goto step 1
End // when the user exits from proposed
Grid.

21. BENEFITS OF GRID
COMPUTING:
Grid computing enables organizations to
aggregate resources within an entire IT infrastructure no
matter where in the world they are located. It eliminates
situations where one site is running on maximum capacity,
while others have cycles to spare.
Grid computing enables companies to access and
share remote databases. This is especially beneficial to the
life sciences and research communities, where enormous
volumes of data are generated and analyzed during any
given day.
Grid computing enables widely dispersed
organizations to easily collaborate on projects by creating
the ability to share everything from software applications
and data, to engineering blueprints.
Grid computing can create a more robust and
resilient IT infrastructure better able to respond to minor
or major disasters.
A grid can harness the idle processing cycles that
are available in desktop PCs located in various locations
across multiple time zones. For example, PCs that would
typically remain idle overnight at a company's Tokyo
manufacturing plant could be utilized during the day by its
North American operations.
V. CONCLUSION
Grid computing was once said to be fading out but due
to the technological convergence it is blooming once again
and the Intranet Grid we have proposed adds a milestone
for the Globalization of Grid Architecture, which, leads to
the hasty computing that is going to conquer the world in
the nearest future. By implementing our proposed Intranet
Grid it is very easy to download multiple files very fast and
no need to worry about the security as we are
authenticating each and every step taking place in our Grid
and in particular user to access the database. Further
implementations could be carried out in the nearest future.

REFERENCES
[1].P Plaszczak, R Wellner, Grid computing, 2005, Elsevier/Morgan
Kaufmann, San Francisco
[2]Francesco Lelli, Eric Frizziero, Michele Gulmini, Gaetano Maron,
Salvatore Orlando, Andrea Petrucci and Silvano Squizzato. The many
faces of the integration of instruments and the grid. International Journal
of Web and Grid Services 2007 Vol. 3, No.3
[3]Fran Berman, Geoffrey Fox, Anthony J. G. Hey Grid computing, 2005,
Elsevier/Morgan Kaufmann, San Francisco
[4]Pawel Plaszczak, Richard Wellner
Jos Cardoso Cunha, Omer Rana
Lizhe Wang, Wei Jie, Jinjun ChenFundamentals of Grid Computing:
Theory, Algorithms and Technologies BY Frdric
[5]Advances in grid computing -- EGC 2005: European Grid Conference,
... Peter Sloot
[6]Grid computing: the new frontier of high performance computing-
Lucio Grandinetti
[7] Distributed data management for grid computing-Michael Di Stefano
[8]Grid Technology and Applications: Recent Developments-G. A.
Gravvanis, H. R. Arabina,
[9]Advances in grid and pervasive computing: third international-Song
Wu, Laurence Tianruo Yang, Tony Li Xu
ISBN: 978-81-920575-7-6 :: doi: 10. 74187 /ISBN_0768
NANOTECHNOLOGY THE NEW UPGRADE FOR COMPUTERS

A.RAMYASREE
Affiliated to Anna University, Chennai.
B.E Final Year CSE,
MAILAM ENGINEERING COLLEGE. Mailam.
Tamil Nadu, India.

Abstract The Next Big Thing Is Really Small - K. Eric
Drexler, Founder and Chairman, Foresight Institute, and
Predictions, observing the rise of the human dependence on
the sector of Nano Technology. Nano technology is not just
studying every field in a larger scope but deriving an
enormous development in improvable areas like usage of
NVRAM & larger memory storage in it, Use of the
nanomagnetism initiative which provides an interdisciplinary
framework to help stage the next advance in complex
materials research that can be used in Computer sciences
Application of nanocrystals & nanocomputer, Molecular and
Quantum and Nano Computing and Nanomechanical
computing, Usage of Graphene chips instead of silicon.

I. INTRODUCTION
A basic definition: Nanotechnology is the engineering
of functional systems at the molecular scale.
This covers both current work and concepts that are
more advanced. In its original sense, 'nanotechnology'
refers to the projected ability to construct items from the
bottom up, using techniques and tools being developed
today to make complete, high performance products.
Driven by rapid technological advances within the past
two decades, computing and high-speed networkings have
emerged as powerful tools for science and are even
changing the ways in which modern science is conducted.
DOE is a national leader in the scientific computing field
supporting fundamental research in advanced scientific
computing, applied mathematics, computer science, and
networking. The DOE computational infrastructure
provides world-class, high-performance computational and
networking tools that enable scientific, energy,
environmental, and national security research.
More than 2400 scientists in universities, federal
agencies, and U.S. companies use DOE-(DEPARTMENT
OF ENERGY-Washington D.C) funded high-performance
computers. Research communities that benefit from these
resources include structural biology; superconductor
technology; medical research and technology development;
materials, chemical and plasma sciences; high energy and
nuclear physics; and environmental and atmospheric
research.
II. NANO COMPUTING

Quantum Nano computing and quantum computers are
in advanced stages of development due to recent
nanotechnology advances in these areas. These computers
promise to be so fast and powerful that our modern
supercomputers will be little more than toys in years to
come. The problem is that their sheer capacity for
processing data means that they will likely be able to
readily crack even the most advanced encryption methods
through brute force.
Computers play a pivotal part in nanotechnology as
they allow us to develop atomic models and foresee the
Nano devices we desire. Furthermore, as nanotechnology
evolves, new, more powerful computers will emerge,
which will lead to breakthroughs in nanotechnology and
yet to more powerful computers. Hence we will have a
cycle of scientific discovery unprecedented in human
history. It will be the singularity.

A. CONSEQUENCES:

Because of the singularity, it is shortsighted to discuss
the applications of nanotech, but here's an attempt: In
computer science, we would have nanocomputer that
would open the door for androids and artificial
intelligences. Nanotech would develop an army of
molecular robots and nanodevices that would allow us to
completely dominate Nature. We now dominate it at a
macroscopic level; we would then dominate it at a
microscopic level, too. With molecular medicine we would
eliminate many, if not all, diseases with intelligent
molecules capable of repairing and changing our cells. Just
for curiosity, my first idea towards ending aging, inspired
by the legendary TV series Transformers which we used to
watch growing up, was to use microscopic devices capable
of turning our neurons into metal ones with a far greater
strength and therefore impossible to kill.
In synergy with other technologies such as genetic
engineering, the changes each of us could pursue would be
staggering; from having a skin invulnerable to bullets to
augmenting our physical and psychological capacities.
Industrial, agricultural, and environmental applications are
almost unimaginable at this stage: eradicating hunger,
creating molecular superconductors at room temperature,
developing nanomachines that eliminate pollution, etc.
a
a
r
b
i
t
b
p
o
s
i
b
h
t
e
o
t
d
I
t
n
r
d
u
c
o
n
r
n
N
t
d
r
r
t
r
Importantl
and colonizati
amazing mol
reproducing.
being started
instructions to
these molecul
behind nanote
properly code
of complexity
scouts in sp
inherited pow
bacterium wit
hours, if allow
the weight of
enough mass/
outposts and
theory but it
dealing with.
There are
I also mentio
theorized by
nanotechnolog
replicating ai
dioxide into
unleashed and
cover the eart
on earth. R
nanotechnolog
review of th
nanotech are
Nanotech will

B. FAST
WITH NANOT
Smaller, l
electrical fail
inaccessible
research Lab
"Tiny" me
the width of
differently th
researchers h
reactive, exhib
to create mat
resistant to fri

ly, we would
ion. Each nan
ecular engine
Remember th
as a tiny c
o blossom in
lar assemblers
ech is to creat
ed, can give ris
y. Thus nano
ace explorati
wer of nanot
th a generation
wed to grow,
the earth. Thi
/energy has t
life throughou
shows the d
some dangers
on elsewhere.
Ralph Merkl
gy. Basically
irborne nano
graphite. S
d, in as little as
th, block the s
Robert Freita
gy, offers an
he dangers o
e another re
l forever chang
TER, LIGHTE
TECHNOLOG
lighter compu
lures sending
limbo mark t
at Chicago on
eans billionth
a human hai
han their larg
have learned t
bit new electr
terials that ar
iction and wea
d be capable o
nomachines wo
eering achiev
hat every sin
cell containin
nto an adult b
s are like a hum
te such tiny na
se to machine
odevices woul
ion. To give
technology, j
n time of 20
reach a popu
is means that
the potential
ut the galaxy.
dimension of
inherited to n
For example
le as a doom
y, the grey
odevice that
Should such
s a few days,
sun, and even
as Jr., a w
optimistic but
of nanotech.
eason for sp
ge humanity.
ER COMPUTE
GY.

uters and an en
hours of on-sc
the potential
n tiny ferroele
s of a meter,
ir. These nano
er bulk coun
that they are
ronic propertie
re stronger, t
ar than bulk m
of space conq
ould be capab
vements and
ngle adult hu
ng all the ge
being. In a se
man egg. The
anodevices th
s with a high
ld make exce
e an idea of
just think th
minutes can i
ulation 4000 t
a nanodevice
to create pla
. Of course th
the powers w
nanotechnolog
e, the grey cl
msday offsprin
cloud is a
catalyzes ca
h nanodevice
a solid wall w
ntually destroy
world expert
t well-docume
The danger
pace coloniza
ERS POSSIBLE
nd to worries a
creen work int
result of Arg
ectric crystals.
or about 1/5
omaterials be
nterparts. Arg
more chemi
es and can be
tougher and m
materials.
quest
ble of
even
uman
netic
ense,
goal
hat, if
level
ellent
f the
hat a
in 48
times
with
anets,
his is
we're
gy, as
loud,
ng of
self-
arbon
e be
would
y life
t on
ented
rs of
ation.
E
about
to an
onne

500th
ehave
onne
cally
used
more
tim
In
dia
dia
so
co
sta
wh
ha
fer
fie
sim
ge
sy
on
ch
me
or
an
ho
ap
me
in
ma
po
ba
an
ph
of
W
no
ac
fer
vo
inh
co
co
of
da
Fig: Shows
mes smaller in
n comparison t
ameter, the N
ameter.

III
a.RAM & NV

RAM Ra
omeone enters
omputer. It ca
andard comm
hile powered
ave created
rroelectric ma
eld and retain
Ferroelectric
milarly to ferr
enerally conta
ymmetry cause
ne or more of
hange this pola
echanical to e
r electrical ene
A strong ex
nd minus pole
old their orien
pplied electric
emory, repres
the other.
Because the
ade with them
ower failure.
ack-up batterie
nd lighter. Th
hones.
Achieving s
f the computer
Companies
Westinghouse E
on-volatile RA
chieve commer
Back then,
rroelectric sin
oltage to swi
herent in the
oncept, made t
The usage o
omputers as it
f Data Storage
ata mining and

a picture of N
n diameter tha
to a human ha
Nano fibers
. COMPUTIN
VRAM. (Nonvo
andom Acces
s information
an be written
mercial technol
by electricity
and are stud
aterials that ca
any changes.
c materials s
romagnetic ma
ain iron co
es spontaneou
f their axes. T
arity. Ferroele
electrical ener
ergy to optical
xternal electric
es of ferroelec
ntation until
c field. Thus,
senting "zero"
crystals do no
m would not
Laptop comp
es, permitting
here would b
such permanen
r industry.
such as AT
Electric made
AMs (NVRAM
rcial use.
NVRAMs
ngle crystals,
itch their po
then recently
them impractic
of NVRAM is
may lead fur
e and unearths
d Data storage
Nano fibrils th
an convention
air which is c
are 1,000 ti
NG APPLICAT

olatile RAM)
ss Memory
or gives a c
to as well as
logy - holds
. Argonne ma
dying nanosc
an be altered
so called, beca
aterials even th
onsist of crys
us electrical po
The application
ectric crystals
rgy the piezo
effects.
cal field can
ctric polarizati
forced to cha
they can be
in one orient
ot revert spon
be erased sh
puters would
them to be m
be a similar
nce is a long
T&T, Ford, I
e serious eff
M) in the 195
were based
, which requ
olarity. This,
y devised row
cal.
s a good optio
rther towards t
s new concept
.
hat are 10 to 1
nal textile fibe
ca. 80,000 nm
imes smaller
TIONS
is used wh
command to t
read but - w
its content on
aterials scienti
cale crystals
by an electri
ause they beha
hough they do
stals whose l
olarization alo
n of voltage c
can also chan
oelectric effec
reverse the p
ion. The cryst
ange by anoth
coded as bina
tation and "on
ntaneously, RA
hould there be
no longer ne
made still smal
impact on c
g-standing drea
IBM, RCA a
forts to devel
50s, but could
on expensi
uired substant
and cross t
w matrix addr
on in the field
the developme
ts in the arena
100
ers.
m in
in
hen
the
with
nly
ists
of
ical
ave
on't
ow
ong
can
nge
ct
lus
tals
her
ary
ne"
AM
e a
eed
ller
cell
am
and
lop
dn't
ive
tial
alk
ess
d of
ent
a of
d
a
o
c
a
s
r
i
c
s
f
e
m
w
t
h
l
A
r
r
o
f
C
t
s
t
m

Working o
density memo
amount of me
of informatio
capacity. It's
available, but
some compan
research withi
a.SMART
Scientists
improve smar
cards but co
substantial inf
for use by doc
emergency. U
memories do
will not wear

Current sm
memory.
the Colorado
higher capaci
lasting memo
Argonne since
research topic
research cente
of Energy (DO
Nanomate
The effort
frontier wher
Chemists wor
the smallest
scientists wor
together as we
Materials
100 nanomet
molecular clu
Fig: A
on the nanosca
ories with fas
emory needed
n) - or even
not clear ho
t competition
nies claim they
in less time.
CARDS
are using the
rt cards. These
ontain ferroele
formation, suc
ctors, pharmac
Unlike magnet
not come in
out.
Fig: A S
mart cards carr
Argonne rese
Springs, Colo
ity card with
ory. Nanoma
e the 1980s. T
cs worldwide
ers are being p
OE), with one
erials challeng
t to understand
re physics,
rk with atoms
particles to
rk from large
e approach the
behave differe
ters (billionth
sters at this si
AN NVRAM
ale changes th
ster speeds an
d to store one
gigabyte (on
ow soon such
is heavy, sta
y will have the
eir expertise i
e are the size a
ectric memor
ch as its beare
cists and even
tic strips on c
contact with

SMART CAR
ry about 250 k
earchers are c
o., Symetrix C
h a more flex
aterials have
They are now
e. Several na
planned by the
likely to be bu
ge researchers
d ultra-small m
chemistry an
s and molecu
larger ones
er materials d
e nanoscale.
ently in the ra
hs of a me
ize may be dif
is. It means hi
nd megabyte
million chara
ne billion byt
h capacity wil
akes are high,
e first fruits of
in ferroelectric
and shape of c
ry that can c
r's medical his
n paramedics i
credit cards, t
their readers
RD
kilobytes of
collaborating
Corp. to devel
xible and lon
been studie
one of the ho
anoscale mate
e U.S. Departm
uilt at Argonn
s:
materials is on
nd biology m
ules, moving
s, while phy
down. They c
ange of size b
eter. Atomic
fferent colors
igher
(the
acters
es) -
ll be
, and
f this
cs to
credit
carry
story
in an
these
s and
with
lop a
nger-
d at
ottest
erials
ment
ne.
n the
meet.
from
ysical
come
elow
and
from
the
mo
pr
an
sc
un
mu
ce
Na
we
co
the
W
im
ox
su
ma
na
Na
me
fea
no
the
me
wh
pr
Th
na
be
ma
us
to
co
re
tog
us
me
of
re
of
sm
tec
bu
sc
co
im
int
int
e same eleme
ore chemical
roperties. Obje
nd tougher th
ience that is th
nderstood.
Nanoscale f
ultilayer capac
ell phones, ap
ano-electro-m
Nature likes
e learn more
onstruction at
e nanoworld
We should fin
magine yet.
It is synerg
xide materia
uperconductors
agneto resistiv
Ferroelectric
anomaterials,
ano crystallin
edical, transpo
atured in logo
ovel nanosize
ere are as ma
etals, etcIn
here two or m
roximity effect
he possibilitie
anotechnology
b.Closeness
Such proxim
ecause the ma
agneto-resista
sed in comput
read individu
opper, iron an
sistance in ma
gether. IBM
sed this to
echanisms. Th
f phenomena.
search that DO
f $1.2 million
mall clusters,
chnologies - s
ulk materials fr
We want to
ales and are
ommercially v
mportance from
to technologic
The nano
terdisciplinary
ents or compo
lly reactive
ects assemble
han their bulk
he foundation
ferroelectrics
citors, which
pplication in m
mechanical syst
s to put things
about nanosc
the nanoscale
differently an
nd new mate
gistic with a
als, including
s and thermal
ve materials.
cs are just o
Gibson said.
ne diamond fi
ortation, indu
os Nano Scie
effects have
any or more
n addition, th
more of the abo
ts, which can b
s for creating
y are endless.
breeds materi
mity effects
aterials are s
ance, a phenom
ter hard drives
ually, but inte
nd chromium
agnetic fields b
and the magn
create ultra
he Nano-comm
The research
OE recently a
n a year. It in
, and carrie
such as semic
from a micro-s
know how pr
finding new
viable. Nano
m how quickly
cal application
omagnetism
y framework t
ounds in bulk
and display
ed from them
k counterparts
n for the techn
allow us to
could be used
motors to po
tems.
s together in c
cience - when
e - we will be
nd create nove
erials, things
broader stren
g related h
l barrier coatin
ne of the ho
Argonne's w
films for micr
strial and aero
ence Summit.
been found i
in magnets, s
here are comp
ove are combi
be dominant o
new useful m
ial changes:
changes in m
o close sh
menon discove
s. Tiny magne
erleaved nanol
show substan
because the la
netic recordin
a sensitive h
munity looks
h initiative in
approved for f
ncludes atoms
es forward
conductors - b
structural view
roperties chang
effects, som
oscience draw
y we've been a
ns."
initiative
to help stage t
k. They becom
new electron
can be strong
s. However, t
nology is still n
develop bet
d in even smal
ower micro- a
certain ways.
n we can cont
able to engine
el combinatio
we can't ev
ngth in compl
high-temperatu
ngs and colos
ot new areas
work with ult
ro machines
ospace uses w
Although ma
in ferroelectri
superconducto
posite materia
ined, introduci
on the nanosca
materials throu
material behav
how up in gia
ered in 1988 a
etic bits are ha
layers of coba
ntial changes
ayers are so clo
ng industry ha
hard-drive re
at a wide ran
nanomagneti
funding at a r
s, molecules a
some existi
by understandi
w.
ge at the smal
me of which
ws some of
able to turn the
provides
the next advan
me
nic
ger
the
not
tter
ller
and
As
trol
eer
ons.
ven
lex
ure
ssal
in
tra-
for
was
any
ics,
ors,
als,
ing
ale.
ugh
vior
ant
and
ard
alt,
in
ose
ave
ead
nge
ism
rate
and
ing
ing
ller
are
its
ese
an
nce
in complex materials research. It takes a broad approach,
working with materials that fall from around one micron
(one millionth of a meter) in size to less than 10
nanometers. As the scale decreases, the dominant physics
changes, and new materials, properties and applications
emerge.
Computer world might one day be based in magnetic
properties instead of electrical. This might make it possible
to build computers with architectures that could be
restructured depending on the task of the moment. The
same machine could be configured like a Macintosh for
tasks that a Mac operating system performs best and like a
PC when Windows OS is preferable.
Also possible could be magnetic configurations that
would not be limited by binary logic, making them more
like the human brain. This is far away, but promising.
Studies on the nanoscale could lead to better bulk
magnets and more efficient motors with consequent
savings in the use of fossil fuels. It may also become
possible to incorporate magnetic molecules in polymers,
creating plastics that could be used where traditional
magnets cannot, for example in certain corrosive
environments.

IV. MOLECULAR COMPUTING
Molecular computing uses switches made of a single
protein molecule. Proteins "can self-assemble complex
arrays of molecules on a surface [nanometer-scale
patterns]....Many proteins do have charge-transfer and
charge-switching properties, which could...be harnessed to
provide some aspects of the information processing
capability of a semiconductor device. Longmuir-Blodgett
films - thin films of lipids - are known to be an essential
part of the electrical properties of nerve cells, and can be
made quite readily in the laboratory. Nerve-cell proteins
are inserted into the lipid film which alters the film's ability
to let ions pass depending on what other ions are present or
on the electric field to which they are exposed. This has
progressed to the stage of building films, putting proteins
into them, and demonstrating the electrical characteristics
of the protein, which is similar to the position with
transistors in 1930
C. QUANTUM COPMPUTERS
Seth Lloyd of Los Alamos National Laboratory
proposes computers "made of weakly coupled quantum
systems arranged in one-, two-, or three-dimensional
arrays. Computations are performed in response to
sequences of light pulses imposed upon the entire array
simultaneously. Lloyd shows that by an appropriate choice
of materials and architecture, and appropriate choices of
pulses, each subunit in the array will change to a new
quantum state in a manner that depends both on its
previous state and upon the pulse sequence. He presents a
simple proof that such an array could carry out the full
range of logic operations required of any computer."
For any given pair of quantum states in the set of
allowed states of such a device, there are sequences of
pulses which, if applied at the proper times, will switch the
device from one state to the other. Since the physical
conformation of an object is determined by its quantum
state, the conclusion is inescapable that an appropriately
designed quantum "computer" can be used as a quantum-
mechanical micromanipulator.
The quantum states of an effective manipulator would
need to correspond to useful configurations of some kind
of molecular arm. Manipulators would be designed
specifically for that purpose rather than for carrying out
general computations. Both types of devices would,
however, be based on the same general principles and
would be controlled by sequences of light pulses.

V. NANOMECHANICAL COMPUTERS

Even with a billion bytes of storage, a Nanomechanical
computer could fit in a box a micron wide, about the size
of a bacterium. Although mechanical signals move about
10
5
times slower than the electrical signals in today's
machines, they will need to travel only 10
-6
as far, and thus
will face less delay."

VI. NEXT GENERATION NANOTECHNOLOGY
COMPUTER MEMORY MADE OF GRAPHENE:
Carbon comes in many different forms, from the
graphite found in pencils to the world's most expensive
diamonds. In 1980, we knew of only three basic forms of
carbon, namely diamond, graphite, and amorphous carbon.
In recent times Graphene was also discovered which
caused a lot of fuzz and interest.
Experiments with Graphene have revealed some
fascinating phenomena that excite nanotechnology
researchers who are working towards molecular
electronics. It was found that Graphene remains capable of
conducting electricity even at the limit of nominally zero
carrier concentration because the electrons don't seem to
slow down or localize. This has to do with the fact that the
electrons moving around carbon atoms interact with the
periodic potential of graphemes honeycomb lattice, which
gives rise to new quasi particles that have lost their mass,
or 'rest mass' (so-called mass less Dirac fermions). This
means that Graphene never stops conducting.
For computer chips, 'smaller and faster' just isn't good
enough anymore. Power and heat have become the biggest
issues for chip manufacturers and companies integrating
these chips in everyday devices such as cell phones and
laptops. The computing power of today's computer chips is
provided mostly by operations switching at ever higher
frequency. This physically induced power dissipation
represents the limiting factor to a further increase of the
capability of integrated circuits. Heat dissipation of the
latest Intel processors has become a widely discussed
issue. By the end of the decade, you might as well be
f
a
s
c
d
w
m
c
d
c
f
t
f
a
t
t
c
c
e
m
n
t
t
c
o
Diamondoid
nanotube
feeling a rock
after 2010, co
surface of th
continues to
devices, manu
ways to com
management.
carbon nanotu
devices might
HEAT is
computer an
fundamental w
thermal distre
a)
increase o
lifetimes
Keeping
faster spin
b)
requires m
c)
several re
d.Deg
much as
problem a
devices.
e.Incr
complex
approach
To reduc
finned devic
aluminum or c
to pull therma
transfer it into
Using mic
carbon nanotu
chips have pro
effectively as
material for
nanotubes are
than any other
Discovere
thick sheet of
consist of she
other to form

d

Super nanot
ket nozzle th
omputer chips
he sun itself
churn out s
ufacturers hav
mbat the pe
New research
ubes (CNTs)
t provide a sol
becoming on
nd semicondu
ways the elec
ss:
Shortened
of 10C in ope
in half. I
devices cool
nning fans, wh
Consumer
more cooling f
Reduced
ecalls and prod
graded perform
90% of powe
affecting batte
reased build
heat sinks an
hes are driving
ce high temp
ces made of
copper are a
al energy awa
o the surroundi
cro fin structur
ube arrays m
oven that nano
copper the
thermal mana
e more flexible
r cooling mate
d only in 200
f carbon. Exis
ets of graphen
a solid mater
tube

Nanot
han touching a
s could feel li
f. As the ele
smaller and
ve been challe
ersistent prob
h suggests that
as heat sink
ution to this p
e of the most
uctor design.
ctronics indus
d product li
erating temper
Increased op
increasingly r
hich use more
r acceptance
fans which cre
reliability
duct failures d
mance With
er is lost throu
ery life on all
costs Eve
nd fans, and
g up costs.
peratures, toda
f conductive
attached to the
ay from the m
ing air.
res made of al
mounted to th
otubes can dis
e best known
agement appl
e, resilient, an
erial available.
04, graphene
sting forms o
ne, either bond
rial like graph
tube pipe
a chip. And
ike the bubbly
ectronics indu
slimmer por
enged to find
blem of the
t the integratio
ks into electr
problem.
t critical issue
. There are
stry faces grow
ifetimes - E
rature cuts pro
perating costs
requires more
electricity.
e More
eate more nois
There have
due to heat.
h RF amplifie
ugh heat, a ser
l kinds of wir
r-larger and m
even more ex
ay's heat sink
metal such
e back of the c
microprocessor
ligned multiw
he back of sil
ssipate chip he
n, but most co
lications. And
nd 10 times lig
.
is a flat one-a
f carbon basi
ded on top of
hite, rolled up
soon
y hot
ustry
rtable
new
ermal
on of
ronic
es in
six
wing
Every
oduct
s
e and
heat
se.
been
ers as
rious
reless
more
xotic
ks
h as
chips
r and
alled
licon
eat as
ostly,
d the
ghter
atom
cally
each
p into
ca
na
ba
M
of
op
arbon nanotu
anotube as a
all shaped ful
Minister Fulleri
So Graphen
f replacing the
perations of th

VII. M
can be st
operations.
Speed of Co
improves te
indirectly a
base manag
of electron
capable of
nominally z
direct the
structure, a
copies. An
ten thousan
of one hund
be develope
built upon s
less time.
purpose of
field into a
live more c

VIII. STE
researc
boomin
great c
as nano
deep
Biotech
applica
overloa

[1] BC Cran
Global Abund
London, Engla
ubes (think
Graphene cyl
lerenes., just
ine
e can be used
e silicon chip
he Computer.
MERITS OF N
Through N
ored and ca
New concep
omputers can
The use o
echnically the
and also serve
gement.
The Use of
nic devices.
conducting e
zero carrier co
A nanocom
disassembly
and then dire
advanced asse
nd moving par
dred atoms.
Androids a
ed thereby dev
Excellent a
so that they ha
Thus Nano
future mecha
a larger biosco
omfortable lif
EPS FOR DEV
TECHN
The govern
ch and develop
South As
ng economies
chance of bein
otechnology.
Students sh
sectors such
hnology.
Artificial
ations must be
aded Servers.
REF
ndall, Nanotech
dance, The MIT
and. 1996.
of a single
linder), or fol
like the Carb
d for not only
p but also for
NANOTECHN
NVRAM large
an be used
pts regarding
be developed
of Molecular
areas of genet
es the purpose
f Graphene can
It also Gra
lectricity even
oncentration.
mputer system
y of an obj
ect the assem
embler could
rts, each conta
and Artificial
veloping Nano
and multiface
andle more co
o technology
anisms and the
ope through w
fe.
VOLOPEMEN
NOLOGY
nment should
pment of Nano
ian Countrie
s such as Indi
ng first to deve
hould show m
h as nanot
Intelligence
e used instead
FERENCES
hnology: Molecul
T Press: Cambri
e-walled carb
lded into soc
bon C6 or Bu
the sole purpo
r better and f
NOLOGY
amount of d
for large sc
the operation
.
nanotechnolo
tical engineeri
e of Large D
n reduce the h
aphene rema
n at the limit
m will be able
ject, record
mbly of perf
have as many
aining an avera
intelligence c
o robotics.
eted servers c
omplexity with
serves well, t
e develops eve
which a man c
NT OF NANO
lay emphasis
otechnology
es which a
ia China have
elop a field su
more interest
technology a
and Andro
d of relying up
lar Speculations
idge, Massachuse
bon
cer
uck
ose
fast
data
cale
nal
ogy
ing
Data
heat
ains
of
to
its
fect
y as
age
can
can
hin
the
ery
can
O
on
are
e a
uch
on
and
oid
pon
on
etts;
[2] Douglas Mulhall, Our Molecular Future: How Nanotechnology,
Robotics, Genetics, and Artificial Intelligence Will Transform Our
World. Prometheus Books, Amherst, New York, Hard Cover, 392
pages, 2002.
[3] Cristina Buzea, Ivan Pacheco, and Kevin Robbie (2007),
"Nanomaterials and Nanoparticles: Sources and Toxicity".
Biointerphases 2: MR17.
[4] N. Taniguchi (1974), On the Basic Concept of Nano-
Technology. Proc. Intl. Conf. Prod. London, Part II British Society
of Precision Engineering.
[5] Fritz Allhoff, Patrick Lin, Daniel Moore, What is nanotechnology and
why does it matter? from science to ethics, pp.3-5, John Wiley and Sons,
2010.
[6] Rodgers, P. (2006). "Nanoelectronics: Single file". Nature
Nanotechnology.
[7] Smith, Rebecca (August 19, 2009). "Nanoparticles used in paint could
kill, research suggests". London: Telegraph. Retrieved May 19, 2010.
ISBN: 978-81-920575-7-6:: doi: 10. 74390 /ISBN_0768
A Bounded Time Impending Hit Detection Method for Tele-manipulators

J. K. Mukherjee,
CnID, Bhabha Atomic Research Centre,
Mumbai, 400085, India

Abstract: An effective workspace modeling method that
ensures fast robot-part interaction sensing, for forewarning
operator at master end of man in loop type tele-manipulator
system, in time bound manner has been developed. The
technique minimizes run time computation and eliminates
dependence on boundless search based detection to ensure real
time performance in realistic tele-robot application.
Underlying modeling method and performance are presented.
I. INTRODUCTION
Tele-controlled Robotic arms working as slave agents
under human operator control, are versatile remote working
tool in hot cells which very frequently contain parts and
structures other than manipulated components in work
environment. Bilateral manipulators give force feedback on
contact only [1]. Under no contact state velocity based
damping is also offered [2]. But it does not depend on
workspace neighbourhood. Limited perception of
workspace makes erroneous operation highly probable. Hits
may occur. Making the system forewarn the operator
about robot approaching obstacle object is a promising way.
Modelling method for creating system reaction by on line
estimation of closeness to obstacle must work in real time.
Machine actuated warning and opposing action to operator
while approaching towards obstacles are effective in
implementing safe working. Techniques for avoiding
impending hit with obstacles are built around detection of
interaction between the object models and robot model. The
detection process rebuilds the state of the remote robotic
arm in this model space by monitoring the tele-robots joint
parameters. Though for very simple objects with few
surfaces the methods work, for situations with complex
objects it fails. Use of CAD based part models in forming
workspace model permits automation and so is a natural
choice. Though these are effective in graphic perception
development for operator[3], pose the problem of
indeterminate time demand in computing interference with
a moving robot. The present approach attempts at
alleviating this drawback.
II. PROBLEM STATEMENT
A robot arm moving in its workspace shown as R
(figure 1) may be approaching towards an object B. By
tracking its dynamics a motion vector V can be
computed. The range of sensitivity is defined by euclidian
distance from which the robot must be able to sense the
approaching obstacle. Therefore I V I of the vector is
limited to this range. Now if V intersects any surface of
object B , the robot has sensed the object B. It must be
possible to find the nearest intersection within this range.
Consider a workspace with object A, B and C that are
cylinders of circular and polygon cross section. Surface
based occupancy model is made from CAD models or
from range sensed data(3). It comprises triangular surface
segments.
The process requires each triangular surface to be
considered for computing its unit normal, computing a
vector parallel to the normal from a given R, computing its
intersection with surface, finding length of this vector and
comparing these lengths for all surface patches to find the
shortest of all. This entire process has to be repeated. For
simplified depiction 2D plan view are shown in fig 1c.
Segments 1-2, 2-3 , 3-4 etc. represent planar surface
patches of object B. Robot R has distances from these
surfaces indicated by L
1
to L
4
. Surface based occupancy
model of workspace is formed as indicated in figure (2). On
line robot state determination gives location R in
workspace. Interaction of robot now needs to be computed
with the workspace model. But computation of various L
needs search on all surfaces in the workspace model
containing thousands of surfaces for finding the nearest
surface. Reduction of computation is not easy. An
interesting case is depicted in figure.1-c. The nearest
surface

Figure 1 a. Workspace in 2D; b. search zone along motion vector ; c.
distance to obstacle.
patch 2-3 is formed by nodes 2,3 which are relatively at
large distance from R. The surface formed by nearer node
set 5-6 has distance L 0 where L 0 > L2 and L3. A node
distance based criteria must not be used to decide
candidature of a surface patch for computing nearest object
distance. Problem of finding minimum L escalates with
increase of part count in workspace and also with fine
surface definition of contoured bodies ( fig 3). It is evident
that obstacle distance computing task can not be completed
in time bound manner. as number of surfaces are
indeterminate. Heuristics based search space reduction are
unreliable. Unforeseen hits occur.

Figure 2. Modality of surface model approach

Figure 3. CAD models using bounding surfaces and planar surface
triangles on them for defining object
Fig.4a. A typical workspace depicted in picture. B
shows surface triangles and their normals formed from the
range sample data of workspace in a .
In order to solve this problem, planar barrier surfaces
are artificially placed as boundaries around object clusters
and are used in computation [4]. The drawback is that user
has to place these barriers. Improper barrier placement may
result in severely compromised workspace situations where
free space for robot movement is too constraining. The
present approach attempts at alleviating this drawback
III. MODIFIED WORKSPACE MODELLING
A. Voxel based modeling:
This approach treats the remote workspace as a
Cartesian co-ordinate 3D volume. The entire volume is
represented as three dimensional array of equal sized
cubes with integer coordinate locations (X,Y,Z) [5]. The
cube volume element is referred as 'Voxel' in further
descriptions. A voxel at location (X,Y,Z) is V
(XYZ)
and
represents cube with diagonal nodes X
i
,Y
j
,Z
k
and X
i+1
, Y
i+1

Z
i+1
as shown in figure 5a.A grid position in workspace
where no object exists is represented by free voxel. A voxel
through which a triangular surface passes is boundary voxel
(figure 5B). Voxels inside object are occupied. Using mesh
version (fig 4b) a voxel model of workspace is made (fig
5c). For CAD models, data representing the 3D shape of
object is associated to workspace geometry as per its pose
and position in work space. Triangular surface segments in
CAD models as well as mesh coded surfaces [4] from
range based volume occupancy models [5] are
computationally intersected by horizontal planes at heights
matching Z intervals of Cartesian coordinate system
starting from bottom to top of workspace.

Fig.5a voxel definition; b, surface voxel; c. workspace model

B. Benefit of the new approach :
Sensing interaction of R with obstacles along V needs
to be relooked with workspace model in voxel form. The
surfaces of generalized cylinder (fig 5) appear as voxels at
corresponding coordinate locations. For simplicity fig 6a
shows 2D view at some Z value containing R and the
pentagon section. Moving along towards object, voxels are
encountered along vector V . Figure 6b shows the case
with voxelised workspace in 3D. Test of voxel V
(x,y,z)
with
(x,y,z) along V indicates property of space as occupied or
free. Starting from R first encounter with occupied
property voxel gives shortest distance of hit along trajectory
V . Test is bound along V for a finite distance dictated by
hit prospecting range irrespective of surfaces and so avoids
unbound response time syndrome. Real time performance
synchronized with robot is feasible
Figure 6. Hit detection in voxel modelled workspace
IV. EXPERIMENTAL IMPLEMENTATION
An experimental system comprising voxel modeller
(figure 7) working on principal referred above, a spherical
robot arm typically used in tel-manipulation in large cells
(figure 8a) , an electronic system (figure 9) capable of
joint tracking and torque application to motors and
kinematics model (figure 8c) based position estimator for
robot arm tip was set up to test the approach. The shaded
segment keeps a model ready. At run time the robot joint
state is tracked and using kinematics model its spatial
location in workspace is determined. By computation
explained in earlier section a force signal is applied to
torque motors depending on closeness of impending hit to
dissuade the operator.

Figure 7. experimental system with modified modelling
A. Robot Arm : For experiments a prototype arm has
been developed. Figure 8-a. shows the same in laboratory.
The kinematics equivalent is shown in figure 8-b. The joint
sensing is achieved through high resolution encoder with
5000ppr on motor shaft. The position is tracked by high
speed counters over span of up to 2^31 count. The arm is
manually operated in experiments with 1200 mm reach
with sideward swing u +/40
o
and orthogonal tilt | +/-
30
o

Figure 8 a. Robot arm in laboratory; b Equivalent DH
representation for the arm ; c. joint tracking and torque
control module
B. Force Actuator: Actuators used in the arm are high
precision ac servomotors (with mechanical reducers),
capable of generating moderate torque (4.8 N-m). Torque
demand computed by the reaction estimation module is
used to develops opposing force by commanding the motor
with a torque demand signal (refer figure 8-c).
C. Robot state tracker: R
(x,y,z)
is computed using
current joint parameters u, | and L received from joint
sensors of robot .The module uses transforms indicated in
figure 9 .

Figure 9.
V. EXPERIMENTAL RESULT
For the experimental results presented here, voxel size
is 1cm X 1cm X 1cm, workspace extents are 150cm X 100
cm X 80cm, the sensitivity of hit is restricted to a
maximum distance of 20 cm and minimum of 3cm, search
interval along
Figure 10 a. Linear function for D based opposing
force b. transfer function programmed for the torque
transducer
V is 0.8cm. Hence maximum IVI =20 cm, minimum
IVI = 3 cm The toque demand is programmed as per linear
function in figure 10-b
The system is run in a modelled environment with
voxel models of obstacle object. (refer figure 11). The
system

Figure 11 right-voxel modelled workspace equivalent of the CAD
model shown on left. Mark the presence of voxelized obstacle block .
continuously tracks the robot arm joints and computes
position of virtual sensor at the lower tip of the arm by
using kinematics model and current robot joint status
u, | and oL.
Impending hit detection is done along vector V and is coded as
parameter D. The opposing force control is applied as function of D
shown in figure 10. The parameter D itself is estimated by interaction
assessment module that works on method described in figure 12 .

Figure 12. Method for prospecting hit
The experimental results have been recorded with axis
assignments such that the motion of suspended arm parallel
to wall and ground surface is considered X motion and one
normal to wall is Z motion. Toque profile of the motors
are programmed as per details in figure 10-b.The high
fidelity A/D conversion has 12 bit range. The high speed
data recording is done at 50 samples per second for of all
the robot axes, computed position of robot lower end tip
and torque demand computed from estimated parameter D
for the given work space scenario. Note that only current
robot joint status u, | and L are acquired from the robot
control system and torque resulting from the real time
interaction detection is sampled from the instantaneous
torque level signal provided by the control system after 12
bit A/D conversion Results for typical work space
scenarios (figure 11 and 14 ) appear in figure 13 and 15 .

Figure 13. Impending hit detection result for workspace
with block object on right as depicted in figure11.

Figure 14 . Voxelised workspace with the stick model
robot arm, block object on right and cylinder on left.

Figure15. Variation of torque demand for force
feedback creation with movement of robots lower tip.
VI. DISCUSSIONS ON PERFORMANCE
The results shown in figures 13 and 15 establish
fidelity of hit prospecting in accordance with the sensor
range and onset of interaction response occurs at 20 cm
from the obstacles modelled in voxel form. The motion is
executed manually as would be done naturally by an
operator of tele-manipulated robot. This is visible in results
as position gaps. The torque feed-back opposing moves
nearing obstacles are correct for both sides of operations.
The computation requirements being bounded to limited
numbers were successfully completed synchronously with
arm movement and ordered data were produced. The
cluttering at very close vicinity of objects arises from
digitisation errors in voxel formation and operator arm
oscillations as he faces high opposition force and arm
receives unintended motion. Since the onset of reaction
from obstacle is early, the operator need not reach close
vicinity of obstacle and cluttering does not hamper normal
working.

ACKNOWLEDGEMENTS
The author is thankful to Shri G.P. Srivastava, Director,
E&I Group for encouragement. Thanks to Shri Gaurav
Baluni for extending support in experiments.

REFERENCES
[1] H. Kazerooni et. all. A Controller Design Framework for Telerobotic
System IEEE Transactions On Control Systems Technology , Vol. I ,
No. 1, March 1993

[2] John E. Speich, Kevin Fite ,Michael Goldfarb, A Method for
Simultaneously Increasing Transparency and Stability Robustness in
Bilateral Telemanipulation Proceedings of IEEE International
Conference on Robotics & Automation, Sanfransisco 2000

[3] J. K. Mukherjee , Computer Aided Remote Graphics Environment
for Computer Assisted Robot Training and Control (CARE-CAT)
Proceedings of International Conference on Mechatronics Technology
2000.Singapore

[4] J. K. Mukherjee et al, An Observer Based Safety Implementation in
Tele-Robotics Proceedings of International Conference Trends In
Measurements and Automation TIMA-2011 Chennai, India

[5] J. K. Mukherjee , AI Based Tele-Operation Support Through Voxel
Based workspace Modeling and Automated 3D Robot Path
Determination Proceedings of IEEE International Conference
TENCON-2003 Banglore India

ISBN: 978-81-920575-7-6 :: doi: 10. 74194/ISBN_0768

GESTURE BASED COMMUNICATION A GESTURE HUMANOID

G S L K CHAND
DEPARTMENT OF ELECTRONICS AND COMMUNICATION ENGINEERING
ST.MARYS ENGINEERING COLLEGE,CHEBROLU,GUNTUR DISTRICT,
ANDHRA PRADESH,INDIA
Abstract-- Nowadays new interaction forms are not
limited by Graphical User Interfaces (GUIs) making
Human Computer Interaction (HCI) more natural. The
development of humanoid robots for natural
interaction is a challenging research topic. By using
gesture based humanoid we can operate any system
simply by gestures. The inter-human communication
is very complex and offers a variety of interaction
possibilities. Although speech is often seen as the
primary channel of information, psychologists claim
that 60% of the information is transferred non-
verbally. Besides body pose, mimics and others gestures
like pointing or hand waving are commonly used. In
this paper the gesture detection and control system of
the humanoid robot CHITTI is presented using a
predefined dialog situation. The whole information
flow from gesture detection till the reaction of the
robot is presented in detail.

KeywordsHand gesture, computer vision, HCI, marking
menu,UBIhand
I. INTRODUCTION
Ubiquitous, embedded computing, e.g., in domestic

environments requires new human-computer interaction
styles that are natural, convenient and efficient.
Keyboards and mice are still the most common and
most used interfaces between humans and computer
systems, no matter if it is just a desktop PC, a
notebook or a robot. However there is an increasing
interest in developing additional interfaces such as
speech recognition, handwriting, gesture recognition
and emotion detection, since these ways of
interaction may bring more naturalness into the
human-computer interface. In addition most people
tend to feel more comfortable if they can interact
with computers and robots in the same manner
humans are communicating with each other. The
interesting thing is that psychologists claim that
over 60% of interaction signals are transferred non-
verbal [5].This is one of the motivations for the
work presented in this paper. Our main goal is to
establish a communication interface through gesture
recognition between a human and the robot CHITTI,
that stands for Robot Human Inter- action Machine.
The robot must recognize some gestures made by a
human and respond by speaking or making some
body movements.
Traditionally there are some techniques for
gesture recognition based on the shape
of the hand or based on the movement of the human
arm and hand. Also there is no tight definition for a
gesture. Gestures can be viewed as a non-verbal
interaction and may range from simple static
signals defined by hand shapes, actions like
pointing to some objects, waving a hand or more
complex movements to express ideas or feelings
allowing the communication among people [8].
Thus for recognizing gestures it is necessary to
find a way by which computers can detect dynamic
or static configurations of the human hand, arm,
and even other parts of the human body. Some
methods used mechanical devices to estimate hand
positions and arm joint angles such as the glove-
based approaches. The main drawback of this
interface is that, besides being expensive, the user
must wear an uncomfortable glove, having a lot
of cables to connect the device to the system that
restricts the workspace to a small area and limits
the movements of the user. Therefore, one of the
best options to overcome the disadvantages of the
glove-based methods and to implement less
restricted systems is the usage of computer vision
to detect and track hands and arms.
Basically, the computer vision approaches
concentrate on recognizing static hand shapes
(pose gestures) or interpreting dynamic gestures
and motion of the hands (temporal gestures). For
the static gesture approaches the focus is to
identify a gesture by the appearance of the hand,
silhouettes, contours, 2D or 3D models, while the
methods that consider dynamic gestures are
concerned about motion analysis . There are also
works like shown below that evaluates both pose
gestures and temporal gestures.
Our approach, like many others, aims to support
the interaction to a robot and control some
movements. But besides that we intend to improve
the interaction and communication with a robot,
allowing the usage of gestures and dialogs. As an
example of some related work, in hand shapes are
used to control a walking robot through gestures
that indicate commands as stop, go forward, etc.
Their system has four modules: a hand detection
using skin segmentation, hand tracking, a hand-
shape recognizer based on contours and the robot
controller. In an integration of gaze and gestures
is used to instruct a robot in an assembling task.
Basically, pointing hands are detected through
skin color and splines for contour descriptions. In

[13] an attention model for humanoid robots is
defined using gestures and verbal cues. Human
motion and gestures are captured using markers
attached to the subject body.
In this work we focus on an appearance based
method for recognizing gestures that will help
improve the interaction with a robot allowing the
usage of gestures and dialogs. Our approach also
includes a hand tracking module, thus the user can
make different sequences of gestures while the
robot keeps looking at the users hand, without
having to process the entire image every time in
order to detect a hand and then identify a
gesture. Not only by hand but also nodding of head
etc. The preliminary results are encouraging and
for future work we expect to take the motion
history into account for recognizing more complex
and temporal gestures.
This paper is organized as follows. The first
one presents the marking menus applied for
detecting the hands of the user. The second one
describes
the pie and marking menus and the following section
explains recognition algorithms. The integratio
into the control system of the robot is described while
finally we presented the conclusions and future
work. This also includes the:
a.. Marking menus for gesture control

A fundamental concern in all kinds of gestural
control is what the command set should be. Exactly
what hand posture
and movements should be used? A possible strategy
is to base the command set on a menu system. The
language is determined by menu layout and
organization, and can be made culturally neutral and
self-explanatory. Gestures can be kept relatively
simple. The assumption here is that Pie- and marking
menus are especially well suited for the purpose,
because they offer a possibility forusers to develop
the skill to work with no feedback from the menus.
b.. Pie- and Marking Menus
They are pop-up menus with the alternatives
arranged radially, often used in pen-based interfaces.
Because the gestures (marks) are directional users
can learn to make selections without
looking at the menu items. With expert users, menus
need not even be popped up. Hierarchic marking
menus are a
development of pie menus that allow more complex
choices. The shape of the path, rather than the series
of distinct menu choices, can be recognized as a
selection. If the user, e.g., a novice, works slowly, or
hesitates, the
underlying menus can be popped up to provide
feedback.

c. The Prototype

Following we chose a scenario for the first prototype
that is well known to most: remote control of appliances
in
a domestic environment. A hierarchic menu system
for controlling the functions of a TV, a CD player, a
VCR, and a lamp is under development and some initial
user trials have been performed. The prototype has been
set up similar to a home environment in an open lab
/demo space at CID.
In order to maximize speed and accuracy, gesture
recognition is currently tuned to work only against a
uniform background within a limited area,
approximately 0.5 by 0.65 m in size, at a distance of
approximately 3 m
and under relatively fixed lighting conditions.
D. The Recognition Algorithms
The computer vision system for tracking and
recognizing the hand postures that control the menus
is based on a combination of multi-scale color feature
detection, view based hierarchical hand models and
particle filtering. The
hand postures or states are represented in terms of
hierarchies of multi-scale color image features at
different scales, with qualitative inter-relations in
terms of scale, position and orientation. In each
image, detection of multiscale color features is
performed. The hand postures are then
simultaneously detected and tracked using particle
filtering, with an extension of layered sampling
referred to
as hierarchical layered sampling. To improve the
performance of the system, a prior on skin color is
included
in the particle filtering. white ellipses show detected
multi-scale features in a complex scene and the
correctly recognized hand posture is superimposed in
gray.
c.. UBI HAND
It is a wireless control of home applicances via hand
gestures by using humanoid It works as given in
below flow chart
d. FLOWCHART WORKING
4

A. INTEGRATION
The experiments of the presented approaches have
been performed on the humanoid robot CHITTI .The
robot consists of a humanoid upperbody and head
with 24 degrees of freedom: 7 for the movement of
the body and neck, 6 for eye movement and 11 for
emotional expressions.

Fig. 2. Screenshot of the Camera

group with GestureDetector and
CamShift module. The image shows the
tool mcabrowser which can be used for
run-time analysis of the control
architecture.

Open
Hand
Close
d
H d
Positive
Open
d
0.917 0.00 0.
Closed
Hand
0.00 0.850 0.
00
Positiv 0.00 0.00 0.883
V 0.00 0.00 0.050
L 0.00 0.050 0.
00
Figure 3 shows an interaction between

the robot and a human during the
experiments.

IV. CONCLUSION
Future work concerning the gesture based
interaction will focus on the integration of
verbal and non-verbal interaction signals into
multimodal dialog situations.

ACKNOWLEDGEMENT
We thank mainly our parents who supported not
only by finanicially but also technically. My
sincere thanks to my sister who helped me a lot
in this project

REFERENCES

[1]. Bretzner, L., Laptev, I. & Lindeberg, T. (2002) Hand Gesture
Recognition using Multi-Scale Colour Features,
[2]. Hierarchical Models and Particle Filtering. Submitted to
5th Intl. Conf. on Automatic Face and Gesture
[3]. Recognition.
[4]. Callahan, J., Hopkins, D, Weiser, M. & Shneiderman,
[5]. B. (1988) An Empirical Comparision of Pie vs. Linear
Menus. Proceedings of CHI88, pp. 95-100.
[6]. Freeman, W.T. & Weissman, C.D. (1994) Television Control by
Hand Gestures. In 1st Intl. Conf. on
[7]. Automatic Face and Gesture Recognition.
[8]. Guimbretire, F. & Winograd, T. (2000) FlowMenu: combining
Command, Text and Data Entry. Proceedings of UIST2000,
pp. 213-216.
[9]. Kurtenbach, G. & Buxton, W. The Limits of Expert
Performance Using Hierarchic Marking Menus. Proceedings of
CHI94, pp. 482-487.
[10]. Jacob Fraden, Handbook of Modern Sensors: Physics, Designs,
[11]. and Applications, Springer publications 2010 edition
[12]. K. Mianowski, N. Schmitz, and K. Berns.
Mechatronics of the humanoid robotroman. In
Sixth International Workshop on Robot Motion
and Control (RoMoCo), Bukowy Dworek, Poland,
June 11-13 2007.
[13]. V. I. Pavlovic, R. Sharma, and T. S. Huang.
Visual interpretation of hand gestures for
human-computer interaction: A review. In IEEE
Transactions on Pattern Analysis and Machine
Intelligence, volume 19, pages 677695, 1997.
[14]. F. Quek. Gesture, speech and gaze cues for
discourse segmentation. In IEEEConference on
Computer Vision and Pattern Recognition, pages
247254, 2000.
[15]. D. J. Sturman and D. Zeltzer. A survey of glove-
based input. In IEEE Computer Graphics and
Applications, volume 14, pages 3039, 1994.
[16]. 11. S. Waldherr, S. Thrun, and R. Romero. A
gesture-based interface for human-robot
interaction. In Autonomous Robots, volume 9,
pages 151173, 2000.


Part II
ICCA 12
Volume 4
ISBN: 978-81-920575-7-6:: doi: 10. 74397 /ISBN_0768
Analysis of Data losses in Sensor Networks Used in Harsh Environment

Sunita B. K

Lecturer, Computer Science Department
Vidhya Vardhaka college of Engineering

Abstract Wireless sensor networks are expected
to be deployed in harsh environments characterized
by extremely poor and fluctuating channel conditions.
Transmission errors are inherent in wireless
communications because of the instability of wireless
channels resulting due to channel fading, time-
frequency coherence, inter-band interference, etc,
and thus receiving unreliable message packets Among
a wide range of sensor network applications, many of
them such as in the process industry, require reliable
data communications such that data packets can be
delivered to the destination without loss even under
harsh conditions. However, existing reliable
transmission techniques either are too costly for
resource-constrained sensor networks or have limited
capabilities for achieving reliable data
communication under such harsh conditions.
In this paper we present a survey on data loss in
WSN used in harsh environment. Describe the
sources of data loss and focus on available
communication channel and study on performance of
various strategies on Error Control Codes We also
provide a brief summary and comparison of existing
Communication Channel
Keywords - Wireless sensor network; Error Control
Code; BER; power consumption; Forward Error
Correction.
I. INTRODUCTION
Wireless communications has seen a grand
development over the last thirty years. There are large
numbers of various applications, which inducing more
and more complicated technical challenges. Some of
these challenges include higher data rates, ubiquity of
wireless devices, trend towards faster, smaller and
cheaper hardware and frequency congestion, sensors
have limited battery power, limited transmission rate, and
much reduced other capabilities.. Many application
scenarios are mission critical and correct delivery of data
in a timely fashion is paramount and therefore may take
precedence over energy efficiency, bandwidth utilisation,
and other considerations. To be precise we can define
performance as the following:
1. The ability to accurately and effectively sense the
desired data
2. The ability to reliably deliver this data to its
destination
3. The ability to deliver data within the necessary
time bounds differing kinds of errors and the amount of
delay typically suffered by their use. We also identify
several other issues of importance with regard to real
time systems
The ability to accurately and effectively sense the
desired data is dictated by the abilities of the sensors,
their mode of operation and their physical deployment
related wireless channel .The wireless radio channel is a
shared, naturally limited resource which has a
characteristic that varies with time, frequency and space.
Radio waves are affected by various phenomena which
can cause errors and affect the quality of
communications. In this Paper we have discussed
different physical and statistical models to characterize
the wireless channels encountered in harsh
communication environment. Also it examines the causes
of data loss in sensor networks and the strategies used to
attain reliable data transmission. There are several
techniques available to increase data reliability, e.g.,
error control codes and retransmissions, but to select the
best method it is essential to have some knowledge of the
characteristics of the wireless channel. In addition, many
communication systems are evaluated using computer
simulations of Monte-Carlo type. In these cases a
mathematical model of the wireless channel is needed.
The cost of implementing these techniques has been
measured primarily in terms of energy consumption. An
underlying problem, disregarded by previous research, is
that reliability mechanisms can create delays which cause
problems for delay intolerant applications. Such a
problem area has not previously been defined in WSN
research. We intend to investigate various reliability
strategies employed in wireless sensor networks and
discuss their applicability to differing kinds of errors and
the amount of delay typically suffered by their use. We
also identify several other issues of importance with
regard to real time systems
The rest of the paper is organised as follows. Section
II discusses the causes of data loss in sensor networks.
Section III discusses the different models for wireless
sensor network channels. Section IV discusses reliability
strategies to prevent data loss and the effects of
implementing the reliability strategies discussed on data
delivery delay. Section V discusses the Energy model for
wireless sensor network. Section VI concludes the paper.
II. CAUSES OF DATA LOSS
A strategy to ensure reliability in a sensor network
has a number of obstacles to overcome. These sources of
data loss include:
1. Losses in the wireless channel due to interference
and physical effects
2. MAC layer losses due to collisions
3. Losses due to node related errors or failures
4. Losses due to routing failure
The wireless channel is inherently lossy and
experiences both bit and burst errors caused by
interference from other wireless devices, and multipath
effects caused by the environment.
In addition the wireless signal may be absorbed or
blocked by certain objects. This problem is compounded
by the fact that the wireless channel is not constant over
time [3], [4]. It is often assumed that errors in the
wireless channel occur as a statistical distribution of bit
errors. However, in wireless networks it has been
observed that errors can be bursty in nature and can
effect significant areas of the network for a period of
time before disappearing [5]. During these periods it is
often impossible to send or receive from any of the
affected nodes.
Accessing the MAC layer may lead to errors in some
particular cases. It is often assumed, incorrectly, that
node connectivity is bi-directional and this is sometimes
not the case. In these particular instances MAC layer
errors can potentially occur. Contention based MAC
protocols, whether using collision avoidance or not, falls
foul of this condition via the hidden terminal problem.
Typically, collision avoidance schemes assume bi-
directional connectivity while CSMA schemes explicitly
ignore the problem. Results from [3], [4] suggest that bi-
directional connectivity assumptions are unrealistic.
Other MAC protocols which coordinate with other nodes
in order to automatically configure themselves may also
be susceptible to errors caused by asymmetrical
connectivity. This is particularly true where automatic
clustering and synchronization is involved.
A node may experience a failure or may simply fail
to forward a message for a number of reasons. Hardware
failure, battery failure, destruction of the node, etc., is not
uncommon among sensor nodes and this can result in
lost messages. Aside from this, constraints on memory
and buffer space can cause a packet to be dropped. For
example, certain radios have a single buffer space and if
an incoming packet is not handled before the next packet
arrives one of the packets must be dropped. In general,
nodes with a heavy routing load may be able to hold a
finite queue of messages in memory awaiting
transmission. Problems arise when traffic loads within
the network and on a particular node, are heavy and/or
the node experiences difficulty accessing the
transmission media. Thus dropped packets can occur. A
route may fail for any of the reasons described above.
Link errors, periods of congestion and node errors can
cause a designated route to become disconnected. Should
a node find itself unable to forward a message towards its
destination it will very likely drop the packet after a
period of time has expired. In this case an alternative
route is necessary and if a message is to be successfully
delivered, one must either already be formed or created
as needed.

III. MODELS OF WIRELESS SENSOR
CHANNELS

We need models for the wireless channels to evaluate
different communication protocols using computer
simulations. For this reason, we need to develop methods
to describe their different properties mathematically.
Usually, describing all phenomena which can cause
errors is too a complex problem. Since the received
signal strength determines the performance of a system,
it is vital to know the path loss and the fading statistics.
Since the received signal strength is a random variable,
we can never be sure that communication will be
successful. We describe the fading margin by cumulative
distribution function (cdf). By it, we can calculate the
probability that the received signal power is lower than a
certain value. It is common practice in communication
theory to model a channel with three different
components: multipath fading, path loss and noise.
Typically, this formula is used to model a channel:
y(t) = a(t) b(d) x(t) + n(t)
Here a(t) represents fading - a multiplicative random
variable, b(d) path loss, which can be deterministic and
n(t) is noise.
3.1. Fading and Multipath Models
On a short-distance scale we can notice fluctuations
comparable with one wavelength called small-scale
fading. Slow fading can be caused by shadowing, where
large obstruction obscures the main signal path. If the
observation was made over long distances, it will give us
more slowly varying view. Such large-scale fading is
explained by the slow loss of the received signal power.
Large-scale propagation models predict the mean signal
strength, while the small-scale propagation models
predict the rapid fluctuations. Depending on the Line Of
Sight (LOS) component and the number of Non Line Of
Sight (NLOS) components, different models are used.
Additive White Gaussian Noise (AWGN) represents
communication in free space, where only a strong LOS
component exists. In most realistic cases, there are also
one or more NLOS components. The Rician distribution
models a LOS component and several NLOS. The K-
factor of the Rician distribution describes the strength of
the LOS component as compared to the NLOS, e.g. when
K ! 1, we get AWGN and for K = 0, there is no LOS and
the fading has Rayleigh distribution. Some important
fading models are outlined below.
3.2. Noise Model
Additive white Gaussian noise (AWGN) is always
present in the real life because it comes from many
natural sources like earth and the Sun. It comprises John-
Nyquist, shot noise, black body radiation. It can be used
alone for satellite and deep space communication links,
or with other models, but then it only reconstructs the
background noise (in terrestrial environment) or the
electron movements in components, which can be
modelled as AWGN due to the central limit theorem.
3.3 Binary Erasure Channel
The binary erasure channel (BEC) can transmit one
of two symbols, 0 and 1. The bit can get erased in the
channel, but if the receiver receives the bit, it must be
correct. the probability of erasure is , while the
probabilities of getting the correct bit are 1-. An IP
network can be modelled as a BEC. Either a packet
arrives or then it is correct, i.e. it passed the error
detection check, or it is wrong and then dropped, i.e.
erasure. small height. Total path loss consists of the free
space path loss PL0, multi screen loss along the
propagation path Lmsd and attenuation from the last
roof-edge to the mobile station, called rooftop-to-street
diffraction and scatter loss, Lrts, PL = PL0 + Lrts +
Lmsd:
IV. RELIABILITY STRATEGIES
There are at present a number of different reliability
strategies that can be used in combination or ndividually.
These can be classified by the particular problem area
that they attempt to address. In this section we have
considered data transmission techniques related to link
layer data
A number of techniques exist which address the need
for reliability at the link layer. Link layer reliability
mechanisms can be effective in increasing the reliability
of a single link but fail to eliminate the effects of route
failure and only partially remove the effects of node error
and failure. In this regard additional mechanisms are
needed. Forward error correction (FEC) relies on
redundant data being appended onto the existing data. Bit
errors that occur can be detected and corrected [5], [6].
Various types of codes exist with various properties. In
general the stronger the code that an FEC uses the more
bit errors it can successfully detect and correct. In
general, the stronger the FEC the more redundant bits are
needed thus increasing transmission time. The use of
variable FEC strength depending on packet importance
was investigated in [6]. Recent research has shown that
the use of error correcting codes is of limited use in
sensor networks [5] since errors are typically bursty in
nature. In essence [5] describes that where burst errors
occur the number of redundant bits necessary for
correction is prohibitively expensive. Since, burst errors
are common in wireless sensor networks it is reasonable
to conclude that FECs are of limited value. If FECs are
used the additional redundant data takes additional time
to encode, transmit and additional time to decode. In
general, the stronger the FEC the more redundant bits are
needed thus increasing transmission time. However this
time is usually relatively small. The time taken to encode
and decode FEC codes depend on the abilities of the
sensor node and the complexity of the FEC code used.
Since processing power on sensor nodes is relatively low
this may take quite a long time. However, on the Dsys25
it is possible to add stackable hardware layers, one of
which is an FPGA which can be used to greatly speed the
encoding and decoding of FEC codes

Figure 1 FEC delay summed over number of hops
The time taken to encode, transmit, receive and
decode an FEC can be described as follows:
Where Encode FEC and DecodeFECis the time taken to
encode and decode respectively, and T
x/Rxdata+header+FEC
is the time taken to transmit the data, the header and
the additional bits for the FEC respectively to the
receiver. Consider Fig. 1. We evaluate the total
transmission time for a message over a variable number
of hops using Equation 1. The time taken to transmit and
receive a packet without addition FEC codes is 4
milliseconds (typical transmission time on Dsys25 node).
The additional time take to transmit and receive the FEC
is 4ms. MAC access is considered constant and takes 16
ms (typical MAC access time on Dsys25 node using
contention based MAC protocol under medium traffic).
The times taken to encode and decode are described in
Fig.1 and are 8ms and 32ms respectively. Note that in the
following figures (Fig. 1 to 4) the value for the MAC
access time is considered to be constant and the fore the
resultant graphs appear linear whereas in a real network
they will be more random and affected by changes in
traffic cause by longer packet size and retransmission
etc.. ARQ protocols may use multiple repeat
transmissions to ensure correct data delivery between
two nodes. Obviously this will have a significant effect
on the delay incurred. However note that ARQ protocols
are inherently resilient to burst errors since the repeat
messages are temporally displaced from each other.
Repeating a message can also help to ameliorate the
effects of both node errors and, in some cases, MAC
errors. The time taken to successfully send a packet when
using repeat link layer transmissions is as the following
probability:

where pdrop is the probability that a packet is lost on
the wireless link, n is the nth transmission sent, and
TimeOut is the time a node waits for an ACK to be
received. The same values for T x/Rx and [MACaccess]
as used in Fig. 1are used (4 and 16ms).The timeout takes
6ms. The delay for a varying number of retransmissions
is considered.


Fig 2 High quality or alternate route delay

Fig 3 Delay due to routing

Fig 4 Delay incurred by end to end methods using RTT

V. ENERGY MODEL FOR WIRELESS
SENSOR NETWORK

Energy consumed at node level can be computed as
follows:

Where
E
total
: Total energy consumed in the network
E
enc
: Energy consumed by the encoder at the first
node
E
TX
: Energy consumed in transmission by all nodes
E
RX
: Energy consumed in receiving the data by all
nodes
except first one

Where
m: Number of hops
Nb: Total number of bits transmitted
Etx: Energy consumed in transmitting a single bit
from a
node
Erx: Energy consumed in receiving a single bit at a
node
The term Etx can be represented as [9]:

Where, Ete is the power consumption at transmitter
electronics is the path loss component usually varies
between 24 with = 3 being a typical value when
scattering is considered [8].
Eta is the power consumption of transmitter amplifier
and can be given as:

Where is the desired SNR at the ith
receiver, NF
RX
is noise figure at receiver, N
0
is the
thermal noise, BW is the bandwidth of channel noise,
is wavelength, Gant is antenna gain,
amp
is the
transmitter efficiency and Rbit is the raw channel rate in
bps. If the channels have low noise, this scheme may
completely recover the original data with the additional
advantage of low energy consumption. Research has to
be done on this area by implementing these models in
harsh environment.
VI. CONCLUSION
Reliable data transmission in harsh environment
depends on the wireless channels and data transmitted
from the sensor nodes are vulnerable to corruption by
errors induced by these noisy channels Hence it is
necessary to analyse and model wireless channel to
provide a proper error control scheme to reduce the bit
error rate (BER). Due to the stringent energy constraint
in sensor networks, it is vital to use energy efficient error
control scheme.
In this paper, the wireless channel is analysed and the
characteristics of wireless channels change over time,
space, and frequency in ways that are difficult to capture
entirely with theoretical models It has been clearly
shown that the needs of sensor network applications with
real-time constraints and demands are not considered by
current methodologies. Previous research has considered
energy as the primary cost and de facto unit of currency
in sensor networks research. We believe that if
applications with real-time constraints are to be
developed a radical shift of focus is needed that
encompasses the demands of these applications, namely
reliability of data delivery and the timeliness of data
delivery.
REFERENCES

[1] C. Intanagonwiwat, R. Govindan, D. Estrin. Directed Diffusion:
A Scalable and Robust Communication Paradigm for Sensor Networks.
Proc. of the 6th Annual Conference on Mobile Computing and
Networks, 2000
[2] F. Stann and J. Heidemann. RMST: Reliable Data Transport
in Sensor Networks. Proc. of the IEEE Internationatl Workshop on
Sensor Net Protocols and Applications, 2003
[3] A. Woo, T. Tong, D. Culler. Taming the underlying
challenges of reliable multihop routing in sensor networks. Proc.
of ACM Sensys03, Los Angeles, CA, USA, 2003
[4] D. Ganesan,B. Krishnamachari, A. Woo, D. Culler, D.
Estrin, S. Wicker. Complex behavior at scale: An experimental
study of low power wireless sensor networks. Tech. Report
UCLA/CSD-TR 02-0013, Computer Science Dept., UCLA,
2002
[5] A. Willig and R. Mitschke. Results of Bit Error
Measurements with Sensor Nodes and Casuistic Consequences
for Design of Energy- Efficient Error Control Schemes. Proc. of
the IEEE European Workshop on Wireless Sensor Networks,
2006
[6] E. O. Elliott. Estimates of error rates for codes on burst-
noise channels. In Bell System Technical Journal, volume 42,
page 19771997, 1963
[7] E. O. Elliott. Estimates of error rates for codes on burst-
noise channels. In Bell System Technical Journal, volume 42,
page 19771997, 1963
[8] P. Chen, B. ODea and E. Callaway, Energy Efficient
System Design with Optimum Transmission Range for Wireless
Ad Hoc Networks, IEEE International Conference on Comm.
(ICC 2002), vol. 2, pp. 945-952, 28 April - 2 May 2002
[9] Yang Liu. Performance improvement of wireless
communications using frequency hopping spread spectrum. In
International Journal of Communications

ISBN: 978-81-920575-7-6 :: doi: 10.73879/ISBN_0768
ACM #: dber.imera.10.73879
A 3D Stacked Mesh NoC For Reliable Inter-Layer Communication And Congestion
Reduction
K.A. Karthigeyan

PG Scholar
Department of Electronics and Communication Engg
Veltech Multitech Dr.Rangarajan Dr,Sakunthala Engineering College
Chennai,India

Abstract- The increasing viability of 3D silicon integration
technology has opened new opportunities for chip
architecture innovations. One direction is in the extension of
2D mesh based chip multiprocessor architecture into three
Dimensions. We present an efficient architecture to optimize
system performance, power consumption, and reliability of
stacked mesh 3D NoC is proposed. Stacked mesh is a feasible
architecture which takes advantage of the short inter-layer
wiring delays, while suffering from inefficient intermediate
buffers. To cope with this, an inter-layer communication
mechanism is developed to enhance the buffer utilization,
load balancing, and system fault-tolerance. The mechanism
benefits from a congestion-aware and bus failure tolerant
routing algorithm for vertical communication.

KeyWords: Networkon-Chip(NoC), 3D Integration, Fault
Tolerance, congestion reduction.

I INTRODUCTION

Ever increasing requirements on electronic
systems are one of the key factors for evolution of the
integrated circuit technology. Continuous technology
scaling has made it possible to integrate billions of
transistors on a single chip[1]. Thus the entire system with
hundreds of components can be integrated on a single chip,
which is known as a Multiprocessor System-on-Chip [2].
At such integration levels, communication plays a major
role in the design and
performance.
One outcome of higher integration levels is that
interconnection platforms are replacing the shared
buses.Networks-on-Chip (NoCs) are proposed to be used
in complex SoCs for inter-core communication because of
scalability, better throughput and reduced power
consumption [3].
On the other hand, increasing the number of cores
over a 2D plane is not efficient due to long interconnects.
With the emergence of viable 3D integration technologies
opportunities exist for chip architecture innovations to
enhance system power/performance characteristics [4]. In 3D
integration technologies, multiple layers of active devices are
stacked above each other and vertically interconnected using
through-silicon vias (TSVs) [5]. As compared to 2D designs,
3D ICs allow for performance enhancements even in the
absence of scaling because of the reduced interconnect lengths
[6]. In addition to this clear benefit, package density is
increased significantly, power is reduced due to shorter wires,
and system is more immune tonoise [7].
One of the well-known 2D NoC architectures is
the 2D Mesh. This architecture consists of an mn mesh of
switches interconnecting IP blocks placed along with them.
The straightforward extension of this popular planar
structure is 3D Symmetric NoC by simply adding two
additional physical ports to each router; one for Up and one
for Down [8]. Despite simplicity, this architecture has two
major inherent drawbacks. Firstly, it does not exploit the
beneficial attribute of a negligible inter-wafer distance in
3D chips, because in this architecture, inter-layer and intra-
layer hops are indistinguishable. Secondly, a considerably
larger crossbar is required as a result of two extra ports [9].
The Stacked (Hybrid NoC-Bus) mesh architecture
presented in [10] is a hybrid architecture between the
packet switched network and the bus architecture to
overcome the mentioned 3D Symmetric NoC challenges. It
takes advantage of the short inter-layer distances, around
20m, that are characteristics of 3D ICs [4]. It integrates
the multiple layers of 2D mesh networks by connecting
them with a bus spanning the entire vertical distance of the
chip.As the inter-layer distance for 3D ICs is small, the bus
length will also be smaller; approximately (n-1) * 20m,
where n is the number of layers.
This makes the bus suitable for interlayer
communication in vertical direction. By using the stacked
mesh architecture, six-port router is required instead of
seven ports for typical 3D NoC router and vertical
communication is just one hop away to any destination
layer.
In this paper, we address the routing issues and
buffer utilization to enhance the overall system power and
performance of existing stacked mesh . In addition, the
proposed architecture increases tolerance against single bus
failure architectures.

Thre
evolved t
scaling by
descriptio
the 3D I
shown th
efficient,
considera
3D NoCs
For each
ports are
55 cross
crossbar
of ports,
much hig
The s
router ha
proposed
Its basic p
Time-Div
used as
vertical c
architectu
consumpt
architectu
two buffe
input buf
later, th
congestio
W
enhance
communi
architectu
traverse p
makes it
hops per
degrades
architectu
vertical
crossbar
performan
deals wit
decision
congestio
addition,
tolerate a
U
resources
tolerance
mesh arch
is high-th
other han
mesh arch
of reduce
mesh
II REL
ee(3D) dime
to deal with t
y stacking act
on of the cha
ICs is provid
hat 3D ICs
but when t
ation, the stat
s are extension
NoC router
needed result
sbar for the 2D
power increas
the power co
gher than a 2D
solution to the
as been prop
architecture
purpose was c
vision Multipl
a communica
communicatio
ure is efficien
tion and lat
ure is that eac
ers: the source
ffer as shown
he output b
on-aware inter-
We improve
the throughp
ication resou
ure due to tw
packets throu
less power e
packet for v
system t
ure deals wi
communicatio
improves
nce. Adaptive
th load balan
making ro
on in 2D NoC
the proposed
single link fa
Utilizing the
s, we enhance
for inter-lay
hitecture. Sym
hroughput but
nd, architectur
hitectures are
ed throughpu
architecture
LATED WOR
ensional inte
the limitations
tive silicon lay
allenges faced
ded in [4]. T
are power a
the 3D NoC
tistics are qui
n to the 2D N
of mesh topo
ting a 77 cr
D mesh archit
ses quadratica
onsumption fo
router [11].
e power consu
osed by Li
is stacked m
cache coheren
le Access (dT
ation pillar. D
on and 66 r
nt enough in
tency. The
ch packet is t
e output buffe
in Fig. 1. As
buffer hinder
-layer commu
the architec
put by usin
urces. We c
wo major asp
ugh two stage
fficient. Seco
vertical comm
throughput.
ith both issu
on without
the system
e vertical rou
ncing. In [12
outing algori
architectures
dynamic rout
ilure.
e available
our approach
yer communic
mmetrical 3D
t not power
res like True
power efficie
ut. We reinfo
without
RK
egrated circ
s of interconn
yers. The deta
d to manufact
The authors h
and performa
is taken un
ite different. T
NoC architectu
ology, two ex
ossbar instead
ecture. Since,
ally with num
or a 3D route
umption for a
et al. [10]. T
mesh architectu
nce. The dyna
TDMA) bus w
Due to one
routers, propo
terms of po
issue with
traversed thro
er and destinat
s we will disc
rs implement
unications.
cture to furt
ng the availa
complement
pects. First is
crossbar, wh
ond is to use
munication, wh
Our propo
ues. Single
using two-st
m power
uting mechan
2], an enhan
ithm to av
was proposed
ting approach
communicat
to deal with f
cation of stac
NoC architect
efficient. On
NoC or stac
ent at the expe
orce the stac
adding ex
cuits
nect
iled
ture
have
ance
nder
The
ure.
xtra
d of
the
mber
er is
3D
The
ure.
amic
was
hop
osed
wer
this
ugh
tion
cuss
ting
ther
able
the
s to
hich
two
hich
osed
hop
tage
and
nism
nced
void
d. In
can
tion
fault
cked
ture
the
cked
ense
cked
xtra
c
th

n
d
m
th
la
r
c
in

R
p
T
p
u
d

2
3

communication
hroughput, fau
In this arch
nodes are dif
dTDMA pillar
must be provi
he vertical lin
ayers. An ex
outer for the
channel has
ndistinguishab
In the static
RXYZ is the
packet to the d
This particula
performance
utilization. Th
different comm
1. Bus is busy
2. Bus is free b
3. Bus is busy
is full:
n resources
ult-tolerance, a
III MOTIV
hitecture, rou
fferent, as an
r (vertical lin
ided to enable
nks with the
xtra physical
e vertical com
its own ded
ble from the ot
c XYZ routing
source, whic
destination RX
ar architecture
efficient bec
his can be j
munication sce
y but input buf
but input buff
y and also the i
and enhan
and power eff
VATION
uters connecte
n interface b
nk) and the N
e seamless in
2D network
channel is ad
mmunication.
dicated buffe
ther channels.
g algorithm, c
h needs to s
XYZ+1 as show
e is neither
ause of inef
justified by
enarios as follo
ffer of RXYZ+
fer of RXYZ+
input buffer of
nce system
ficiency.

ed to pillar
between the
NoC router
ntegration of
k within the
dded to the
The extra
ers, and is

consider that
send a data
wn in Fig. 2.
power nor
fficient bus
considering
ows.
+1 is free:
1 is full:
f RXYZ+1

I
It is
vertical b
We show
obstruct r
section, w
Initially,
reduce
implemen
an adapt
AdaptiveZ
F
for intra-
algorithm
routing a
routing p
the XY
identically
layer co
algorithm
As c
communi
way for th

IV PROPOSE
s not efficient
bus without co
wed that the
routing adaptiv
we approach th
the buffer o
power con
nting an adap
tive inter-lay
Z is applied.
For the sake o
-layer commu
m is used. Not
algorithm is
olicy. Therefo
routing is p
y as shown in
ommunication
m is elaborated
can be seen
ication, the fir
he vertical com
ED ARCHITE

t to connect a
onsidering the
extra output
vity and load b
he problem in
on output por
nsumption a
ptive routing
yer routing a
of simplicity, i
unication a st
te that the pro
not dependen
ore, in ZXY ro
projected in
n Fig. 3 with th
n. The Ada
d in Algorithm
from Fig. 3
rst bus pillar
mmunication i
ECTURE
a NoC router t
eir characterist
buffer even
balancing. In
n different pha
rt is removed
and to ena
algorithm. Th
algorithm ca
it is assumed
tatic XY rout
oposed inter-la
nt to intra-la
outing algorith
different lay
he adaptive in
aptiveZ rout
m 1.
3, for inter-la
available on
is used.
to a
tics.
can
this
ases.
d to
able
hen,
lled
that
ting
ayer
ayer
hm,
yers
nter-
ting
ayer
the

a
th
w
n
a
n
h
m
w
th
b
b
c
a
c
is
V
2
s
c
tr
o
p
d
th
s
V D
To deal wi
architecture is
he input buff
which needs t
node R101. Ro
availability of
node R100. Als
have the same
mesh architect
without virtual
he proposed a
buffers as com
bus architectu
communication
architecture.
The o
communication
s used to rec
VC concept. T
21 multiplex
ignaling wir
consumption d
raversal path
optimization d
power consum
In ad
deadlock, but
he buffer is
ystem power a
DEADLOCK

ith deadlock,
used as show
ffer of router
to be transm
outer R101 is
dTDMA bus t
so, consider th
e situations. In
ture with adap
l channel, the
architecture, w
mpared to stac
ure [10], we
n scheme
output buffer
n was remove
eive the data
There is very sm
xer, one 12
res. Reductio
due to remova
h of packet
due to routing
mption of this s
dition, the ad
it also improv
used in a ri
and performan
AVOIDANCE
typical virtu
wn in Fig. 4. C
R001 contain
mitted to the
already wait
to deliver a pa
hat the node R
n case of typ
ptive vertical
ere will be a d
with the same
cked mesh wi
modify the
to support
from the rou
ed. An extra i
packets and
mall area over
de-multiplex
on in dynam
al of the buff
and also st
adaptivity do
small extra log
dded VC not
ves the throug
ght way and
nce efficient.

E
ual channel
Consider that
ns a packet,
neighboring
ting for the
acket at the
R000 and R100
ical stacked
routing and
deadlock. In
e number of
ith dTDMA
inter-layer
the VC
uter for bus
input buffer
support the
rhead of one
xer and few
mic power
fer from the
tatic power
ominates the
gic.
only avoids
ghput. Now,
makes the

communi
is a critic
physical c
TSVs can
stacked k
number o
TSVs inc
reliable
considera
explore h
routing al
when the
destinatio
For fau
adaptive
adding an
wait sig
condition
wait sig
arbiter do
signal. T
introducin
architectu
It is ass
enough re
bus link
services d
When the
which ind
bus is n
temporari
wait sig
wait sig
layer and
inter-laye
VI FAUL
Through Si
icationlinks fo
cal design issu
components, t
n fail. A failed
known-goodd
of dies to be
crease the cost
inter-layer c
ably mitigate th
how the avai
lgorithm to av
ere are other p
on pair.
ult tolerance, t
inter-layer co
ny extra wire
gnal acts as
ns. In case of
gnal is perm
oes not serve
Thus, routing
ng the fau
ure.
sumed that the
esources (test
is broken or
due to therma
e bus arbiter
dicate that ther
not able to
ily or perma
gnal accordin
gnal, it will r
d in future, w
er communic
LT TOLERAN

ilicon Via,
or dies in the v
ue in 3D integr
the fabrication
d TSV may ca
dies to be di
e stacked incr
t and decrease
communicatio
hese issues. In
ilable signals
void these pat
paths between
the existing si
ommunication
e. As explaine
a congestion
any fault on t
manently asser
e any request
adaptivity is
ult toleranc
e dTDMA bus
units) to sign
the bus is no
al problem or
receives any
re is a fault on
provide the
anently, it ass
ngly. When ro
reroute the pa
will not deriv
cation to the
NCE
which provi
vertical directi
ration. Like ot
n and bonding
ause a numbe
iscarded. As
reases, the fa
the yield [13]
on scheme
n this section,
can enable
ths (faulty bus
n the source
ignaling used f
n is used with
ed in section
n flag in nor
the dTDMA b
rted 0 and
t raised by r
not affected
ce in exist
is equipped w
nal that either
t able to prov
r any other fa
of such sign
n the bus link
communicat
serts 0 on
outer checks
ackets within
ve any traffic
e correspond

ides
ion,
ther
g of
r of
the
iled
]. A
can
we
the
ses)
and
for
hout
IV,
rmal
bus,
bus
req
by
ting
with
the
vide
ault.
nals,
and
tion
the
the
the
for
ding
d
w
m
s
m

a
p
p
d
c
e
r
n

e
in
d
b
w
p
s
n
p
d
a
r
m
dTDMA bus.
will be used ev
modification
ection IV. Bu
mechanism nee
Consider tha
a packet to n
paths are fault
proposed routi
dTDMA bus c
002 should b
current situatio
either of nod
outing algorit
normally accor
Now consid
exactly below
nter-layer com
destination but
be rerouted to
where the pa
packet will be
ituation, when
node 122. F
packet can only
dTDMA bus c
122, which i
algorithm requ
outed to the n
minimal path.
Normally, th
ven in case of
in routing
ut there are f
eds to be mod
at the node 00
ode 112. Th
ty as sown in
ng algorithm
connecting the
be tried but t
on. So, the pa
des 010 or
thm. Then th
rding to the pr
der another si
w/above its d
mmunication,
t the bus link
the neighbori
acket is curre
delivered to th
n the node 1
For proposed
y be routed th
connecting the
is faulty. In t
uires modifica
nodes 110 or
he minimal p
f faulty links
algorithm di
few cases, w
dified.
00 in Fig. 6 ne
he vertical lin
Fig. 6. Acco
in section IV,
e nodes 000
that bus link
cket will be r
100 accord
he packet wil
roposed algori
ituation, if th
destination no
it can be deliv
is broken, the
ing node withi
ently residing
he destination
20 contains a
d routing alg
hrough the vert
e nodes 120
this situation,
ation. The pac
r 020 followi
path routing
without any
iscussed in
when routing
eeds to send
nks on two
rding to the
, the vertical
, 001 and
is faulty in
outed to the
ding to the
ll be routed
ithm.
he packet is
ode and by
vered to the
e packet will
in the layer,
g. Then the
n. This is the
a packet for
gorithm, the
tical
, 121 and
the routing
cket will be
ing the non-

VII CONCLUSION

In this paper, an enhanced architecture for 3D
stacked mesh NoC was proposed to enhance system
performance, reduce power consumption, and
improve the system reliability. To this end, a
congestion aware adaptive interlayer communication
algorithm was introduced. To deal with deadlock, an
appropriate VC architecture was used with same
number of buffers as compared to the existing
stacked mesh architectures. In addition, the
congestion signal triggered by the bus arbiter was
used to deal with fault tolerance. This enabled
avoiding a faulty vertical bus in a possible path for a
number of source-destination pairs in the event of a
bus failure.

REFERENCES
[1] S. Borkar, Designing reliable systems from unreliable
components: The challenges of transistor variability and
degradation, IEEE Micro, Vol. 25, No. 6, 2005, pp. 10-16.
[2] A. Jerraya and Wayne Wolf, Multiprocessor Systems-on-
Chips, Morgan Kaufmann, 1st edition, October 12, 2004.
[3] A. Jantsch, and H. Tenhunen, Network on Chip, Kluwer
Academic Publishers, 2003.
[4] A. W. Topol et al., Three-Dimensional Integrated Circuits,
IBM J. Research and Development, Vol. 50, No. 4/5, July-Sept.
2006, pp. 491-506.
[5] V. F. Pavlidis and E. G. Friedman, 3-D Topologies for
Networks-on- Chip, in Proc. of IEEE Trans. Of VLSI Systems.,
Vol. 15, No. 10, 2007, pp.1081-1090.
[6] B. S. Feero and P. P. Pande, Networks-on-Chip in a Three-
Dimensional Environment: A Performance Evaluation, IEEE
Transactions on Computers, Vol. 58, No. 1, 2009, pp. 32-45.
[7] R. S. Patti, Three-Dimensional Integrated Circuits and the
Future of System-on-Chip Designs, in Proc of IEEE, Vol. 94, No.
6, 2006, pp. 1214-1224.
[8] L. P. Carloni et al., Networks-on-Chip in Emerging
Interconnect Paradigms: Advantages and Challenges, in Proc. of
International Symposium on Networks-on-Chip, 2009, pp. 93-109.
[9] J. Kim et al., A novel dimensionaly-decomposed router for
on-chip communication in 3D architectures, in Proc. of
International Symposium on Computer Architectures, 2007. pp.
138-149.
[10] F. Li et al., Design and Management of 3D Chip
Multiprocessors Using Network-in-Memory, in Proc. of
International Symposium on Computer Architecture, 2006, pp.
130-141.
[11] R. S. Ramanujam and B. Lin, A Layer-Multiplexed 3D On-
Chip Network Architecture, IEEE Embedded Systems Letters,
Vol. 1, No. 2, 2009, pp. 50-55.
[12] P. Lotfi-Kamran et al., EDXY A low cost congestion-
aware routing algorithm for network-on-chips, Journal of Systems
Architecture, Vol. 56, No. 7, 2010, pp. 256-264.
[13] A.-C. Hsieh et al., TSV Redundancy: Architecture and
Design Issues in 3D IC, in Proc. of International Conference on
Design, Automation, and Test in Eroupe, 2010, pp. 166-171.

c
a
S
t
e
a
a
a
f
c
i
G
c
p
f
a
a
b
m
s
L
l
M
e
t
m
Proc. of th
Volume 1. Co
ISBN: 978-8
ra

Abstract -
communicative
and that is
SMARTPHON
to end mobile c
every manufa
and innovation
As a resu
attractive featu
adaptability to
factors contrib
comparative s
iPhone, Symb
Google Androi
Keywo
Os,Android

The aim
comparative s
point of view
features, by b
and limitation
aspects of thes
The major
based softwar
market by
smartphones.A
Linux-based p
launch like it
Mobile, Symb
etc.
The battle
technological
mobile OS.
he Intl. Conf.
opyright 20
81-920575-7-
ACM #: dbe
M
Rashmi Pr
MCA Dep
SRMG
Lucknow
ashmi1330@
- Mobile pho
e devices to so
s why they
NES. Smartpho
communication
cturer is focu
ns in this field.
ult there are
ures. The succ
o third party a
bute in its suc
study of some
ian, Blackberr
id.
rd- Apple
d and Window
I. IN
of this report
study of differe
w. A compar
bringing the d
ns together, th
se mobile OS
r industry is o
re developing
about 50%
Android, the
platform whic
ts competitor
bian OS, Palm
e for market
developments
on Computer
012 Techno Fo
6 :: doi: 10. 73
er.imera.10. 73
Mobile Op
rajapati
artment
GPC
w, UP
@gmail.com
ones have em
oftware intensi
y are altern
ones are provid
n across the wo
using on mobi
many mobile
cess of any plat
applications. O
ccess. This pap
e of the key
ry OS, Windo
e iPhone,S
ws Mobile
NTRODUCTIO
t is to provid
ent Mobile OS
rison between
different Mob
his paper wil
also.
occupied by S
g company wh
% especially
flagship mob
ch is getting
rs like the iP
m OS, and R
t share is m
s and innovat
Applications
orum Group, I
3886/ISBN_0
3886
perating S
merged from
ive devices lik
natively calle
ding solution fo
orld. In presen
le operating s
e OS having
tform depends
Of course many
per is intended
mobile OS:
ows Mobile O
Symbian,Black
ON
de a review
S platform fro
n all core pla
bile OS advan
l provide tech
Symbian, a Lo
hich is leadin
in PDAs
bile platform,
ready for a w
honeOS, Win
RIM Blackberr
making a pat
tions in the fi

ndia.
768
System: A

mere
ke PCs
ed as
or end
nt time
system
many
on its
y more
d to a
Apple
S and
kberry
and a
om our
atform
ntages
hnical
ondon
ng the
and
, is a
wider-
ndows
ry OS
th for
eld of
II.
play
And
In p
plat
tow
the
And
Sea
des
Cal
aug
plat
ker
upd
fixi
are-
A Compar

. THE BA
.

To occupy m
yers like Sym
droid are unve
present scenar
tform. May b
wards a truly o
sophistication
Android, Inco
dy Rubin (co
ars (once VP
sign and interf
lifornia, Unite
gust 2005.Afte
tform for mo
rnel.
Since its orig
dates, each add
ing bugs of pr
-
rative An
Ruch
MCA
SR
Luc
ruchirashuk
ATTLE FOR M
market and to m
mbian, RIM B
eiling their ow
rio customer p
be a needs gr
open source p
n on their own
A. A
orporation was
o-founder of
at T-Mobile
face developm
ed States. It
er that a team
obile devices
ginal release, A
ding some new
revious releas
nalysis
hira Shukla
A Department
RMGPC
cknow, UP
kla89@gmai
MARKET SH
monopolize th
Blackberry, Ap
wn application
prefers a simpl
row end user
platform and t
n.
Android OS
s founded in O
Danger), Ric
e), and Chris
ment at WebTV
was acquired
m led by Rub
having the p
Android has s
w and innovat
e. Recent rele
t
l.com
HARE
he industry ma
pple-iPhone, a
n stores over n
le but innovat
r will be turn
they will exte
October, 2003
ch Miner, Ni
White (head
V), in Palo Al
d by Google
bin developed
power of Lin
een a number
tive features a
eases of Andro
ajor
and
net.
ive
ned
end
by
ick
ded
lto,
in
d a
nux
r of
and
oid
clair having a new web browser and support for
HTML5 and W3C geolocishaving some smart features like
digital Zoom, color effects and more.
clair also includes support for animated backgrounds,
voice controls, buttons to open the menu, new weather
application and improved Email and Phonebook
applications.
Froyo introduced speed improvements with JIT
optimization and the Chrome V8 JavaScript engine, and
added Wi-Fi hotspot tethering and Adobe Flash support.
Honeycomb was announced at the 2011 Google I/O on
10 May 2011. - To allow honeycomb devices to directly
transfer content from USB devices.
Honeycomb is "an incremental release that adds several
new capabilities for users and developers". Highlights
include optimization for a broader range of screen sizes;
new "zoom-to-fill" screen compatibility mode; capability to
load media files directly from the SD card; and an extended
screen support API, providing developers with more precise
control over the UI.
b.Symbian OS
Symbian,on June 24, 2008 was acquired by
Nokia.Previously it was owned by Ericsson, sonyericsson,
Panasonic and Samsung. Symbian
OS offers a high level of integration of personal
information management with communication (PIM).It
provides many services like navigation, games,
associatelibraries; music playbacks etc.It has some hardware
requirements-

X86 architecture Monitor: 1024 x 768pix
el screen, Hard drive: 200 MB of free disk space, 16
bit color Memory: 512MB, CPU: 1.2GHz processor.
Some of the releases of Symbian are-

1) Epoc Release 5: In 2000, it was released with
Bluetooth successor code named conon.It also
supported Unicode.
2) Symbian OS 6.0 and 6.1 ( ER6 ):It was
released in 2002.In this release Bluetooth was added
for exchanging data over short distances creating
Personal Area Network(PAN).It is used
in Nokia 9210.
3) Symbian OS 7.0 and 7.0s:
This version added EDGE support and IPV6.
4) Symbian OS 8.0:
In this version of Symbian OS, there are no majo
r evolutions, has shared some APIs to support 3G.
5) Symbian OS 9.1:
The change for this version is the change of versi
on 1.2 for the Bluetooth has version 2.0 where the di
fference is the introduction of an Enhanced Data Rat
e (EDR) for faster data transfer. The nominal Rate of
EDR is about 3 megabits per second although the pr
actical data transfer rate is 2.1 megabits per second.
al television broadcasts in DVBH and
ISDBT formats and also location services.
B. Windows Phone
Microsoft captured a great attention at Mobile World
Congress by introducing Windows Phone 7 Series, such as
WP7S, or WinPho and represents a completely new
approach to interacting with your phone. Instead of icons for
Apps on the WP7 screen uses hubs, which are centers for
any typeof information, and all the hubs seem to
interconnect to one another. For example, the People hub
contains your contacts, synced from and linked with all your
social networks.

The tight integration of social networks such as Twitter,
Facebook, and MySpace, may present another challenge.
Microsoft had to create a whole subset of instructions for
each social network and build it into Windows Phone 7
Series at the core level.
Windows Mobile is an excellent multi-tasking platform
and it has no problem with tasks running in the background
because WinMo automatically changes memory allocation
based on the needs of each application, and if the system
memory starts to get stretched thin, the OS shuts down
background tasks automatically.

Windows Mobile OS is designed from the ground up to
integrate with both Exchange Servers in the corporate
world, and with Outlook environments on the desktop. All
aspects of desktop Outlook commingle happily with Mobile
Outlook on the Windows Mobile side of things.
C. Apple iPhone OS:
Apple has created a PC-grade Operating system for a
Smartphone. Currently, Apples iPhone is the longing of the
modern man. Top-notch product design, an perceptive user-
interface as well as fancy multimedia internet based services
may contribute to its unique market position.
The first Apple version was introduced on January 9,
2007.Its successor, the iPhone 3G was revealed on July 11,
2008. iPhone 3G has the following hardware specifications:
Device dimensions/weight: 4.5 x 2.4 x
0.48 inch / 4.7 ounces
Display size: 3.5 inch
Resolution: 480 x 320 pixel at 163 ppi (i.
e. aspect ratio 3:2)
Input devices: Multi-touch enabled screen
interface plus 4 buttons (Home, Sleep/Wake,
Volume up/down, Ring/silent)
Connectors and input/output: 30-pin dock
connector, 3.5mm stereo headphone

Apple iPhone OS also provides an explicit
Security framework (Security. Framework)
provides interfaces for managing certificates,
public and private keys, and trust policies that you
can u
appli

A
powe
multi
and p
be us
a ver
its ex
the m
oppo
peop

B
devel
multi
that
hand
track
touch
hand
B
suppo
and,
comp
with
Micr
R
its o
April
Black
2010
2011
Curv
(9810
(9900
devic
(No L
Blackberry
Im
Black
use to guarant
ication manag
Another key a
ered graphics
iple sensors
proximity sen
sed, and it rapi
ry profitable g
xpected for the
market as mo
ortunities to ge
les minds for
a..BlackB
BlackBerry
loped by Re
itasking and s
have been a
dhelds, predo
kball, and mo
h screen for it
dheld devices.
BlackBerry O
ort for corpo
more recently
plete wireless
Novell Group
rosoft Exchang
RIM develops
perating syste
l 2010 RIM
kBerry OS 6
0. The softwar
making its a
ve (9350/93
0/9850/9860)
0/9930). RIM
ces would not
Legacy Devic
y OS Addition
mproved prod
kBerry applica
o
o
Black
Black
o
Savin
o
openi
tee the securi
es.
aspects of the
s, touch inte
like accelerom
nsor, which w
idly shown ho
gaming platfo
e iPhone to co
ore players ta
et in, so should
r a long time to
Berry OS
mobile op
esearch In M
upports specia
adopted by R
ominantly the
ost recently,
s BlackBerry
Os best know
orate email, t
y, a subset of M
s activation an
pWise email
ge etc.
and releases u
em to suppor
announced t
6.0 version, i
re was legally
appearance on
60/9370), B
and B
M announced
t be updated t
e Support).
nal Features su
ductivity & co
ation :
ZIP A
Faste
kBerry Sma
kBerry Playbo
Enha
ng Attachment
Faste
ing
ity of the data
e iPhone: Op
erface, as we
meter, light s
were all availa
ow it could get
orm. For the f
ontinue its asce
ake notice for
d the iPhone s
o come.
perating sy
Motion .It pro
alized input de
RIM for use
e track wh
the track pad
line of Smartp
wn for its n
through MID
MIDP 2.0, it a
nd synchroniz
, Lotus Domin
updated versio
rt each devic
to release the
in the 3rd q
released in A
the the Black
Blackberry
Blackberry
that any pre
o BlackBerry
uch as:
onnectivity wi
Attachment Su
er pairing betw
artphone an
ook
anced Suppor
ts
er attach
a your
penGL
ell as
sensor
able to
t to be
future,
ent on
r their
stay in
ystem,
ovides
evices
in its
heel ,
d and
phone
native
P 1.0
allows
zation
no or
ons of
e. On
e new
quarter
August
kberry
Torch
Bold
evious
OS 7
ith the
upport
ween a
nd a
rt for
hment
II
the
Win
ofte
the
app
ava
inte
win
app
to f
goo
mu
may
abo
app
def
But
bac
acc
bac
ove
of B

II. A COM
Android com
market. Its
ndows mobile
en been critici
When we go
key features
plications at th
ailable, multita
egrate with
ndows mobile
b. Multitas

Android is
plications at th
foreground pr
od user exper
ultitasking.
Windows mo
y be a bit su
out multi-task
plication clos
fault when exit
t the OS has
ckground, it a
cording to the n
ckground tasks
Imp
er WEP netwo
Add
Blackberry OS
o
App
o
Pictures
MPARISON B
WINDOWS
me very well in
market share
e has come w
ized for being
for a compari
to be compare
he same time
asking, the us
computers
and android o
sking
very well s
he same time
rocesses as mu
rience. Andro
obile are also
rprising becau
king. The r
e option. Ap
ted, which is v
s no problem
automatically
need of applic
s automatically
proved Wi-F
orks
ditional Multi
S:
Pinch to
Portrait
s App
BETWEEN AN
S MOBILE
n a short time
is growing a
with recent up
a dated Smart
son between m
ed are the abi
e, the number
ser interface a
desktop etc
on these featur
suited for ru
seamlessly. I
uch they dese
oid OS does
good in multi
use they have
eason is tha
pplications do
very confusing
ms with tasks
changes me
cations and the
y.
Fi connectiv
imedia Featu
o Zoom in Vid
Support
NDROID AN
e and penetrat
at a rapid pa
pdates but it h
tphone platfor
mobile platfor
ility to run ma
r of applicatio
and the ability
c.Lets compa
res:
unning multip
It gives attenti
erve to provide
a good job
itasking and t
e been criticiz
at it lacks
o not close
g for some use
running in t
emory allocati
e OS shuts dow
vity
ures
deo
in
ND
ted
ace.
has
rm.
rms
any
ons
y to
are
ple
ion
e a
in
this
zed
an
by
ers.
the
ion
wn
In-fact Android OS does the same thing. When
a user leaves an application and starts another the
first one keeps running in the background.
Windows mobile has a big advantage over
Android in the area of user-installed applications.
Windows mobile allows them to be installed in
either system memory which is much larger or in
external memory cards. It supports a large number
of applications to be installed.
b. Available Applications
Android is promising over 10,000+ applications
available and too in a very short time which is
indeed a great achievement. While Android has a
very few applications that one can find on a
iPhone, yet they are good to meet users
requirement.
Windows in somewhat lacking in this feature
especially in apps store games which is not
surprising because store is new to the market.
Android web browser is claimed to be much
better than the latest version of mobile Internet
explorer and it is accurate claim.
C. User Interface
Windows mobile interface is older than
Android and for that Android interface is better
than it. Android interface looks more modern and
is easy to interact with. Both interfaces are simple
Icon based in which the icons are organized in
launchers.
The two platforms can be compared more fairly
on the basis of third party alternatives. Android is a
very capable interface for third party interface shell
such as HTC sense.HTC has long produced its own
interface for Windows Mobile phones, too, and the
latest Sense interface, such as found on the HD2, is
amazing. It does for Windows Mobile what it does
for Android adds a highly customizable face on
the OS that makes it work the way the user prefers.
IV. COMPARATIVE STUDY OF IPHONE OS,
ANDROID OS AND WINDOW OS
Comparison of iPhone OS, Android OS and Windows
OS can be done on the basis of following points:

a.. Kernel-
iPhone kernel is OS X mobile which allows Apple to
reuse a lot of the mature code that is already proven to work
on the Mac. Android kernel is Linux where as Windows
kernel is CE6. CE 6.0 can run as many as 32,000
simultaneous processes, each with 2 GB of virtual memory
space. The functionality of an application is generally
unaffected by changes to the kernel.

b..Multitasking-
Android and iPhone both provide multitasking but
iPhone also provides push navigation service with
multitasking. Multitasking in windows OS is limited but it
provides push navigation service.

c.. Proprietary cloud-
Sync service-iPhones service Mobile Me is paid while
in Android it is Google apps which are mostly free.
Windows provides service My Phone 2.0 which is
unconfirmed and free.
.d.Home Screen-
iPhone has limited customization support while
Windows starts screen with customizable tiles.

e. Skin able OS-
iPhone OS and windows OS is not skin able or it is very
limited in this feature while Android OS is skin able.

THE BLACKBERRY OS AND SYMBIAN
Comparison between Blackberry OS and Symbian can
be done on these points:
a. Ease of use-
The interface of blackberry OS is wholly thumbwheel
driven and user may find it easy to use though it might be
slow. There is a built-in keyboard which accelerates some
operations such as composing E-mails. Blackberry OS may
not be the most efficient one but it is easy to use.
Symbian OS greatly depends on the phone for its ease of
use feature. For example in Nokia 9300 it is very easy to
navigate but on the other side in Nokia 6682 the interface is
confusing and difficult for data entry.
b. E-mail-
Blackberry excels in this area. It is not only the E-mail
applet which is addictive for its streamlined efficiency but
also the push method of e-mail delivery. Instead of being
retrieved manually messages appear like a magic on device.
Symbian is as versatile as other platform having no
innovative features. If you want something a little more
robust, you can choose from several push e-mail solutions,
including BlackBerry Connect. The operating system also
supports Lotus Notes and Microsoft Exchange for
maximum compatibility in the corporate world.
c. Multimedia-
Blackberry OS includes built-in music player, integrated
cameras and the latest featuring video recording facilities.
The Symbian operating system is pretty adept at
multimedia, with integrated support for audio and video
playback and recording, though not all those capabilities are
realized on the devices themselves. The Nokia 9300, for
instance, comes with both an MP3 player and the mobile
version of RealPlayer, which enables playback of
RealAudio, Real Video, and MP3 files. But it cannot
capture audio or video.

V. CONCLUSION
In Conclusion, all mobile operating systems presented
itself as a highly compelling investment on todays mobile
market, they proved there self to be capable to reach many
different type consumers in many new and innovative ways,
as well as to raise its own Business. Besides all the
advantages of Android over Windows Mobile, there are
some users who are comfortable with Windows Mobile due
to its stability and their enterprise solution. Further,
Microsoft is going to release its Windows Mobile 7, which
can be much better than the previous version.

Our analysis shows that each Mobile Operating System
may have strengths and shortcomings; the most secure
platform for the Business use is the Blackberry platform and
the iPhone has advanced towards being enterprise-class.
Symbian brings its limitations when it comes to
development. As earlier explained youll need a differen
t
SDK for every device.You will also have a different set
of tools in each SDK this will limit the porting of program
s you right for a Symbian Smartphone.
It will work on all devices with the same SDK but on oth
ers it will give problems.

Right now if you are looking for something that is at the
tip of technology trend, Android is what you need. It has
great future if they keep their innovation cycle running
continuously

VI. REFERENCES
[1]. Symbian OS C++ for Mobile Phones by Richard Harrison
[2]. htts://en.wikipedia.org/wiki/Mobile_operating_system
[3].http://www.appleinsider.com/articles/10/08/12/android_based
_smartphone_shipments_leapfrog_apples_iphone.html
[4]. Android Issue 3748: Add support for partition/block device
encryption

ISBN: 978-81-920575-7-6 :: doi: 10. 73893/ISBN_0768

Design of DAODV MANET Routing Protocol

G. Ilanchezhiapandian

Research Scholar , Department of CSE,
Sathyabama university,chennai

Dr.P.Sheik Abdul Khader

Professor & HOD, Department of Computer Applications,
B.S.Abdur Rahman University,Chennai

Abstract - Ad-hoc networking is a concept in computer
communications, which means that users wanting to
communicate with each other form a temporary network, without
any form of centralized administration. Each node participating in
the network acts both as a host and as a router. The study of
distributed network and mobile ad hoc network (MANET) are
currently two hotspots in distributed computing and mobile
communication researching domain. We build up a distributed
overlay network on top of MANET's physical infrastructure, to
effectively integrate distributed networks advantage on
sustaining highly dynamic network into the design of MANET
routing protocol. We deploy passive MANET routing algorithms
with innovative distributed routing mechanisms. We propose a
new kind of MANET routing protocol named distributed
Computing based ad-hoc on demand routing (DAODV). Our
work indicates that DAODV has an improved routing
performance in comparison with the popularly used AODV
protocol.

Keywords- MANET, DAODV, distributed network.

I. INTRODUCTION

In ad-hoc networks, communications are done over
wireless media between stations directly in a distributed
fashion without the help of wired base station or access
points. With the popularity of ad-hoc networks, many
routing protocols have been designed for route discovery
and route maintenance. They are mostly designed for best
effort transmission without any guarantee of quality of
transmissions. Some of the most famous routing protocols
are Dynamic Source Routing (DSR) Ad-hoc On Demand
Vector (AODV) Optimized Link State Routing protocol
(OLSR), and Zone Routing Protocol (ZRP). On the other
side, Distributed network is currently a hotspot in the
international network researching domain. Different kinds
of distributed network models such as Napster, Freenet and
Gnutella have already been popularly used. However, each
of them has some open problems [2], [3]. To solve these
problems, structured DISTRIBUTED overlay networks
(SON) such as CAN, Chord [4], Pastry and Tapestry are
proposed to provide a kind of mechanism to improve
network scalability [5]. All of these SON algorithms have a
common ground that they all assign keys to data and nodes
through a hashing function, and at the same time they use
these keys to store and maintain data which are shared in
the network [6].

II.OVERVIEW OF MANET AND DISTRIBUTED
NETWORKS
DISTRIBUTED networks are popular Internet applications
that enable users to share computing resources. Today,
DISTRIBUTED, Web browsing, and email can be put
together in the same class of importance. Napster is the
tool probably most responsible for the popularization of
DISTRIBUTED networks due to its music sharing support.
So far, several Napster clones and general file sharing
systems have been created, such as KaZaA, BitTorrent,
and eMule. DISTRIBUTED networks are composed of
software nodes (peers) installed on the edges of the
Internet, usually in inexpensive user hosts along diverse
administrative domains. Each peer communicates with
other peers directly or indirectly to share local resources
(e.g., memory, files, processing power), forming robust
distributed systems over the Internet. The DISTRIBUTED
communication model clearly differs from the client-server
model. In client-server, servers store resources accessed by
clients, by contrast, in DISTRIBUTED each peer is itself a
repository that shares its resources with other peers, thus
forming environments where resources are distributed
among peers acting as both servers and clients.
Client-server addressing is based on the Internet Domain
Name System (DNS), while DISTRIBUTED networks
employ addressing schemes independent from DNS to
index peers, users, and resources. Client-server messages
are routed at the network layer by Internet routers, but
DISTRIBUTED messages are routed at the application
layer by routing peers.
Having routing and addressing built on top of Internet
protocols, DISTRIBUTED networks are quite dynamic
environments where new peers can join the network when
they are started in user hosts, and leave when users
shutdown the peers. This dynamic nature makes
DISTRIBUTED networks both unpredictable and scalable:
u
p
e
a
i
D
w
a
t
a
s
F
A
l
c
A
c

unpredictable
peers that ho
expansion of
addition of ne
During th
improvements
Due to its po
without any c
as the art of n
the research o
always taken
similarities ar
Figure 1).
A. Dynamic N
Nodes in D
leave the netw
change freque
And the mob
can make net
[8].
because resou
ost them are
f the networ
ew peers.
he last cou
s are made in t
ossibility to c
entral manage
networking wi
of DISTRIBU
as two sepa
re seen betwe
Network Topolo
DISTRIBUTE
work, which c
ently with tim
bility characte
twork topolog
(a). A
urces may be
e down; scal
rk capacity r
uple of ye
the research o
create and or
ement, MANE
ithout a netwo
UTED network
arated areas. B
een these two
ogy
D networks r
causes the net
me. It is the sam
er of mobile n
gy changes mu
Architecture of
unavailable i
lable because
requires only
ears tremen
f ad hoc netw
rganize a netw
ET is character
ork [7]. Curre
k and MANET
But the follow
technologies
randomly join
twork topolog
me with MAN
nodes in MAN
uch more freq
f MANET
if the
e the
y the
ndous
orks.
work
rized
ently,
T are
wing
(see
n and
gy to
NET.
ANET
quent

co
co
wi
co
th
de
as
in
to
bo
ro
b.M
bo
ca
ca
wi

DI
tha
DI
ph

(b). Arch
Fig. 1 Sim
A. Multi-h
Nodes in b
onnect with ea
onnections in D
ith physicall
onnections in
e radio transm

B. Decent
Both DISTR
ecentralized st
what is seen
the network
each other. T
ottleneck cause
bustness of th
Multi-Identity
Nodes in DI
oth as a client
an route and t
an also provid
ith each other
II. DE
a. Architect
The architec
ISTRIBUTED
at DAODV
ISTRIBUTED
hysical networ
hitecture of DI
milarities of MANE
hop Connectio
both MANET
ach other via m
DISTRIBUTE
ly unlimited
MANET are
mission range.
tralized contro
RIBUTED net
tructure. Ther
in the Client/
are all peers a
The network w
ed at the centr
he network.
y of Nodes
ISTRIBUTED
and a server,
transfer messa
de resource do
directly [9].
ESIGN OF DA
ture of DAOD
ture of the D
D computing
V protocol
D overlay netw
rk of the MAN
ISTRIBUTED

ET and DISTRIB
on
T and distrib
multi-hop rout
ED networks a
range, wh
via wireless l
ol
twork and M
re is no centra
/Server networ
and can comm
will not suffer t
ral server, whi
D network an
as well as a ro
ages independ
ownloading an
AODV PROT
DV Model
DAODV mod
technology. F
is integrat
work which i
NET.
D Network
BUTED networks
buted networ
ting. Hop-to-h
are via TCP lin
hile hop-to-h
links, limited
MANET have
al administrati
rk model. Nod
municate direc
the paralysis a
ich improves t
nd MANET
outer. Each no
dently. And th
nd communic
TOCOL
del is based
Figure 2 sho
ted with t
s built upon t

rks
hop
nks
hop
by
e a
ion
des
ctly
and
the
act
ode
hey
ate
on
ows
the
the

Fig. 2 Architecture of DAODV Model

a.Basic Design
DAODV uses the node naming mechanism used in Chord.
It adopts a consistent hashing algorithm [10], [11]. Each
node is assigned an m-bit node identifier (NID), which is
obtained by hashing the node's IP address. Thus we are
able to specify the location of the node in the hash ring.
When a node joins the network for the first time, a new
NID ranging from 0 to (2m-1) is automatically assigned
based on the adopted hashing algorithm. To simplify the
routing algorithm, we have adopted the discontinuous
naming mechanism used in Chord. The node state is
maintained in the DAODV routing table (PRT) which is
similar to the routing table of AODV. The difference is
that, DAODV uses the NID for the destination node and
the next hop, whereas the later uses the nodes IP address.
A. Route discovery Algorithm
The DAODV route discovery algorithm discovers the
route from the source node to the destination node
reactively. The route discovery is triggered by the mobile
nodes under any of the following, and an update is required
in the PRT table.

i. A node initiates the route discovery procedure.
ii. When a node receives the route query message
(RREQ), it queries the PRT, if the route is null or out dated,
the route discovery method is invoked.
iii. When a route issues a route error message (RERR) to
the node which initiated the data transmission. The
initiating node starts the route query method.
A description of the route discovery algorithm is given
below:
1) Node M checks for RREQ in cache. If the source
identifier and the request identifier is the same, the request
is discarded.
2) M checks in the Route Record whether RREQ
message contains NID(M). If found, the message is
discarded.
3) M checks whether the destination node is present in
the PRT entries. If found go to step 5, otherwise to step 6.
4) M checks whether the destination node is M itself. If
true, go to step 5, otherwise to step 6.
5) M checks whether the route is bi-directional. If it is,
the route reply message (RREP) is send to the initiating
node S directly. Otherwise, the route discovery method is
invoked.
6) M adds itself to the route record and broadcasts
RREQ message to the neighbors.
B. PRT Querying Algorithm
The PRT querying algorithm queries the PRT and picks
up the Next Hop Ring Interval (NRI) which is the closest
to D. Any outdated entries are updated. The algorithm is
depicted as follows:
1) When a node M initiates or receives a RREQ
message, it compares the destination node identifier of the
RREQ with NID(M). If they are matched, M will return
RREP to querying initiating node S.
2) M will search into the PRT and compares the
destination node identifier (D) with the PRT entries to find
the NRI to which D belongs for message transfer.
3) M triggers the route discovery for the destination
node and waits for the response. If the response if received
the PRT is updated, otherwise a failure message is send.
c.DAODV Route Update and Maintenance
Route maintenance is required due to the nodes
mobility. In MANETs, a nodes position changes with time,
this leads to dynamic network topology. A destination
node may become out of reach due to normal or abnormal
departure or has moved out of the transmission range.
When this occurs, the route updating algorithm is triggered
reactively. When there is a failure in the data transmission,
the node which initiated the packet transfer will issue an
overdue message to the related nodes. Thus the PRT
entries are outdated and the route update is triggered. If the
route discovery process ends in a failure, it means that the
destination node has moved out of range. Therefore the
destination node is marked as invalid.
Due to the high dynamic character, mobile nodes may
join and leave the network at any time. To save the route
maintenance cost, when a node joins the network, DAODV
will not initialize its PRT. On the contrary, the PRT is
constructed only on demand.
DAODV nodes will only overhear the source route
related to the next hop node in its local PRT. By doing this,
DAODV inherits all the advantages of AODV at the same
time limits. By the use of Phello messages in DAODV,
mobile nodes broadcast phello messages reactively.
Therefore the information about the neighbor nodes are
collected only when there is a need for route discovery. If
the destination node is one of the neighbors, the route
discovery process stops and RREP is returned to the source.
Consequently, the overload of nodes in DAODV is
reduced

III. SIMULATION RESULTS
File
Discovery
Service
Information
Administration
Service
File
Transfer
Service
Communication
Security
Service
DISTRIBUTED Computing Based MANET Routing
Algorithm
DISTRIBUTED Overlay Network (Node Naming and
Indexing)
MANET Physical Network

s
s
m
p
t
s
n
a
a
r
v
o
o
p
p

e
a
i
o
P

f
T
D
D
d
s
t
i
l
Simulation
space. The
simulation sc
movement du
pause time (P
the model. At
static for a PT
node and mov
arrives, the no
a node as th
repeated till t
values of PT.
on DAODVs
of nodes to tr
packets are 12
packets per se
A. Pack
Packet del
estimating a r
adapts to the c
it provides a r
of Packet Deli
Pause Time.
Fig. 3
The probab
failure decrea
The success
DAODV is m
DAODV has
delivery perf
structured DIS
the caching m
its problem in
low packet de
B. Avera
is done with
nodes are
cope. And e
uring the sim
PT) is used to
t the beginnin
T. Then it will
ves toward it
ode will keep
he next dest
the simulation
To simulate
s performance
ansmit CBR p
28 bytes packe
econd
ket Deliver Rat
liver ratio i
routing protoc
change of netw
route discover
iver Ratio betw
Comparison
DA
bility of link f
ases when the
rate and stab
much better th
improved th
formance by
STRIBUTED
mechanism ado
n providing ou
eliver ratio.
age end to En
fifty nodes in
randomly di
ach node pe
mulation period
o define the m
ng of simulat
l randomly ch
t with a certai
static for a PT
tination. This
n ends. We ch
the network s
e, we randoml
packets to eac
ets with a tran
tio
is an impor
ols performa
work topology
ry. Fig 3 depic
ween DAODV
of PDR betwe
AODV
failure betwee
e mobility of
bility of packe
han AODV. T
he node disco
y taking ad
routing mech
opted in AOD
ut-of-time netw
nd Delay (AED
n a 1000m*10
istributed in
erforms a Br
d. The traditi
mobile charact
ion, a node k
hoose a destina
in speed. Wh
T and then ch
process wil
hoose for diffe
services influ
ly choose 20 p
ch other. The C
nsmission rate
rtant criterion
ance in how w
y and how effi
cts the compar
V and AODV
een AODV an
en nodes and r
f nodes decre
et transmissio
The reason is
overy and pa
dvantages of
hanisms. How
DV is the orig
work status an
D)
000m
the
rown
ional
er of
keeps
ation
hen it
hoose
ll be
ferent
uence
pairs
CBR
e of 4
n in
well it
cient
rison
with
nd
route
eases.
on in
that
acket
the
ever,
in of
nd its
fa
co
DA

me
ro
DA
bu
M
ac
c.N
de
th
ma
we

Fig

dir
Th

Average end
st route dis
omparison of
AODV and A
Fig. 4 Co
AODV emp
echanism and
ute when the
AODV impro
uilding up a
MANETs phy
ccelerate the ro
Node Storage
Figure
ecreases the ro
e same time
aintenance ov
ell.
g. 5 Compari
The cached r
rect proportio
hat is, for a M
d-to-end delay
covery is d
f Average E
AODV with Pau
omparison of AED
loys broadca
d consumes l
e current path
oves the rou
a DISTRIBU
ysical topolog
oute discovery
Overhead (NS
5 depicts
oute storage o
e it decrease
verhead relate
ison of NSO betw
routes for a m
on to the num
MANET with N
y estimates ho
done. Figure
End-to-End
use Time.
D between AODV

asting based
longer time t
h has failed. O
ute discovery
UTED overla
gy and intro
y process.
SO)
that DAOD
verhead for ea
es the route
ed to these c
ween AODV and D
mobile node i
mber of nodes
N nodes, the c
ow efficient a
4 depicts t
Delay betwe
V and DAODV
route discove
to re-discover
On the contra
y efficiently
ay network
ducing PRT
DV effectiv
ach node, and
discovery a
cached routes
DAODV
in AODV are
in the netwo
cached routes
and
the
een

ery
r a
ary,
by
on
to
ely
d at
and
as

in
ork.
for
e
s
r
w
D
f
d
c
b

t
r
c
a
D
s
a

D
l
o
D
D
h
s
o
s
r
i
each node is
structured D
relationship o
with the num
DAODV effe
for each node
discovery an
cached routes
C. Rout
Figure 6 d
between DAO
Fig. 6 Com
The numbe
the number o
route mainten
cached routes
and decrease
DAODV impr
structured DI
and limiting th

DAODV
DISTRIBUTE
logical names
operating in
DAODV ha
DISTRIBUTE
highly dynam
scalability. Be
optimizing me
structured DIS
routing efficie

[1] Mengkun
improving search
s O(N). On t
DISTRIBUTED
of cached rou
mber of node
ectively decrea
e, and at the s
nd maintenanc
as well.
ing Overhead
epicts the com
ODV and AOD
mparison of RO
er of route pa
of cached rout
nance and upd
s consume mu
e the perfor
roves the rout
ISTRIBUTED
he number of
IV.
seamlessly
ED overlay ro
pace with tho
a physical
as inherited
ED networkin
mic network
esides these, D
echanisms to
STRIBUTED
ency.
REFE
n Yang and Zon
h efficiency in
the contrary,
D routing
utes for each
es is O(logN)
ases the route
same time it d
ce overhead
d (RO)
mparison of R
DV with Pause
O between AO
ackets in AOD
tes for each n
date processes
uch of the ne
rmance of A
ting performan
D overlay net
cached routes
CONCLUSIO
integrates th
outing protoco
se of MANET
namespace.
d all the
ng technolog
ks and main
DAODV intro
solve the deto
overlay netw
ERENCES

ngminm Fei, A
unstructured pee
by implemen
mechanism,
node in DAO
). By doing
e storage over
decreases the r
related to t
Routing Over
e Time.
ODV and DAO
DV is huge du
node is huge.
s caused by t
etwork bandw
AODV. Whe
nce by introdu
twork techno
.
ON
he functions
ols operating
T routing proto
By this me
advantages
gy in sustai
ntaining netw
oduces a serie
ouring problem
work and impr
A novel approa
er-to-peer netw
nting
the
ODV
this,
rhead
route
these
rhead

ODV
ue to
The
these
width
ereas
ucing
ology
s of
in a
ocols
eans,
of
ining
work
es of
m of
roves
ach to
works,
Jou
No
and
Th
pp
net
20
Ka
pee
on
bet
net
No
po
Te
to
Sel
13
"Is
fut
Co
for
Co
37
WC
Co
im
NO
Sym
urnal of Paralle
ovember, 2009
[2] J. Xu, "On
d network diamet
he Conference on
. 2177-2187.
[3] S. Sen and J
tworks," IEEE/AC
04, pp. 219-232.
[4] I. Stoica, R
aashoek, F. Dabe
er lookup protoco
Networking, Vol
[5] J. Xu, A. K
tween routing t
tworks," IEEE Jo
o. 1, Jan 2004, pp
[6] S. Jain, R. M
tential of DHT-b
chnologies and Sy
[7] S. Jiang, Y. L
variable topologi
lected Areas in C
56.
[8] D. Cavalcan
ssues in integrat
turistic heterog
ommunications, V
[9] C. H. Foh a
rmula for one-dim
onference on Com
42.
[10] A. Duran
CNC 2004 - I
onference, Vol. 5,
[11] J. Mischke
mplementation for
OMS 2004 - I
mposium, Vol. 9,

el and Distribute
the fundamental
ter in peer-to-peer
Computer Comm
J. Wang, "Analy
CM Transactions

R. Morris, D. L
ek, and H. Balak
ol for internet ap
l. 11, No. 1, Feb 2
Kumar, and X.
table size and
ournal on Selected
. 151-163.
Mahajan, and D. W
based overlays,"
Systems(USITS 20
Liu, Y. Jiang, and
es for routing sch
Communications,
nti, D. Agrawal, C
ting cellular netw
geneous wirele
Vol. 12, No. 3, Jun
and B. S. Lee, "A
mensional MANE
mmunications, V
and C.-C. Shen,
IEEE Wireless
No. 1, March 20
e and B. Stiller,
a highly scalabl
IEEE/IFIP Netw
No. 1, Apr 2004,
d Computing, V
tradeoffs betwee
r networks," IEEE
munications, Vol. 2
yzing peer-to-peer
s on Networking,
Liben-Nowell, D.
krishnan, "Chord:
pplications," IEEE
2003, pp. 17-32.
Yu, "On the fu
network diame
d Areas in Comm
Wetherall, "A study
4thUSENIX Sym
003), March 2003.
d Q. Yin, "Provisi
hemes in MANET
Vol. 22, No. 7,
C. Cordeiro, B. X
works, WLANs,
ess network,"
ne 2005.
A closed form n
ETs," ICC 2004 -
ol. 27, No. 1, Ju
"Mobile ad hoc
Communication
04, pp. 114-119.
"Efficient protoc
le peer-to-peer se
work Operations
, pp. 411-424.
Volume 69 Issue
en routing table s
E INFOCOM, 200
22, No. 1, Mar 20
r traffic across la
Vol. 12, No. 2, A
. R. Karger, M.
A scalable peer
E/ACM Transacti
undamental tradeo
eter in peer-to-p
munications, Vol.
y of the performa
mposium on Inter
.
oning of adaptabi
Ts," IEEE Journal
Sep 2004, pp. 13
Xie, and A. Kum
, and MANETs:
IEEE Wire
network connectiv
- IEEE Internatio
une 2004, pp. 37
c P2P file sharin
s and Network
col specification
earch infrastructu
and Managem
11,
size
03 -
003,
arge
Apr
. F.
r-to-
ions
offs
peer
22,
ance
rnet
ility
l on
347-
mar,
: A
less
vity
onal
739-
ng,"
king
and
ure,"
ment

ISBN: 978-81-920575-7-6 :: doi: 10. 73900/ISBN_0768
Modeling Impact of Varying Data Rates for Various Spread Spectrum
Techniques on Load in WLAN Ad-Hoc Mode

Rajan Vohra
Lecturer, Department of
Electronics Technology
Guru Nanak Dev University
Amritsar, Punjab, India

Abstract:-.Wireless Local Area Network (WLAN) uses
high frequency radio waves rather than wires to
communicate and transmit data. A wireless local area
network links two or more devices and gives users the
mobility to move around within a local coverage area and
still be connected to the network. Wireless data connections
have inherently high bit error rates, low bandwidth and long
delays. In this paper OPNET Modeler 9.1 network simulator
has been used to do the comparative modeling and
simulation of the load for different data rates (1, 2, 5.5,11
mbps) and various techniques (FH, DS, IF) and concluded
FH technique is better in 1 Mbps scenario and best in 11
Mbps scenario than DS and IF techniques.[7.8]

Keywords: WLAN, Load, OPNET, FH, DS, IF.

1. INTRODUCTION

Wireless LANs are becoming popular the world over
due to ease of installation, and the ability to offer wireless
access to their users often for free [1]. The IEEE (Institute
of Electrical and Electronic Engineers) 802.11
specifications used the 2.4 GHz frequency band and
initially supported a maximum data rate of 1 to 2 Mbps .
The 802.11b specification increased the performance to
11 Mbps in the 2.4 GHz range. It encompasses the
physical (PHY) and the media access control (MAC)
layers of the ISO seven-layer network model [2]. Wireless
local area networks (WLANs) based on the IEEE 802.11
standard are one of the fastest growing Wireless access
technologies in the world today. These provide an
effective means of achieving wireless data connectivity in
homes, public places and office and also enable people on
the move to communicate with anyone, anywhere at
anytime with a range of multimedia [1-3]. The WLAN
media is error prone and the bit error rate (BER) is very
high compared to the BER of wired networks. In addition,
Carrier Sensing is difficult in wireless networks because a
station is incapable of listening to its own transmissions in
order to detect a collision.
There are two operation modes defined in IEEE 802.11
(a) Infrastructure Mode

In the infrastructure mode above, the wireless network
consists of at least one AP (access point) connected to the
wired infrastructure. All the wireless stations are connected to
the AP. An AP controls encryption on the network and also can
route the wireless traffic to a wired network (same as a router).
We can assume an AP as similar to base station used in
cellular networks.
(b) Adhoc Mode
AP is not required in Ad-Hoc mode, the wireless station is
connected to each other directly without using an AP or any
other connection. The topology is useful to set up a wireless
networks quickly. Ad-Hoc mode is also called peer to peer
mode or (IBSS) Independent Basic Service Set.

This paper deals with the load optimization techniques that
can be effectively used for a WLAN based on the network
simulator, OPNET Modeler 9.1[3].

II. MODEL OUTLINE
In our model, we used two workstations between which,
we transfer information/data through wireless mode with the
help of different techniques and data rates that are available
in OPNET MODELER 9.1. We have made 4 scenarios and
in each scenario 4 different data rates for 3 different
techniques (Frequency Hopping, Direct Sequence and
Infrared) have been employed for our simulation and through
these scenarios we compare load parameter and evaluate the
performance at different data rates for various techniques.
Table I shows the WLAN traffic generation parameters
for both the scenario. Operation Mode used in the scenario is
serial (ordered), start time (seconds) are uniform (100,110),
repetition pattern in the scenarios is serial, number of
repetitions is constant and rest of the parameters used are
listed in table below:
Table II shows the parameters, which are used in
different scenarios for simulation. Parameters indicate that
we use PHY characteristics as Direct sequence, Frequency
Hopping and Infrared, buffer size (bits) 1024 k for all
scenarios. Bandwidth used for the WLAN in scenario 1 and
scenario 2 is 1 Mbps and 2 Mbps respectively and for
scenario 3 and 4 it is 5.5 and 11 Mbps respectively.
t
t
O
h
p
s
p
M

I

In Our mo
transfer infor
techniques usi
OPNET MOD
having differe
performance i
scenarios have
100*100m WL
performance u
Mbps respectiv

WLAN scena

TABLE I. W

A
D
Re
Ope
Start T
Inter-r
(se
Numbe
II. SIMULAT
odel, we used
rmation/data
ing variable da
DELER 9.1. H
ent data rates
is evaluated
e been tested
LAN model fam
under different
vely.
ario at different te
Wireless LAN tra

Attribute

Duration

epeatability

eration Mode

Time (seconds)

repetition Time
conds)

r of Repetitions
TION EVALU
two workstati
wirelessly f
ata rates which
Here we have
s for various
under these
d in campus e
mily. In this pa
t data rates of
echniques (FH, D
affic generation pa

Val

End of sim

Once a
Time

Serial (o

uniform (

constan

constan
UATION
ions in which,
for the vari
h are available
made 4 scena
techniques.
scenarios. Th
environment w
aper, we verify
1, 2, 5.5, and
DS, IF) and data ra
arameters
lue
mulation
at Start
e
ordered)
(100,110)
nt (300)
nt (30)
we
ious
e in
arios
The
hese
with
y the
d 11

ates.

rate
obs
nod
tech
Mb
Dir
100
data
aro
the
IR
(bit
and
term
Mb
in
sce
ban
Ch
Th
R
R
Repetitio

We have obs
es for various
served that at
des is comp
hniques(DS,FH
bps scenario F
rect Sequence
0,000(bits/sec)
Figure
a rate, there is
und 350,000 (
same time FH
technique,
ts/sec) for FH.
When we com
d FH techniqu
ms of Bits/sec
bps scenario w
1 Mbps scen
nario. Reason
ndwidth gives
nodes using in

Parameter
(s)
Address
WLAN
bandwidth
(bps)
PHY
haracteristics
Slot time
(s)
Packet
Reception-
Power
hreshold (W)
Short
Retry Limit
(Attempts)
Long
Retry Limit
(Attempts)
Buffer
size(bits)
Max
Receive
Lifetime
(secs)

on Pattern
served the res
techniques. F
2 and 5.5 Mb
paratively sa
H and IF) bu
Frequency Hop
e and Infrared
) for the period
5 shows that f
s drastic incre
(bits/sec) for t
H and DS tech
110,000 (bit
.
mpare scenario
ues are better t
c in case of 11
which shows th
nario is less a
n for this is
s less load on t
n the scenario
Scen
ario 1 ari
Auto Assign
1
Mbps Mb
DS,F
H,IF H,
1024000
0.5

Serial
sults of Load
From figure 3
bps Load (bit
ame for all
ut figure 1 su
pping techniq
d Techniques
d 10-60 minut
for IF techniqu
ease in Load o
the period 0-60
hniques give b
ts/sec) for D
o for 1Mbps a
than IR. The v
Mbps scenari
hat the load o
as compared t
working of
the
o.
Scen
io 2
Sce
ario 3
ned Au
2
bps
5.5
Mbps
DS,F
,IF
DS,
H,IF
5.0 E-05
7.33 E-14
7
4
0
at different d
3 and 4 we ha
ts/sec) at WLA
l the differ
uggests that at
que is better th
by 50,000 a
tes.
ue with 11 Mb
on WLAN nod
0 minutes, but
etter results th
DS and 50,0
and 11 Mbps, D
value of Load
io is more than
on WLAN nod
to the 11 Mb
scenario at l
n Scen
ario 4
uto Assigned
5 11
Mbps
F DS,F
H,IF
1024000
0.5
data
ave
AN
ent
t 1
han
and
bps
des
t at
han
000
DS
d in
n 1
des
bps
ess

Fig 1: Load a

TABLE II: S

Fig 2: Load a

at 1 Mbps in diffe
Simulation scenari

at 2 Mbps in diffe
erent techniques
io parameters
erent techniques

for
resu
and
tech
bes
wan
req
resu
tech

Fig 3: Load at 5

Fig 4: Load at 1

This paper in
different data
ult of the Loa
d 4 there is
hniques. FH t
st in 11 Mbps
nt to work
quired less Lo
ults we will
hniques.

5.5 Mbps in differ
1 Mbps in differe
IV. CON
nvestigates th
a rates and tec
ad is comparat
s drastic cha
technique is b
scenario than
with high b
oad on WLA
opt FH tech
rent Techniques
ent techniques
NCLUSION
e Load on th
chniques. For
tively same b
ange in Loa
better in 1 Mb
n DS and IF te
bandwidth en
AN nodes the
hnique over

he WLAN nod
scenario 2 and
but in Scenario
ad for differ
bps scenario a
echniques. If
nvironment a
en by observ
the DS and
des
d 3
o 1
ent
and
we
and
ved
IF
REFERENCES

[1] Walid Hneiti and Naim Ajlouni, Performance Enhancement of
Wireless Local Area Networks,Proceedings of IEEE ICTTA06, 2nd
International Conference on Information CommunicationTechnologies:
from Theory to Applications, Damascus, Syria, vol. 2, pp. 2400- 2404,
April 2006.

[2] L.Bononi, M. Conti, and E. Gregori,Runtime Optimization of
IEEE 802.11 Wireless LANs Performance, IEEE Transactions on
Parallel And Distributed Systems, vol. 15, pp. 159 172.January 2004.

[3] Mohammad Hussian Ali and Manal Kadhim Odah,Simulation
Study 0f 802.11b DCF Using OPNET Simulator, Eng. & Tech. Journal,
Vol. 27,No6 , pp. 1108-1117. 2009.

[4] Ms. Amardeep Kaur1, Dr. Sandip Vijay2, Dr.
S.C.Gupta3 Performance Analysis and Enhancement of IEEE
802.11 Wireless Local Area Networks Vol. 9 Issue 5
(Ver2.0),pp.130.January 2010.

[5] Q. Ni, I. Aad, C. Barakat, and T. Turletti, Modeling and
Analysis of Slow CW Decrease for IEEE 802.11 WLAN. 14th IEEE
Proceedings on Personal, Indoor and Mobile Radio
Communications,IEEE PIMRC 2003,vol. 2, pp. 1717-1721. Beijing,
September2003.

[6] Elena Lopez-Aguilera, Martin Heusse, Franck
Rousseau, Andrezej Duda and Jordi Casademont, Evaluating
Wireless LAN Acess Mothods in Presence of Transmission
Errors.INFOCOM 2006.Barcelona, pp. 1-6, Spain,April 23-29, 2006.

[7] Rajan Vohra; R.S.Sawhney; Manju Sharma:
Performance Enhancement of WLAN by Buffer Size Optimization using
OPNET National conference on Advance Computing and
Communication Technology (ACCT-10) organized by Vaish College of
Engineering, Rohtak, 25-26 June,2010 pp 495-499

[8] Rajan Vohra; R.S.Sawhney; Shikha: WLAN Metrices
enhancement using Buffer Size Optimization International Journal of
P2P Network Trends and Technology- Volume 1 Issue 2 -2011 Page 22-
25 (ISSN: 2249-2615)


V
A
Proc. of th
Volume 1. Cop
ISBN: 978-8
Vertic

Abstract - Se
networks is on
wireless netwo
this seamless m
Vertical Hand
this continuity
schemes are a
handover dec
which provid
seamless mobi
Keywords - Ver
DVHD,C-DVH

Seam
any system w
better QoS du
another. It
telecommunic
protocols whi
A messag
the serving
indication th
communicatin
All additiona
wireless netw
network to t
received the
the neighbor
radio signals
neighboring w
minimum t
communicatio
network to th
after the neig
wireless chan
handover requ

II. VERT
In han
Handover (H
(VHO) as sh
process when
network with
Wi-Fi to Wi-
access point i
connectivity
normally to

e Intl. Conf. o
pyright 201
1-920575-7-6
ACM #: dber
cal Hand
ervice continu
ne of the chal
orks. Handove
mobility while
dover has bee
y. In this pape
analyzed and th
cision scheme
e less process
ility over heter
rtical Handove
HD.
I. INT
mless handove
with mobility
uring process
is a met
cations system
ich are differe
ge is sent from
wireless ne
hat the mo
ng with the n
al information
work is also s
the mobile st
mobile statio
ing wireless
s between t
wireless netwo
threshold th
on transactio
he neighboring
ghboring wire
nnel for the m
uest.
TICAL HAND
HAN
ndover there a
HHO) and an
hown in Fig.
n the mobile
the same tech
Fi. Vertical h
in a network
it uses to a
on Computer A
12 Techno For
6 :: doi: 10. 73
r.imera.10. 739
over Deci
Assi
S
uity in hetero
llenges in the
er technique i
switching in b
en normally u
r the vertical
he clustered D
has been eval
sing delay an
rogeneous netw
r decision, clus
TRODUCTIO
er is a fundam
y. It is the att
migration fro
thod used
m to transfer
ent from each
m a mobile sta
etwork, whi
bile station
neighboring w
n relating to
sent from the
tation. Using
on receives ra
network. Th
the mobile
ork is measure
hen a han
n from the
g wireless net
less network
mobile station
DOVER AND
NDOVER
are two types,
nother is Ve
1.Horizondal
e user switch
hnology and s
handover refer
changing the
access a supp

Applications
rum Group, In
907/ISBN_07
907
ision Sche
Ar
istant Professo
SCMS School
Karukutty
ogeneous wire
fourth genera
is used to pro
between netwo
sed for provid
handover deci
Distributed vert
luated as the
d hence prov
works.
stering, DVHD,
ON
mental concern
tempt to prov
om one domai
in a wire
r messages us
other.
ation to a use
ch includes
is capable
wireless netw
the neighbo
serving wire
this informa
adio signals fr
he quality of
station and
ed. If this mee
ndover of
serving wire
twork is execu
has establishe
n pursuant to
HORIZONTA
one is Horizo
ertical Hando
handover is
hing between
ame network
rs to a node o
mode or typ
porting netw

ndia.
768
emes in H
.
rchana Sreej
or, Departmen
l of Engineeri
y, Cochin, Ke
eless
ation
ovide
orks.
ding
ision
tical
one
vides
T-
rn in
vide
in to
eless
sing
er to
an
of
work.
ring
eless
ation
from
the
the
ets a
the
eless
uted
ed a
the
AL
ontal
over
the
the
like
or an
e of
work,
wir
Wi
acc
pro
tech
Thu
bas
we
use
ano
sele
term
und
freq
oth
for
swi
Han
dec
pha
crit
sec
to
net
Ser
pha
the
wit
wh
con
Heterogen
ith
nt of Compute
ing and Techn
erala., India

Fig.1 Ho

support node
reless device
ireless LAN
cess. It is c
ovides better
hnology whic
us the Laptop
sed on the av
say that ver
er to automat
other to maint

When a
ected for Com
minal needs t
der both situ
quency bands
her important c
mobility is
itching. Hand
ndover Initi
cision, Handov
ase the hand
teria value lik
cond phase ca
decide which
twork and ex
rvice (QOS)
ase which is
e neighbor ne
th this ,QOS
hich network
nnection. The
neous Wir
er Science
nology
orizontal versu
e mobility. Fo
like laptop m
and a cellula
clear that W
speed rates
ch provides ub
user can swit
ailability and
tical handove
tically switch
ain communic
Vertical Han
mmunication i
to have a dua
uations such
s and various
challenge whi
s the automa
dover mechan
ation, Syste
ver execution
dover process
ke signal stren
alled the Syste
h mobile use
xchanges info
provided by
the Handove
twork QOS a
decision mak
the mobile
e last phase i
reless Netw
us Vertical Ha
or example, a
might be able
lar technology
Wireless LAN
compared to
biquitous netw
itch using thes
d necessity. In
er will provid
h from one
cation ultimat
ndover mechan
it is necessary
al mode card
h as WLAN
s modulation
ile using a ver
mation aspects
nisms have a
em Discover
n. In the Hand
s was modif
ngth, link qua
em discovery
er discovers
ormation abo
these networ
er decision ph
and the mobi
aker makes th
e user has
is the Hando
works
andover
well equipped
to use both a
y for interne
N connectivity
o the cellula
work coverage
se two device
n simple term
de or help the
technology to
tely.
nism has been
y that a mobile
that can work
N and UMTS
schemes. The
rtical handove
s in network
a four phases
ry, Handove
dover Initiation
fied by some
lity etc. In the
phase is used
its neighbor
ut Quality o
rks. The third
hase compare
ile users QOS
he decision to
to direct the
over execution

d
a
et
y
ar
e.
s
ms
e
o
n
e
k
S
e
er
k
s.
er
n
e
e
d
s
of
d
s
S
o
e
n
ISBN: 978-81-920575-7-6 :: doi: 10. 73914/ISBN_0768
An Algorithm for Dynamic Search in Unstructured Peer-to-Peer
Overlay Network
Ashish S Poste
Computer Engineering Department
PICT
Pune, India

Abstract Peer-to Peer (P2P) computing network has
emerged as a model for resource sharing over the internet.
Designing efficient search algorithm is a key challenge in
unstructured peer to peer network (P2P).Efficient resource
discovery is the first step towards the realization of
unstructured P2P network. In this work, we present a detailed
overview of recent and existing search methods for
unstructured Peer-to-Peer networks. Most unstructured P2P
networks utilize flooding based resource discovery which
assumes no knowledge about the network topology, and thus
can lead to number of redundant message generation. Random
Walk (RW) search generates a fixed amount of query messages
at each hop but would take longer time to search. Researchers
have presented number of alternative search algorithms which
take advantage of the context under which Flooding and RW
performs well. We propose a searching algorithm, which will
first construct a minimum spanning tree like overlay network
over the physical network. The search algorithm will resemble
as flooding for some small number of hops on the original
physical network, after the initial flooding for certain hops it
will behave like a random walk on the constructed overlay
network and forward the query to limited set of neighbor
depending on the knowledge gained from previous search
attempts.Problem occurring due to the peers getting
disconnected is also handled effectively.
Keywords-component; Peer-to-Peer network, Network
Topology,
Searching Technique.
I. INTRODUCTION
In the past few years, emergence of P2P applications such
as Napster, Gnutella, and KaZaA have shown their good
support for content lookup and sharing, so their popularity and
number have greatly exploded. A P2P system usually involves
thousands or millions of nodes and resources in the network.
The P2P system forms a fully-distributed, cooperative network
design, where nodes do not require any supervision. A good
search scheme has to arrive at two goals: (1) high search
performance, which tries to deliver high quality service (e.g.,
large number of search results and low query response time) to
users, and (2) low cost, which reserves system resources to
sustain a large number of users. Early systems, such as Napster,
use a central server to store indices of participating peers. This
centralized design and practice give rise to the concerns of
performance bottleneck and single point of failure. Instead of
maintaining a huge index in a centralized system, a
decentralized system such as Gnutella distributes all searching
and locating loads across all the participating peers. Though the
decentralized approach addresses the overloading and
reliability issues, and is promising to build a highly scalable
P2P system, its success is heavily dependent on an efficient
searching strategy employed to locate the required resource.
Reaching out to a large scope of peers is a fundamental
procedure in an unstructured P2P network, because there are no
controls and no accurate information on network topologies
and locations of desired files. Thus, the system scalability is
directly affected by the efficiency of the broadcasting
mechanism.
Search algorithms in unstructured P2P networks can be
classified into two categories: breath first search (BFS) method,
and depth first search (DFS) method. These two types of search
algorithms tend to be inefficient, as they have the disadvantage
of generating too much load on the system [1], [2], or not
meeting users requirements of minimum search time and
performance [3].Flooding, which belongs to BFS-based
methods, is the default search algorithm for Gnutella network
[4], [5].In a blind search, nodes do not store any information
regarding file locations. In informed approaches, nodes locally
store metadata that assist in the search for the queried objects.
The shortcoming of most informed methods is the maintenance
cost of the indices after peers join/leave the network or update
their collections. In most cases, these events lead to generation
of update messages resulting in network traffic. Current P2P
search algorithms can be roughly divided in two categories:
The ones that exhibit good performance at high cost (either
during the search or the metadata updates), and those that are
bandwidth efficient but exhibit varying performance. The
important observation is that popular P2P networks display a
highly dynamic behavior, with most users connecting for small
periods of time and then leaving the system.

II. RELATED WORK
To address the searching problem in unstructured P2P
network many researcher and practitioner have proposed a lot
of methods. Today, the most popular P2P applications operate
on unstructured networks. In these networks, peers
joining/leaving a network is dynamic, the location of the
documents is not controlled by the system and no guarantees
for the success or the complexity of a search are offered.
_____________________________________________
INPUT:
_____________________________________________
N(u) is the 1 hop neighbor set of node u. It is defined
as:
N(u) ={v / (u, v) E }.
E->set of edges in graph.
n->number of nodes/peers.
_____________________________________________
Algorithm:
_____________________________________________
1. Each node u collects its 1 hop N(u) neighbor
information.
2. Each node u compute the MST(N(u)) of all nodes
N(u) including u itself.
3. For each edge (u,v) MST(N(u)) , node u tells
node v about this edge selected for MST.
4.Node u keeps an edge (u, v) if
(u,v) MST(N(u)) or (u,v) MST(N(v)) .
5.
n
i
i=1
MST(G) MST(N(u )).
Gives complete spanning

tree.

Flooding is an example of blind search, a message received by
a peer will be forwarded to all its neighbor except the node
from which the message arrived, but flooding would produce
redundant messages which will affect bandwidth utilization
adversely .Researcher have modified the flooding method to
achieve the advantages of flooding while reducing the
redundant messages greatly. Unlike pure flooding searching
techniques like Expanding ring [6] (or iterative deepening[7]),
Random periodical flooding[8] takes more dynamic factors into
consideration to reduce the flooding horizon while maintaining
search quality. Searches are
initiated with increasing TTLs until enough responses are
received. Modified BFS [7], Directed BFS [9] sends query to
neighbor satisfying certain specific heuristics. These are some
examples of modified flooding techniques.
To eliminate the use of flooding and improve search
efficiency some researcher have provided certain Interest-based
locality [4],[5],[14] states that if peer A has a particular piece of
content that peer B is interested in, then it is likely that A will
have other pieces of content that B is also interested in. Based
on this assumption, a peer selects others that have previously
provided results to its queries as peers with common interests
and creates
Shortcut links to these peers.
Super-peer based search strategy are provided in research
work like [21].A super-peer is a peer which acts as a
centralized server for some subset of its clients. Many such
Super-peers are connected to each other and forward the query
among them if they dont find a match within their own client
peers.
Some researchers have presented methods in which the peer
will learn about the location of resource in query through
knowledge updating from previous search results. A number of
search methods depending on the knowledge learned are:
Adaptive probabilistic search (APS) [10],[11], Biased RW
[12],Routing index[13],Local indices[14],Intelligent search[15]
Redundant message generation due to the loops in the
network will still be a problem in most of the above mentioned
search strategy. So construction of tree like overlay network
will be beneficial for reducing the redundant messages due to
the loops in the network [17], [18], [19].
Our scheme will construct an location aware minimum
spanning tree like overlay network and then apply the searching
algorithm on this overlay network. Tree like structure is created
by classifying the neighbor of every peer into a set of
FLOODING neighbor and NONFLOODING neighbor.
FLOODING neighbor are those to whom the peer can send the
message, while NONFLOODING neighbor are those to whom
message cannot be sent in MST. Unstructured P2P network are
so dynamic that a peer in the route of search may get
disconnected making the entire downstream node unreachable.
In this scenario our search will detect the node getting
disconnected and inform all 1 hop neighbor of the peer. The 1
hop neighbor of the disconnected peer will then get connected
to one of the NONFLOODING neighbor.

III. PROPOSED SEARCHING ALGORITHM
In this section, we provide the details of the proposed
search algorithm. Section A presents the operations of
Construction of Tree-like Overlay Network, section B presents
the operations of knowledge-based searching algorithm, and
section C present the operation on peer getting disconnected.
A. Construction of Tree-like Overlay
Network
Location-aware topology construction explores the
topology mismatch problem between the overlay network and
the underlying physical network by constructing an overlay
minimum spanning tree (MST) among each peer and the peers
within 1 hop distance from them. Proposed algorithm does not
need to know the information of the entire network. It just
needs to know the peers between 1 hop distance and the cost of
routing between them. We will construct the MST for every
peer and its neighbor eliminating the redundant loops between
the networks. Each peer will query its neighbor and construct
the
Neighbor cost table which will be required for MST
construction. Once the MST is constructed the query message
will be forwarded within the MST instead of forwarding the
message to its entire neighbor. During the construction of MST
a set of FLOODING and NONFLOODING neighbor is
maintained for each peer who will direct the peer to forward the
query message in MST. Construction of MST is done using a
general algorithm i.e. PRIM or KRUSKAL.
Consider a random graph G(V,E) with n nodes and e edges,
that has average degree d where V is set of nodes, E is set of
e
d
F
a
t
r
o
i
f
r
T
a
f
T
k
f
K
a
r
__________
INPUT:
__________
S->Queryin
r->queried r
TTL->Time
M->queryin
t->threshold
original net
__________
Algorithm:
1.SearchAlg
2. If (Pee
information
3. S returns
4. Elseif (h>
Stop forw
5. elseif (h<
h++
S forward
network.
6.elseif (h>
h++
Node ha
chosen to fo
edges. For a gi
distance from u

Algorithm

B.
Overlay N
Searching
Flooding and
advantage of c
the nodes with
redundant mes
of at least 95 p
initial hops b
forward the q
redundant mes
TTL value, on
amount of peer
forwarding the
TTL value can

Algorithm
Combining
knowledge-bas
further improv
Knowledge-ba
about whether
result in a hit/
____________
____________
ng node/peer.
resource.
e-to-live param
ng message.
d selected for
twork.
____________
gorithm(S, r, T
er S has
n about the loc
s a hit, Stop for
>TTL)
warding messa
<=t)
ds the query t
t)
aving height p
orward the me
iven node u, le
u.
for constructio
Knowledg
Network.
algorithm wi
Random walk
coverage i.e. q
hin a small num
ssages. Usually
percent. Cover
but later it d
query to one o
ssages initially
nce the messa
r, each individ
e message in th
n be done by pe
for Searching
g the search
sed search suc
ve the perform
ased search alg
forwarding qu
/success. Each
___________
___________
meter.
r forwarding t
___________
TTL, t)
the required
cation of resou
rwarding the m
age M.
to its entire ne
probability of
essage M to.
et N(u) be the s
on of MST:
ge-base Search
ill use the co
k perform wel
query message
mber of hops, b
y a TTL of 7 c
rage growth rat
degrades [19].R
of its neighbo
y flood the m
age are availa
dual peer will d
he MST. Selec
erforming expe
in Unstructure
h strategy d
ch as SPUN
mance of the s
gorithm will b
uery message to
peer will mai
____________
____________
the message M
____________
d information
urce )
message M.
eighbor on orig
f success wil
set of nodes at
hing Algorithm
ontext under
ll. Flooding h
e reaches to m
but generates a
can achieve cov
te is high for c
Random Walk
or reducing a
messages with
able to consid
do a random w
cting the appro
eriments.
ed P2P network
described with
[16], APS [10
searching algo
build the know
o particular pee
intain a probab
____
____
M in
____
n or
ginal
ll be
_
A
1
i
u
2
N
1 hop
m on
which
as the
most of
a lot of
verage
certain
k will
lot of
small
derable
walk by
opriate
k:
h the
0] can
orithm.
wledge
er will
bilistic
tabl
prob
pee
Var
fun
to
upd
P2P
mes
whi
the
and
So
Onl
mai
algo
pee
num
grow
netw
sear
dow
sch
leav
NO
inta
get
mai
nod
Nod
Sup
leav
this
NO
____________
Algorithm:
1. Every node
informing its
unavailability.
2. For each pe
Peer gets c
NONFLOODI
le for each of i
bability of hit
er is done whe
rious update f
ction, a linear
update the kn
date function ha

C.
in constructed
If we build a
P overlay netw
ssages. Howev
ich will be add
MST in a para
d its entire 1 ho
there is no ne
ly local inform
intaining the M
orithm will flo
er network and
mber of peer it
wth rate is not
Finally due t
work peers le
rch route get
wnstream node
eme the node
ving. The neig
ONFLOODING
act. Peers joini
connected to
intaining the M
Node S will co
de E, F in its F
de G will conta
ppose node S o
ves the network
s case node G w
ONFLOONG lis

Reconstruction
____________
e u getting di
s 1 hop
.
eer/node in nei
connected to
ING set of nei
its neighbor an
. Updating of
en a hit/miss
functions like
function and a
nowledge tabl
as been evalua
Addressing t
d MST.
spanning tree
work, we can
ver, the spanni
dressed in our s
allel fashion w
op distance nei
eed to obtain th
mation will b
MST is therefor
oods the query
d when the me
t does a rando
a problem in o
to the dynami
ave/join at an
disconnected
e unavailable u
e informs its e
hboring peers
G neighbor thus
ing the network
o any one of
MST.
ontain node G
FLOODING li
ain node S and
originates a que
k making node
will get connec
st i.e. S or F
F
n of MST after
____________
isconnected se
neighbor N(
ighbor set N(u
one of its
ighbor.
nd for each res
the knowledg
is returned fr
Flat update
hop count fun
le. The perfor
ated in APS [10
the Problem of
that connects a
n eliminate a
ing tree has so
scheme. As we
where we consi
ighbor for each
he whole netw
be sufficient. C
re not a problem
y message in
essage has reac
om walk on M
our scheme.
ic nature of u
ny time. If a n
d which will
until the tree
entire neighbo
will get conne
s maintaining t
k will not be a
the peer of t
in its NONFLO
ist as shown in
d F in its NON
ery message m
e G unavailable
cted to one of
Fig. 1

r a node gets di
____________
ends a messag
(u) about i
u)
neighbor fro
ource to infer
ge table for ev
rom its neighb
function, a s
nction can be us
rmance of th
0].
f Peers leaving
all the peers on
ll the redund
ome disadvanta
e are constructi
ider only the p
h and every pe
work informati
Constructing a
m. Initially as
original peer-
ched to suffici
MST. So covera
unstructured P
node gets on
make the ent
is rebuilt. In o
oring peer bef
ected to one of
he MST structu
a problem as th
the network s
OODING list a
n fig 1. Simila
NFLOODING l
meanwhile node
e for querying.
the node from
isconnected:
_
ge
its
om
the
ery
bor.
step
sed
ese
g
n a
dant
age
ing
peer
eer.
on.
and
the
-to-
ient
age
P2P
the
tire
our
fore
f its
ure
hey
still
and
arly
list.
e E
. In
m its

Let S be the system represents Peer-to-Peer network.
i i S={G, A, F, s,sh, N(u),FL,NFL}
Let G(V,E) be the graph representing the network
topology.
1, 2, 3..... m let E={ } where m=number of edges in the netw e e e e
1 2 , , 3..... n let V={ } where n=number of vetex/peers in the v v v v
1 2 r Let F={f ,f ,f } where r=number of resources.
Let A be the adjacency matrix representation of the
network/graph.
Aij=weight if(i,j) E
= 0 otherwise.
Where weight is the network delay between any two
peers i and j.
N(u) is the 1 hop neighbor set of node u. It is defined as:
N(u)={v / (u,v) E }
Set of neighbor of vertex u within 1 hops.
For each vertex i and neighbor within distance of 1
hops construct MST.
1 ij i F (i,A ) MST(T).
Where Ti is tree rooted at vertex i.
1 Function F takes the input as vertex i and adjancey matrix
rooted at vertex 'i'.
i = 1 2..... f FL {v v v } .
i FL is set of flooding vertex i.e. to which vertex 'i' is
connected
i 1 2 nf NFL = {V ,V ....V }.
i NFL is set of non flooding vertex i.e. to which vertex 'i' is
not directly connected
1 2 n MST(G)=MST(T )UMST(T ).........UMST(T )
Let s=peer querying for some resource in P2P network
f= queried resource.
M=query message.
3 F (s,f,M) Forwards the message to its neighbor.
4 F (s,f,M) Forwards the message to some of its neighbor
over MST.
Function 4 F forwards the query message to one of its

neighbor depending on the probability table that it will
return a hit/success.
h=number of hops.
sh=Selected Hop count.
3 ( Dynamicsearch=F S,F,m) if h<=sh
= 4( , , ) F S F m if h>sh
IV. MATHEMATICAL MODE FOR THE SYSTEM V. CONCLUSION
In this paper, we have proposed the reconstruction of
overlay topology during the event of peer getting disconnected
from the P2P network. Topology construction algorithm
presented uses only the local information available hence
improving the performance. The search algorithm is a
combination of Flooding and Random Walk strategy and uses
the context under which they perform well. Selecting the
threshold value for flooding network will greatly impact the
performance of the system. The search algorithm is further
combined with the knowledge learned from the previous search
attempts thus improving the probability of query returning a hit
by directing the query in right direction. The problem of node
getting disconnected and making all downstream nodes
unavailable in MST is also addressed in the scheme.
REFERENCES
[1] K. Sripanidkulchai, the Popularity of Gnutella Queries and Its
Implications on Scalability, white paper, Carnegie Mellon Univ.Feb.
2001.
[2] M. Jovanovic, F. Annexstein, and K. Berman, Scalability Issues in
Large Peer-to-Peer Networks: A Case Study of Gnutella, technical
report, Laboratory for Networks and Applied Graph Theory, Univ. of
Cincinnati, 2001.
[3] B. Yang and H. Garcia-Molina, Improving Search in Peer-to-Peer
Networks, Proc. 22nd Intl Conf Distributed Computing Systems
(ICDCS 02), pp. 514, July 2002.
[4] G. Kan, Gnutella, Peer-to-Peer Harnessing the Power of
Disruptive Technologies, OReilly, pp. 94122, 2001.
[5] RFC-Gnutella 0.6, http://rfc-gnutella.sourceforge.net/developer/
testing/index.html, 2008.
[6] V. Kalogeraki, D. Gunopulos, and D. ZeinalipourYazti, A Local
Search Mechanism for Peer-to-Peer Networks, Proc.ACM CIKM
Intl Conf. Information and Knowledge Management (CIKM 02),
pp. 300307,Nov. 2002.
Networks, Proc. 22nd Intl Conf.Distributed Computing Systems
(ICDCS 02), pp. 514,July 2002.
[8] Z. Zhuang, Y. Liu, L. Xiao, and L.M. Ni, Hybrid Periodical
Flooding in Unstructured Peer-to-Peer Networks, Proc. 32nd Intl
Conf. Parallel Processing (ICPP 03), pp. 171-178, Oct. 2003.
Search Mechanism for Peer-to-Peer Networks, Proc.ACM CIKM
Intl Conf. Information and Knowledge Management (CIKM 02),
pp. 300307,Nov. 2002.
[10] D. Tsoumakos and N. Roussopoulos, Adaptive Probabilistic Search
for Peer-to-Peer Networks,Proc. Third Intl Conf. Peer-to-Peer
Computing (P2P03), pp. 102-109, Sept. 2003.
[11] D. Tsoumakos and N. Roussopoulos, Analysis and Comparison of
P2P Search Methods, Technical Report CS-TR-4539, UMIACSTR-
2003-107, Dept.of Computer Science, Univ. of Maryland, 2003.
[12] Y. Chawathe, S. Ratnasamy, L. Breslau, N. Lanham, and S. Shenker,
Making Gnutella-Like P2P Systems Scalable, Proc.ACM
SIGCOMM 03, pp. 407-418, Aug. 2003.
[13] A. Crespo and H. Garcia-Molina, Routing Indices for Peer-to-Peer
Systems, Proc. 22nd Intl Conf.Distributed Computing Systems
(ICDCS 02), pp.23-32, July 2002.
Networks, Proc. 22nd Intl Conf. Distributed Computing Systems
(ICDCS 02), pp. 514,July 2002
Search Mechanism for Peer-to-Peer Networks, Proc. 11
th
Intl Conf.
[16] Information and Knowledge Management (CIKM 02), pp. 300-307,
Nov. 2002.
[17] D.M Rasanjalee Himali and Sushil K. Prasad, SPUN: A P2P
Probabilistic Search Algorithm based on Successful Paths in
Unstructured Networks. 2011 IEEE International Parallel &
Distributed Processing Symposium, DOI 10.1109/IPDPS.2011.316.
[18] Yunhao Liu, Member, IEEE, Li Xiao, Member, IEEE, Xiaomei Liu,
Location awareness in unstructured peer-to-peer systems, IEEE
Transactions on Parallel and Distributed Systems, VOL. 16, NO. 2,
pp.163 174, February 2005.
[19] Qiuyan Huo, Jingyu Chen, Xuezhou Xu, Yachuan Zhou, Xin Liu A
Location-aware Efficient Contentbased Searching over Unstructured
P2P Network, IEEE International Conference on Network Computing
and Information Security,pp.183 - 187, May 2011.
[20] S. Jiang, L. Guo, X. Zhang, and H. Wang, LightFlood: Minimizing
Redundant Messages and Maximizing Scope of Peer-to-Peer .
Search, IEEE Trans. Parallel and Distributed Systems, vol. 19,no. 5,
pp. 601-614, May 2008
[21] KaZaA Homepage, http://www.kazaa.com .
[22] B. Yang, H. Garcia-Molina, Designing a Super-peer Network,
Proceedings of the 19th International Conference on Data Engineering
,March 2003

ISBN: 978-81-920575-7-6 :: doi: 10. 73921/ISBN_0768
Mobile Disaster Management System Using Android
Technology

Miss. Deepti B. Shelar
Lecturer in Shah and Anchor
Kutchhi Engg College
Mumbai - 88, India

Abstract In todays scenario use of smartphones is
gaining interest in people, the disaster management system was
implemented as a smart phone application using Googles
Android operating system. The disaster management system
Android application determines the optimum route along
different geographical locations that the volunteers and
rescuers need to take in order to serve the most number of
people and provide maximum coverage of the area in the
shortest possible time. Genetic algorithm was applied for
optimization and different parameters were varied to determine
the most optimum route.
KeywordsAndroid, Disaster management, Genetic
algorithm, Travelling salesman problem, Smart phone
Application, Wireless mobile technology

I. INTRODUCTION
Many countries and territories are at risk of natural
hazards due to earthquakes, floods, tropical cyclones, and
landslides. Due to this mortality rate index (MRI) is
increasing. It has claimed the lives of thousands of
people and damaged billions worth of property. Though
we cannot stop these natural disasters but we can
definitely find solution to overcome it faster. Even with
the recurrent calamities, there is still the absence of an
efficient disaster management system that will help the
people in times of calamities.
II. DISASTER MANAGEMENT
The process of disaster management involves four
phases: mitigation, preparedness, response, and recovery.
attempt to reduce disaster risks is called as
mitigation. development of an action plan for an
upcoming disaster comes under preparedness. The
response phase consist of the mobilization of services
and relief when disaster strikes and the recovery phase is
the restoration of the affected area to its previous state.
A.USE OF SMART PHONES:
A smartphone is a high-end mobile phone that
combines the functions of a personal digital assistant
(PDA) and a mobile phone. Today's models typically also
serve as portable media players and camera phones with
high-resolution touch screens, web browsers that can
access, and properly display, standard web pages rather
than just mobile-optimized sites, GPS navigation, Wi-Fi
and mobile broadband access. The term smartphone is
usually used to describe phones with more advanced
computing ability and connectivity than a contemporary
feature phone, although the distinction can be vague and
there is no official definition for what constitutes the
difference between them. The definitions also shift over
time since many phones that are considered feature phones
today can have capabilities that exceed those of phones that
had been promoted as smartphones in the past.
B.ANDROID MOBILE DEVELOPMENT ENVIRONMENT:
Android is a software platform and operating system for
mobile devices Based on the Linux kernel. It is developed
by Google and later the Open Handset Alliance (OHA).
Allows writing managed code in the Java language.
Android is not a single piece of hardware; it's a complete,
end-to-end software platform that can be adapted to work
on any number of hardware configurations. Everything is
there, from the boot loader all the way up to the
applications. Android uses Linux for its device drivers,
memory management, process management, and
networking. It supports wireless communications using:
GSM mobile-phone technology, 3G , 802.11 Wi-Fi
networks.
Android is a multi-process system, in which each
application (and parts of the system) runs in its own
process. Most security between applications and the
system is enforced at the process level through standard
Linux facilities, such as user and group IDs that are
assigned to applications. Additional finer-grained
features ar
mechanism th
operations tha
per-URI perm
cific pieces of
GENETIC A
algorithm (G
rocess of natur
sed to gen
and search p
e larger class
generate soluti
ques inspired b
mutation, selec
intains a popu
blem at hand
plying a set of
tic algorithm,
s or the geno
date solutions
s) to an optimi
ns. Traditiona
ngs of 0s and
evolution usu
generated in
In each gen
n the popul
re stochastica
(based on t
and possibly
on. The new
of the algori
hen either a m
duced, or a sat
he population.
ximum numbe
y or may not
find appli
l science, engi
g, mathematic
IV. SYSTEM
e travelling sa
ic algorithms
ed disaster
Droid (MDD
MyDisasterDro
nd ListView.
s based on Go
of the people i
re provided
hat enforces
at a particula
missions for g
data.
ALGORITHM
GA) is a searc
ral evolution.
erate useful
problems. Gen
s of evolution
ions to optimi
by natural ev
ction, and cros
ulation of can
d and makes
f stochastic ope
a population
otype of the
s (called indiv
ization problem
lly, solutions
1s, but other e
ually starts fro
dividuals an
neration, the
lation is ev
ally selected
their fitness)
randomly mu
population is
thm. Common
maximum numb
tisfactory fitn
If the algorith
er of generatio
t have been
ication in
ineering, econ
s, physics and
M IMPLEMENT
alesman probl
s to generate
management
) was imple
id, it show
MapView sho
ogle Maps wh
in need, their c
through
restrictions o
ar process ca
granting ad-ho
ch heuristic th
This heuristic
solutions
netic algorithm
nary algorithm
ization problem
volution, such
ssover. A gene
ndidate solutio
s it evolve
erators.
of strings (ca
genome), w
viduals, creatu
m, evolves tow
are represente
encodings are
om a populatio
nd happens
fitness of e
aluated, mul
from the cu
), and mod
utated) to for
s then used in
nly, the algor
mber of generat
ess level has
hm has termin
ons, a satisfac
reached. Gen
bioinforma
nomics, chemi
d other fields
TATION
lem as basis a
a solution,
system nam
emented. Up
ws two view
ows the locati
hile the ListVi
corresponding

a
n
an
oc
hat
c is
to
ms
ms
ms
as
etic
ons
by
alled
which
tures,
ward
ed in
also
on of
in
every
ltiple
urrent
dified
rm a
n the
rithm
tions
been
nated
ctory
netic
atics,
istry,
and
an
med
pon
ws:
ion
iew
g

Figure1: BBlock Diagram o
Algorithm
Implementatio
of the Genetic
on

Figure
Figure 3: M

e 4: MyDisasterD
MyDisasterDroid
Droid ListView
d MapView
w

g
p
c
tr
o
c
s
fo
o
m
c
a
ti
p
in
n
u

V.
(TSP

Determinin
geographic loc
problem wher
coordinates an
ravelling sales

The travell
Given a finite
of travel betwe
cheapest) way
tarting point
formally descr
objective of fin
minimum cost
cities or nodes
at one node, v
ime each, and
problem with t
ncrease on th
number of citi
using optimiza
. TRAVELLIN
P)
ng the most op
cations is simi
rein geograph
nd the rescuer
sman.
ling salesman
e number of ci
een each pair
of visiting all
. The trave
ribed as a pe
nding the path
t) on an und
to be visited.
visits all other
d finally retur
the travelling
he number o
ies increase.
ation technique
NG SALESM
ptimum route
lar to the trave
hic locations
s or volunteer
problem is st
ities and the d
of them, find
l the cities and
elling salesm
ermutation pr
h of the shorte
directed graph
The travelling
r nodes succes
rns to the sta
salesman prob
of possible ro
However, this
es.
MEN PROBL
along differen
elling salesma
represent cit
rs represent th
tated as follow
distance (or co
d the shortest
d returning to t
man problem
roblem with t
est length (or t
h that represen
g salesman sta
ssively only o
arting node. T
blem is the rap
outes when t
s can be solv

EM
nt
an
ty
he
ws:
ost)
(or
the
is
the
the
nts
arts
one
The
pid
the
ved

VI. CONCLUSION
In this paper we tried to implement rescue and
disaster management operation using
MyDisasterDroid was implemented in an Android-
based mobile phone. Using genetic algorithm, an
optimum route along the given geographic locations
was determined.

REFERENCES

[1] United Nations International Strategy for
DisasterReduction Secratariat (UNISDR), Mortality
RiskIndex, 15 June 2009
[2] Asia Pacific Telecom Research
Ltd.,Telecommunications in the Philippines, 1 June 2009
[3] Damon P. Coppola, Introduction to International
Disaster Management, Elsevier Inc., 2007.
[4] Bartel Van de Walle, Gerd Van Den Eede, and Willem
Muhren, Humanitarian Information Management and Systems,
Mobile Response: Second International Workshop on Mobile
Information Technology for Emergency Response 2008,
Bonn,Germany, May 29-38, 2008, Revised Selected Papers,
Springer, 2009

ISBN: 978-81-920575-7-6 :: doi: 10. 73928/ISBN_0768
Handover Techniques For 4G Networks

Piyali Dutta
PG student, Computer Science Department
Datta Meghe College of Engineering. Airoli
Navi Mumbai, India

Abstract The next generation 4G networks are represented
by heterogeneous environment with different access network
technologies that vary in bandwidth, latency and cost. 4G
networks are known for their seamless connectivity between
existing networks like GSM, wireless LAN, Bluetooth and
others. Seamless Connectivity in such networks entirely
depends on efficient handover mechanisms and multiple
mobility management mechanisms. In this paper, we discuss
different challenges facing 4G mobility. We propose an user
zoning handover technique based on the IEEE802.21
protocol, where the approach connects users to a preset
network upon entering a zone, guided by the MIIS server. We
also discuss on vertical handover for real-time video traffic.
Keywords- handover; 4G mobility; user zoning;
IEEE802.21; vertical handover.
I. INTRODUCTION
Advanced mobile communication systems have mainly
increased due to the needs of the increasing user
requirements. Many organizations such as IEEE,
Electronics and Telecommunications Research Institutes
have been working towards 4G networks to provide
seamless connectivity anywhere and anytime. The 4G
networks which is heterogeneous in nature where there
would be multiple service providers, equipped with varied
technologies offering varied services for the benefit of the
users. 4
th
generation systems are being developed to
support greater bandwidth services, higher data rates,
improved quality and seamless global mobility as
compared to previous systems.
Handover is the transfer of an on-going call or data
session from one cell to another. Handovers occur due to
the movement of the mobile user from one area to another
area. Handovers are used to prevent an on-going call to be
disconnected. If we dont use handovers then whenever a
user leaves the area of a particular cell then its on-going
call is immediately disconnected.
The process of handover requires a number of
parameters like what is the handover scheme we are using
and how many channels are free. In the handover process
we should also keep the QoS up to the standard. In 4G
networks, we find different handover techniques.
Handovers between two base stations (BSs) of the same
system is called Horizontal handover. Horizontal handover
involves a terminal device to change cells within the same
type of network to maintain service continuity. Horizontal
handover between two BS, under same foreign agent (FA)
is known as Link-layer handover. In Intra-system
handover, the horizontal handover occurs between two BSs
that belong to two different FAs and both FAs belong to
the same system and hence to same gateway foreign agent
(GFA). Vertical handover may be referred to a process of
transferring call connected to a network/data session from
one channel connected in a cell to the core network of
another. A suitably equipped device may be able to use
both technologies at a time, the high speed Wireless LAN
and cellular technology. Vertical handover refers to a
network node changing the type of connectivity it uses to
access a supporting infrastructure, usually to support node
mobility.
IEEE802.21 [1] is a standard that was created in order
to facilitate heterogeneous handovers between IEEE 802
and non IEEE802 access networks regardless of the
network features. The IEEE 802.21 standard defines the
MIH (Media Independent Handover) mechanism to enable
the optimization of handover.
In this paper, we present a brief overview 4G general
architecture. Finally, we propose a user zoning handover
technique based on the IEEE 802.21 protocol and vertical
handovers between UMTS and WiMAX networks.
II. CHALLENGES IN MOBILITY MANAGEMENT
There are several challenges and issues regarding
mobility in 4G networks:
A. Location Management
Location management helps to locate the users
instantaneous point of attachment for connection delivery
purposes. Location management is divided into two main
functions:
1) Location Registration: Location registration keeps
the network aware of changes at the point of attachment of
each individual user among the network.
2) Call Delivery : Call delivery is used at time of a
connection request and in order to deliver the connection to
the specified user. The network is queried for the user
location profile so the current user position is retrieved.
B
C
c
a
m
c
t
a
i
s
e
v
v
D
i
d
p
c
E
p
u
h
a
u
c
n
o
c
l
p
b
l
O
a
I
t
p
r
m
M
r
W
A
B. The Choic
This issu
Connected pr
connected to t
architecture o
multiple diffe
connected ove
C. Mobility C
Context is
to characteriz
applications a
information ab
situation. Con
exhibiting te
values about
values.
D. Paging
Paging is s
information o
data sessions.
paging is req
cells can be co
E. Moving Gr
The movi
prominent com
users form a
hierarchies wh
a cluster, wit
unified netwo
cluster in a
network and s
Most of th
on cell forma
cells, responsi
limits. The c
physical resou
being structu
limited to a
Overlapping c
A universa
access request
IP networks
technologies,
provided part
Service Doma
routing and b
model are mai
Management
represents he
WiMAX, GP
All of these
ce Of Technolo
ue is associa
rinciple which
the best availa
of 4G [6] net
erent access t
er.
Context Manag
defined as an
e the situation
and networks
bout themselv
ntext informat
emporal char
past, presen
sending essent
f a user in ord
Due to powe
quired to be a
ombined toget
roup Formatio
ing formation
mponent of fu
a temporary
hile traveling.
thin which u
ork, functions
different beh
sever overhead
he wireless co
ation structure
ible for contac
ells capacity
urces over whi
ured over. Ac
certain numb
coverage is use
III. GENER
al common int
ts from variou
face differe
services and
tly by mobil
ain. The Netw
bit pipe mana
inly related to
(MM) suppo
eterogeneous
RS/UMTS, an
are connecte
ogy
ated with th
h implies that
able connectio
tworks is bas
technology fo
gement
ny information
n of an entity
s will be req
ves, their surro
tion can be s
racteristics, r
nt as well as
tial messages
der to be used
er limitations o
as efficient as
ther to form on
ons
n of a multi-
uture networks
moving clus
Such groups
users are usua
such as hando
havior to resu
d.
mmunication
es. The netwo
cting the mob
y is defined b
ich communic
ccording to t
ber of users t
ed in 4G syste
RAL ARCHITEC
terface is prov
us types of m
ent access s
roaming prot
e devices the
work Domain
agement. Chal
o the QoS, Sec
ort. The archi
networks. It
nd EUTRAN
ed to IP net
he Always
t a user has t
on at all time.
sed on combi
or the user t
n that can be
y, where termi
quired to pro
oundings and
static or dyna
regarding con
s predicted fu
to update loca
d to route calls
of mobile dev
possible. Sev
ne paging area
-user session
s. Several netw
ster and netw
of users are c
ally joined in
over will deal
ult a reductio
systems are b
ork is divided
bile units withi
by the numbe
cation channel
that, each ce
to host at a t
ems.
TURE
vided to satisfy
mobile systems
systems, term
tocol. Services
emselves and
n is responsibl
llenges facing
curity and Mob
itecture in Fi
contains WL
N access netw
twork through
Best
to be
. The
ining
o be
used
inals,
ovide
their
amic,
ntext
future
ation
s and
vices,
veral
a.
is a
work
work
alled
nto a
with
on in
based
into
in its
er of
ls are
ell is
time.
fy the
s. All
minal
s are
d the
le of
g this
bility
ig. 1
LAN,
orks.
h an
ev
ne
ha
me
ha
pr
rad
rep
an
Ot
Ne
ca
lay
ne
inv
se
de
pr
ne
av
wh
pr
fu
(M
thr
se
(M
(M
flo
ev
up
low
de
of
pr
aid
ar
volved packet
etwork. To pe
andover proc
echanisms: ha
andover execu
Figu
The handove
rocedures whe
dio measurem
ports provide
n urgent hand
therwise, it w
ew link disco
an be realized
yers. During
etworks cove
volve a quer
rver. In case
evice must a
roceeding. If
etwork, QoS c
vailability chec
hich was not i
rocedure is ba
unction (MIHF
MN) and the n
ree types of
rvice (MIES)
MICS); and (3
MIIS). Each of
ow direction w
vents such as l
pper network l
wer layer fro
evices and sca
f commands o
roviding the
des the hando
e handled by
core (EPC) w
erform a med
cess is perfo
andover initiat
ution.
ure 1 : A Heterog
er initiation in
ere each device
ment reports
information a
dover if spec
will send peri
very is also p
d with specific
g the hando
erage scannin
ry for inform
of access to
authenticate it
the device
context must
ck. Finally, ha
included by th
ased on the
F). The MIHF
network node p
f services: (1
; (2) Media in
3) Media ind
f the services
with certain r
link deteriorat
layers. MICS t
om higher one
anning for ava
ordered by M
mechanisms
over decision.
MIIS such as
which is an evo
dia independen
formed in th
tion, handover
geneous network a
nvolves old lin
e has to be co
. These mea
about the nee
cific threshol
odic informat
performed in t
c triggers from
over preparat
ng takes pla
mation from
o new network
tself to the
is authenticat
be transferred
andover execu
he IEEE 802.2
media indepe
exists in both
protocol stack
1) Media ind
ndependent co
dependent info
provided by
responsibilitie
tion and link
transfers contr
es. Configura
ailable networ
MICS. MIIS is
to retrieve i
. Static link l
s channel info
olved 3GPP co
nt handover, t
hree stages
r preparation a
architecture
nk configurati
onfigured to se
asurements a
ed of performi
lds are cross
tional messag
this stage whi
m available li
tion, addition
ace which m
an informati
k resources, t
network befo
ted at the n
d for a resour
ution takes pla
21. The handov
endent handov
the mobile no
k. It provides t
dependent ev
ommand serv
ormation serv
the MIHF ha
s. MIES repo
unavailability
rol commands
ation of netwo
rks are examp
s responsible
information th
layer paramet
ormation and t
ore
the
or
and

ion
end
and
ing
ed.
ges.
ich
ink
nal
may
ion
the
fore
new
rce
ace
ver
ver
ode
the
ent
vice
vice
as a
orts
y to
s to
ork
ples
of
hat
ters
the
m
(
q
n
c
p
z
a
t
p
a
t
w
c
a
t
t
a
M
a
t
medium acces
(AP).
IV.
The hando
query from
networks in t
commands to
policies and
zoning approa
access networ
the favor of a
perform the h
apply user zo
target zones h
which the zon
Fig
We sugg
concerning sp
are held, whe
to enter this
traffic over all
aims to optim
MIIS privileg
amount of tra
the three stage
ss control (MA
USER ZONING
over preparatio
the MIIS s
the vicinity o
owards the M
networks prio
ach proposes
rks covering t
a certain netw
handover tow
oning concept
has to be prov
ne is located.
gure 2 : User zoni
est updating
pecific zones
ere a large num
zone generat
l available net
mize the hando
ges, where it w
affic on each a
es that handov
AC) address o
G HANDOVER
on stage incl
server concer
of the MN, ad
MN, in addi
ority order. O
amending the
the surroundin
work. The MN
wards this netw
t over MIH,
vided by the M
ing handover proc
g the MIIS
such as stadi
mber of audie
ting a huge a
tworks. The us
over procedure
will be capabl
available netw
ver procedure
of the access p
APPROACH
ludes: informa
rning the ac
dditional scan
ition to char
Our proposed
e priority orde
ngs of the M
N will be force
work. In orde
information a
MIIS server u
cedure flow.
server data
iums when ga
ences are expe
amount of ur
ser zoning con
e by extending
e of balancing
work. Fig. 2 sh
has to go thro
point
ation
ccess
nning
rging
user
er of
MN in
ed to
er to
about
under

abase
ames
ected
rgent
ncept
g the
g the
hows
ough:
ini
pr
A.
tow
thr
wh
co
me
(M
thr
de
Up
de
Po
a.
in
me
de
oth
the
ch
W
rea
to
(M
de
do
pr
pe
pr
b.A
req
vic
inc
pr
inc
c.T
ind
ne
fro
ne
tra
the
req
su
ch
un
req
(G
es
itiation, prep
rocess, importa
The user zon
The mobile
wards the n
resholds is u
hich is in o
onnection is e
essage to
MIH_configure
resholds for t
evice generate
pon receiving
evice has to se
oS (MIH_conf
After that,
formation
easurements.
evice status at
her significan
e wireless d
hanges to the
When deteriora
ach unaccepta
the WiMAX
MIH_link_goin
etails about th
own and the re
recision. Now
erforming the
reparation stag
At this stage,
questing info
cinity of the
cludes inform
resent around
cludes: netwo
The MIIS serv
dicating forcin
etwork instead
om. This ste
etwork availab
affic managem
e handover
quired to perf
urrounding. Th
heck on candid
nit will skip a
quired measu
GPRS in ou
tablishing a co
aration, and
ant changes in
ning handover
unit is pre-co
network. The
usually set by
our case the
established, th
be sent
e_link_REQ).
the connection
es a report an
g the configu
end a confirma
figure_link_CN
the device
messages
The WiMAX
t all time. As
nt parameters i
evice will ge
e PoS (MIH_
ation in conn
able levels, the
X PoS indicat
ng_down_IND
he time perio
eason of this
the WiMAX
handover an
ge starts.
a query to th
ormation reg
mobile devic
mation about
d the wireles
rk status, avai
ver will send a
ng the device
d of suggesting
p aims to o
bility, cost, se
ment and will
process. The
form additiona
here will be no
date networks
all regular rep
urements reg
ur example).
onnection with
execution. I
n the link quali
r procedure wo
onfigured to g
pre-configur
y the PoS (P
e WiMAX P
he PoS issues
to the
This mess
n parameters u
nd sends it b
uration messag
ation message
NF).
starts to se
stating sig
X PoS will b
far as QoS re
in case of cros
enerate a rep
_linkparamete
nection quali
e mobile unit
tes the need
D). This me
od at which t
links weaknes
X PoS is aware
nd this is whe
he MIIS serve
garding acces
ce. The MIIS
t different a
ss device. T
ilability, and s
a message to t
to connect to
g multiple net
ptimize servi
eamless mobi
reduce the tim
e wireless un
al scanning for
o need of reso
. In this techn
porting actions
garding the
Then, imm
h the new radi
n the initiati
ity are reporte
orks as follow
generate trigg
ration of the
Point-of-Servi
PoS. When t
a configurati
mobile no
sage states t
under which t
ack to the Po
ge, the wirel
e to the WiMA
end a period
gnal streng
be aware of t
equirements a
ssing threshol
port to indic
ers_report_IND
ty continues
sends a messa
for a handov
essage includ
the link will
ss with a certa
e of the need
en the handov
er might be s
ss networks
server databa
access networ
This informati
signal strength
the WiMAX P
o a certain pre
tworks to choo
ice in terms
ility, and urg
me consumed
nit will not
r networks in t
ource availabil
nique, the mob
s and just rep
target netwo
mediately sta
io network.
ion
ed.
s:
gers
ese
ce)
the
ion
ode
the
the
oS.
ess
AX
dic
gth
the
and
lds,
ate
D).
to
age
ver
des
go
ain
of
ver
ent
in
ase
rks
ion
h.
PoS
eset
ose
of
ent
d in
be
the
lity
bile
port
ork
arts
d.The WiMAX PoS sends an information request (MIH-
Get-Information.REQ) message. MIIS sever receives the
request and then sends back a special message based on
PoA (Point-of-Attachment) location where user zoning is
applied, containing the target network along with its
characteristic. The mobile unit now will only scan for this
specific network rather than scanning for a list of available
networks. MIIS will choose this network based on multiple
aspects such as resource availability, QoS, signal strength
within the zone, capability to maintain active sessions, and
the types of services offered.
F. At this point, the WiMAX PoS will send an order to the
mobile unit to start scanning for the concerned PoA
recommended by the MIIS server. A signal strength
measurement message has to be sent back to the WiMAX
PoS.
G. The PoS sends a confirmation to the MIIS server stating
the readiness of the wireless device to execute handover.
H. The WiMAX PoS communicates with GPRS PoS and
send a (MIH_N2N_HO_Query_Resourse) request and
Reply. The GPRS PoS has to reserve required resources.
Resources have to be guaranteed in both core and target
networks. This is when the execution stage begins and the
WiMAX PoS orders the wireless device to start the
handover by indicating the actions over the old and new
link (MIH_Net_HO_commit.REQ). Old link resources are
not needed anymore so it can be released and an IP
connection is now established over the GPRS network
which has to be reported back to the WiMAX PoS
(MIH_Net_HO_Commit.RSP) then to the GPRS PoS
(MIH_N2N_HO_complete.REQ).
I. Now, the WiMAX has to respond
(MIH_N2N_HO_complete.RSP). Then, WiMAX releases
resources if not already done and data starts flowing
through GPRS Network.
V. VERTICAL HANDOVERS FOR REAL-TIME VIDEO TRAFFIC
BETWEEN UMTS AND WIMAX NETWORKS
Internetworking approach of the converged next
generation multi-access network can make the best use of
the advantages of the different wireless networks and
eliminate their weaknesses. UMTS, WiMAX and WLAN
networks can complement each other in term of
geographical coverage and QoS.
One of the possible scenarios where the above three
heterogeneous networks can be used is defined here. While
the vehicle moves on the road, the wider area is covered
only with UMTS. The metro zones are covered
additionally with WiMAX, while hotspots have also WiFi.
Different access networks may be ranked by the operator
(if they are owned by single one) or by the mobile user, for
given services. As the user in the vehicle approaches the
city, besides UMTS, WiMAX (IEEE 802.16e) becomes
available, so vertical handover between UMTS and
WiMAX can occur. Vertical handover among these
different networks can take place when QoS is poor for the
current wireless access network, or due to service
management either by the network provider or by the user.
For the purpose of analysis of impact of vertical handovers
between UMTS and WiMAX for video traffic, we use the
following three streaming video technologies: HDTV
(High Definition TV), MPEG-4 and H.261. Each of these
applications requires different bandwidth. Each wireless
technology has a limit on the maximum supported user
velocity. Also, user mobility influences the behavior of the
wireless access network and the video stream transmitted
over such network.
The aim of this work is to investigate the reasons for
behaviors like poor vertical handover latency, and to
improve the vertical handovers in the case of streaming
video traffic, and hence to reduce the latency, to reduce the
number of vertical handovers and throughput gap between
UMTS and WiMAX networks.
Vertical handover process can be divided in three
phases: handover decision, radio link transfer and channel
assignment. Handover decision deals with the selection of
the target point of attachment and the time of the handover.
Radio link transfer is the process of developing links to the
new point of attachment. Channel assignment phase
arranges the allocation of channel resources. Because IEEE
802.21 standard supports two types of vertical handovers,
handovers initiated by the network and handovers initiated
by mobile terminals, events related to handovers can be
originated at the MAC (Medium Access Control) or MIHF
(Media Independent Handover Function) layer located in
the node or at the point of attachment to the network.
Figures 3 and 4 represent a vertical handover process
between UMTS and WiMAX networks. When mobile
node (MN) enters the WiMAX cell from UMTS link is
detected upon reception of DL_MAP message from BS.
MAC layer implement synchronization messages
DL_MAP and DCD messages for downlink
synchronization and UL_MAP and UCD messages for
uplink synchronization. After the synchronization in Figure
3 initial ranging and registration occurs in the layer 2
vertical handover process.
4
c
t
d
r
t
r
a
i
a
h
F
W
q
m
m
s
Figure 3: La
Layer 3 ve
4. When a lay
created by the
that MIH trig
discover the I
replies to the
the ND (Nei
receives the R
a new addres
interface, bec
available for
handover proc
Figure 4 : Laye
WiMAX.
Vertical h
quantitatively
measuring th
maximum han
session main
ayer 2 vertical han
W
ertical handove
yer 2 connecti
e MAC and se
ggers a RS (R
IP prefix of th
RS message
ighbor Discov
RA (Router Ad
ss and redirec
cause the Wi
application
cess is comple
er 3 vertical ha
handover proc
measured un
he number o
ndover delays,
ntained over
ndover process be
WiMAX.
er process is d
ion is establis
nt to the MIH
Router Solicit
he new link.
according the
very) module
dvertisement)
cts the flow o
iMAX interfa
traffic. With
eted.
andover process
cess can be
nder various u
of handovers
, and the overa
a specific
etween UMTS an
described in Fi
shed, a Link U
H module [6]. A
tation) messag
The access ro
e rules define
e. When the
message, it b
on to the WiM
ace is consid
this the ver
between UMTS
qualitatively
usage scenario
, the mean
all throughput
mobility pat

nd
igure
UP is
After
ge to
outer
ed by
MN
uilds
MAX
dered
rtical

S and
and
os by
and
t of a
ttern.
Fr
re
be
ha
tim
ne
ini
co
of
Re
de
me
mo
av
de
tec
an
vid
fo
ne
ov
as
uti
Th
en
in
ha
W
fo
fu
co
pr

[1]
[2]
[3]
[4]
[5]
[6]
[7]
requent vertica
sources. So, d
eneficial for
andover is co
me, a handove
eeded. This kin
Handover la
itiation of t
ompletion. Han
f the VHD
eduction of th
elay-sensitive
easured with
obile terminal
vailable access
esirable.
In this pape
chnique base
nalysis on vert
deo streams
cus on vertica
etworks. The
verhead and ti
well as allow
ilization by ba
his approach
ntering a zone
formation ab
andover, small
WiMAX netwo
r the mobile u
urther analysi
onducted to ev
roposed techni
] The IEEE802.
] H. Brendt, T
Wiley, West S
] S. Hussain, Z
Issues in 4G
France 2006.
] 3GPP TS 23
Services and
Evolution: Ar
3GPP Technic
] I.F. Akyildiz
management
IEEE Wireles
pp.1628.
] K. Taniuchi, Y
A. Dutta, D. B
independent
IEEE Commu
] S.J. Yoo, D.
Mechanism fo
al handovers c
decreasing the
better effici
onsidered unn
er back to the
nd of handove
atency is the
the vertical
ndover latenc
(vertical h
he handover l
multimedia
the data rat
ls on the netw
s network with
VI. CO
er, we presen
ed on the IE
tical handover
in heterogene
al handovers b
user zoning
ime consumed
wing the MII
alancing loads
connects use
guided by the
bout target z
ler the throug
rks for differe
users with vid
is, validation
valuate and co
iques.
REFER
.21 Standard, http
Towards 4G Tec
Sussex, England, 2
Z. Hamid and N
G Heterogeneous
3.402 (2007) 3G
d Systems Asp
rchitecture Enha
cal Specification 2
z, J. Xie, S. M
in next-generati
ss Communicatio
Y. Ohba, V. Faja
Baker, M. Yajnik
handover: featur
unications Magazi
Cypher, and N
or Seamless Hand
can cause was
e number of ha
iency of the
ecessary whe
original point
ers should be m
e time duratio
handover p
y is related to
handover dec
latency is ver
sessions. The
te that is tra
work. Vertical
h higher throu
ONCLUSION
nted a user z
EEE802.21 p
s done with di
eous wireless
between UMT
g technique a
d while perfor
S server to op
s over the ava
ers to a prese
e MIIS server
zones. Also
hput gap betw
ent velocities b
deo streams. A
n and simul
onfirm the per
RENCES
p://www.ieee802.o
chnologies: Servi
2008.
. Khattak, Mob
Networks, Int
GPP Technical S
pects; 3GPP Sy
ancements for N
23.402 v 1.1.0, 20
Mohanty, A s
ion all-IP-based
ons Magazine, V
ardo, S. Das, M.
k, D. Famolari,
res, applicability
ine, Vol.47, No.1,
N. Golmie, Pred
dovers in Heterog
stage of netwo
andovers is ve
e resources.
en, after a sh
of attachment
minimized.
on between t
rocess and
o the complex
cision) proce
ry important
e throughput
ansmitted to t
l handover to
ughput is usua
zoning handov
protocol and
ifferent real tim
s scenario, w
TS and WiMA
aims to redu
rming handov
ptimize netwo
ailable networ
et network up
r which provid
in the Verti
ween UMTS a
better is the Q
As a future wo
lations will
rformance of t
org/21
ices with Initiat
bility challenges
terSense 06, N
Specification Gr
ystem Architect
Non-3GPP Acces
007.
survey of mobi
wireless system
Vol.11, No.4, 20
Tauil, Y.-H. Che
IEEE 802.21: me
y, and realizatio
, 2009, pp.11212
dictive Link Trig
geneous Handove
ork
ery
A
hort
t is
the
its
xity
ess.
for
is
the
an
ally
ver
an
me
with
AX
uce
ver,
ork
rks.
pon
des
ical
and
QoS
ork,
be
the
tive,
and
Nice,
oup
ture
ses,
ility
ms,
004,
eng,
edia
on,
20 .
gger
ers,
Wiley Wireless Communications and Mobile Computing, Vol. 9,
No.5, May 2009.
[8] Wikipedia.com;Google.com.


ISBN: 978-81-920575-7-6 :: doi: 10. 73949/ISBN_0768
SMS Length Expansion and Security Approach
an A Mobile Device
Deepak S. Manglani
Lecturer in G.V.Acharya Engg.
College
Abstract Ever since their invention, mobile phones have
revolutionized the regular communication methods. Theyve
shrunk the world and have brought all of us closer.
Exchanging a texted conversation between a Tx and a Rx via
mobile phone turned out to be a very successful service.
Such a communication was termed as Short Message
Service, commonly abbreviated and known as SMS. The
flipside to the SMS is the limitation to the maximum number
of characters, i.e., 160. Every 161
st
character happens to be a
new SMS. This increases cost and complexity. Here, in this
paper, weve proposed ways to compress and encrypt the
SMS communication. We believe this holds substantial
content in bringing about a new social impact to the
development of larger content in SMS messaging.
Keywords- Tx Transmitter, Rx Receiver, SMS Short
Message Service, GSM - Global System for Mobile
communications, TDMA Time Division Multiple Access,
GPRS General Packet Radio System, CDMA Code Division
Multiple Access, DES Data Encryption Standard, IDEA -
International Data Encryption Algorithm, AES Advanced
Encryption Standard.

I. INTRODUCTION
In Mobile Phone Communication, SMS is the
tremendous service, with 10 billion messages sent around
the world daily. SMS makes it possible to send and
receive short text messages to and from mobile phones.
SMS became a popular communication tools by
individuals and business. SMS are used in day-to-day
applications like banking, mobile payment, recharge and
mobile commerce, e-ticket, money transfer, etc. [1] In
1985, Friedhelm Hillebrand of Bonn, Germany and a
dozen others had been laying out the plans to standardize
a technology that would allow cellphones to transmit and
display text messages. Because of tight bandwidth
constraints of the wireless networks at the time, each
message would have to be as short as possible. [2]
The maximum number of characters for a single
normal SMS is only 160 characters. The SMS security is
also very crucial to protect the sensitive information.
People sometimes exchange confidential information such
as passwords or sensitive data. SMS technology suffers
from risks such as security vulnerabilities and
unauthorized access. [3]
Current standard SMS technology is non-secure. It can
be intercepted along the transmission path at multiple
locations (aggregators, carriers, infrastructure providers,
or tower operators) especially when using Internet to
submit your messages.
Our approach to this problem in this paper is to
propose a method that can be used to compress and to
encrypt SMS messages that are about to be sent. In our
approach, the physical underlying GSM architecture
remains unchanged.
This paper is structured as follows: Section II
highlights GSM System Architecture. Section III provides
SMS Structure. Section IV explains compression
technique in SMS and Section V shows a method secure
the SMS. Conclusion is presented in Section VI.
II. GSM SYSTEM ARCHITECTURE
In GSM system, the mobile handset is called Mobile
Station (MS). A cell is formed by the coverage area of a
Base Transceiver Station (BTS) which serves the MS in
its coverage area. Several BTS together are controlled by
one Base Station Controller (BSC). The BTS and BSC
together form Base Station Subsystem (BSS). The
combined traffic of the mobile stations in their respective
cells is routed through a switch called Mobile Switching
Center (MSC). Connection originating or terminating
from external telephone (PSTN) are handled by a
dedicated gateway Gateway Mobile Switching Center
(GMSC). The architecture of a GSM system is shown in
figure 1.
In addition to the above entities several databases are
used for the purpose of call control and network
management. These databases are Home Location
Register (HLR), Visitor Location Register (VLR), the
Authentication Center (AUC), and Equipment Identity
Register (EIR).

(
c
w
w
s
d
a
e
e
e
a
E
m
n
t
w
o
`
a
m
r
s
Home Loc
(such as user
current locatio
with the netw
who are bein
stored in HLR
data like loc
authentication
encryption. T
equipments an
equipments.
All the m
assigned uniq
Equipment I
manufacturer
number is sto
the IMSI (I
which is store
of the user. A
SIM is inserte
``real'' telepho
and is stored i

SMS is a
mobile comm
received simu
SMS exists a
such as GSM,
The SMS P
Figure
cation Registe
r profile) as
on) informatio
work. A VLR
ng serviced c
R for faster ac
cation of the
n information o
The EIR sto
nd can be used
mobile equip
que id called
dentity) and
and registered
ored in the EIR
nternational
ed in the Subs
A mobile statio
ed into an equ
one number is
in SIM. [4].
III. SM
communicatio
munication sy
ultaneously w
as a subset o
, TDMA, GPR
Packets in des
1.
er (HLR) store
well as tem
on about all th
stores the dat
currently. It in
ccess as well
e user. The
of the user suc
ores stores
d to prevent ca
pments in G
d IMSI (Inter
is allocated
d by the servi
R. The users
Module Sub
scriber Identit
on can be use
uipment with
s different fro
MS STRUCTURE

on service sta
ystems; it ca
with voice, da
f the larger w
RS and CDMA
sign are shown
es the perman
mporary (such
e users registe
a about the us
ncludes the d
as the tempor
AUC stores
ch as the keys
data about
alls from a sto
GSM system
rnational Mob
d by equipm
ice provider. T
are identified
bscriber Ident
ty Module (SI
ed only if a v
valid IMSI. T
om the above
E
andardized in
an be sent
ata and fax ca
wireless doma
A. [5].
n in the Figure

nent
h as
ered
sers
data
rary
the
s for
the
olen
are
bile
ment
This
d by
tity)
IM)
alid
The
ids
the
and
alls.
ains
e 1.
tr
g
c
m
c
d
is
w
a
th
d
r
v
T
(
u
c
4
r
c
S
S
a
m
fo

SMS applic
ransport them
gateway that c
centre. It provi
mechanism for
characters from
The SMS i
directly but de
s a network el
which SMS is
available. SMS
he network. M
destination cel
eceive or tran
voice call is in
TCP/IP networ
GMSC). [6]

In this sec
used to com
compression e
40 percent. Th
atio of total c
characters of th
Compres
SMS
Efficien
SMS
Our appro
application for
manufacturers
for mobile app

Figure 2
cations are on
m. An SMS s
connects to a m
ides a point-to
r transmitting
m mobile hand
is store and fo
livered via an
lement in the
stored until th
SC guarantees
Messages are s
ll phone beco
nsmit an SMS
n progress or n
rks via a Gate
IV. C
ction, we des
mpress the
fficiency of th
he efficiency
characters in a
he original SM
ssion = T
ncy
oach to this
r SMS Compr
have adopted
plication deve
2.
ly as good as
service is hos
mobile service
o-point and br
short message
dsets.
orward service
n SMS Center
mobile teleph
he destination
s delivery of
stored in the n
omes availabl
message at an
not. SMSCs co
eway Mobile S
OMPRESSION

scribe the pro
SMS length
he message m
is obtained b
a compressed
MS.
Total character
Total chara
problem is
ression. Sever
d Java as their
elopers. To ce
the networks t
sted on an SM
e providers SM
roadcast-to-po
es up to 160 t
e, and is not s
(SMSC). SM
hone network,
device becom
text message
network until t
le, so users c
ny time, wheth
ommunicate w
Switching Cen
oposed techni
h. The aver
must be more t
y calculating
SMS to the to
rs in Compres
acters in Origi
to develop
ral mobile dev
platform offe
ertain extent J

the
MS
MS
oint
ext
ent
SC
, in
mes
by
the
can
her
with
nter
ique
rage
than
the
otal
ssed
inal
an
vice
ered
Java
a
v
s
s
m
i
o
t
T
s
c
s
t
T
u
c
u
c
applications a
vendors. Wit
significantly t
sending a lo
message to se
it and by send
original, it wil
Many SM
to solve the
These method
so-called wo
compressed m
senders. In or
to type as muc
160 character
The abbreviat
used and inter
While SM
communicatio
using the
confidentiality

I
Mobile d
and this me
designed app
several secu
integrity and
key assets th
This sec
formal descr
specification
and decrypt
station is th
attacker cou
the mobile s
Hence our a
messages us
readable. Th
plaintext to
and diffusion
DES alg
encryption a
and financia
small by curr
can be searc
block cipher
Massey in 1
cipher, PES
originally c
developed to
plaintext and
128-bit key.
are portable
th this applic
the number o
ong message.
end, the applic
ding the compr
ll reduce the c
S compression
limitation of
ds include hum
ord or phrase
messages are
der to save th
ch information
rs per SMS b
tions utilized
rpreted by the
MS is one of
on, businesses
SMS channel
y issues.
I. ENCRYPT
devices have
eans that the
propriately. F
urity services
d availability a
hat are protecte
ction highligh
ription of a c
n for its messa
tion functions
heft or mispl
ld gain acces
tation. This th
approach to th
sing IDEA a
he algorithm
64-bit block
n.
gorithm has
algorithm and
al applications
rent standards
ched in appro
r designed b
1991. It is a
(Proposed En
called IPES
o replace DE
d cipher text
The algorith
between dev
cation, the u
of single mess
When the
cation can be
ressed messag
cost.
n methods hav
f 160 charact
man interpretiv
e abbreviation
e manually
e cost, many S
n as possible w
by using vario
in the messa
senders and re
the most pre
s and banks ar
l considering
TION & DECRY

limited comp
security mec
or a secure S
s are needed
are the key ser
ed [7].
hts our resear
cryptosystem
age, key, ciphe
s. The main
lacement and
s to the data
hreat applies t
is problem he
algorithm to
m encrypts a
of cipher tex
been a pop
d is used in m
s. However, it
s and its entire
oximately 22 h
by Xuejia La
a minor revis
ncryption Stan
(Improved
ES. IDEA ope
blocks and i
hm structure
vices of diffe
users can red
sages required
user has a
used to comp
ge, shorter than
ve been develo
ters per mess
ve compressio
n, in which
keyed-in by
SMS senders
within the lim
ous abbreviati
age are popul
eceivers.
eferred mode
re conservativ
the security
YPTION
puting capabil
chanisms mus
SMS environm
d. Confidentia
rvices and also
rch work give
that includes
er text, encryp
risk to a mo
as a result,
content stored
to SMS messa
ere is by enco
make them n
64-bit block
t using confu
pular secret
many commer
ts key size is
e 56 bit key sp
hours. IDEA
ai and James
ion of an ea
ndard). IDEA
PES) and
erates with 64
s controlled b
has been cho
erent
duce
d for
long
press
n the
oped
sage.
on or
the
the
tend
mit of
ions.
larly
es of
ve in
and
lities
st be
ment,
ality,
o the
es a
s the
ption
obile
the
d on
ages.
ding
non-
k of
usion
key
rcial
s too
pace
is a
s L.
arlier
was
was
4-bit
by a
osen
e
p
b
k
p
th
s
d
k
s
u
c
e
b
k
s
d
such that wh
encryption pr
The 64-bi
bit sub-blocks
bit key. Sinc
required for t
of 52 (= 8 x 6
generated fro
partitioned in
directly used
bit key is th
positions, afte
partitioned int
as the next
procedure is
key sub-block

The priva
establishing th
path, since me
be decrypted w
key can be
public/private
han AES sinc
ufficient proc
devices and SM
key would hav
erver applica
unacceptable v
The compu
ciphered text
encryption. Th
bit key sub-blo
key sub-block
ub-blocks mu
decryption in o
hen different
ocess is identi
it plaintext blo
s six 16-bit ke
ce a further fo
the subsequent
6 + 4) differen
om the 128-b
nto eight 16-b
as the first ei
hen cyclically
er which the r
to eight 16-bit
eight key s
repeated until
ks have been g
Figure
ate/public ke
he foundation
ssages encryp
without the p
distributed
key exchang
ce the messa
cessing power,
MS in genera
ve to be progra
ation. This
vulnerability to
utational proce
is essentially
he only differe
ocks used for d
used during e
ust be used
order to revers
key sub-bloc
ical to the decr
ock is partition
ey are generat
four 16-bit ke
t output transf
nt 16-bit sub-b
bit key. The
bit sub-blocks
ight key sub-b
y shifted to
resulting 128-b
t sub-blocks to
sub-blocks. T
l all of the re
generated.
3.
ey technique
for a secure
pted with the p
private key. H
without risk
e can be sign
age size is lar
, which is not
al. This means
ammed into bo
scenario wo
o reverse engin
ess used for d
y the same a
ence is that ea
decryption is t
encryption. In
in the rever
se the encrypti
ks are used,
ryption proces
ned into four
ted from the 1
ey-sub-blocks
formation, a to
blocks have to
e 128-bit key
s which are t
blocks. The 1
the left by
bit block is ag
o be directly u
This cyclic s
quired 52, 16
e is ideal
communicati
public key can
Hence, the pub
k. However,
nificantly slow
rge and requ
t ideal for mob
s the initial A
oth the client
ould present
neering.
decryption of
as that used
ach of the 52
the inverse of
addition, the
rse order dur
ion process. [8

the
ss.
16-
128-
are
otal
o be
y is
then
128-
25
gain
used
shift
6-bit

for
ions
nnot
blic
, a
wer
uires
bile
AES
and
an
the
for
16-
f the
key
ring
8]
c
p
D
s
o
a

In this pap
character limi
SMS service,
phenomena, w
SMS Structur
Decryption tec
The prop
security in SM
of view, th
alongwith sec

Figure 4
V. C
per, we've tri
itation concept
ever since its
we studied th
re, SMS Packe
chniques with
osed method
MS communic
his technique
urity.

4.
CONCLUSION
ied to elimina
t that has been
s introduction
he GSM Syst
et, Compressio
h regards to the
d also provid
cation. From th
ensures co

ate the maxim
n surrounding
n. To achieve
tem Architectu
on, Encryption
e SMS.
des end to
he end user po
ost effectiven

mum
the
this
ure,
n &
end
oint
ness

[1] Tarek M M
Ahmed M
For Securi
And Securi
[2] http://teck.i
only.html [
free and op
[3] Johnny Li-C
End-To- En
South Afri
[4] http://www
Indian Insti
[5] N.J Croft A
Pad To Sec
Computer
University
[6] Sun
http://Dev
[7] Otto Kols
Enhanceme
Conference
[8] Internationa

REFERE
Mahmoud, Bahgat
Mahfouz, Hybrid
ing SMS, Interna
ity (IJCSS) 2010.
in/why-sms-messa
Source: http://te
pen source softwar
Chang Lo, Judit
nd Protocol For
ca 2008.
w.it.iitb.ac.in/~saty
itute of Technolog
And M.S Olivier
cure Short Messa
Security Archi
Of Pretoria, Sout
Developer
veloper.Sun.Com/
si, Teemupekka
ents Proceedings
e On System Scien
al Data Encryption

ENCES

t A. Abdel-Latef,
d Compression E
ational Journal O

ages-are-limited-t
ck.in - The techn
re as well]
th Bishop, J.H.P
Secure SMS, Un
yajit/seminar/node
gy, Bombay]
r, Using An App
aging Service (SM
itectures (ICSA)
th Africa 2005.
Network
/Mobility/Articles
a Virtanen, M
s Of The 37th
nces.
n Algorithm - Auth

, Awny A. Ahme
Encryption Techn
Of Computer Scie
to-160-characters-
nology blog prom
P Eloff Smssec:
niversity Of Pret
e14.html [Sou
proximated One-T
MS), Information
) Research Gr
SMS Art
s/Sms
MIDP 2.0 Secu
Hawaii Internati
hor: How-Shen Ch

ed &
ique
ence
-
motes
: An
toria,
urce:
Time
And
roup,
ticle,
urity
ional
han..
ISBN: 978-81-920575-7-6 :: doi: 10. 73956/ISBN_0768
Effect of topology on routing performance in partially mobile ad hoc networks

Ningrinla Marchang ,Nibedita Shial
Department of Computer Science and Engineering North Eastern Regional Institute
of Science and Technology Nirjuli, Itanagar - 791109, Arunachal Pradesh, INDIA.
ningrinla@yahoo.co.in, nibedita7865@gmail.com

A mobile ad hoc network (MANET) is a network of mobile
nodes which communicate with each other without the help
of any fixed infrastructure. Links
between nodes are created and broken as the nodes move
within the network. Thus mobility
of the nodes adversely affect routing performance
parameters such as throughput and end-to-end delay. It is
desirable to improve the performance and various ap-
proaches proposed toward this goal include modifying the
routing protocol or reinforcing the network with powerful
backbone nodes. Without relying on such elaborate
approaches, which are time-consuming and costly, it may be
required to improve routing performance. The challenge is to
improve on routing performance making use of only the
existing facilities in a MANET. We propose that one way of
achieving this goal is topology design. In this paper, we
investigate the effect of topology on routing performance in a
partially mobile ad dhoc network. In other words, by keeping
some nodes stationary in some fashion, we examine the effect of
the topology of these nodes on the performance. We have
presented our simulation results for the same.

Keywords: Mobile ad hoc network (MANET);
Topology; Routing performance.

I INTRODUCTION

A mobile ad hoc network is a network of wireless
mobile devices, which can be set up on the fly. The
network is created, managed and organized purely by the
nodes themselves without the assistance of any centralized
third party or fixed infrastructure. A node can be a
laptop, a PDA, a mobile phone, or any mobile device
with the ability to communicate with other devices. In a
MANET, a node behaves as a host as well as a router.
Since there is no requirement of a fixed infrastructure, a
MANET can be set up at any place and at any time as
needed. For instance, MANETs can be set up in
emergency situations (e.g., rescue mission), at places
where wired networks cannot be used (e.g., outdoor
conferences), or at places where the cost of setting up wired
networks is prohibitive (e.g., a remote hilly resort).
A MANET is an attractive solution for the com-
munication needs of a community or area where there no
fixed infrastructures. An example of such a community
would be a rural village without any means of electronic
communication within the village. However, for a
MANET to support voice communication, performance
parameters such as


end-to-end delay and jitter should be within ac- ceptable
limits. It is seen that delays beyond about
230 milliseconds in a voice call causes the human ear to
detect choppiness in the call.
It would be interesting to determine whether routing
performance can be improved by making the network
partially mobile by keeping some of the nodes stationary.
If it could be done so, we have found a simple and cost-
effective solution. Therefore, one would not need to rely
on other elaborate approaches such as investing on more
powerful nodes as backbone nodes or modifying the
routing protocol. We suggest that in certain situations,
one may not have the time nor the resource required for
such elaborate solutions. For instance, during a rescue
operation for which a MANET is set up, if it were found
that the delay and jitter of the network are such that no
effective communication were possible, the simplest and
quickest solution would be to keep some nodes stationary
albeit in some arrangement. This paper presents the
analytical results that were arrived at after extensive
simulation from which we can infer about the feasibility
of such a solution.
The following are some work which are related in some
way that is found in the literature. Topol- ogy and mobility
considerations in mobile ad hoc networks are presented in
[1]. The authors studied various characteristics concerning
the links and routes that make up a mobile ad hoc
network. One important finding was that due to rapid rout-
ing changes brought about by a highly dynamic topology,
one must select a reasonable mobility setting in order to
achieve some reasonable level of performance. The
impact of topology in robust routing on wireless mesh
networks is investigated in [2]. In this work, the authors
explored the impact of node distribution on routing perfor-
mance in a wireless mesh network. They found that
network performance using robust routing can be
improved by using a distribution which places a greater
density of nodes toward the outside of a network. In [3],
the authors studied the problem of designing the optimal
topology
for backbone-based hybrid wireless ad-hoc net works.
The problem is formulated as a generalized optimal
geometric Stiener network problem with minimum number
of Steiner nodes. They proposed a multiple-step solution
approach and compared several heuristic algorithms.
Another work in which design considerations for a sparse
network are discussed is [4]. The authors also defined a
metric called reachability that is useful in evaluating
design trade-offs in sparse networks. All the above work
are concerned with networks in which either all nodes are
mobile or all nodes stationary. Our proposed approach
deals with a network in which some nodes are mobile
while the others are stationary.
Topology control of ad hoc networks ([5]-[14]) is
another area in which topology of the network is
determined by controlling the communication range of the
nodes. Although topology design is also done here, the
main aim here is to reduce the overall energy consumption
of the nodes while satisfying some constraints such as
maintaining connectivity. Besides, topology control is
generally carried out for static ad hoc networks.
The paper is organized as follows. In section I, we
give a brief introduction and related work. Section II
presents the topology design. In section III, we give
the simulation results which is followed by the conclusion
in section IV.

II TOPOLOGY DESIGN

Considering that the nodes in a MANET are mobile,
there seem to be little or no meaning concerning the
topology design of the network. How- ever, in our present
context, our interest is in a partially mobile ad hoc
network in which some nodes are kept stationary while
the rest are mobile. Thus, by topology design, we mean
design of the the topology of the network formed by the
set of stationary nodes. For instance in Fig. 1, we show a
partially mobile MANET. A filled circle denotes a mobile
node whereas an empty circle denotes a stationary node.
The stationary nodes
are arranged in a grid within the MANETs net-
work area represented by the box. An edge be-
tween any two stationary nodes denotes that the nodes
are within communication range of each
other. The stationary nodes are the same as the
mobile nodes in every respect such that they have the same
processing capacity, transmission range and battery power
as the mobile nodes. Moreover, a stationary node behaves
both as host and router like any other node. Due to the way
the stationary nodes are uniformly laid out, thereby
providing good connectivity, it is intuitive to expect a higher
degree of routing performance than when all the nodes are
mobile.
Another way of placing the stationary nodes
is by randomly scattering them in the MANETs network
area. Since, they are randomly placed, one cannot
ascertain about the effect the topology design would have
on the networks connectivity. We evaluate the impact of
topology on routing performance and we present the
simulation results in the following section.

III Simulation and Performance
Evaluation

Simulations were done using the ns2.32 simulator
[15], where the underlying MAC protocol used is 802.11.
The routing protocol we have used is AODV [16]. We
have taken a deployment of
50 nodes in an area measuring 1500m 1500m.
A percentage of the 50 nodes are kept stationary and the
rest are mobile. The radio range of each node is 250m.
The simulation is run for varying maximum speeds of the
mobile nodes and a pause time of 0 second. Every point in
the graphs is an average of the results obtained from
simulating
20 different topologies using 20 connections for a period
of 500 seconds. Each data packet is of size 128 bytes and
the traffic type used is constant bit rate at the rate of 4
packets per second. The movement model used for the
mobile nodes is the random waypoint model [17].
To compare the routing performance of the case
Figure 1. Grid deployment of stationary
nodes; the empty and filled circles denote
stationary and mobile nodes respectively

when all nodes are mobile with that when some nodes are
stationary, we simulated AODV in the situations when
varying number of nodes out of the 50 nodes are made
stationary. We consid-
ered the cases when 0(0%), 5(10%), 10(20%),
15(30%), 20(40%) and 25(50%) nodes are sta-
tionary. First we simulated for the case when the stationary
nodes are arranged in a grid. We placed the nodes in a grid
in such a way that the horizontal as well as the vertical
distance between any two nodes is 250m. Moreover, the
grid is placed around the middle of the network area.
Second, we placed the stationary nodes randomly in the
network area using the CMU topology generation utility
that comes along with ns2.32 simulator.
We have considered six performance measurement
parameters:

(a) Packet Delivery Ratio (PDR): Ratio of the total
number of data packets received by the destinations
to the total number of data packets sent.

(b) Jitter: Average Variation in the received time of data
packets.
J
i
t
t
e
r

(
m
s
)

P
a
c
k
e
t

D
e
l
i
v
e
r
y

E
n
d
-
t
o
-
E
n
d

d
e
l
a
y

1

0.9

100
0

90
0

80
0

70
0

60
0

No Grid

Grid-5
Grid-10
Grid-15
Grid-20
Grid-25

0.8

0.7
500

400

300

0.6
200

100

0.5

0.4

No Grid

Grid-5
Grid-10
Grid-15
Grid-20
Grid-25

0
0 2 4 6 8 10 12
14
Maximum
speed (m/s)

0 2 4 6 8 10 12 14
Maximum
speed (m/s)

Figure 2. Grid deployment; Packet delivery
ratio Vs. Maximum speed

0.04

0.035
Figure 4. Grid deployment; End-to-end
packet delay Vs. Maximum speed

(c) End-to-end Delay: Average of the delay (received
time minus sent time) of every data packet.

(d) Route Frequency: The number of route requests
generated per second.

(e) Routing Load: The number of control packets
transmitted (including the ones that are forwarded)
for every data packet received.

0.03

0.025

0.02

0.015

0.01

0.005

0
No Grid

Grid-5
Grid-10
Grid-15
Grid-20
Grid-25

0 2 4 6 8 10 12 14
Maximum
speed (m/s)
(f) Average Throughput: The total amount of data
received by all nodes per second.

Fig. 2 shows the comparison of packet delivery
ratio(PDR) when the stationary nodes are placed in the
form of a grid. The label No Grid denotes the plot when
all the nodes are mobile (i.e., there are no stationary
nodes). Similarly Grid-5 denotes the plot when there
are 5(10%) stationary nodes among the 50 nodes. The
other labels carry similar meaning. It is seen that as the
percentage

Figure 3. Grid deployment; Jitter Vs. Maxi-
mum speed
of stationary nodes increases the packet delivery
ratio increases significantly. This is as expected as the
connectivity in the network increases with the increase in
the size of the grid of stationary nodes. It is also noticed
that the PDR decreases as the mobility increases, which is
as expected.
R
o
u
t
i
n
g

L
o
a
d

R
o
u
t
e

F
r
e
q
u
e
n
c
y

T
h
r
o
u
g
h
p
u
t

(
M
b
)

0.1

3

2.5

No Grid

Grid-5
Grid-10
Grid-15
Grid-20
Grid-25
0.08

0.06

2

1.5

0.0
4

0.0
2

No Grid

Grid-5
Grid-10
Grid-15
Grid-20
Grid-25

1

0.5
0
0 2 4 6 8 10 12
14
Maximum
speed (m/s)

0
0 2 4 6 8 10 12 14
Maximum
speed (m/s)

Figure 5. Grid deployment; Route frequency
Vs. Maximum speed

2

Figure 7. Grid deployment; Average throughput
(Mb/sec) Vs. Maximum speed

Fig. 3 shows the jitter with respect to maximum speed for
varying percentage of stationary nodes. It is seen that the
jitter reduces as the no. of stationary nodes increases.
This is expected as the network becomes more well
connected and stable as the no. of stationary nodes
increases. The end-to-end delay comparison is presented
in Fig.
4. The delay decreases rapidly as the The end-to-

1.5

1

0.5

0
No Grid

Grid-5
Grid-10
Grid-15
Grid-20
Grid
-25

0 2 4 6 8 10 12 14
Maximum
speed (m/s)
end delay comparison is presented in Fig. 4. The delay
decreases rapidly as the percentage of stationary nodes
increases. When 50% of the nodes are stationary, the delay
becomes lower than even
100ms. Fig. 5 shows the the route frequency for
varying number of stationary nodes. It is seen that the
number of route requests generated per second
decreases as the number of stationary nodes increases.
When more number of nodes are stationary and they are
deployed in grid, the number of link breaks reduces.
Consequently, the number of route requests per second
reduces since a

Figure 6. Grid deployment; Routing load Vs.
Maximum speed
link break potentially results in establishment of
new routes. Fig. 6 illustrates the comparison of the
routing load. As expected, the routing load reduces as the
number of stationary nodes increases. When there are more
number of stationary nodes forming a grid, the routes are
more stable and so
J
i
t
t
e
r

(
m
s
)

P
a
c
k
e
t

D
e
l
i
v
e
r
y

E
n
d
-
t
o
-
E
n
d

d
e
l
a
y

1000

90
0

1

0.9

800

700

600

0.8

0.7

0.6

0.5

0.4

No
Random
Rando
m-5
Rando
m-10
Rando
m-15
Rando
m-20
Rando
m-25
50
0

40
0

30
0

20
0

10
0

0

No Random

Random-5
Random-10
Random-15
Random-20
Random-25

0 2 4 6 8 10 12
14
Maximum
speed (m/s)

0 2 4 6 8 10 12 14
Maximum
speed (m/s)

Figure 8. Random deployment; Packet delivery
ratio Vs. Maximum speed

0.05
Figure 10. Random deployment; End-to-end
packet delay Vs. Maximum speed

the number of control packets generated and forwarded is
less as compared to when there are less number of
stationary nodes. Finally, the throughput increases as the
number of stationary nodes increases (Fig. 7). This is
expected since the PDR also increases as the number of
stationary nodes increases.
The second part of the simulation involves the random
deployment of the stationary nodes in-

0.0
4

0.0
3

0.0
2

0.0
1

0

No Random

Random-5
Random-10
Random-15
Random-20

Random-25

0 2 4 6 8 10 12
14
Maximum
speed (m/s)
stead of arranging them in a grid. The above
six performance parameters were used to evaluate this
scenario. Fig. 8 to Fig. 12 show the performance for
varying number of stationary nodes, which are deployed
randomly. We observe that there seem to be no notable
impact on the performance w.r.t. all the performance
parameters. Thus we conclude that improve in
performance can clearly be achieved by deploying the
stationary nodes in a grid whereas there is more or less
no effect on the performance when they are deployed
randomly.
Figure 9. Random deployment; Jitter Vs.
Maximum speed

IV Conclusions

We have presented the impact of topology of
stationary nodes in a partially mobile ad hoc
R
o
u
t
i
n
g

L
o
a
d

R
o
u
t
e

F
r
e
q
u
e
n
c
y

T
h
r
o
u
g
h
p
u
t

(
M
b
)

0.1

3.5 0.08

3
0.0
6

2.5

2

1.5

0.0
4

0.0
2

No Random

Random-5
Random-10
Random-15
Random-20
Random-25

1

0.5
No Random

Random-5
Random-10
Random-15
Random-20
Random-25

0
0 2 4 6 8 10 12
14
Maximum
speed (m/s)

0
0 2 4 6 8 10 12
14
Maximum
speed (m/s)

Figure 11. Random deployment; Route fre-
quency Vs. Maximum speed

3.
5

3

Figure 13. Random deployment; Average
throughput (Mb/sec) Vs. Maximum speed

network. First, we considered the case when the
stationary nodes are arranged in a grid. Second, we
also investigated the scenario when the stationary nodes
are deployed randomly. We observed that the routing
performance can be increased significantly by increasing
the number of the stationary nodes in the grid topology.
This would mean that instead of adopting elaborate ap-
proaches such as modifying the routing protocol
or deploying powerful and expensive backbone

2.5

2

1.5

1

0.5

0
No Random
Random-5

Random-10
Random-15
Random-20
Random-25

0 2 4 6 8 10 12 14
Maximum
speed (m/s)
nodes, the routing performance of a MANET can be
enhanced simply by making some nodes stationary and
arranging them in a grid. How- ever, we found that
randomly distributing the stationary nodes in the
network area does not help in improving the routing
performance. Thus, we conclude that to enhance the
performance, stationary nodes must not be just stationary
but must be arranged is some regular fashion so that
it helps in improving the connectivity of the
Figure 12. Random deployment; Routing load
Vs. Maximum speed
network.

Acknowledgment: This work is supported by Re-
search Promotion Scheme of AICTE, India (project no.
8023/BOR/RID/RPS-233/2008-09, 31/3/2009).

References

[1] B. Ishibashi and Raouf Boutaba, Toplogy and
mobility considerations in mobile ad hoc
networks, Elsevier Ad Hoc Networks, 3 (2005)
762-776.

[2] J. Wellons and Y. Xue, The Impact of Topol- ogy in
Robust Routing on Wireless Mesh Networks,
Proc. Mobile Computing and Communications
Review, 2010, pp. 46-48.

[3] Y. Xin, T. Guven, M. A. Shayman, Topology design
for wireless ad-hoc networks with backbone
support, Proc. Conference on Information Sciences
and Systems, 2005.

[4] S. Perur and S. Iyer, Sparse Multi-hop wireless for
voice communication in rural India, Proc. 12th
National Conference on Communications, NCC
2006, New Delhi, India; pp. 534-538, January 2006.

[5] M. Cardei, J. Wu and S. Yang, Topology control
in ad hoc wireless networks using cooperative
communication, IEEE Transactions on Mobile
Computing, vol. 5, no. 6, June 2006.

[6] A. Rahman and P. Gburizynski, MAC- assisted
topology control for ad hoc wireless networks,
International Journal of communication systems,
2005;00;1-6.

[7] S. Agarwal, S. Krishnamurthy, R. Katz, and S. Dao,
Distributed power control in ad-hoc wireless
networks, Proc. PIMRC, 2001.

[8] T. Kwon and M. Gerla, Clustering with power
control, Proc. MILCOM, vol. 2, pp. 1424-1428,
1999.

[9] R. Ramanathan and R. Rosales-Hain, Topology
control of multihop wireless networks using
transmit power adjustment, Proc. of IEEE
INFOCOM 2000, pp. 404-413, Tel Aviv, Israel,
March 2000.
[10] L. Li, J. Y. Hlapern, P. Bahl, Y. Wang, and R.
Watenhofer, Analysis of a con- based topology
control algorithm for wireless multi-hop
networks, Proc. ACM Sym- posium on Principle of
Distributed Comput- ing (PODC), 2001.

[11] N. Li, J. C. Hou, and L. Sha, Design and analysis
of an MST-based topology control algorithm,
Proc. INFOCOM, 2003.

[12] V. Rodoplu and T. Meng, Minimum energy mobile
wireless networks, IEEE Journal of Selected
Areas in Communications, 17(8), pp. 1333-1344,
1999.

[13] R. Wattenhofer, L. Li, P. Bahl, and Y.- M. Wang,
Distributed topology control for wireless multihop
ad-hoc networks, Proc. INFOCOM 2001, pp. 1388-
1397, 2001.

[14] Z. Huang, C. Shen, C. Srisathapornphat, and C.
Jaikaeo, Topology control for ad hoc networks
with directional antennas, Proc. IEEE Int.
Conference on Computer Communications and
Networks, pp. 16-21, 2002.

[15] The Network Simulator - ns2,
www.isi.edu/nsnam/ns/.

[16] C. E. Perkins and E. M. Royer, Ad-hoc On-
Demand Vector Routing, Proc. 2nd IEEE
Workshop on Mobile Computing Systems and
Applications, pp. 90-100, New Orleans, LA,
February 1999.

[17] D. B. Johnson and D. A. Maltz, Dynamic Source
Routing in Ad Hoc Wireless Net- works, Mobile
Computing (T. Imielinski and H. Korth, editors,
Kluwer Academic Publishers, pp. 153-181, 1996.
V
c
f
a
w
b
a
a
d
f
t
S
w
a
t
s
c
m
p
c
s
o
F
l
c
T
t
(
e
f
n
W
a
P
w
d
t
p
a
F

a
c
t
d
m
Proc. of the
Volume 1. Cop
ISBN: 978-81
A
Abstr
communication
fading. Orthog
and orthogona
which are m
become widely
against freque
advantages, O
drawbacks; hi
for an adaptive
the channel, an
Single carrier
which utilizes
and frequency
technique that
same overall s
carrier frequ
modified form
promising te
communication
similar throug
overall comple
FDMA is the
lower than tha
candidate for
Term Evolutio
the 3rd Genera
(3GPP LTE).
In th
explain how SC
frequency dom
needed an alte
We present it
analyze the eff
PAPR. Among
we find that
dependent sch
than interleav
performance o
also explain th
FDMA and OF

On the le
are depicted
converted to
then in the fr
desired locati
must then be c
e Intl. Conf. on
pyright 2012
1-920575-7-6
ACM #: dber.i
A S
ract- Broa
ns suffer from
gonal frequenc
al frequency di
multicarrier co
y accepted pri
ncy selective fa
OFDM and O
gh peak to-ave
e or coded sche
nd high sensitiv
frequency divi
single carrier
y domain equ
t has similar
structure as th
ency division
m of Orthog
echnique for
ns in future c
ghput perform
exity as OFDM
peak-to-averag
at of OFDMA
the uplink mu
on of cellular
ation Partners
is paper, we gi
C-FDMA utiliz
main equalizat
ernative of such
ts basic conce
fects of subcar
g the possible
localized FD
heduling (CDS
ved FDMA (
of IFDMA is b
he similarities a
FDMA.
ft hand side o
in the time
the frequency
frequency dom
on in the ove
converted bac
n Computer A
2 Techno Foru
:: doi: 10. 739
imera.10. 739
Study on S

adband w
m multipath
cy division mu
ivision multiple
ommunication
imarily becaus
ading channels
OFDMA suff
erage power ra
eme to overcom
vity to frequen
ision multiple
r modulation
ualization at
performance a
hose of an OFD
n multiple ac
gonal FDMA
r high da
cellular system
mance and es
MA. A principa
ge power ratio
. SCFDMA is
ultiple access s
systems unde
hip Project Lo
ive an overview
zes single carri
tion technique
h an advance O
ept on which
rrier mapping
subcarrier ma
DMA (LFDM
S) results in
(IFDMA). How
better than th
and main differ
of this Figure t
e domain. Th
y domain usin
main they are
erall carrier b
ck to the time
Applications
um Group, Ind
963/ISBN_076
63
SC-FDMA
Ch
R
Singh
wireless m
frequency-sele
ultiplexing (OF
e access (OFDM
techniques,
se of its robust
s .Despite the m
fer a numbe
atio (PAPR), a
me spectral nu
ncy offset.
access (SC-FD
at the transm
the receiver
and essentially
DMA system S
ccess (SCFDM
(OFDMA),
ata rate up
ms. SC-FDMA
sentially the
al advantage of
o (PAPR), whi
currently a st
scheme in the L
er consideratio
ong Term Evolu
w of SC-FDMA
ier Modulation
es simply. Wh
OFDMA techn
it works. We
on throughput
apping approa
MA) with cha
higher throug
wever, the P
at of LFDMA
rences between
the data symb
he symbols a
ng an FFT, a
e mapped to t
bandwidth. Th
domain in ord
dia.
68
A Advanc
handra Shek
Research Scho
hania Univers
obile
ective
FDM)
MA),
have
tness
many
r of
need
lls in
DMA)
mitter
is a
y the
Single
MA),a
is a
plink
A has
same
f SC-
ich is
trong
Long
on by
ution
A.We
n and
y we
nique.
also
t and
aches,
nnel-
ghput
PAPR
A. We
n SC-
Ke
Pe

In
co
sy
the
nu
ph
ac
Co
Or
do
(S
co
sy
sy
FD
do
ac
alt
wh
po
ge
ols
are
and
the
hey
der
b
p
e Access S
khar
olar
sity, Raj.

eywords:
erformance Me
I.
n order to
ommunications
ystems, the 3rd
e Long Ter
umerous differ
hysical layer (
ccess schemes
ode Division
rthogonal Fre
ownlink and S
SC-FDMA) for
SC-FD
ombines the lo
ystems - such
ymbol time a
DMA is an ex
omain equaliz
ccess. SC-FD
ternative to O
here lower PA
ower efficienc
eneration are s

Figur
to have the c
An alternative
Transform Sp
Single-ca
division mult
linearly preco
guard interva
between bloc
eliminate tim
propagation)
Scheme fo
Channel-Depe
easures, SC-FD
SINGLE-CAR
MULTI
transition fr
s systems to m
d Generation P
rm Evolution
rences betwee
(PHY), specif
. While its pa
n Multiple A
equency Div
Single-Carrier
r its uplink.[1]
DMA is a
ow PAR chara
as those used
and flexible f
xtension of sin
zation (SC/F
MA has dra
OFDMA, espe
APR greatly b
cy. [2][3][4].T
shown in Figur
re1: SC-FDM
cyclic prefix i
e name for S
pread OFDM (
arrier FDMA
tiple access s
oded OFDMA
als with cycl
cks of symb
me spreadin
among the bl
or 4G
endent, IFDM
DMA Vs. OFDM
RRIER FREQ
IPLE ACCES

from today's
meet the need
Partnership Pro
n (LTE) spe
en these gener
fically in the
arent generatio
Access (CDM
ision Multipl
r Frequency-D
].
hybrid transm
acteristics of si
d for GSM and
frequency all
ngle carrier m
FDE) to acco
awn great att
ecially in the
enefits the mo
The principles
re.
MA signal gene
nserted prior
C-FDMA is D
(DFT-SOFDM
(SC-FDMA)
scheme. SC-F
A scheme. Just
lic repetition
bols in view
ng (caused
ocks. In OFD
DMA, LFDM
MA.
QUENCY-DIV
S (SC-FDMA
3rd genera
ds of 4
th
gene
oject (3GPP) h
ecification. A
rations are cha
modulation a
on relied on v
MA), LTE
lexing (OFD
Division Mult
mission sche
ingle-carrier t
d CDMA - w
ocation of O
modulation with
ommodate m
tention as an
e uplink comm
obile terminal
behind SC FD
eration
to transmissio
Discrete Four
M).[5]
) is frequenc
FDMA can as
t like in OFD
are introduc
w to efficien
by multi-pa
DM, Fast Four
MA, PAPR,
VISION
A)
ation (3G)
eration (4G)
has released
Among the
anges in the
and multiple
variations of
implements
M) for its
tiple Access
eme which
transmission
with the long
OFDM. SC-
h frequency
multiple-user
n attractive
munications
in terms of
DMA signal
on.
rier
cy-
s a
DM,
ced
ntly
ath
rier
transform (FFT) is applied on the receiver side on each
block of symbols, and inverse FFT (IFFT) on the
transmitter side. In SC-FDE, both FFT and IFFT are
applied on the receiver side, but not on the transmitter
side. In SC-FDMA, both FFT and IFFT are applied on
the transmitter side, and also on the receiver side. In SC-
FDMA, multiple access is made possible by inserting
silent fourier-coefficients on the transmitter side before
the IFFT, and removing them on the receiver side before
the IFFT.
What makes SC-FDMA a single carrier modulation
is the fact that there is just one carrier. PAPR of
SC/FDMA is 2dB smaller compared to OFDMA.
Single Carrier Frequency Division Multiple Access
(SC-FDMA)
A new single carrier multiple access technique
Utilizes single carrier modulation and frequency
domain equalization.
Has similar performance and essentially the same
overall structure as those of OFDMA system
Has low PAPR because of its inherent single
carrier transmitter structure.
An attractive alternative to OFDMA, especially in
the uplink communications where lower PAPR
greatly benefits the mobile terminal in terms of
transmit power efficiency.
The basic uplink transmission technique is single-
carrier transmission (SC-FDMA) with cyclic prefix to
achieve uplink inter-user orthogonality and to enable
efficient frequency domain equalization at the receiver
side

1. History:

Over the past fifteen years, the bit rates achieved in
cellular and local area wireless communication systems
have increased steadily. The earliest digital cellular
systems, North American TDMA and GSM employed
time division multiple access. Second generation
CDMA systems and the two third generation cellular
systams all use direct sequence spread spectrum for
multiplexing and muliplu access The highest bit
rates in commercially deployed wireness systems are
achieved by means of Orthogonal requency Division
Multiplexing (FDM) in wipeless LANs bas%d on the
IEEE 802.11a and IEEM 802.11g standards. The next
advance in cellular systems, under investigation bi the
Third Generation Partnership Project (3GPP), also
anticipates the adoption of OFDA to achieve higher bit
rates. Now in fourth Generation we use SC-FDMa to
achieve maxhmum performance with a single
carrier.[6].`

3. Why Needed

Today multi-carries transmission And OFDM are meanw
to allow for very high overall transmission bandwidth
while still being robust to signal corruption due to radio-
channel frequency selectyvity. However, a drawback of
OFDM modulation, as well as any kind of multi-carrier
transmission, ir the large variations in the instantaneous
Power of the transmitted signal. Such power variations
imply a reduced power-amplifier efficiency and higher
power-amplifier cost. This is especially critical for the
uplink, due to the high importance of low mobile-
terminal power consumption and cost. Several methods
have been proposed on how to reduce the large power
variations of an OFDM signal. However, most of these
methods have limitations in terms of to what extent the
power variations can be reduced. Furthermore, most of
the methods also imply a significant computational
complexity and/or a reduced link performance.[7] Thus,
there is an interest to consider also wider-band single-
carrier transmission as an alternative to multi-carrier
transmission, especially for the uplink, i.e. for mobile-
terminal transmission. One of such single-carrier
transmission scheme can be implemented using DFT-
spread OFDM.

II. SC-FDMA/DFT-spread OFDM:

DFT-spread OFDM
DFT-Spread OFDMA Mapping of spread symbols , not
original symbols to subcarriers.
SC-FDMA can be regarded as discrete Fourier
transform (DFT)-spread OFDMA, where time domain
data symbols are transformed to frequency domain by
DFT before going through OFDMA modulation.
DFT-spread OFDM (DFTS-OFDM) is a
transmission scheme that can combine the desired
properties for uplink transmission i.e.
Small variations in the instantaneous power of the
transmitted signal (single carrier property).
Possibility for low-complexity high-quality
equalization in the frequency domain.
Possibility for FDMA with flexible bandwidth
assignment.
Due to these properties, DFTS-OFDM has been
selected as the uplink transmission scheme for LTE,
which is the long-term 3G evolution.

5. How SC-FDMA Work?

The transmitter of an SC-FDMA system converts a
binary input signal to a sequence of modulated
subcarriers. At the input to the transmitter, a baseband
modulator transforms the binary input to a multilevel
sequence of complex numbers xn in one of several
possible modulation formats. The transmitter next groups
the modulation symbols {xn} into blocks each containing
N
s
f
s
o
t
s
s
i
s
t
a
i
a
x
p
t
r
o
p
w
a
t
B
i
v
F
D
s
p
i
p
f
r
o
e
g
t
c
r
c
t
c
t
b
t
a
s
D
D
m
t
u
N symbols. T
subcarriers is
frequency do
symbols. It th
of the M (>
transmitted. I
symbols per
simultaneous
interference. Q
symbol sequen
the set l X (
amplitudes, w
in OFDMA, a
amplitudes to
xm then are
performs two
transmission.
How
remaining inp
of the IDFT
properties, i.e
with a bandw
assuming a sa
the nominal b
BW =M/N *
instantaneous
varied, allow
Furthermore,
DFT outputs
shifted in the f
It ins
prefix (CP) in
inter-block
propagation.
filtering opera
reduce out-ofb
of the last par
each block fo
guard time be
the CP is long
channel, or ro
response, then
copy of the la
time linear c
convolution. T
the channel c
between the
transmitted da
a point-wise
samples. The
DFT of the re
DFT of the c
more sophis
technique can
used pulse sha
The first step i
to perform a
omain repres
hen maps each
> N) orthogo
If N = M/Q a
r block, the
transmissi
Q is the bandw
nce. The resul
(l = 0, 1, 2
where N of the
an M-point ID
a complex tim
e transmitted
other signal p

wever, if M
puts to the IDF
T will be a
e. a signal wit
width that depe
ampling rate
bandwidth of
fs. Thus, by
bandwidth of
wing for flex
by shifting th
are mapped,
frequency dom
serts a set of s
n order to pro
interference
The transmi
ation referred t
band signal en
rt of the block,
or a couple of
etween succes
ger than the m
oughly, the le
n, there is no
ast part of the
convolution in
Thus transmit
can be model
e channel im
ata block, whi
multiplicatio
en, to remove
eceived signal
channel impu
sticated frequ
n be impleme
aping filter is t
in modulating
an N-point DF
sentation Xk
h of the N DFT
onal subcarrie
and all termi
e system c
ons withou
width expansi
lt of the subca
, M-1) of com
e amplitudes a
DFT transform
me domain sig
sequentially.
processing op
is smaller th
FT are set to
signal with
th low power
ends on M. M
fs at the outp
the transmitte
varying the b
f the transmitt
xible-bandwid
he IDFT inpu
the transmitte
main.[8]
symbols referr
ovide a guard
(IBI) due
tter also per
to as pulse sha
nergy. In gener
, which is add
f reasons. Fir
ssive blocks.
maximum dela
ength of the c
o IBI. Second
e block, it con
nto a discret
tted data prop
led as a circu
mpulse resp
ch in the frequ
on of the D
e the channel
can simply b
ulse response
uency domai
ented. One of
the raised-cos
g the SC-FDM
FT to produce
of the inp
T outputs to o
ers that can
inals transmit
can handle
ut co-chann
ion factor of t
arrier mapping
mplex subcarr
are non-zero.
ms the subcarr
gnal xm . Ea
The transmit
perations prior
han N and t
zero, the outp
single-carri
r variations, a
More specifical
put of the IDF
ed signal will
block size M t
ted signal can
dth assignme
uts to which t
ed signal can
red to as a cyc
time to preve
to multipa
rforms a line
aping in order
ral, CP is a co
ded at the start
rst, CP acts as
If the length
ay spread of t
channel impu
d, since CP is
nverts a discr
te time circu
pagating throu
ular convoluti
ponse and t
uency domain
DFT frequen
l distortion, t
e divided by t
point-wise or
in equalizati
f the common
ine filter.
MA
e a
put
one
be
N
Q
nel
the
g is
rier
As
rier
ach
tter
r to
the
put
ier
and
lly,
FT,
be
the
be
ent.
the
be
clic
ent
ath
ear
r to
opy
t of
s a
of
the
ulse
s a
ete
ular
ugh
ion
the
n is
ncy
the
the
r a
ion
nly
The m
a multi-carrie
reduced varia
implying the
efficiency.

DFT: Discrete
IDFT: Inverse
CP: Cyclic Pr
PS: Pulse Sha
DAC: Digital-
RF: Radio Fre
ADC: Analog
Figure2: Tran
FDMA/DFTS

Two approach
1 Distribu
(IFDMA)
2. Localize
Ther
subcarriers am
(LFDMA), e
subcarriers to
of an LFDMA
the system b
distributed SC
terminal are
realization o
main benefit o
er transmission
ations in the i
possibility fo
e Fourier Tran
e Discrete Fou
refix
aping
-to-Analog Co
equency
g-to-Digital Co
nsmitter and R
S-OFDM
6. Apportioni
hes to apportio
uted SC-FDM
d SC-FDMA
re are two
mong termin
each termina
o transmit its
A transmissio
bandwidth. Th
C FDMA in w
spread over
of distributed
of DFTS-OFD
n scheme suc
instantaneous
for increased
nsform
urier Transform
onversion
onversion
Receiver Struct
ing Subcarrier
oning subcarri
MA or Inter
(LFDMA)
approaches t
nals. In locali
al uses a s
symbols. Thu
on is confined
he alternative
which the subc
the entire si
d SC-FDMA
DM, compared
ch as OFDM,
transmit pow
power-amplif
m
ture of SC-
rs:
iers
rleaved FDM
to apportioni
ized SC-FDM
set of adjac
us the bandwid
to a fraction
e to LFDMA
arriers used by
gnal band. O
is interleav
d to
, is
wer,
fier

MA
ing
MA
ent
dth
of
is
y a
One
ved
F
e
m
a
t
a
l
l
l
m
(
m
c
I
e
s
I
s
s
C
c
s
p
s
d
p
l
s
T
o
P
d
I
F
(
u

W
FDMA (IFD
equidistant fro
In o
mapping mod
allocated over
the unused su
are occupied b
localized subc
localized sub
localized FDM
mapping mod
(DFDMA). T
mode with eq
called Interle
IFDMA is a
efficient in th
strictly in the
IDFT.
From
subcarrier ma
static and chan
CDS assigns
channel frequ
scheduling m
provides freq
signal is sp
distributed m
performance.
localized sub
significant mu
Two subcarri
other.
Distributed
PAPR in case
Localized:
dependent sch
IFDMA but si
Figure3: Subc
(3 users, 12 s
user).
With respect t
System thr
DMA) where
om each other
other way, in
de, DFT out
r the entire ban
ubcarriers, wh
by the DFT ou
carrier mappin
bcarrier mapp
MA (LFDMA
de of SC-FD
The case of M
quidistance bet
eaved FDMA
special case
hat the transm
time domain
m a resource
apping metho
nnel-dependen
s subcarriers
uency respon
methods, dist
quency divers
read over th
mapping, CD
By contrast,
bcarrier map
ulti-user divers
ier mapping
d: Frequency
of IFDMA.
Frequency
heduling, hig
imilar PAPR t
carrier allocati
subcarriers, an
7. Performa
to the two maj
roughput
e occupied
[9].
n the distrib
tputs of the
ndwidth with
hereas consecu
utputs of the i
ng mode. We
ping mode of
A) and distrib
DMA as dist
M = Q N for
tween occupie
(IFDMA) [
of SC-FMDA
mitter can mod
n without the u
e allocation
ods are furth
nt scheduling
to users ac
nse of each
tributed subc
sity because
he entire ba
DS incremen
CDS is of gr
pping becaus
sity.
have advanta
y diversity,
selective gain
gher PAPR th
to the case of D
ion methods f
nd 4 subcarri

ance Measures
jor performanc
subcarriers a
buted subcarr
input data a
zeros occupyi
utive subcarri
input data in t
will refer to t
f SC-FDMA
buted subcarr
tributed FDM
r the distribut
ed subcarriers
10], [11], [1
A and it is ve
dulate the sign
use of DFT a
point of vie
her divided in
(CDS) metho
ccording to t
user. For bo
carrier mappi
the transmitt
andwidth. W
ntally improv
reat benefit w
se it provid
ages over ea
lower
n with chann
han the case
DFDMA.

for multiple us
iers allocated
s:
ce indicators,
are
rier
are
ing
ers
the
the
as
rier
MA
ted
s is
2].
ery
nal
and
ew,
nto
ds.
the
oth
ing
ted
With
ves
with
des
ach
nel
of
sers
per
p
u
u
b
b
u
PAPR

7.1 System Th
While PAPR
information
indicator of
throughput in
information sy
With respect
determines th
against freq
information i
Therefore it o
On the other
multi-user d
selective fadin
portion of the
transmission
user diversity
dispersed tra
dependent sc
requires the s
function of
subcarrier as
frequency resp
For e
measures wit
channel depen
engineering sy
design parame
the lowest P
Therefore equ
use their judg
specific needs
many users e
better off with
system with a
7.2 PAPR ana

The PAR is d
block (one O
signal power
distribution is
instantaneous
distributions h
variations of O
SC-F
OFDMA, w
transformed t
through OFD
users stems
different subc
the case of OF
is a single
compared to
multicarrier si
hroughput:
is a major c
throughput is
system perf
n SC-FDMA d
ymbols are ap
to immunity
hroughput),dis
quency selec
is spread acr
offers the adva
r hand, LFD
diversity in
ng if it assign
e signal band
characteristics
y relies on
ansmitters.[13]
cheduling (C
system to mon
frequency fo
ssignments to
ponses of all t
each configur
th static subc
ndent schedul
ystems there a
eters and perf
PAPR tend
uipment design
gment to find t
s. For exampl
each transmitt
h IFDMA, wh
a few high-bit-
alysis of SC-F
defined as the
OFDM symbol
r. It should
s not the sam
transmit
have often bee
OFDM.
FDMA can b
where time
to frequency d
DMA modulati
from the fa
carriers in the
FDMA. Becau
carrier signa
the case of
ignal.
concern in por
s an even m
formance. A
depends on th
pplied to subca
to transmissio
stributed SC-F
ctive fading
ross the enti
antage of freq
MA can pot
the presence
ns each user to
where that us
s (high chann
independent
] It also re
CDS) of su
nitor the chan
or each termi
o changes i
the terminals.
ration, we pre
carrier assignm
ling. We find
are complex tr
formance. Con
to have low
ners and syste
the best trade
le, we find tha
ting at a mod
hile LFDMA w
-rate users.
DMA:
peak power w
l) normalized
d be noted
me as the dis
power. Hist
en used to illu
be regarded
domain data
domain by DF
ion. The orth
act that each
frequency do
use the overal
al, PAPR is
OFDMA wh
rtable termina
more import
As in OFDM
he way in whi
arriers.
on errors (whi
FDMA is rob
g because
re signal ban
quency diversi
tentially achie
e of frequen
o subcarriers in
ser has favorab
nel gain). Mu
fading amo
equires chann
ubcarriers. CD
nnel quality a
inal, and ad
in the chann
esent throughp
ments and w
that as in oth
radeoffs betwe
nfigurations w
wer throughp
em operators c
off to meet th
at a system w
derate bit rate
works better in
within one ID
d by the avera
that the PA
stribution of t
torically, PA
ustrate the pow
as DFT-spre
a symbols a
FT before goi
ogonality of t
h user occup
omain, similar
ll transmit sign
inherently l
hich produces
als,
ant
MA,
ich
ich
ust
its
nd.
ity.
eve
ncy
n a
ble
ulti-
ong
nel-
DS
s a
apt
nel
put
with
her
een
with
put.
can
heir
with
e is
n a
FT
age
AR
the
AR
wer
ead
are
ing
the
pies
r to
nal
ow
s a
High PAPR requires expensive and inefficient
power amplifiers with high requirements on linearity,
which increases the cost of the terminal and also drains
the battery faster.SC-FDMA solves this problem by
grouping together the resource blocks in such a way that
reduces the need for linearity, and so power
consumption, in the power amplifier. A low PAPR also
improves coverage and the cell-edge performance.
In this section, we analyze the PAPR of the SC-FDMA
signal. We assume that the total number of subcarriers is
M = Q N, where N is the number of subcarriers per
block. The integer Q is the maximum number of
terminals that can transmit simultaneously. For
distributed subcarrier mapping, we consider the case of
IFDMA with subcarriers equally spaced over the system
bandwidth.
The peak power of any signal x(t) is the
maximum of its squared envelope [x(t)]^2. However, for
a continuous random process, max [x(t)]^2 could be
unbounded. Even for random process with discrete
values where max [x(t)]^2 is bounded, it may occur at
very low probability which is not very useful in practice.
Rather, the distribution of [x(t)]^2 is more useful and we
describe it with the complementary cut-off probability.
The PAPR is defined as the ratio of peak power
to average power of the transmitted signal in a given
transmission block. Without pulse shaping, that is, using
rectangular pulse shaping, symbol rate sampling will
give the same PAPR as the continuous time domain case
since an SC-FDMA signal is modulated over a single
carrier.
To evaluate PAPR of individual system
configurations, we have simulated the transmission of
105 blocks of symbols. After calculating PAPR for each
block, we present the data as an empirical CCDF
(Complementary Cumulative Distribution Function). The
CCDF is the probability that PAPR is higher than a
certain PAPR value PAPR (Pr {PAPR > PAPR}).
We also relate the PAPR to RF transmit power amplifier
efficiency. In an ideal linear power amplifier where
linear amplification is achieved up to the saturation
point, maximum power efficiency is achieved when the
amplifier is operating at the saturation point. To prevent
distortion in the presence of PAPR, transmit power
backoff is needed to operate the power amplifier in the
linear region. SC-FDMA signals indeed have lower
PAPR than OFDMA signals. Also, LFDMA incurs
higher PAPR compared to IFDMA but, compared to
OFDMA; it is lower, though not significantly. Another
noticeable fact is that pulse shaping significantly
increases the PAPR of IFDMA. A pulse shaping filter
should be designed carefully in order to limit the PAPR
without degrading the system performance. In general,
IFDMA is more desirable than LFDMA in terms of
PAPR and power efficiency.

8. Channel-Dependent Scheduling (CDS) for Uplink SC-
FDMA:

CDS assigns subcarriers to users according to the
channel frequency response of each user. A key question
of CDS is how we should allocate time and frequency
resources fairly among users while achieving multi-user
diversity and frequency selective diversity. To do so, we
introduce utility-based scheduling where utility is an
economic concept representing level of satisfaction. we
consider two different utility functions: aggregate user
throughput for maximizing system capacity and
aggregate logarithmic user throughput for maximizing
proportional fairness [14],[15]. The objective is to find
an optimum chunk assignment for all users in order to
maximize the sum of user utility at each transmit time
interval (TTI). If the user throughput is regarded as the
utility function, the resource allocation maximizes rate-
sum capacity ignoring fairness among users. Therefore,
only the users near the base station who have the best
channel conditions occupy most of the resources. On the
other hand, setting the logarithmic user data rate as the
utility function provides proportional fairness.
In considering the optimization problem of CDS
for multicarrier multiple access, it is theoretically
possible to assume that the scheduler can assign
subcarriers individually. Allocating individual
subcarriers is, however, a prohibitively complex
combinatorial optimization problem in systems with 256
subcarriers and on the order of ten terminals transmitting
simultaneously. Moreover, assigning subcarriers
individually would introduce unacceptable control
signaling overhead. In practice, the units of resource
allocation are chunks, which are disjoint sets of
subcarriers. As a practical matter chunk-based
transmission is desirable since the input data symbols are
grouped into a block for DFT operation before subcarrier
mapping. With regards to chunk structure, there is a
restriction on chunk selection for IFDMA such that all
assigned subcarriers should be equidistant in order to
maintain the lowest PAPR [16].
Even with subcarriers assigned in chunks,
optimum scheduling is extremely complex for two
reasons:
The objective function is complicated, consisting of
nonlinear and discrete constraints dependent on the
combined channel gains of the assigned subcarriers; and
There is a total transmit power constraint for each
user.
Instead of directly solving the optimization
problem, a sub-optimal chunk allocation scheme can be
used for both IFDMA and LFDMA to obtain most of the
benefits of CDS. For LFDMA, a greedy chunk selection
method can be applied where each chunk is assigned to
the user who can maximize the marginal utility when
occupying the specific chunk. For IFDMA, the benefit of
multi-user diversity can be achieved by selecting users in
order of the estimated marginal utility based on the
average channel condition over the entire set of
subcarriers. The users with higher channel gains may
occupy a larger number of chunks than users with lower
channel gains [17].
The advantage of channel dependent scheduling
over round robin scheduling increases as the number of
users increases. This is because the scheduler selects the
closer users who can transmit at higher data rate. If there
are more users, the possibility of locating users at closer
distance to the base station increases. As a result, the
CDS achieves significant improvements for both IFDMA
and LFDMA. In the case of logarithmic rate utility, the
CDS gain stops increasing beyond approximately 32
users. With 32 users, maximizing logarithmic rate utility
can increase system capacity by a factor of 1.8 for
LFDMA and 1.26 for IFDMA relative.
To static scheduling. the CDS scheme based on
the logarithm of user throughput as a utility function
provides proportional fairness whose gains are shared
among all users, whereas the CDS gains are concentrated
to the users near the base station when the user
throughput is considered as the utility function.
For static subcarrier assignment (round robin
scheduling), a system with users each transmitting at a
moderate data rate is better off with IFDMA while
LFDMA works better in a system with a few high date
rate users. Due to the advantages of lower outage
probability and lower PAPR, IFDMA is an attractive
approach to static subcarrier scheduling. For channel-
dependent scheduling, LFDMA has the potential for
considerably higher data rate. The results show that CDS
increases system throughput by up to 80% relative to
static scheduling for LFDMA but the increase is only
26% for IFDMA. The scheduling gains in LFDMA can
be exploited to reduce power consumption and PAPR by
using power control to establish a power margin instead
of increasing system capacity.

9. SC-FDMA Vs. OFDMA:

Comparison study between SC-FDMA and OFDMA can
be done in two ways
Similarities
Dissimilarities
Block Diagram.

9.1 Similarities:
Block-based modulation.
Divides the transmission bandwidth into smaller
subcarriers.
Channel equalization is done in the frequency
domain.
CP is added to overcome IBI (Inter-Block
Interference) and to convert linear convolution of the
channel impulse response to circular one.
SC-FDMA can be regarded as DFT-precoded or
DFT-spread OFDMA.

9.2 Dissimilarities:
The only difference is the presence of the DFT in SC-
FDMA. For this reason SCFDMA is sometimes referred
to as DFT-spread or DFT-precoded OFDMA. In terms of
data detection at the receiver, OFDMA performs it on a
per-subcarrier basis whereas SC-FDMA does it after
additional IDFT operation. Because of this difference,
OFDMA is more sensitive to a null in the channel
spectrum and it requires channel coding or power/rate
control to overcome this deficiency. Also, the duration of
the modulated time symbols are expanded in the case of
OFDMA with parallel transmission of the data block
during the elongated time period whereas SC-FDMA
modulated symbols are compressed into smaller chips
with serial transmission of the data block, much like a
direct sequence code division multiple access (DS-
CDMA) system.

9.2.1 Channel equalization:
OFDMA: Per subcarrier detection.=>A Null in
channel spectrum severely degrades the system
performance. >Channel coding or power/rate control is
required
SC-FDMA: Detection after IDFT. >SNR is
averaged out over entire bandwidth.

9.2.2 Time domain symbol duration:
OFDMA: Modulated symbol duration is longer than
input data symbol duration.
SC-FDMA: Modulated symbol duration is shorter than
input data symbol duration.

9.2.3 PAPER:
OFDMA: Parallel transmission of data over multiple
carriers. High PAPR.
SC-FDMA: Serial transmission of data over single
carrier. =>Low PAPR.

9.3 Data Symbols Diagram:
F
t

i
d

a
c
m
i
a
E

I
P
a

u
c
s
u
f
p
Figure 4: C
transmitting a
10. Ad
Single carrier
in which the
domain, which
SC-FDMA ha
alternative to
communicatio
mobile termin
is currently a
access scheme
Evolved UTR
Low PAP
Low sens
I would like
Professor Dr.
and encourage
SC-FDMA is
uplink commu
carrier FDMA
scheme that i
uplink. SC-FD
frequency d
performance a
omparison o
a series of QPS
dvantage of SC
FDMA (SC-F
data symbol
h results in a m
11.
as drawn gre
o OFDMA,
ons where low
nal in terms of
a working ass
e in 3GPP Lo
RA.
PR
sitivity to carri
ACKNOWL
e to thank m
Mohd. Huss
ement.
12. CON
a promising
unication in fu
A (SC-FDMA
is currently b
DMA utilizes
domain equa
and essentially
f OFDMA a
SK data symbo
CFDMA over O
FDMA) is a v
s are modula
more power ef
Usage:
eat attention
especially
wer PAPR gre
f transmit pow
sumption for
ong Term Evo
ier frequency o
LEDGEMENT
my official s
sain for all th
NCLUSION:
technique fo
future cellular
A) is a new
being adopted
single carrier
alization, and
y the same ov
and SC-FDM
ols
OFDMA:
variant of OFD
ated in the tim
fficient schem
as an attracti
in the upli
eatly benefits t
wer efficiency
uplink multip
olution (LTE),
offset
T
supervisors a
heir help, advi
or high data r
systems. Sing
multiple acce
d in 3GPP LT
modulation a
d has simi
verall complex

MA
DM
me
me.
ive
ink
the
. It
ple
or
and
ice
ate
gle
ess
TE
and
ilar
xity
b
p
p
n
w
as those of
benefittial for
power efficien
With
configuration,
choices that a
is shown tha
PAPR compa
focused on
throughput an
Among the po
find that loc
dependent sch
than interleav
performance o
of LFDMA.
Whe
to control ad
narrower diff
terms of PAP
that pulse shap
in the case
significant im
filter should b
PAPR withou
Work in prog
channel estim
of SC-FDMA
information a
channels linki
Channel estim
and changes
the performan
of the modula
subcarriers to
.

[1]. R. van
Communications
[2]. Hyung
Single Carrier
Vehicular Techno
[3]. H. Ek
Parkvall, J. Torsn
Long-Term Evol
2006, pp. 38-45.
[4]. E. Da
Karlsson, M. L
Evolution - Rad
IEEE Vehicular
Melbourne, Austr
[5]. Long
website, www.3g
[6]. Hujun
A Broadband Wi
2006 (IEEE): pp.
[7]. Jianw
and Randall A. B
OFDMA s
r the mobile t
ncy and manu
hin a spe
, there are m
affect perform
at SC-FDMA
ared OFDMA
the effects
nd peak-to a
ossible subcar
calized FDMA
heduling (CDS
ved FDMA (IF
of IFDMA is
n we conside
djacent chann
ference betwe
PR performanc
ping increases
of raised-co
mpact on PAPR
be designed ca
ut degrading th
gress is an in
mation error on
A. Effective sc
about the freq
ing terminals
mation errors a
in channel pr
nce of CDS by
ation techniqu
users
REFER
n Nee and R. Pra
, Artech House, 2
g G. Myung, Jun
FDMA for Upl
ology Magazine,
kstrm, A. Furu
ner, and M. Wahl
lution, IEEE C
ahlman, H. Eks
Lundevall, and
io Interface Con
r Technology
ralia, May 2006.
Term Evolution o
gpp.org/Highlight
n Yin and Siavash
ireless Access Te
14. doi:10.1109
ei Huang, Vijay
Berry, Downlink
ystem.SC-FD
terminal in te
facturing cost
ecific SC-F
many design
mance in a com
A signals inde
A. In this p
of subcarrie
average power
rrier mapping
A (LFDMA)
S) results in hi
FDMA). How
better by 4 to
r the pulse sh
nel interferen
een LFDMA
ce. Another n
s PAPR and th
osine pulse
R of IFDMA.
arefully in ord
he system perf
nvestigation o
n the throughp
cheduling depe
quency respon
to an SC-FDM
are caused by
roperties. The
y causing inco
ue and incorrec
RENCES
asad, OFDM for W
2000.
nsung Lim, and
link Wireless T
vol. 1, no. 3, Sep
uskr, J. Karlss
lqvist,Technical
Commun. Mag., v
strm, A. Furus
S. Parkvall,Th
ncepts and Perfor
Conference (VT
of the 3GPP radio
ts/LTE/LTE.htm.
h Alamouti (Augu
chnology". IEEE
9/SARNOF.2006.
y G. Subramania
k Scheduling and
MA is grea
erms of transm
.
FDMA syst
and operation
mplex manner
eed have low
paper, we ha
er mapping
r ratio (PAPR
g approaches,
) with chann
igher throughp
wever, the PAP
o 7 dB than th
haping necessa
nce, we find
and IFDMA
noticeable fact
hat roll off fac
shaping has
A pulse shapi
der to reduce t
formance.
of the impact
put performan
ends on accur
nse of the rad
MA base statio
noisy estimati
e errors degra
orrect adaptati
ct assignment
Wireless Multime
David J. Goodm
Transmission, IE
. 2006, pp. 30-38
son, M. Meyer,
Solutions for the
vol. 44, no. 3,M
skr, Y. Jading,
he 3G Long-Te
rmance Evaluatio
TC) 2006 Spri
o technology, 3G
ust 2007). "OFDM
E Sarnoff Symposi
.4534773.
an, Rajeev Agraw
Resource Allocat
atly
mit
em
nal
r. It
wer
ave
on
R).
we
nel-
put
PR
hat
ary
d a
in
t is
ctor
a
ing
the
of
nce
rate
dio
on.
ion
ade
ion
t of
edia
man,
EEE
.
S.
3G
Mar.
, J.
erm
on,
ing,
GPP
MA:
ium,
wal,
tion
for OFDM Systems, IEEE Transactions on Wireless Communication,
Vol. 8, No. 1,January 2009.
[8]. Ellina Foutekova,Patrick Agyapong, and Harald Haas
Channel Asymmetry in Cellular OFDMA-TDD Networks, EURASIP
Journal on Wireless Communications and Networking, October 2008.
[9]. U. Sorger, I. De Broeck, and M. Schnell, Interleaved
FDMAA New Spread-Spectrum Multiple-Access Scheme, Proc.
IEEE ICC 98, Atlanta, GA, pp.10131017, June 1998.
[10]. R. Dinis, D. Falconer, C. T. Lam, and M. Sabbaghian, A
Multiple Access Scheme for the Uplink of Broadband Wireless
Systems, Proc. IEEE GLOBECOM 04, vol. 6, Dallas, TX, Dec. 2004,
pp. 3808-3812.
[11]. U. Sorger, I. De Broeck, and M. Schnell, Interleaved
FDMA - A New Spread-Spectrum Multiple-Access Scheme, Proc.
IEEE ICC 98, Atlanta, GA, Jun. 1998, pp. 1013-1017.
[12]. M. Schnell and I. De Broeck, Application of IFDMA to
Mobile Radio Transmission, Proc. IEEE 1998 International
Conference on Universal Personal Communications (ICUPC '98),
Florence, Italy, Oct. 1998, pp. 1267-1272.
[13]. Li-Chun Wang, and Anderson Chen, Optimal Radio
Resource Partition for Joint Contentionand Connection-Oriented
Multichannel Access in OFDMA Systems, IEEE Transactions on
Mobile Computing, Vol. 8, No. 2, February 2009.
[14]. G. Song and Y. Li, Cross-layer Optimization for OFDMA
Wireless Networks-PartII: Algorithm Development, IEEE Wireless
Commun., vol. 4, pp. 625634, Mar.2005.
[15]. Z. Han, Z. Ji, and K. Liu, Fair Multiuser Channel
Allocation for OFDMA Networks Using Nash Bargaining Solutions
and Coalitions, IEEE Trans. on Commun.,pp. 13661376, Aug. 2005.
[16]. R. Dinis, D. Falconer, C.T. Lam, and M. Sabbaghian, A
Multiple Access Scheme for the Uplink of Broadband Wireless
Systems, Proc. IEEE GLOBECOM 04, vol.6, Dallas, TX, pp. 3808
3812, Dec. 2004.
[17]. J. Lim, H.G. Myung, and D.J. Goodman, Proportional Fair
Scheduling of Uplink Single-Carrier FDMA Systems, to be presented
at The 17th Annual IEEE International Symposium on Personal, Indoor
and Mobile Radio Communications (PIMRC06), Helsinki, Finland,
Sep. 2006.
[18]. www.google.com
[19]. swww.wikipedia.com
ISBN: 978-81-920575-7-6 :: doi: 10.73872/ISBN_0768
ACM #: dber.imera.10.73872
Simulation of AODV Routing Protocol

Ashish Chaurasia
Asst.Professor
Gyan Ganga Institute of Technology & Sciences , Jabalpur

Abstract - The growth of laptops and 802.11/Wi-Fi wireless
networking have made MANETs a popular research
topic since the mid- to late 1990s. Many academic papers
evaluate protocols and abilities assuming varying degrees of
mobility within a bounded space, usually with all nodes
within a few hops of each other and usually with nodes
sending data at a constant rate. Different protocols are then
evaluated based on the packet drop rate, the overhead
introduced by the routing protocol, and other measures.
In this paper we are measuring the throughput of
AODV protocol in different scenarios.

.Keywords: MANET, Wi-Fi, routing, throughput, wireless
network

I. INTRODUCTION

A Mobile Ad hoc NETwork (MANET),
sometimes called a mobile mesh network, is a self-
configuring network of mobile devices connected by
wireless links. Each device in a MANET is free to move
independently in any direction, and will therefore
change its links to other devices frequently. Each must
forward traffic unrelated to its own use, and therefore be
a router. The primary challenge in building a
MANET is equipping each device to continuously
maintain the information required to properly route traffic.

Such networks may operate by themselves or may be
connected to the larger Internet.MANETs are a kind of
wireless ad hoc networks[1][6] that usually has a
routeable networking environment on top of a Link Layer
ad hoc network. They are also a type of mesh network,
but many mesh networks are not mobile or not wireless.
1.1 Mobile Ad-hoc Network
MANET [2][3][4] is a kind of wireless ad-hoc network
and it is a self-configuring network of mobile

routers (and associated hosts) connected by wireless links
the union of which forms an arbitrary topology. The
routers, the participating nodes act as router, are free to
move randomly and manage themselves arbitrarily; thus,
the network's wireless topology may change rapidly and
unpredictably. Such a network may operate in a
standalone fashion, or may be connected to the larger
Internet. Mobile ad hoc network is a collection of
independent mobile nodes that can communicate to each
other via radio waves. The mobile nodes can directly
communicate to those nodes that are in radio range of
each other, whereas others nodes need the help of
intermediate nodes to route their packets. These
networks are fully distributed, and can work at any place
without the aid of any infrastructure. This property
makes these networks highly robust.

Figure 1.1. MANET of 14 Mobile Nodes

II. AD-HOC ON DEMAND DISTANCE
VECTOR ROUTING (AODV)

Ad hoc On-Demand Distance Vector (AODV)
routing is a routing protocol for MANETs and other
wireless ad-hoc networks [3][7]. It is a reactive routing
protocol, meaning that it establishes a route to a
destination only on demand. In contrast, the most
common routing protocols of the Internet are
proactive, meaning they find routing paths
Parameter Value
Number of nodes 7
Simulation Time 15sec
Pause Time 5ms
Environment Size 1000*1000
Transmission Range 250m
Traffic Size VBR(VariableBit Rate)
independently of the usage of the paths. AODV is, as
the name indicates, a distance-vector routing protocol.
AODV avoids the counting-to-infinity problem of
other distance-vector protocols by using sequence
numbers on route updates, a technique pioneered
by DSDV. AODV is capable of
both unicast and multicast routing.

In AODV[7][8]the network is silent until a
connection is needed. At that point the network node
that needs a connection broadcasts a request for
connection. Other AODV nodes forward this message,
and record the node that they heard it from, creating an
explosion of temporary routes back to the needy node.
When a node receives such a message and already has
a route to the desired node, it sends a message
backwards through a temporary route to the requesting
node. The needy node then begins using the route that
has the least number of hops through other nodes.
Unused entries in the routing tables are recycled after
a time.

When a link fails, a routing error is passed back to a
transmitting node, and the process repeats.

Figure 2.1.: Working of AODV

III. SIMULATION & RESULTS

3.1 a Performance Metrics

Throughput The ratio of the number
of data packets sent and the number of data
packets received.
b.Simulation Parameters

We have taken one On-demand (Reactive)
routing protocols, namely AODV.. For all the
simulations, the number of traffic sources was taken
for 5 and 7 nodes, the maximum speed of the nodes
was set to 20m/s and the simulation time was fixed as
10s.

Scenario 1
Parameter Value
Number of nodes 5
Simulation Time 10sec
Pause Time 5ms
Environment Size 1000*1000
Transmission Range 250m
Traffic Size VBR ( Variable Bit
Rate)
Packet Size 512bytes
Packet Rate 5packets/s
MaximumSpeed 20m/s
Simulator ns2.31

Table 3.1: Scenario 1 for implementation of AODV

Figure 3.1: X Graph of throughput (AODV) for 5 nodes

Scenario 2
PacketSize 512bytes
PacketRate 5packets/s
MaximumSpeed 20m/s
Simulator ns2.31

Table 3.2: Scenario 2 for implementation of AODV

Figure 3.2: X Graph of throughput (AODV) for 7 nodes

IV. CONCLUSION

Throughput gain is maximum when AODV
protocol used for less nodes and decreases as the node
increases.

REFERENCES
[1]. Ian D. Chakeres and Elizabeth M. Belding- Royer.
"AODV Routing Protocol
Implementation Design." Proceedings of the International
Workshop on Wireless Ad Hoc Networking (WWAN),
Tokyo, Japan, March
[2]. Elizabeth M. Belding-Royer. "Report on the
a. AODV Interop." UCSB Tech Report 2002-18,
June 2002.
[3]. Elizabeth M. Royer and Charles E.
a. Perkins. "Transmission Range Effects
on AODV Multicast
Communication." To Appear in ACM
Mobile Networks and Applications
special issue on Multipoint Communication in
Wireless Mobile Networks.
[4]. Samir R. Das, Charles E. Perkins, Elizabeth M.
Royer and Mahesh K. Marina.
"Performance Comparison of Two On-demand Routing
Protocols for Ad hoc Networks." IEEE Personal
Communications Magazine special issue on
Ad hoc Networking, February 2001, p. 16-28.
[5]. Elizabeth M. Royer and
Charles E.
[6]. Perkins. "An Implementation Study of the
AODV Routing Protocol." Proceedings of the IEEE
Wireless Communications and
Networking Conference, Chicago, IL,
September 2000.
[7]. Samir R. Das, Charles E. Perkins,
and Elizabeth M. Royer. "Performance Comparison of
Two On-demand Routing Protocols for
Ad Hoc Networks."
Proceedings of the IEEE Conference on
Computer Communications (INFOCOM), Tel Aviv,
Israel, March 2000, p. 3-12.
[8]. Elizabeth M. Royer and
Charles E.
[9]. Perkins. "Multicast Operation of the Ad hoc
[10]. On-Demand Distance Vector
Routing Protocol." Proceedings of
MobiCom '99, Seattle, WA, August 1999, pp. 207-218.
[11]. Charles E. Perkins and
Elizabeth M.
[12]. Royer. "Ad hoc On-Demand Distance Vector Routing."
Proceedings of the 2nd IEEE Workshop on
Mobile Computing Systems and Applications, New
Orleans, LA, February
[13]. 1999, pp.
90-100.
ISBN: 978-81-920575-7-6 :: doi: 10. 73970/ISBN_0768
A MODIFIED CLUSTER BASED COOPERATIVE SPECTRUM SENSING
ALGORITHM USING DEMPSTER-SHAFER THEORY

Prof. Dr. A. V. Kulkarni, AliAsger S. Gabuji, Nikhil A. Agarwal, G. Pawan Kumar.
Dept. of Electronics & Telecommunication,
Pad. Dr. D. Y. Patil Institute Of Engineering & Technology,
Pune, India.

Abstract In cognitive radio systems, secondary users can be
coordinated to perform cooperative spectrum sensing so as to
detect the primary users more accurately. A variety of factors
such as hidden node problem, deep fading & shadowing
deteriorate the sensing performance of the cognitive users.
Cluster based sensing method addresses these problems to a
great extent. In this method the locations of all the users are
fixed where secondary users are separated into a few clusters
and the most favorable user in each cluster is selected to report
to the common receiver. The decisions are made by the
common receiver comparing with a single threshold value
where all the favorable users have equal priority in decision
making. In this paper we have proposed a weight based
decision making algorithm using double threshold technique
considering dynamic locations of the users. Numerical results
show that the sensing performance is improved significantly as
opposed to conventional method.
Keywords Dempster-Shafer Theory, Weighted clusters,
Collaborative decision making.
I. INTRODUCTION
Cognitive radio is a paradigm for Wireless Communication
in which either a network or wireless node changes its
transmission or reception parameters to communicate
efficiently avoiding interference with licensed or unlicensed
users. This alteration of parameters is based on the active
monitoring of several factors in the external and internal
radio environment, such as radio frequency spectrum, user
behavior and network state [1].
Todays wireless networks are characterized by a fixed
spectrum assignment policy. However, a large portion of the
assigned spectrum is used sporadically and geographical
variations in the utilization of assigned spectrum ranges
from 15% to 85% with a high variance in time. The limited
available spectrum and the inefficiency in the spectrum
usage necessitate a new communication paradigm to exploit
the existing wireless spectrum opportunistically. This new
networking paradigm is referred to as NeXt Generation (xG)
Networks as well as Dynamic Spectrum Access (DSA) and
cognitive radio networks.
The xG network functions are divided as spectrum
management, spectrum mobility and spectrum sharing.
From the architecture description, cognitive radio networks
correspond to networks where a node needs to sense its
surroundings to determine the proper free spectrum,
possibly with help from other cognitive nodes, and
simultaneously respecting the priority of the primary users.
[11].
The major motivation for this is the heavily underutilized
frequency spectrum. The development is being pushed
forward by the rapid advances in software defined radio
technology which enable a spectrum agile and highly
configurable radio transmitter/receiver.
II. SYSTEM MODEL
Distributed Spectrum Sensing (DSS)
Carrying out reliable spectrum sensing is a challenging
task for a CR. In a wireless channel, signal fading can cause
received signal strength to be significantly lower than what
is predicted by path loss models. The effect of fading can
result in the hidden node problem. [8] The hidden node
problem in the context of CR networks can be described as
an instance in which a secondary in a CR network is within
the protection region of an operating incumbent but fail to
detect the existence of the incumbent. Besides the hidden
node problem, it is also possible for a secondary to falsely
detect an incumbent because of noise or interference in the
wireless environment. Recent research results indicate that
these problems can be addressed by requiring multiple
secondarys to cooperate with each other in spectrum
sensingi.e., DSS. In DSS, each secondary acts as a
sensing terminal that conducts local spectrum sensing. The
local results are reported to a data collector (or fusion
center) that executes data fusion and determines the final
spectrum sensing result.[2][3] The application of DSS
requires that the distance between any two sensing terminals
is small relative to their distances from a primary transmitter
(typically a TV transmitter). In an ad hoc CR network, this
means that when a node needs to conduct spectrum sensing,
it becomes a data collector and collects local sensing reports
from neighboring nodes. The data are exchanged in a
control channel. The control channel can be either a
common control channel shared by all users (such as in the
MAC protocol for CR networks) or a local group control
channel that is used only by neighboring nodes for
communications.[4]

To facilitate our discussion, we model the DSS process as
a parallel fusion network. In this N
0
is a data collector,
N
i
(i = 0, 1, 2, . . . ,m, where m is the number of N
0
s
neighboring sensing terminals) denotes one of N
0
s sensing
terminals (N
0
is both a data collector and a sensing
terminal), y
i
represents the incumbent signal received at N
i
,
and u
i
is the local spectrum sensing report that N
i
sends to
N
0
. The output u is the final sensing decision, which is a
b
i
f
h
b
s
t
b

a
t
m
o
b
f

e
t

b

w
s
r
i
m
s
T
d
m

binary variab
incumbent sig
fusion proble
hypothesis tes
by H
0
and H
1
simplify the d
that spectrum
band. [2][3]
a.Decision fus
It re
threshold valu
m + 1 needs to
or equal to th
occupied, i
band is determ
i.e., u
Depe
fusion can hav
These rule
environments.
threshold tec
[5] [7]
b.Dempster-Sh
Here we set
which make
strengths, rec
reliable than a
implementatio
method in wh
strength value
The decisions
decision algor
making.
III
ble - a one
gnal, and a ze
em therefore
sting problem
1
. Correspond
discussions, th
m sensing is c
sion:
equires the dat
ue that is no l
o be specified
he threshold, t
.e., u = 1 an
mined to be fa
u = 0 and H
0
is
ending on the
ve several vari
A thresh
fusion rul
a value
rule,
and a va
fusion rul
es can only
. Therefore w
chnique based
hafer Theory -
t up two thresh
local decisi
ceived at all
all the other m
on a Weight b
hich lowest w
es falling betw
s are made
rithms both, pr
I. ALGORIT
e denotes th
ero denotes it
can be rega
with two hyp
dingly, each u
i
he following d
carried out in
ta collector to
ess than one a
d. If the sum of
then the final
d H
1
is accep
allow,
s accepted.
value of the
iants.
hold value of
le,
of (m+1) is
alue of (m+1/
le.

be formul
we have impl
d on Demps
-

holds for energ
ions dependin
the CRs. Thi
methods. To m
based techniq
weight is assi
ween the two
sequentially u
roposed as Co
THM / FLOW

he presence
ts absence. Th
arded as a b
otheses repres
i
is also binar
discussion ass
a single spe
o sum up all u
and no greate
f u
i
s is greate
sensing decis
pted; otherwis
threshold, de
f one is an
an AND f
/2) is a Maj
ated in noi
lemented a d
ster-Shafer th

gy detection at
ng on the
is method is
make it suitab
que is added t
igned to the
thresholds [5
using hard &
ollaborative de
WCHART
of an
he data
binary
sented
ry. To
sumes
ectrum
u
i
s. A
r than
er than
sion is
se the
cision
OR
fusion
ority
iseless
double
heory.
t SUs,
signal
more
ble for
to the
signal
5][12].
& soft
ecision
T
mo
dur
acti

Ass

The
is n
fall

c.Sc
T
clus
freq
CR
The proposed a
dules depict
ring the sensin
ion as specifie
F
sumptions :
Maximu
= 27 dB
Maximu
= 0 to -
BAND I
occupie
BAND
moderat
BAND
occupie
e geographica
named as a cl
ling in the clus
can Test :
The base statio
ster heads for
quency which
Rs. The BS c
algorithm is di
the activities
ng action. Th
ed below.

Fig 1. Flowch
um transmissi
Bm = 500 mW
um reception
-80 dBm = 10
I (eg:- 943.1 M
d frequencies.
II (eg:- 944
tely occupied
III (eg:- 94
d frequencies.
al region is div
luster. A datab
sters is collect
on (BS) condu
r respective c
h is received
calculates the
ivided into six
s undertaken
hus they repre
hart / Modules
on power of
W. [9]
power of a
0 pW. [9]
MHz to 944 M
.
4.1 MHz to
frequencies.
0 to 943 M
. [10]
vided in k sect
base of all the
ted and update
ucts a scan tes
clusters. The B
& re-transm
distance of
x modules. The
by the syst
sent the flow

a cellular pho
a cellular pho
MHz) as spars
946 MHz)
MHz) as heav
tors. Each sec
e cognitive us
ed.
t for designati
BS broadcasts
mitted by all t
each CR usi
ese
em
of
one
one
ely
as
vily
ctor
ers
ing
s a
the
ing
which it calculates the weight of each CR of each cluster by
using the following formula :

Weight[i] = distance[i] * signal strength[i]

The CR having maximum weight is designated as the
cluster head of the respective cluster.

Weights of the individual clusters are calculated as follows :
Average weights of all the CRs in a cluster is calculated
& the cluster having maximum average weight is assigned
the highest priority in decision making.

The spectrum is sequentially scanned by all the CRs in the
respective order between the ranges,
BAND I as sparsely occupied frequencies,
BAND II as moderately occupied frequencies,
BAND III as heavily occupied frequencies.

If all the frequencies are occupied, then the network stays
congested which is a rare case.

Using the Dempster-Shafer Theory & Hard/ Soft decision
concept,
If the signal strength received by the CR <= -70 dBm , CR
sends a '0' to its cluster head
If the signal strength received by the CR >= -10 dBm , CR
sends a '1' to its cluster head
If the signal strength received by the CR lies between -10 &
-70 dBm, CR sends the received signal strength to the
cluster head.
Now the cluster head calculates the % of CRs concluding
primary user is present or absent.

A = [(no. of CRs concluding '0') * 45] +[(no. of CR's
receiving signal strength between (-70 to -45 dBm)) * 10]
P = [(no. of CRs concluding '1') * 45] + [(no. of CR's
receiving signal strength between (-45 to -10 dBm)) * 10]

Decision by Cluster Head
Primary present = [P / (A+P)] * 100 %
Primary absent = [A / (A+P)] * 100 %

These decisions are sent to the BS for analysis.

No. of clusters : k
CRs in a cluster : n
1
, n
2
n
k

Now, x = 1,2,.,k
n
x

Avg wt. of CR[x] = [ d[i] * s[i] ] / n
x

i = 1

Let the clusters after rearranging according to their weights
in descending order be as follows

W
k
, W
k-1
, W
k-2,..
., W
1

Final decision by the BS,

= ( decision of cluster * weight of cluster )

1
R
Wi
* i
i = k
Final decision = _____________
1
i
i = k

where R
Wi
= decision of cluster [i] sent to the BS

The frequency hole scanned is allocated to the CR having
highest priority as per its subscriber authorization plan.

Priority
No.
Category Charges per call
1 (Highest) Gold Extra
2 (Medium) Silver Extra
3 (Lowest) Bronze No Extra charges

IV. SIMULATION

The following representations denote the simulation
results of the modules.

Fig. 2 & Fig. 5 represent cluster formation & cluster head
designation.


u

m
Fig2. Cluste
Fig. 3 & Fig
Station.
Fig 3. D
The Overall
using 100 CRs
Fig. 4 & Fig
making for dif
9%
DE
er Formation
(4
g. 6 represent t
Decision Makin
Efficiency for
s = 91 %
g. 7 represent
fferent numbe
ECISION OF

and Cluster H
clusters)

the decision m

ng For 100 Co

r Sensing on

the efficiency
er of Cognitive

91%
THE BASE S
D
P
N
Head Designat
made by the Ba
ognitive Radio
ne Frequency
y in decision
e Radios.
STATION
DECISION (P.U.
PRESENT/ABSE
NO RESULT

tion
ase
os
Hole
F

.
ENT)
Fig 4. Eff
Fig5. Cluster F
0
20
40
60
80
100
10 20
2%
DEC
fficiency For

Formation and
(8 clu

0 30 40 50
Efficie
CISION OF T
100 Cognitive
d Cluster Hea
usters)
0 60 70 8
ency
98%
THE BASE S
e Radios
ad Designation
80 90
STATION
DECISION (P.U
PRESENT/ABS
NO RESULT

n
U.
ENT)
u
o

b
e
c
t
h
i
F
w

Fig 6. D
The Overall
using 200 CRs
Fig 7
The efficiency
of Cognitive R
In this pape
based coope
encouraging
collaborative
technique by
heads. The l
increasing th
Fast fading w
work where re
[1] J.
ma
0
20
40
60
80
100
10
Decision Makin
Efficiency for
s = 98 %
7. Efficiency F
y of Spectrum
Radios increas
V. CO
er, we have m
erative sens
results. The
decision m
acquiring re
load on the
he efficiency
would be an int
eceiver sensiti
VI.R
III Mitola and
aking software
00 110 120 13

ng For 200 Co

r Sensing on

For 200 Cogni

Sensing incre
ses.

ONCLUSION
modified the c
sing algorith
main base
making using
sults from th
common ch
of sensing to
teresting exten
ivity can also b
REFERENCES
G. Q. Jr. Mag
radios more
0 140 150 160
Efficienc
ognitive Radio
ne Frequency
itive Radios
eases as the nu
conventional c
hm & obt
station per
double thre
he weighted c
hannel is red
o a certain lim
nsion to the p
be incorporate
S
guire, Cognitive
personal, P
0 170 180 190
cy
os
Hole
umber
cluster
tained
rforms
eshold
cluster
duced,
mit[6].
resent
ed.
e radio:
Personal
[
[
[

0 200
Comm
Aug. 1
[2] C. Sun
coope
Procee
Comm
[3] A. Gh
for op
DySPA
[4] S. Hay
comm
Comm
[5] J. Zhu
thresh
in cog
Cogni
(Crow
[6] D. Cab
Implem
radios
Comp
[7] Coope
Detect
Liu Sc
Unive
(UEST
[8] F. F. D
energy
in Pro
(IEEE
3575
[9] Mouly
comm
10] Cogni
Alexa
Engin
11] NeXt
radio
Yeol
Broad
Electr
Techn
12] Colla
with n
B.Eng
munications, IEEE
1999.
n, W. Zhang, and
erative spectrum s
edings of IEEE In
munications, Glasg
hasemi and E. Sou
pportunistic access
AN, Baltimore,MD
ykin. Cognitive ra
munications. IEEE
munications, vol. 2
u, Z. Xu, F.Wang
hold energy detec
gnitive radio, i
itive Radio Orie
wnCom), pp. 1-5, 2
bric, S. M. l\4isha
mentation issues
s. in Proc, of Asilo
puters, pages 772
erative Spectrum
tion and Dempst
chool of Commun
ersity of Electron
TC) Chengdu 611
Digham, M.-S. Al
y detection of unk
oc. IEEE Internati
E ICC 2003), vol.
3579.
y M., Pautet M.
munications, 1992
itive Radio Te
ander, R. David
neering, Indian Ins
generation/dynam
wireless network
Lee, Mehmet
dband and Wirele
rical and Comput
nology, Atlanta, G
aborative spectrum
non-gaussian nois
gg., Lakehead uni
E, Volume 6, Issu
K. B. Letaief, C
sensing in cognitiv
nternational Conf
gow, Scottland, Ju
usa, Collaborativ
s in fading enviro
MD, USA, p.p. 131
adio: brain-empow
E Journal on Selec
23:201220, Feb
g, B. Huang and
ction of cooperati
in Proc. of Inte
ented Wireless N
2008.
ar, arid R. W. Bro
in spectrum sensi
omar Conf on Sig
776, November
Sensing Based o
ter-Shafer Theory
nication and Infor
nic Science and T
1731, China
louini, and M. K.
known signals ov
ional Conference
5, no. 4, May 11
B., The GSM
2.
echniques for G
Koilpillai Depar
stitute of Technol
mic spectrum acce
ks: A survey Ian
C. Vuran *, S
ss Networking L
ter Engineering,
GA 30332, United
m sensing in a cog
se by Tan Feng
versity, 2006
ue 4, Page(s):13 -
Cluster-based
ve radio systems
ference on
une 2007.
ve spectrum sensin
onments, Pro. IE
136, Nov. 2005.
wered wireless
cted Areas in
bruary 2005
d B. Zhang, Dou
ive spectrum sens
ernational Conf.
Networks Comm
odersen.
ing for cognitive
gnals, System and
r 2004.
on Double Thresh
y Hongyu Chen J
rmation Engineeri
Technology of Ch
Simon, On the
er fading channel
e on Communicati
15, 2003, pp.
M system for mob
GSM Band B
rtment of Electr
logy, Madras.
ess/cognitive
n F. Akyildiz, W
Shantidev Moha
aboratory, Schoo
Georgia Institute
d States.
gnitive radio Syst
18,
, in
ng
EEE
uble
sing
on
mun.
d
hold
Jian
ing,
hina
ls,
ions
bile
aiju
rical
Won-
anty
ol of
e of
tem
ISBN: 978-81-920575-7-6 :: doi: 10. 73977/ISBN_0768
Protocol Design for Advanced Security in Wireless Networks

Raghuveer Singh Dhaka
Department of Computer Science & Engg.
Central University of Rajasthan
Ajmer, India

Arvind Dhaka
Department of Computer Science & Engg.
Central University of Rajasthan
Ajmer, India

AbstractThis paper is focused to investigate the security
protocols which are used in wireless network. The Protocols
like WEP, WPA or WEP2 are considered under the IEEE
802.11 standard. Throw the analysis of the protocols we
determine that the advantages and disadvantages. And for the
remaining vulnerabilities we are proposing an effective
solution which is more secure and unpredictable.
Keywords- WEP; WPA; MIC;AES;EAP;IV;WPA-2; MSS.
I. INTRODUCTION
The wireless network security is in a state of chaos.
Everyday a new vulnerability is discovered in almost every
existing protocol. The answer to these problems lies in
enhancing our design methodology in addition to
implementing and testing the protocols. In this paper, a new
mechanism is proposed and discussed. The proposed
mechanism will be compared with a number of exiting
working mechanisms in wireless networks.
II. ISSUES AND SECURITY ATTACKS
In this section, issues and security attacks in wireless
networks security protocols are discussed.
A. Modification
It is an active attack where the content of the original
message are modified by the intruder and then sent it instead
of the actual message.
B. Fabrication
It is an active attack where non original message is
generated by intruder and then sent to the receiver.
C. Interception
It is a passive attack where the encrypted message is
listens and records by the intruder and later on decrypt by
intruder.
D. Brute Force
It is a passive attack where the intruders generates every
possible permutation in the key and try to decrypt the
encrypted message.
E. Maintainability
For pages other than the first page, start at the top of the
page, and continue in double-column format. The two
columns on the last page should be as close to equal length
as possible.
III. WIRELESS SECURITY PROTOCOL
A. Wired Equivalent Privacy (WEP)
WEP was developed to provide wireless security to the
wireless users on WLAN that is equivalent to being on a
wired network. WEP, not being intended to provide a higher
security level than a wired LAN, provides equivalent
security level to that by wired LANs. RC4 Stream Cipher is
used to provide confidentiality and CRC-32 checksum is
used to provide Integrity (honesty).WEP uses a 40 bit key
size 24 bit IV (Initialization Vector).
B. Wired Equivalent Privacy2 (WEP 2)
It is same as WEP. The size of Initialization Vector is
same means 24 Bits. But the key size is expended to 104 Bit
[5].
C. Wi-Fi Protected Access (WPA) Protocol
WPA protocol was designed to provide security to the
Wi-Fi networks. It is considered under the IEEE 802.11i
standard. It was developed by Wi-Fi Alliance. There is TKIP
(Temporal Key Integrity Protocol) which provides more
security to change the dynamic key [10]. The size of
Initialize Vector (IV) is just double in compare to WEP,
which is 48 Bit. And the key size is 128 bit.
D. WPA-2 (Robust Security Networks- RSNs)
MAC is replaced by CCMP and RC4 is replaced by AES.
There are 2 mod, I mode is known as pre shared key mode
and II mode is known as full Extensive Mod/WPA-
Enterprise. 802.11i that has been appeared in 2004 uses the
logic of a Robust Security Network (RSN). This new
standard and architecture adopt the IEEE 802.1X standard
for access control and for encryption it uses Advanced
Encryption Standard (AES) [6]. It uses a pair-wise key
exchange (four way handshake) protocol utilizing 802.1X for
mutual authentication and key management process. 802.11i
allows use of TKIP, but by default RSN uses AES
(Advanced Encryption Standard) and CCMP (Counter Mode
CBC MAC Protocol). It provides a stronger, scalable
solution.

E
R
M
r
d
o
M
o
m
w
TABLE I.

The propo
System (M
Encryption/De
RSA, Blowfis
MSS uses 256
random in the
differentiate b
original name
MSS stores a
of different slo
means that 8
25
which is trying

COMPARISON
SECURIT
IV. PROPO
osed mechani
MSS). MSS
ecryption alg
sh, Elliptic Cu
6 slots. One o
ese slots. Ther
between these
e of the algori
key list. Each
ot algorithm an
56
possible com
g to attack.
Figure 1.
N BETWEEN WEP
TY PROTOCOLS
OSED MECHAN
ism is known
S uses
gorithms RC4
urve, AES, H
of the eight al
e is a 2-bit co
eight algorith
ithm. For the
h message enc
nd a different
mbination req
Slot System.
P, WPA AND WPA
NISUM
n as Multiple
one of
4, Diffie-Hel
HMAC and CM
lgorithms is u
ode which is u
hms in place
se eight algor
crypts with the
key by MSS, w
quires to any h
A2

e Slot
eight
llman,
MAC.
used at
used to
of the
rithms
e help
which
hacker

(KS
MIC
bit
par
256
shu
mo
A.
use
the
App
con
Ext
auth
2-b
bits
one
whi
the
1
are
wil
star
nod
algo
key
hea
MSS header h
S) is first part.
C shuffle Sele
long which is
rt is 8-bit long
6 slots. The th
uffle tables, the
ve quickly the

Fig
MSS Operatio
Different con
er, which mean
user is affe
plication laye
nfiguration fil
tensible Authe
hentication ser
The configura
bit long), whic
s only. Eight k
e of the eight
ich is the shuf
MIC generate

1) Sending P
used to choos
l be shuffled
rting from the
de will encry
orithm for the
y. The node w
ader to messag
has divided in
The slot selec
ector (MICSS)
used to select
g which is used
hird part is 8-b
e shuffle table
e MIC bits with
gure 2. MSS He
on
nfiguration file
ns that the netw
ected. MSS
er. The access
es. The AP
entication Pro
rvices.
ation file cont
ch means that
key lists files
t algorithms. O
ffle table. Its w
es and its lengt
Phase: A slot a
se the user or
with the cho
e first bit unti
ypt the new
e slot that it
will concatenat
ge and then sen

n three parts. T
ctor (SS) is sec
) third part. Th
t between 2
48
k
d to select bet
bit long to sel
es are used ma
h the original
eaders & Payloa
es are used by
work will not b
is going to
s point (AP)
act like a co
otocol (EAP)
tains the 256 s
the slots part
are needed al
One more file
width is the nu
th is 2
64
.
and a key from
the node. The
osen row insi
l the last bit i
payload with
was chosen u
te the SS, KS
nd the messag
The key Selec
cond part and t
he first part is 4
keys. The seco
tween one of t
lect between 2
inly to shuffle
message.
ad.
y MSS for ea
be affected wh
operate on t
contains all t
onversion serv
used to prov
slots (each slot
will need 2*2
lso, one for ea
e is also need
umber of bits th
m the list of ke
e generated M
de the messa
in the MIC. T
h the appoint
using the chos
and MICSS a
ge.
ctor
the
48-
ond
the
256
e or
ach
hen
the
the
ver.
ide
t is
256
ach
ded
hat
eys
MIC
ge,
The
ted
sen
as a
i
M
k
s
The select
is done system
MSS will run
key in the key
slot with the f

a.Receiving
separate the
SS to know
message. Th
data, the re
will use th
shuffle tabl
receiver wil
MIC). The r
part and c
received to
b.Control m
will act as a
a refresh of
all nodes. T
handle the
configuratio
will send a
AP will de
control serv
them to the
itself and s
will receive
packet to th
the new ver
files with th
Figure 3. Shu
tion of the slot
matically, whi
n it will use th
y list of that al
first key in his
g Phase: In th
e SS, KS and
w the algorithm
he appointed k
ceiver will de
e MICSS to
le and will un
ll split the pay
receiver will g
compare the
insure the inte
messages and a
a conversion se
f the configur
The AP must h
conversions.
on files, key li
re-issue cont
liver it to the
ver will issue
e AP. The AP
send a copy to
e the configu
he AP, so that
rsion that it ha
he new version

uffeling Mechan
t and the key a
ich means tha
he first slot an
lgorithm. Seco
list and so on
he receiver sid
d MICSS from
m that will be
key list of this
ecrypt the mes
get the shuff
ndo the shuff
yload to two
generate the M
generated M
egrity of the m

access points o
erver. It has th
ration files for
have a good p
MSS needs s
ists and shuffl
trol packet to
e MSS contro
e the new file
will keep a c
o the appoint
uration files a
t it will replac
ad. The node
n and reset all
nism.
and the shuffl
at the first tim
nd the first po
ond time the s
n until the end.
de, the receive
m the message
used to decry
s algorithm an
ssage. The rec
fling row from
fling, after th
parts (messag
MIC for the me
MIC with the
message.
operations: Th
he authority to
r one node or
processing pow
server to issue
fle tables. Any
the AP, whic
ol server. The
es needed and
copy of the fil
ed node. The
and issue a c
ce the old files
will replace th
the counters.

e data
me the
ointed
econd
.
er will
e. The
ypt the
nd KS
ceiver
m the
hat the
ge and
essage
e one
he AP
o issue
r even
wer to
e new
y node
ch the
MSS
d send
les for
e node
ontrol
s with
he old
Slo
one
hac
slot
nod
per
com
per
bec
mea
app
wil
MS
shu
Tim
refe
bra
nam
num
sen
the
ref
Re
actu
MSS depend
ots and permut
e but four alg
cker or intrude
ts configuratio
de in the netw
rmutation. It
mbined with
rmutation of th
cause if one o
ans, MSS c
pearance of tha
l weaken the
SS has solved
uffling the MIC
TABLE II.

List and num
mes, single-sp
erenced in the
ackets, for exa
me(s) of edito
mber citations
ntence punctua
reference nu
ference [3]
eference [3] w
Number foot
ual footnote a
V. CON
d on Permutat
tation of Shuff
gorithms to en
er needs to dev
ons, key lists
work. MSS doe
depends on t
the permuta
he shuffling tab
f the algorithm
can counter
at algorithm in
strength of th
is the static p
C bits inside th
COMPARISON BET
REFE
mber all bibliog
paced, at the
e text, enclose
ample [1]. W
ors of referenc
s consecutive
ation follows th
umber, as in [
except at th
was the first . .
tnotes separate
at the bottom o
NCLUSION
tion of keys,
fling Tables. B
ncrypt and de
vise a mean to
and shuffling
es not depend
the permutati
ation of the
bles. MSS is a
ms is found to
its effect b
n the configur
he MSS. Anoth
placement of
he original me
TWEEN WEP 2, W
ERENCES
graphical refer
end of you
the citation n
Where appropri
ced books. Th
ely within br
he bracket [2]
[3]do not u
he beginning
.
ely in supersc
of the column
Permutation
By using not on
ecrypt with, a
o guess the ex
g tables for ea
only on the k
ion of the ke
slots and t
also maintainab
o broken by a
by denying t
ration slots. T
her problem th
the MIC bits
ssage.
WPA 2 AND MSS
rences in 9-po
ur paper. Wh
number in squ
iate, include t
he template w
rackets [1]. T
. Refer simply
use Ref. [3]
of a senten
cripts. Place
n in which it w
of
nly
any
xact
ach
key
eys
the
ble
any
the
This
hat
by

oint
hen
are
the
will
The
y to
or
nce:
the
was
cited. Do not put footnotes in the reference list. Use letters
for table footnotes.
Unless there are six authors or more give all authors
names; do not use et al.. Papers that have not been
published, even if they have been submitted for publication,
should be cited as unpublished [4]. Papers that have been
accepted for publication should be cited as in press [5].
Capitalize only the first word in a paper title, except for
proper nouns and element symbols.
For papers published in translation journals, please give
the English citation first, followed by the original foreign-
language citation [6].

REFERENCES

[1] IEEE: IEEE 802.11: Wireless LAN Medium Access Control (MAC)
and Physical Layer (PHY) Specifications (2007),
http://standards.ieee.org/about/get/802/802.11.html.
[2] Sepehrdad, P., Vaudenay, S., Vuagnoux, M.: Discovery and
Exploitation of New Biases in RC4. In: Biryukov, A., Gong, G.,
Stinson, D.R. (eds.) SAC 2010. LNCS, vol. 6544, pp. 7491.
Springer, Heidelberg (2011).
[3] Sepehrdad, P., Vaudenay, S., Vuagnoux, M.: Statistical Attack on
RC4 Distinguishing WPA. Accepted at EUROCRYPT 2011 (2011).
[4] Vuagnoux, M.: Computer Aided Cryptanalysis from Ciphers to Side
channels. PhD thesis, Ecole Polytechnique Federale de Lausanne
EPFL (2010).
[5] Wpa-Wpa2, http://www.lylebackemort.com/blog/2008/5/10/ wpa-
wpa2-as-insecure-as-i-expected (accessed on March 14, 2010).
[6] A. H. D. Lashkari, M.M.S. , Samadi, B., "A survey on wireless
security protocols (WEP, WPA and WPA2/802.11i)," Computer
Science and Information Technology, 2nd IEEE International
Conference on Beijing, ICCSIT 2009, pp.48-52, 2009..
[7] E.J.M.A. Filho, P.N.L. Fonseca, M.J.S. Leitao, and P.S.F. de Barros,
Security versus Bandwidth: The Support of Mechanisms WEP &
WPA in 802.11g Network, IFIP International Conference on
Wireless and Optical Communications Networks, WOCN '07, pp. 1-
5, 2007.
[8] M. Bloch et al., Wireless Information-Theoretic Security, IEEE
Trans. Info. Theory, vol. 54, no. 6, June 2008, pp. 251534.
[9] S. Mathur et al., Radio-Telepathy: Extracting a Cryptographic Key
from an Unauthenticated Wireless Channel, 14th ACM MobiCom,
2008.
[10] TecTarget. (2008). Wi-Fi Protected Access. Retrieved 05 May, 2008,
from http://searchmobilecomputing.techtarget.com/sDefinition/
0,,sid40_gci887323,00.html.

A
t
d
d
h
c
m
a
s
s
w
d
s
A
i
W
P
p
u
B
w
w
s
s

a
I
o
c
c
s
t
Proc. of th
Volume 1. Co
ISBN: 978-8
Raghuve
Department o
Central Uni
Aj
Abstract A
technology the
different secur
data are differ
have to gain ac
communication
medium is no
accessed with
security of wi
since its deve
wireless netwo
different attac
security mecha
Keywords-
I. WLA
A. WLAN Sec
WEP is th
implemented.
WEP by cryp
Protected Ac
passed, RSN
under the full
B. Wired Equ
WEP was d
wireless users
wired network
security leve
security level
1) How W
and every da
Stream ciphe
Initialization
original pack
cipher which
cipher stream
sender can ch
to the receive
he Intl. Conf. o
opyright 20
81-920575-7-6
ACM #: dber
eer Singh Dh
of Computer S
& Engg.
iversity of Raj
jmer, India

wired and w
ey use to tran
rity aspects. M
rent in both te
ccess over the p
n i.e. wires. B
one other than
the help of equ
reless network
elopment. The
orks is in air
ks. Thus it is v
anisms to wirel
WEP; WPA; M
AN SECURITY
curity in Gene
he foremost W
Several wea
tanalysts, and
ccess (WPA)
(also known
IEEE 802.11i
uivalent Privac
developed to
s on WLAN
k. WEP, not b
l than a wi
to that by wir
WEP works: W
ata packet is
er. This key
Vector (IV)
et is then bit
results in an
m is used to en
hose a differen
er. We further
on Computer A
12 Techno Fo
6 :: doi: 10. 73
r.imera.10. 73
A Rev
haka
Science
asthan
wireless netw
nsmit data, th
Mechanisms to
echnologies. In
physical mediu
But in wireless
n air, the radi
uipment availa
ks has always
e transmission
r which mak
very difficult to
less networks.
MIC;ICV;IV;CC
PROTOCOLS A
eral
WLAN securit
aknesses were
d WEP was tak
in 2003, an
n as WPA2)
i standard [7, 9
cy (WEP) Mec
provide wirel
that is equiva
eing intended
ired LAN, p
red LANs.
When WEP is
encrypted se
y is genera
and a 40- b
twise XORed
encrypted pla
ncrypt each ne
nt IV periodica
compute 4-by
Applications
rum Group, In
3984/ISBN_07
984
view of W
WEP, W
Departm
Central
work differs i
ereby implem
o access transm
n wired networ
um used for ne
s networks sin
io frequency c
able in market
been a major
n of informati
kes it vulnerab
o provide up t
CMP;TKIP;MA
AND MECHANI
ty mechanism
e identified i
ken place by
nd then after
is ratified in
9].
chanism
less security t
alent to being
to provide a h
provides equiv
s enabled then
parately with
ated by a
bit WEP key
d with RC4 s
aintext. A dif
ew packet sinc
ally. IV is then
yte Integrity C

ndia.
768
Wireless Se
WPA and
Arvind Dhak
ment of Compu
& Engg.
University of
Ajmer, India

in the
enting
mitted
rks we
etwork
ce the
can be
ts. The
r issue
ion in
ble to
to date
AC.
SMS
to be
in the
Wi-Fi
r time
2004
to the
g on a
higher
valent
n each
h RC4
24-bit
y. The
stream
fferent
ce the
n sent
Check
Val
also
enc

2
RC
enc
resu
stre
suc
WE

A
32
(Cy
"ch
stro
dig
be
mea
Sha
aga
WE
leak
pos
eno
ecurity Pr
d WPA-2
ka
uter Science
f Rajasthan
a
lue (ICV) on
o encrypt IC
cryption mecha
2) Vulnerabi
C4 stream ciph
cryption. This
ults in the nee
eam can be
ccessful attack
EP makes it a f
Another vulner
mechanism w
yclic Redunda
hecksum". CR
ong and thus n
est or hash fu
noted that th
asure in WEP
amir Attack d
ainst the key
EP. Also the i
ked by the f
ssible to recon
ough WEP-enc
rotocols:
Mu
Depa
Cen
the original p
CV with the
anism is show
Figure 1. Wep
lities and fla
her with a 24
short size IV
ed to reuse the
exploited by
k on the ciph
failure to achi
rable aspect o
which is used
ancy Code)
RC is not c
not efficient en
unctions such a
he use of CR
P, which is n
describes a pa
scheduling a
information ab
first byte gen
nstruct the sec
crypted packet
ukul Manmoh
artment of Com
& Eng
ntral Universit
Ajmer, I
packet and app
RC4 cipher
wn in following
p Encryption.
aws of WEP:
4- bit Initializa
V is a weakne
key stream. T
y the attacker
her text. This
ieve the securi
of the WEP is
for the integr
is considered
considered c
nough to be us
as MD5 or SH
C-32 was use
not true. The
assive cipher-
algorithm of R
bout individua
nerated by RC
cret key in WE
ts.
han Meghwa
mputer Scienc
gg.
ty of Rajasthan
India
pend with it. W
r stream. W
g diagram.

WEP uses t
ation Vector
ess of WEP as
This reuse of k
r to perform
vulnerability
ity goals.
the use of CR
rity check. CR
d as a class
cryptographica
sed as a messa
HA [1]. It shou
ed as a secur
Fluhrer Man
-text-only atta
RC4 as used
al key bytes g
C4. It could
EP by analyzi
al
ce
n
We
WEP
the
for
s it
key
m a
of
RC-
RC
of
ally
age
uld
rity
ntin
ack
in
gets
be
ing
The authentication flaw in the IEEE 802.11 has been
shown in previous works. A simple successful
eavesdropping attack against IEEE 802.11also was
demonstrated with the WEP enabled.
Both RC4 and IVs are not used efficiently in the standard
WEP. Because of these reasons we conclude that 802.11
WEP is totally insecure and the reason is nothing but poor
implementation of secure technologies (such as RC4). WEP
was not anticipated to be a good solution for all security
needs.

3) The weaknesses of WEP
a) Direct use of Master keys: The Use of master keys is
not directly recommended. The other temporary keys are
generated by Master keys.
b) Small key sizes: The WEP uses 40 bits of key size,
which has been considered as one of the major flaw in the
WEP. In 1997, 40-bit key size was considered to be
reasonable for some applications. At that time the goal was
to provide protection against the casual eavesdropping, it
was seemed good to provide sufficient protection at that
time. There is no any specification of WEP key sizes other
then 40 bits in the 802.11 standard. But most dealers apply
the key size to 104 or 232 bits. RC4 encryption key includes
a 24-bit IV in both cases. 104 and 232-bit keys are more
defensive against to brute-force attacks in comparison to the
40-bit keys but it is still a problem which is not resolved yet.
c) Not have enough key management: WEP standard
does not specify key management. Without interoperable
key management, keys will be of inferior quality and lives
for a longer time. Thus this is considered as its weakness.
Most of the wireless networks implementing WEP shares a
single WEP key between each and every node on the
network. Same WEP key is used to program the access
points and client stations. The task of changing the keys is
anonymous and difficult.
d) Use of RC4: Implementation of RC4 has been
considered to have weak keys, indicating more correlation
between the key and the output. It is easy job to determine
packets encrypted with weak keys. The first three bytes of
the key are extracted from the IV that is sent unencrypted in
each packet. This weakness can be exploited using a passive
attack.
e) Reused and small sized IV: We found that out of the
available 16 million IV values 9,000 are interesting. They
show the presence of weak keys. The attacker captures
"interesting packets" to find out the weak keys, then
analyzes them and only has to try a small number of keys to
gain access to the network. Because all original IP packets
start with a known value, its easy to know when user has
the right key. To determine a 104-bit WEP key, attacker has
to capture the interesting packets between 2,000 and 4,000.
On a busy network the capture of the 5000 interesting
packets might not cause any difficulty and this can be done
in a very short time period. Many dealers are now
implementing new algorithms that are simply based on no
selection of weak IV. The best solution of this type of attack
is not to use weak IV values.
For a WEP key, despite of the key size, 24- bit long IV
can only provide 16,777,216 different RC4 cipher streams.
This number can be achieved on a busy network in a few
hours and we can reuse the same IV. The RC4 stream cipher
is XORed with the original packet in the WEP and the IV is
sent in the clear format with each packet [3]. If the RC4
stream cipher for a given IV is found, subsequent packets
encrypted with the similar IV can be decrypted by an
attacker. There are maximum 16,777,216 IV values.
Unluckily, WEP doesnt specify such facility that how to
choose or how often to change IVs. Some assumes that IV
starts at zero and increases incrementally for each packet,
and return back to zero after sending 16 million packets.
And some other chooses IVs randomly. The IV chosen
randomly leads to 50% chance of reuse after less than 5,000
packets.
f) Weakness of ICV algorithm: WEP ICV is based on
the CRC-32 algorithm usually implemented to detect noise
along with common errors in transmission. CRC-32 is not
considered as a good choice for cryptographic hash. The
CRC-32 ICV is a linear function of the message. This
means that a message can be encrypted and the ICV be
easily fixed, by an attacker, making the message appear
authentic. To do this an encrypted packet stream is captured
and the destination address of each packet is changed with
the attackers wired IP address, fixing up the CRC-32.These
packets are then retransmitted to the access point. The
packets will decrypt by the access point and after that
forward them to the attacker. IV and ICV based attacks are
not depend on the key size.
g) Easy forging of authentication messages: There are
two types of authentication methods which are defined in
802.11 Standards. One is known as Open System
authentication and another is known as Shared Key
authentication. If the WEP authentication service is enabled,
then the intruders and attackers can guess the WEP key [6,
10, 11]. In Shared Key authentication the concept of shared
WEP key is used to encrypt the challenge. With this comes
the possibility that the challenge and encrypted response can
be observed by an attacker who is monitoring the network
traffic. By monitoring the successful authentication, an
attacker can forge an authentication. Shared Key
authentication limits the ability of an attacker to create a
denial-of-service attack by sending garbage packets
(encrypted with the wrong WEP key) into the network.
WEP is far from an ideal security solution but it is still used.
C. Wi-Fi Protected Access (WPA) Protocol
WPA has been developed to provide more wireless
security solution for 802.11-based networks. WPA is a
replacement over Wired Equivalent Privacy (WEP) [2]. The
industries and organizations have been spending a great deal
of time and money on developing and deploying next-
generation solutions that address growing wireless network
security problems. The Wi-Fi (Wireless Fidelity) Alliance
has created the Wi-Fi Protected Access (WPA) [4]. WPA is a
subset of the 802.11i standard.
An enhancements over WEP: WPA is an enhancement over
pre-existing protocol WEP.WPA technology mainly
includes three improvements over WEP.
a) Improved data encryption: Temporal Key Integrity
Protocol (TKIP) is an improvement of data encryption.
TKIP is a Temporal Key Hash Function. TKIP is an
alternative to WEP. It fixes all the security problems and
doesnt require new hardware. TKIP uses the RC4 stream
cipher as the encryption and decryption processes and all
involved parties must share the same secret key similarly as
like in WEP [11]. This secret key must be 128 bits and is
called the "Temporal Key" (TK). TKIP also uses an
Initialization Vector (IV) of 48-bit. The IV used as a
counter. Even if the TK is shared, all involved parties
generate a different RC4 key stream.
b) User authentication: User authentication is missing
in WEP, through the extensible authentication protocol
(EAP). WEP controls access to a wireless network which is
based on a computers hardware-specific MAC address.
EAP is built on a more secure public-key encryption system
to make it certain to happen that only authorized network
users can access the network.
c) The quality of being honest: Message Integrity Code
(MIC) for TKIP is computed by a new algorithm. Message
Integrity Code (MIC) is computed to determine errors in the
data complacency either due to transfer errors or due to
purposeful alterations [5]. It is a 64-bit MIC that is added to
the Data and the ICV. The ICV is CRC of Data and MIC.
d) What is new in WPA in comparison to WEP?
Direct use of master key in WEP: Master Keys are
never used directly in WPA. A hierarchy of keys is
used, all derived from the master key.
Poor Key Management and updating in WEP:
Secure key management is a planned obsolescence
feature in WPA, so key management isnt an issue
with WPA as compared to WEP.
Reused IV/Less size of IV: The length is increased of
IV from 24 bits to 48 bits. Additionally, IVs are
used as sequence counters for the TSC (TKIP
Sequence Counter), protecting against replaying of
data, which was a major vulnerability in WEP.
Weak IV values are susceptible to attack: A different
secret key is used for each packet, and the way the
key is more quickly with the secret key is more
complex.
Ineffective Message integrity checking: Integrity
protocol CRC-32 was proved ineffective for WEP
message so WPA uses a Message Integrity Check
(MIC) mechanism called, .Michael.
c.Drawback in WPA: The 802.11i standard shows that
keys generated from short passwords are subject to
dictionary attack and a key that is generated from a pass
phrase of less than 20 characters is unlikely to deter
attacks.
D. WPA-2 (Robust Security Networks- RSNs)
802.11i that has been appeared in 2004 uses the logic of a
Robust Security Network (RSN). This new standard and
architecture adopt the IEEE 802.1X standard for access
control and for encryption it uses Advanced Encryption
Standard (AES). It uses a pair-wise key exchange (four way
handshake) protocol utilizing 802.1X for mutual
authentication and key management process. 802.11i allows
use of TKIP, but by default RSN uses AES (Advanced
Encryption Standard) and CCMP (Counter Mode CBC
MAC Protocol). It provides a stronger, scalable solution.

1) Working of WPA-2: WPA-2 uses dynamic
negotiation of authentication and encryption algorithms
between access points and mobile devices. The
authentication schemes are based on 802.1X and Extensible
Authentication Protocol (EAP). We use Advanced
Encryption Standard (AES) for encryption [8]. WPA-2 is
significantly stronger in comparison to WEP and WPA.
However, this protocol would run very poorly on legacy
devices. Only the latest devices have the capability required
to accelerate the algorithms.
II. COMPARISON OF WEP, WPA AND WPA-2
SECURITY PROTOCOLS
A. WEP
WEP uses RC4 cipher. The IV is insecure. It uses 40 bits
of encryption key. There is a concatenation of key and IV
values for per packet are sent over the air, key management
mechanism is not embedded in it, it have no automatic or
periodic key.
B. WPA
It uses a subset of 802.11i. It uses RC4 and TKIP for
encryption cipher mechanism. The key size is 128 bits. It
uses mixed type Encryption key per packet [6]. 802.1x
dynamic key management mechanism. The IV size is 48
bit.802.1x EAP usage for authentication, providing data
integrity and header integrity. The TKIP and IV sequence
mechanism are used to prevent replay attacks.
C. WPA-2
For wireless security WPA-2 seems to be the strongest
security protocol. WPA-2 provides all the advantages of
WPA in addition to stronger encryption through the
implementation of AES, roaming support and CCM
mechanism for data and header integrity [10]. WPA
deployments over current WEP installations provide cost
effective and hassle free shifts where vendors can transit to
the WPA standard through a software or firmware upgrade.
For WPA-2this is not the case. It requires extra hardware
upgrade in order to implement AES [12].
n
c
m
n
n
c
t
TABLE
Wireless ne
not managed
companies. So
must be ident
necessary cou
network. Alt
considered tot
to be a depen
E I. COMPAR
III. C
etworks are u
properly mig
o before instal
tified, evaluate
unter measures
though no s
tally unbreaka
ndable one. It
RISON OF SECURIT
CONCLUSION
used in busines
ght cause ver
lling any such
ed, and based
s must be ins
security syst
able, 802.11i R
suffers none
TY PROTOCOLS
ss WLANs th
ry serious ris
h networks, all
d on the result
talled to secu
tem can eve
RSN security s
of the proble

hat are
sks to
l risks
ts, the
ure the
er be
seems
ems of
old
802
can
tim
sec
min
usa
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]

der mechanism
2.11i WPA-2 i
n rely on until
me being in ter
urity preferre
nimum cost pr
age of WPA is
IEEE: IEEE 80
and Physica
http://standards
Li, L., Zhu, L.
WPA security
Telecommunica
Sepehrdad, P.,
Exploitation of
Gong, G., Stins
91. Springer, H
Sepehrdad, P.,
RC4 Distinguis
Vuagnoux, M.:
channels. PhD
EPFL (2010).
WinPcap: Win
http://www.win
http://www.lyle
insecure-as-i-ex
KrishnamRaju,
K.V.S.V.N: Fo
Using CasperF
information an
India (Septemb
Hung, C.C., W
Key Encryption
Journal of Com
Hiertz, G., Dent
B.: The IEEE 8
6270 (2010).
] A.H.D. Lashkar
protocols (WEP
Information Te
Beijing, ICCSIT
] E.J.M.A. Filho,
Security versu
WPA in 802.
Wireless and O
5, 2007.
] A. D. Potorac an
802.11 WLAN
Control System
ms and protoco
is a wireless s
l its vulnerabi
rms of cost v
ed then WP
referred WEP
recommende
REFER
02.11: Wireless L
al Layer (PH
.ieee.org/about/ge
., Ju, T.: The Re
system. Journal
ations 24(1), 788
, Vaudenay, S.
New Biases in R
son, D.R. (eds.) S
eidelberg (2011).
Vaudenay, S., V
shing WPA. Acce
Computer Aided
thesis, EcolePoly
npcap: The wind
npcap.org/Wpa-W
ebackemort.com/b
xpected (accessed
K.V., Vallikum
ormal Verification
FDR. In: Interna
nd communicatio
er 2010).
Wu, E.H., Wu, C.L
n Scheme for the
mputers 18(4) (200
teneer, D., Stibor
802.11 universe. I
ri, M.M.S. ,Sama
P, WPA and WP
echnology, 2nd
T 2009, pp.48-52
, P.N.L. Fonseca,
us Bandwidth: T
.11g Network,
Optical Communic
nd D. Balan.The
Throughput, Co
ms, vol. 2, pp. 47-5
ols namely WE
security protoc
lities are brou
versus security
A-2 must b
must be emp
d.
RENCES
LAN Medium Ac
HY) Specifica
et/802/802.11.htm
esearch of 802.1
of Nanjing Univ
81 (2004).
., Vuagnoux, M
C4. In: Biryukov
SAC 2010. LNCS
.
Vuagnoux, M.: S
pted at EUROCR
d Cryptanalysis f
ytechniqueFeder
dows packet cap
Wpa2,
blog/2008/5/10/
d on March 14, 20
mari, V., Sandee
n of IEEE802.16
ational Conferen
on technologies,
L., Gau, R.L., Ch
e Next Generation
08).
r, L., Zang, Y., Co
IEEE Communic
di, B., "A survey
A2/802.11i)," Co
IEEE Internatio
, 2009.
, M.J.S. Leitao, a
The Support of M
IFIP Internatio
cations Networks
e Impact of Secur
omputer
52, 2009.
EP and WPA.
col that anybo
ught out. For t
y options, if f
e employed,
loyed, otherw
cess Control (MA
ations (200
ml.
1 Wireless Netw
versity of Posts
M.: Discovery
, A.,
S, vol. 6544, pp.
Statistical Attack
RYPT 2011 (2011
from Ciphers to S
rale de Lausanne
pture library (200
wpa-wpa2
010).
epvarma, N., R
6m PKMv3 Proto
nce on advances
ICT 2010, Ker
hen, Y.C.: A Mu
n Wireless Netwo
osta, X.P., Walke,
cations Magazine
on wireless secur
omputer Science
onal Conference
and P.S.F. de Bar
Mechanisms WE
nal Conference
s, WOCN '07, pp
ity Overheads on
Science
So
ody
the
full
if
wise
AC)
07),
work
and
and
74
k on
).
Side
e
08),
-as-
Raju,
ocol
s in
rala,
ulti-
ork.
,
48,
rity
and
on
rros,
P e
on
p. 1-

and
ISBN: 978-81-920575-7-6:: doi: 10. 74404 /ISBN_0768
Survey of Security Issues in Mobile Ad-Hoc Network

Praful R. Pardhi
Assistant Professor
Shri Ramdeobaba College of
Engineering & Management
Nagpur, Maharashtra

Abstract - An ad-hoc network is a self-configuring network of
wireless links connecting mobile nodes. These nodes may be
routers and/or hosts. The mobile nodes communicate directly
with each other and without the aid of access points, and
therefore have no fixed infrastructure. They form an
arbitrary topology, where the routers are free to move
randomly and arrange themselves as required.
Each node or mobile device is equipped with a transmitter
and receiver. They are said to be purpose-specific,
autonomous and dynamic. This compares greatly with fixed
wireless networks, as there is no master slave relationship
that exists in a mobile ad-hoc network. Nodes rely on each
other to established communication, thus each node acts as
a router. Therefore, in a mobile ad-hoc network, a packet
can travel from a source to a destination either directly, or
through some set of intermediate packet forwarding
nodes.In a wireless world, dominated by Wi-Fi,
architectures which mix mesh networking and ad-hoc
connections are the beginning of a technology revolution
based on their simplicity.
Ad hoc networks [1] date back to the Seventies. They were
developed by the Defense Forces, to comply with a military
framework. The aim was to rapidly deploy a robust, mobile
and reactive network, under any circumstances. These
networks then proved useful in commercial and industrial
fields, first aid operations and exploration missions. Ad hoc
networks, also called peer-to-peer networks, still have a
long way to go in order to be fully functional and
commercial, as it has its defects such as security and routing
which we will discuss further.
Key Words: Mobile Ad Hoc Network, Security, Intrusion
Detection, Secure Routing
I. MANET ENVIRONMENT VARIATIONS
The different MANET environment variations are
listed as follows taking its dynamic topology in to
consideration:
1.1 In MANETS all nodes have identical capabilities
and responsibilities, which are termed as fully
symmetric environment. A mobile ad hoc network
(MANET) is a network comprising wireless mobile nodes
that communicate with each other without centralized

control or established infrastructure. These nodes which
are within each others radio range can communicate
directly, while distance nodes rely on their neighboring
nodes to forward packets. In MANETS every node can be
a host or router. In MANET environment, nodes are free to
join or leave the network at any point of time, resulting in a
highly dynamic network environment compared to wired
network.
1.2 The Asymmetric Capabilities in MANETS include
transmission ranges and radios ranges which may differ.
Battery life, speed of movement and processing capacity
will be different at different nodes.
1.3 Asymmetric Responsibilities include that only
some nodes may route packets in the network or some
nodes may act as leaders for nearby nodes like cluster
head.
1.4 Traffic characteristics may differ in different ad hoc
networks like bit rate, timeliness constraints, reliability
requirements, unicast or multicast or geocast, host-based
addressing or content-based addressing or capability-based
addressing.
1.5 MANETS may co-exist and also co-operate with an
infrastructure based network.
1.6 Mobility patterns may be different like people
sitting at an airport lounge, citywide taxi cabs, military
movements and personal area networks. The performance
of a mobile ad hoc network is dependent on the node
mobility pattern as well as topology, data traffic patterns,
and radio interference.
1.7 Mobility characteristics include speed,
predictability, direction of movement, pattern of
movement, uniformity of mobility characteristics among
different nodes.
II. ROUTING PROTOCOLS
Routing protocols between any pair of nodes within
an ad hoc network can be difficult because the nodes can
move randomly and can also join or leave the network.
This means that an optimal route at a certain time may
not work seconds later. Discussed below are three
categories that existing ad-hoc network routing protocols
fall into:
Table Driven Protocols
On Demand Protocols
Hybrid Protocols
P
i
i
n
m
t
r
o
m
t
c
w
l
t
r
R
o
b
(
(
R
w
u
f
W
d
i
n
n
D
D

2.1 Table D
Table Dr
Proactive Pro
independent o
information t
nodes in the n
move data am
type of protoc
routing loops.
of the networ
may react to c
traffic is affe
could create u
with little dat
limited resou
therefore they
routing soluti
Routing is an
of the exampl
below.
2.1.1 De
(DSDV)
2.1.2 Clus
(CGSR)
2.1.3 Wire
2.1.4 STA
2.2 On De
On Dema
Reactive Prot
when they are
updating of ev
focuses on ro
When a rou
destination fo
it starts a rou
node to the o
node in-betw
Demand prot
Diagram 1: Ad-ho
Driven Routin
iven Routing
otocols, work
of traffic dem
o store the l
network and th
mong differen
col is slow to c
. These protoc
rk and this ca
change in the
ected by the
unnecessary o
ta traffic, Tab
urces such as
y might not
ion for Ad-h
example of a
es under the c
estination Se
ster-Head Gate
eless Routing p
AR
emand Protoco
and Routing
tocols, establis
e required to r
very possible r
outes that are
ute is require
r which it doe
ute discovery p
other until it a
ween has a r
tocols are ge
oc Mobile Routin
ng Protocols
g Protocols
out routes in
mands. Each no
location inform
his informatio
nt nodes in th
converge and
cols keep a co
an be a disad
e network topo
topology mod
overhead. Eve
ble Driven Pr
s power and
be consider
hoc Networks
Table Driven
category of Ta
equenced D
eway Switch R
protocol (WR
ols
Protocols, a
sh routes betw
route data pack
route in the ne
being used o
ed by a sou
es not have ro
process which
arrives at the
route to the
enerally cons
ng Protocols
also known
the backgrou
ode uses routi
mation of oth
n is then used
he network. T
may be prone
onstant overvi
dvantage as th
ology even if
dification whi
en in a netwo
rotocols will u
link bandwid
red an effecti
s. Fisheye St
n Protocol. So
able driven giv
istance Vec
Routing proto
RP)
also known
ween nodes on
kets. There is
etwork instead
or being set u
urce node to
oute informatio
h goes from o
destination o
destination.
sidered effici

as
und
ing
her
d to
This
e to
iew
hey
no
ich
ork
use
dth
ive
tate
me
ven
ctor
col
as
nly
no
d it
up.
o a
on,
one
r a
On
ent
wh
tra
di
co
Pr
an
Pr
ex
giv
(C
Ro
Th
es
rep
the
ha
a
nu
in
(Z
att
ch
ma
ch
au
av
no
ne
co
rep
rem
att
ne
no
wh
ca
se
hen the route
ansfer because
scovery step
ommunication
rotocols more
nd low mobi
rotocol is Dy
xamples under
ven below.
2.2.1
2.2.2 Dyn
2.2
Algorit
2.2
Vector
2.2.5
CBRP)
2.2
discove
2.3 Hybrid R
Hybrid Rou
outing Protoc
hey use distan
tablish the b
port routing in
e topology of
as its own rout
zone radius, w
umber of hop
formation for
ZRP) is an exam
Ad-hoc netw
tacks and de
hallenges of d
ain reasons fo
hannel, insecur
uthority, lack
vailability of re
odes. Generall
etwork, we ex
onfidentiality,
pudiation.
Availability
main operatio
tacks.
Confidential
ever disclosed
Authenticati
ode with which
Integrity gua
hen transferre
Non-repudia
annot deny hav
curity requir
discovery is
e the network
p is low
bandwidth.
suited to larg
ility. An exa
ynamic Sourc
r the category
Associative
namic Source
.3 Te
thm (TORA)
.4 Ad
r routing proto
Cluster-bas
.6 Re
ery Ad Hoc R
Routing Protoc
uting Protoco
ols with On D
nce-vectors fo
best paths to
nformation on
f the network
ting zone, the
which is defin
ps. Each node
r its own zon
mple of a Hyb
III. SE
works are hig
ealing with
developers of
or this difficult
re operating e
of associati
esources, and
y, when con
xamine it unde
authenticati
refers to the
onal at all tim
lity ensures t
to certain use
ion is the abil
h it is commun
arantees that a
d.
ation states tha
ving sent it. A
rements cause
less frequent
k traffic cause
compared t
This makes
e networks w
ample of an
ce Routing.
y of On Dem
e Based Routin
Routing Proto
emporally Ord
d-hoc On-Dem
col (AODV)
sed Routin
elative Dist
Routing Protoc
cols
ols combine
Demand Rout
for more prec
destination n
nly when there
. Each node i
e size of which
ned by a metr
e keeps a reco
ne. Zone Rou
brid routing pr
ECURITY
ghly vulnerab
this is one
these networ
ty are shared b
environment,
ion among n
physical vuln
nsidering the
er the heading
ion, integrity
fact that the
mes despite den
that certain i
ers.
ity of a node
nicating.
a message is n
at the sender o
An ad-hoc netw
ed by its la
t than the dat
d by the rout
to the tota
On Demand
ith light traffi
On Demand
Some of th
mand Protocol
ng (ABR)
ocol (DSR)
dered Routing
mand Distanc
ng Protoco
tance Micro
ol (RDMAR)
Table Based
ting Protocols
cise metrics to
networks, and
e is a change in
in the network
h is defined by
ric such as th
ord of routing
uting Protoco
rotocol.
ble to security
of the main
rks today. Th
broadcast radio
lack of centra
nodes, limited
erability of th
security of
gs; availability
y and non
network mus
nial of servic
information i
to identify th
never corrupted
of the messag
work has extr
ack of prope
ta
te
al
d
c
d
e
ls
g
e
ol
o

d
s.
o
d
n
k
y
he
g
ol
y
n
he
o
al
d
e
a
y,
n-
st
e
is
e
d
e
ra
er
infrastructure and the dynamic relationship between the
nodes in the network. Because of the lack of
infrastructure, accountability is very difficult to
determine as there is no central authority which can be
referenced when it comes to making trust decisions about
other parties in the network. The dynamic relationship
between the nodes leaves very little opportunity for the
nodes to form trust relationships with each other. In an
ad-hoc network, nodes must act as both terminals and
routers for other nodes. Because there are no dedicated
nodes, a secure routing protocol is needed. Multi hop
routing protocols are usually employed. These can lead
to problems due to non-cooperating nodes and denial of
service attacks.
IV. SECURITY ANALYSIS OF MANET

4.1 Particularity of MANET Security
MANET is different from the fixed IP network as well
as the general wireless network, which poses new
challenges to guarantee its security as following: the
wireless communication manner makes the threat of being
attacked suffering from wiretapping greatly improved;
multi-hop routing method may induce a series of security
issues, such as black hole, wormhole and etc; the self
organization of mobile nodes blurs the boundary of
network; the independence of node mobile make the
physical security fragile.
4.2 Security Threats on MANET
The security threats confronted by MANET is the
extension and expansion of that be confronted by wired
network in the wireless field, which mainly comes from
wireless channels and networks. The threats can be divided
into two categories of passive attack and active attack.
4.2.1 Passive attack
Passive attack [13] is essentially to listen or
surveillance the message transmission process to obtain
some secret, which mainly includes two kinds of
wiretapping and traffic analysis. Wiretapping is to intercept
packets being transmitted and access to confidential
information. Traffic analysis is to analyze characteristics of
packet frequency, length and etc to hypothesize the content
of communication.
4.2.2 Active attack
Comparing with passive attack, the passive attack
usually does not change the transmission data, while active
attack will tamper data stream or create false data stream.
The active attack includes four kinds of message replay,
fraud counterfeiting, message tampering and denial of
service.

Message replay: After the attacker intercepted
messages, it will store the message and re-transmit the
message to produce an unauthorized effect. The MANET
is likely to be attacked by message replay, in that the
message is transmitted in the air and can easily be
intercepted.

Fraud counterfeiting: When one entity behaves as
another entity to carry out network activities, the
counterfeiting takes place. Lack of protection or weak
authentication mechanism may cause such security threats,
such as replay identification sequence to make believe a
privileged entity.

Message tampering: After the packet or message
sequence being intercepted, the content is modified, or
deliberately delayed transmission, or a passive change in
the order. This kind of attack is attack on the integrity of
message.

Denial of service: When an authorized entity
cannot access to should be network resources or
emergency operations be denied, the attack of denial of
service occurs. In the MANET, it is easy to achieve service
interruption, so wireless communication can be used to
shield the communications spectrum.
V. MANET THRUST AREAS
The following list of thrust areas shows the
inefficiencies and limitations that have to be overcome in a
MANET environment:
5.1 Limited wireless transmission range: In wireless
networks the radio band will be limited and hence data
rates it can offer are much lesser than what a wired
network can offer. This requires the routing protocols in
wireless networks to use the bandwidth always in an
optimal manner by keeping the overhead as low as
possible. The limited transmission range also imposes a
constraint on routing protocols in maintaining the
topological information. Especially in MANETS due to
frequent changes in topology, maintaining the topological
information at all nodes involves more control overhead
which, in turn, results in more bandwidth wastage.
5.2 Time-varying wireless link characteristics: The
wireless channel is susceptible to a variety of
transmission impediments such as path loss, fading,
interference and blockage. These factors resist the range,
data rate, and the reliability of the wireless transmission.
The extent to which these factors affect the transmission
depends upon the environmental conditions and the
mobility of the transmitter and receiver. Even the two
different key constraints, Nyquists and Shannons
theorems, that govern the ability to transmit information
at different data rates can be considered.
5.3 Broadcast nature of the wireless medium: The
broadcast nature of the radio channel, that is, transmissions
made by a node are received by all nodes within its direct
transmission range. When a node is receiving data, no
other node in its neighborhood, apart from the sender,
should transmit. A node should get access to the shared
medium only when its transmissions do not affect any
ongoing session. Since multiple nodes may contend for the
channel simultaneously, the possibility of packet collisions
is quite high in wireless networks. Even the network is
susceptible to hidden terminal problem and broadcast
storms. The hidden terminal problem refers to the collision
of packets at a receiving node due to the simultaneous
transmission of those nodes that are not within the direct
transmission range of the sender, but are within the
transmission range of the receiver.
5.4 Packet losses due to transmission errors: Ad hoc
wireless networks experiences a much higher packet loss
due to factors such as high bit error rate (BER) in the
wireless channel, increased collisions due to the presence
of hidden terminals, presence of interference, location
dependent contention, uni-directional links, frequent path
breaks due to mobility of nodes, and the inherent fading
properties of the wireless channel.
5.5 Mobility-induced route changes: The network
topology in an ad hoc wireless network is highly
dynamic due to the movement of nodes; hence an on-
going session suffers frequent path breaks. This situation
often leads to frequent route changes. Therefore mobility
management itself is very vast research topic in ad hoc
networks. Mobility-induced packet losses:
Communication links in an ad hoc network are unstable
such that running conventional protocols for MANETS
over a high loss rate will suffer from severe performance
degradation. However, with high error rate, it is very
much difficult to deliver a packet to its destination.
5.6 Battery constraints: This is one of the limited
resources that form a major constraint for the nodes in an
ad hoc network. Devices used in these networks have
restrictions on the power source in order to maintain
portability, size and weight of the device. By increasing the
power and processing ability makes the nodes bulky and
less portable. So only MANET nodes has to optimally use
this resource.
5.7 Potentially frequent network partitions: The
randomly moving nodes in an ad hoc network can lead to
network partitions. In major cases, the intermediate nodes
are the one which are highly affected by this partitioning.
5.8 Ease of snooping on wireless transmissions
(security issues): The radio channel used for ad hoc
networks is broadcast in nature and is shared by all the
nodes in the network. Data transmitted by a node is
received by all the nodes within its direct transmission
range. So an attacker can easily snoop the data being
transmitted in the network. Here the requirement of
confidentiality can be violated if an adversary is also able
to interpret the data gathered through snooping.

VI. CONCLUSION
Studied Ad-hoc networks concepts in depth, we
believe that they will be the future of wireless
networking. It is true that performance suffers as the
number of devices grows and large ad-hoc networks
become difficult to route and manage. However, much
time is being devoted to achieving routing stability, and a
few technical issues need to be solved before they
become common place. The area of ad hoc networks is a
very fast growing area, and due to the vast research in
them, we are seeing these problems disappear and they
are coming into a world of their own.

REFERENCES

[1] C. Perkins and E Royer, Ad Hoc On-Demand Distance
Vector Routing, 2nd IEEE Wksp. Mobile Comp. Sys.
and Apps., 1999.
[2] D. Johnson and D. Maltz, Dynamic Source Routing in
Ad Hoc Wireless Networks, Mobile Computing, T.
Imielinski and H. Korth, Ed., Kluwer, 1996.
[3] IEEE Std. 802.11, Wireless LAN Medium Access Control
(MAC) and Physical Layer (PHY) Specifications, 1997.
[4] B. Schneier, Secret and Lies, Digital Security in a Networked
World, Wiley, 2000.
[5] Y. Hu, A. Perrig, and D. Johnson, Ariadne: A Secure
On-demand Routing Protocol for Ad Hoc Networks,
ACM MOBICOM, 2002.
[6] M. Zapata, and N. Asokan, Securing Ad Hoc Routing
Protocols, ACM WiSe, 2002.
[7] B. Dahill et al., A Secure Protocol for Ad Hoc Networks,
IEEE ICNP, 2002.
[8] Y. Hu, A. Perrig, and D. Johnson, Packet Leashes: A
Defense against Wormhole Attacks in Wireless Networks,
IEEE INFOCOM, 2002.
[9] N. Borisov, I. Goldberg, and D. Wagner, Intercepting
Mobile Communications: The Insecurity of 802.11,
ACM MOBICOM, 2001.
[10] V. Gupta, S. Krishnamurthy, and M. Faloutsos, Denial
of Service Attacks at the MAC Layer in Wireless Ad Hoc
Networks, IEEE MILCOM, 2002.
[11] P. Kyasanur, and N. Vaidya, Detection and Handling of
MAC Layer Misbehavior in Wireless Networks, DCC, 2003.
[12] C. Perkins, and P. Bhagwat, Highly Dynamic Destination-
Sequenced Distance-Vector Routing (DSDV) for
Mobile Computers, ACM SIGCOMM, 1994.
[13] P. Papadimitratos, and Z. Haas, Secure Routing for Mobile Ad Hoc
Networks, CNDS, 2002.
ISBN: 978-81-920575-7-6 :: doi: 10. 73991/ISBN_0768
Survey of Reactive Routing Protocols in Mobile Adhoc Networks

V. Jayalakshmi
Assistant Professor, Computer Applications Department,
Sudharsan Engineering College, Pudukkottai,
Tamilnadu, India

Abstract A Mobile ad hoc networks (MANETs) consist of
a collection of wireless mobile nodes which dynamically
exchange data among themselves without the reliance on a
fixed base station or a wired backbone network. In such
networks, the wireless mobile nodes may dynamically enter
the network as well as leave the network. In the last 15
years, the wireless networking community designed a
number of new routing protocols targeting the various
scenarios of this design space. Any on-demand routing
protocol must utilize some type of routing cache in order to
avoid the need to rediscover each routing decision for each
individual packet. So, caching is an important part of on-
demand routing protocols for wireless ad hoc networks The
objective of this paper is to survey the various reactive
routing protocols which uses the route caching technique
for route repository and compare them. We strive to
uncover the requirements considered by the different
protocols, the resource limitations under which they
operate, and the design decisions made by the authors.

Keywords Ad hoc ; On demand, Route Cache, Routing
Protocols
I. INTRODUCTION
Recently, there has been a tremendous interest in
broadband wireless access systems, including wireless
local area networks (WLAN), broadband wireless access
and wireless personal area networks (WPAN). The
advent of ubiquitous computing and the proliferation of
portable computing devices have raised the importance
of mobile and wireless networking.This domain is a
subject of a huge research and many standardization
activities are undertaken throughout the world, in many
3G/4G related study committees like ITU-R, ETSI
BRAN and IEEE 802. Research prototyping is currently
underway at many research academic and industrial
institutions [1].
In the context of these systems, mobile ad hoc networks
(MANETs) have appeared with specific configurations.
They provide a powerful paradigm for modeling open
self-configuring wireless networks and seem so
appropriate to use in the fourth generation of mobile
networks.
A MANET is an autonomous collection of mobile nodes
forming a dynamic network and communicating over
wireless links. Users are allowed to communicate with
each other in a temporary manner with no centralized
administration and in a dynamic topology that changes
frequently. Due to the limited propagation range of
wireless environment, routes in ad hoc networks are
multihop, and mobile nodes in this network dynamically
establish routing among themselves to form their own
network on the fly [2]. Each participating node acts
both as a host and a router and must therefore be willing
to forward packets for other nodes. Nodes in such a
network move arbitrarily, thus network topology changes
frequently,
unpredictably, and may consist of unidirectional links as
well as bidirectional links. Moreover, wireless channel
bandwidth is limited. The scarce bandwidth decreases
even further due to the effects of signal interference, and
channel fading. Network hosts of ad hoc networks such
as laptops and personal digital assistants operate on
constrained battery power, which will eventually be
exhausted and limited CPU and storage capacity.
MANETs strictly depend on radio links. Actually, a
wireless link is the most variable and unpredictable
communication channel. In addition, ad hoc networks are
vulnerable to attacks and have limited physical security.
The increased possibility of eavesdropping, spoofing and
denial-of-service attacks should be carefully considered.
Because ad hoc networks do not typically allow the same
aggregation techniques that are available to standard
Internet routing protocols, they are vulnerable to
scalability problems. These drawbacks lead to define a
set of underlying assumptions and performance concerns
for protocol design.
Ad hoc networks do not rely on any pre-established
infrastructure and can therefore be deployed in places
with no infrastructure. This is useful in Multicasting has
emerged as one of the most focused areas in the field of
networking. As the technology and popularity of the
Internet have grown, applications that require
multicasting (e.g., video conferencing) are becoming
more widespread. Another interesting recent
development has been the emergence of dynamically
reconfigurable wireless ad hoc networks to interconnect
mobile users for applications ranging from disaster
recovery to distributed collaborative computing. disaster
recovery situations and places with non-existing or
damaged communication infrastructure where rapid
deployment of a communication network is needed [3].
Ad hoc networks can also be useful in conferences where
people participating in the conference can form a
temporary network
without engaging the services of any pre-existing
network [4].
Routing in mobile ad hoc networks is a significant
research topic where various mechanisms have been
already proposed concerning this issue. In fact,
conventional routing protocols are not well suited in ad
hoc environment for several reasons. Firstly, they are
designed for static topology, which is not the case in ad
hoc network. Secondly, they are highly dependent on
periodic control messages; this is in contradiction with
resource-limited ad hoc environment. Moreover, classical
protocols try to maintain routes to all reachable
destinations, which wastes resources.
Another limitation comes from the use of bi-directional
links, which is not always the case in ad hoc
environment. Actually, there is a need for new routing
protocols, adapting to the dynamic topology and the
wireless links limitations. Routing protocols in such
networks should provide a set of features including [4]:
distributed operation, loop freedom, on-demand
based operation, unidirectional link support, power
conservation, multiple routes, efficiency, scalability,
security and quality of service support.
The routing protocols may commonly be categorized as
table-driven and on-demand. Several routing protocols
have used on-demand mechanisms in ad hoc network,
allowing them to dynamically adapt the level of routing
protocol activity required to correctly handle the offered
traffic. Any on-demand routing protocol must utilize
some type of routing cache in order to avoid the need to
rediscover each routing decision for each individual
packet. So, caching is an important part of on-demand
routing protocols for wireless ad hoc networks. This
paper discusses the various reactive routing protocols
which uses the route cache technique for storing the
routes.
This paper is organized as follows. Section II gives the
Overview of the routing protocols in MANETs. Section
III gives a brief survey of the various reactive routing
protocols which uses route cache. Section IV gives the
comparison between these protocols. Finally, section V
provides conclusion of our work.
II. OVER VIEW OF ROUTING PROTOCOLS
The basic routing problem is that of finding an ordered
series of intermediate nodes that can transport a packet
across a network from its source to its destination by
forwarding the packet along this series of intermediate
nodes.
The challenge in creating a routing protocol for ad hoc
networks is to design a single protocol that can adapt to
the wide variety of conditions that can be present in any
ad hoc network over time. The routing protocol must
perform efficiently in environments in which nodes are
stationary and bandwidth is not a limiting factor. Yet, the
same protocol must still function efficiently when the
bandwidth available between nodes is low and the level
of mobility and topology change is high. Because it is
often impossible to know a priori what environment the
protocol will find itself in, and because the environment
can change unpredictably, the routing protocol must be
able to adapt automatically
Many classifications are found in the literature. Many
protocols have been proposed for MANETs. These
protocols can be divided into three categories: proactive,
reactive, and hybrid. Proactive methods maintain routes
to all nodes, including nodes to which no packets are
sent. Such methods react to topology changes, even if no
traffic is affected by the changes. They are also called
table-driven methods. Reactive methods are based on
demand for data transmission. Routes between hosts are
determined only when they are explicitly needed to
forward packets. Reactive methods are also called on-
demand methods. They can significantly reduce routing
overhead when the traffic is lightweight and the topology
changes less dramatically, since they do not need to
update route information periodically and do not need to
find and maintain routes on which there is no traffic.
Hybrid methods combine proactive and reactive methods
to find efficient routes, without much control overhead
III. REACTIVE ROUTING PROTOCOLS A BRIEF SURVEY
MANET is one that searches for and tries to discover a
route to some destination node only when a sending node
generates a data packet addressed to that node. In order
to avoid the need for such a route discovery to be
performed before each data packet is sent, such routing
protocols must cache previously discovered routes
In the following we review several representative
examples of reactive routing protocols which uses route
cache for route repository
A. Dynamic source routing (DSR)

Johnson et al. propose one of the most widely known
routing algorithms, called Dynamic Source Routing [5]
which is an on-demand algorithm and it has route
discovery and route maintenance phases.
Route discovery contains both route request and route
reply messages. In the route discovery phase, when a
node wishes to send a message, it first broadcasts a route
request packet to its neighbors. Every node within a
broadcast range adds their node id to the route request
packet and rebroadcasts. Eventually, one of the broadcast
messages will reach either the destination or a node
which has a recent route to the destination. Since each
node maintains a route cache, it first checks its cache for
a route that matches the requested destination.
Maintaining a route cache in every node reduces the
overhead generated by a route discovery phase. If a route
is found in the route cache, the node will return a route
reply message to the source node rather than forwarding
the route request message further. The first packet that
reaches the destination node will have a complete route.
DSR assumes that the path obtained is the shortest since
it takes into consideration the first packet to arrive at the
destination node. A route reply packet is sent to the
source which contains the complete route from the
source to the destination. Thus, the source node knows its
route to the destination node and can initiate the routing
of the data packets. The source caches this route in its
route cache.

In the route maintenance phase, route error and
acknowledgements packets are used. DSR ensures the
validity of the existing routes based on the
acknowledgements received from the neighbouring
nodes that data packets have been transmitted to the next
hop. Acknowledgement packets also include passive
acknowledgements as the node overhears the next hop
neighbour is forwarding the packet along the route to the
destination. A route error packet is generated when a
node encounters a transmission problem which means
that a node has failed to receive an acknowledgement.
This route error packet is sent to the source in order to
initiate a new route discovery phase. Upon receiving the
route error message, nodes remove from their route
caches the route entry using the broken link.

B. Link availability-based QoS-aware (LBAQ) routing

Yu et al. propose the LBAQ [6] routing protocol based
on node mobility prediction and link quality
measurement. While a node moves, it may experience
varying capacity, reliability and bandwidth availability.
Instead of trying to predict the mobility patterns of the
mobile links, the link availability is incorporated into the
routing metrics to help choose the link with the highest
availability in the route. The expected transmission count
(ETX) along with power consumption is also used as a
route metric. The route metrics used are:

Link availability: probability that two nodes of a link stay
directly connected at a time t
0
+ t provided that they were
connected at time t
0
.

Link quality: the number of retransmissions required to
send a packet on a link between two nodes (ETX). This is
computed by measuring the loss rate of broadcast packets
between the node pairs.

Energy consumption: power consumption at each hop.
Using the metrics above, a combined cost function is
designed based on which the route is constructed.
Di(T
i
)=
0
(E
i
)+
1
(Qi)+
2
(1-L
i
(t)),where
0
,
1
and
2
= 1
0

1
are the weighting coefficients, L
i
and Q
i
are
availability and quality of link i, Ei is the energy
consumed on link i, Ti is the traffic carried on path i, and
Di is the final cost of the route. A source node sends a
RREQ packet during route discovery. When an
intermediate node receives this RREQ packet, it
calculates its own cost from the equation above and
forwards the packet with this new information. The total
cost to the destination is computed when the RREQ
reaches the destination node. At the destination, the node
waits for a fixed number of RREQs before selecting the
route with the least cost. If a link breaks, then an
alternate route is selected for data transmission.

C. Gathering based routing protocol (GRP)
Ahn presents the gathering based routing protocol
[7]which collects network information during route
discovery to be used later by the source node. Initially,
the source node broadcasts a destination query (DQ)
packet which is continuously forwarded towards the
destination. This process is similar to the route discovery
process in DSR or AODV using RREQs. When the DQ
reaches the destination node, a network information
gathering (NIG) packet is forwarded by the destination.
This NIG packet is then propagated over the entire
network collecting information. Every NIG packet is
assigned a sequence number. Nodes receiving a NIG
packet with a new sequence number attach network
information to this packet and forward it along effective
outgoing links (EOLs). EOLs are those links in which the
NIG packets are not received. Once the NIG packet
arrives at the source node, the optimal path is calculated
by the source node based on the collected network
information. Data packets are then transmitted through
this calculated route.

D. Source routing with local recovery (SLR)

Sengul and Kravets [8] start from the observation that
although on-demand routing reduces the cost of routing
in high mobility environments, the route discovery
process, which is typically done through network-wide
flooding, consumes a significant amount of bandwidth.
This is especially expensive if the route discovery must
be repeated due to links broken due to node mobility. To
alleviate this problem, the authors propose bypass
routing, a process which patches a route using local
information acquired on-demand, without the need of
network-wide flooding. The SLR protocol is an
implementation example of the bypass routing approach.
Bypass routing in SLR localizes the reaction to a route
failure and initiates only a local recovery procedure.
Using link state information from the MAC layer, a local
patch on the route is created bypassing the broken link.
By using local recovery with link state information, the
chances of recovering from a broken route increases.
When a route fails, first the local route cache is searched
for an alternate route. If no route is available, bypass
recovery is initiated by querying the neighbors. Based on
replies from these nodes, the path is again reconstructed.
MAC caches provide the connectivity information to
immediate one-hop neighbors and the state is updated by
each node whenever any communication is heard in the
neighborhood. MAC caches also help determine the
validity of the routes in the route caches. Error recovery
is initiated by salvaging used route caches, bypass
recovery and error reporting (see Fig 1). The node looks
for alternative routes to the destination in its cache to
salvage a packet. If the salvaged packet reaches its
destination, th
message from
route is in fac

E. Hint based

Beraldi et
framework w
packets towar
The meta-inf
each node. A
to the destina
value indicat
neighbourhoo
higher probab
one-hop neigh
When a no
will try sendin
has not been
trying various
such neighbo
nodes broadc
then used to c
neighbour. T
calculate the
broadcasting
beacon messa
at most L hop
the protocol.
value is suff
Smaller looka
of lower contr

F. Low overh

Yu et al. [10
information f
is down, the
failed links w
initial main r
destination p
transmitted.
neighboring n
nodes are ac
node on the
candidate nod
used during t
he source rec
m the destina
ct an alternate
Fig. 1. Erro
d probabilisti
al. [9] propo
which uses m
rds the genera
formation is p
A hint, hid, com
ation node d,
ting the chan
od of node d. A
bility with hid
hbours.
ode forwards a
ng to the neig
tried before fo
s alternatives
our exist, the
cast periodic
create a vecto
This time info
e hint values
the control m
ages. Every n
ps around it. T
The authors
ficient for a
ahead values
rol overheads
head dynamic
] repair broke
from overhear
proposed pro
with backup li
route is const
pair along w
While thes
nodes overhe
ctually potenti
main route
des are record
the route reco
ceives an enha
ation stating t
to the broken
or recovery examp
c protocol
ose a probabi
meta-informa
al direction o
provided in t
mputed by no
, is represent
nce of node i
A lower hint v
d = 0 when n
a packet to th
ghbour with th
for the same p
in the order o
packet is di
heartbeat pa
or of time info
ormation vect
. Hints are
messages thro
node receives
This is the loo
show that a
node to gath
also have an
.
c route repairi
en routes dyna
ring the nodes
tocol intellige
inks along the
tructed betwe
which the d
se packets
ear these tran
ial candidates
in case of
ded in the pa
onstruction in
anced route e
that the salva
n route
ple
ilistic forward
ation to forw
of the destinat
terms of hint
ode i with resp
ted by a posi
i residing in
value equates
nodes i and d
he destination
he best hint wh
acket (effectiv
of the hint). If
iscarded. All
ackets which
ormation for e
tor is utilized
disseminated
ough the perio
hints from no
okahead value
small lookah
her correct hi
added advant
ing
amically by us
s. Once the ro
ently replaces
e main route.
een a source
data packets
are forward
smissions. Th
s for replacin
a failure. Th
acket headers
n the future.
error
aged
ding
ward
tion.
ts at
pect
itive
the
to a
d are
d it
hich
vely
f no
the
are
each
d to
d by
odic
odes
e for
head
ints.
tage
sing
oute
s the
An
and
are
ded,
hese
ng a
hese
and
The
rou
pac
MR
arri
sen
bet
use
ma
wit
the
che
so,
cou
hop
tow
ena
RE
S
hig
rea
per
pro
spe
rou
inc
mo
bet
tem
dem
tha
des
O
abo
be

The
res
zon
the
bac
inte
orig
we
app

ute constructio
cket initially, s
REQ is flood
ives at the des
nt back. The
tween the sou
ed during rou
ain route goes
th the hop cou
e main route
ecks to see if
it sends a re
unt value is c
p count is sm
wards the dest
ables nodes c
EPQ and reply
IV. CO
Source routing
gh interest for
asons for this.
rceived as e
otocols from
ecific to ad
uting is bes
arnations of t
obility of the
tween a sourc
mporary in nat
mand, but the
at the efficien
sign considera
One of the m
out the handli
approached fr
Impro
protoc
dynam
bypas
Choos
stable
future
LABQ
consid
challe
e novel nature
earchers to b
ne of network
eory and flo
ckground).
erdisciplinary
ginated routin
have an ap
proach.
on uses a ma
similar to a RR
ed throughou
stination, the m
MRRP keep
urce and dest
ute repair whe
s down. A re
unt of that no
failure. A n
it has a saved
epair reply (R
compared with
maller, it pro
tination, else t
closer to the
with a valid b
OMPARISON OF
g protocols ha
the ad hoc co
First, while ta
extensions or
the wired do
hoc networ
st adapted t
he ad hoc net
nodes is so
ce and a des
ture. Not only
probability o
ncy of respond
ation.
main improvem
ng of recover
rom several di
ving the spee
cols of this
mic route repa
s routing).
sing paths whi
: LBAQ [6]
link availabil
Q [6] extend
dering the Qu
nge.
e of source ori
branch out fro
king (which
ow optimizat
A number
approaches h
ng. In the pape
pproach based
ain route req
RREQ packet i
ut the networ
main route rep
ps track of t
tination. This
en a particula
epair query (R
ode is forward
node receivin
d route to the
REPR); other
h its own hop
opagates the
the packet is
destination t
backup route.
F THE PROTOC
ave been a res
ommunity. Th
able driven pr
r adaptations
domain, reacti
rks. Furtherm
to the most
tworks: situati
high that any
stination will
y that routes a
of a route fail
nding to failur
ement directio
ry of failed ro
irections:
ed of rebuildi
class are l
air [10] , SL
ich are predic
] (based on
lity).
ds upon the
uality of ser
iginated routin
om the tradit
implies relia
tion as the
r of inno
have been app
ers surveyed i
d on probab
quest (MREQ
in AODV. The
k and once i
ply (MRRP) i
the hop coun
hop count i
ar link on the
REPQ) packe
ded in case o
ng the REPQ
destination. I
rwise, the hop
p count. If it
REPQ packe
dropped. Thi
to receive the
COLS
search topic o
ere are severa
rotocols can be
s of existing
ive routing i
more, reactive
t challenging
ions where the
y route found
inevitably be
are created on
ling is so high
res is a majo
ons have been
outes. This can
ing the routes
low overhead
LR [8] (using
ted to be more
prediction o
baseline by
rvice Network
ng encouraged
tional comfor
ance on graph
mathematica
ovative and
plied to source
in this section
bilistic routing
Q)
e
it
s
nt
s
e
et
of
Q
If
p
ts
et
s
e
of
al
e
g
s
e
g
e
d
e
n-
h
or
n
n
s:
d
g
e
of
y
k
d
rt
h
al
d
e
n,
g

TABLE I
COMPARISON OF MULTICAST PROTOCOLS
Protocol Multiple Routes Route Metric Route Rebuilding
Communication
Over head
DSR Yes Shortest path or next
available
New route
And notify
High
Yu et al No Shortest Path or backup
route
Local Backup Medium
LABQ Yes Link availability
Quality , and Energy
used
Same as DSR or
local repair
High
GRP Yes Shortest Path Backup route High

SLR Yes Shortest Path or next
available path
New route and
notify source
High
Beraldi
et al.
Yes Hint value Local broadcast High

V. CONCLUSIONS
In this paper, we have reviewed and compared several
representative protocols. While different classes of protocol
operate under different scenarios, they usually share the
common goal to reduce control packet overhead, maximize
throughput, and minimize the end-to-end delay. The main
differentiating factor between the protocols is the ways of
finding and/or maintaining the routes between source
destination pairs. The development of the ad hoc routing
protocols over the last 15 years is an example of one of the
most systematic explorations of a design space in the history
of computer science. Although, clearly, newer protocols have
built upon the earlier ones, we cannot identify a single best
protocol. Almost all the protocols we discussed in this paper
have their own sweet spot deployment scenarios and
performance metric combinations where they outperform their
competitors. From the point of view of the practitioner, this
creates a serious problem. To deploy an ad hoc network with
an optimal performance, it requires a very careful analysis of
the scenario and its requirements, and the appropriate choice
of the routing protocol from the dozens applicable in the
context

REFERENCES
[1] http://www.ietf.org/html.charters/manet-charter.html

[2] J. Broch, D. Maltz, D. Johnson, Y. Hu, J. Jetcheva. A performance
comparison of multi hop wireless ad hoc network routing protocols,
MOBICOM 98
. G. Pei, M. Gerla, X. Hong. LANMAR: landmark routing for large scale
wireless ad hoc networks with group mobility, Proc. Of IEEE/ACM
MobiHoc 2000, Boston, MA, August 2000.
[1] Tony Larsson, Nicklas Hedman. Routing protocols in wireless ad hoc
networks A simulation study, Master Thesis, Stockholm ericsson
switched lab. 1998.
[5] D.B. Johnson, D.A. Maltz and J. Broch, DSR: the dynamic source
routing protocol for multi-hop wireless ad hoc networks, C.E. Perkins,
Editor, Ad Hoc Networking, Addison-Wesley (2001), pp. 139172
(Chapter 5).
[6] M. Yu, A. Malvankar, W. Su and S. Foo, A link availability-based
QoS-aware routing protocol for mobile ad hoc sensor networks.
Computer Communications, 30 18 (2007), pp. 38233831.
[7] C. Ahn, Gathering-based routing protocol in mobile ad hoc networks.
Computer Communications, 30 1 (2006), pp. 202206. Article
[8] C. Sengul and R. Kravets, Bypass routing: an on-demand local
recovery protocol for ad hoc networks. Ad Hoc Networks, 4 3 (2006),
pp. 380397.
[9] R. Beraldi, L. Querzoni and R. Baldoni, A hint-based probabilistic
protocol for unicast communications in MANETs. Ad Hoc
Networks, 4 5 (2006), pp. 547566.
[10] C. Yu, T.-K. Wu and R. Cheng, A low overhead dynamic route
repairing mechanism for mobile ad hoc networks. Computer
Communications, 30 5 (2007), pp. 11521163.

ISBN: 978-81-920575-7-6 :: doi: 10. 73998/ISBN_0768
Node Classification for Energy Efficient and Reliable Hierarchical WSN Protocol

Ning Sun
Dept. of Computer Science
Chungbuk National University
Cheongju, Korea

Sang-ho Lee
Dept. of Software
Chungbuk National University
Cheongju, Korea

AbstractEnergy Efficiency is an important factor to design a
wireless sensor network (WSN). Meanwhile, data reliability is
another important issue for WSN. Both of them can be
obtained by designing an efficient MAC or routing protocol. In
this paper, we proposed an energy efficient and reliable
clustering hierarchical MAC and routing protocol for WSNs.
According to the different functions nodes make in the
network, we creatively classify sensor nodes into five
categories. By adopting different MAC and routing strategy to
different category, our proposed mechanism reduces energy
consumption and provides data reliability in a maximum.
Keywords- WSN, wireless sensor network, energy efficiency,
reliability, cluster, hierarchical, MAC, multi-path routing
protocol
I. INTRODUCTION
A WSN can be organized by hundreds to thousands of
sensor nodes. The sensor nodes collect data through sensing
and monitoring from their environment and finally send to
the base station for furthering processing. WSNs can be
applicable in many domains, such as military activities,
control and automation, environment monitoring, healthcare,
security and surveillance, smart home and building, vehicle
tracking and detection, etc.
As most sensor nodes have little cost and size with
constrained processors, limited memory size, low bit rate
communications and low energy capability. Due to this,
energy saving is always a remarkable issue and gets much
attention for a WSN protocol design. Hundreds of MAC and
routing protocols have been proposed and deployed until
now with the purpose of saving energy.
Most energy efficient routing protocols applied
hierarchical structure such as cluster based structure as in [1].
Hierarchical structure could greatly reduce number of direct
communications between sensor nodes and base station,
eventually obviously decrease energy consumption in the
whole network and prolong network lifetime. Accounting for
the advantages of this method, our paper used clustering
hierarchical structure too.
In the view of MAC protocol, energy waste mainly
comes from the following sources: idle listening, data
collision, overhearing and control packet overhead, etc.
Energy efficient MAC protocols achieve energy savings by
controlling the radio to reduce energy waste caused by these
sources. MAC protocols for WSNs can be categorized into
two main groups: schedule-based and contention-based
protocols. In our proposed mechanism, according to different
phases, we utilized different MAC protocols to reduce the
energy consumption flexibly.
In our paper, data reliability is also considered, which is
defined as the successful end-to-end data-delivery ratio, has
been another important issue in WSNs. Major solutions [6,8]
for data reliability of WSN are based on the multi-path
routing paradigm, which provides each sensor with
alternative paths. In the phase of data transmission from CHs
to BS, we performs multi-path algorithm both in routing and
MAC protocols.
In order to realize the above strategies, we classify sensor
nodes into several categories, according to their different role
in different phase. For each category, it executes alternative
routing and MAC protocol and also some specific functions,
such as data aggregation, energy alarm, radio control,
schedule generation, etc.
The remainder of this paper is organized as follows:
Section 2 presents the related works. Section 3 describes the
classification of sensor nodes and explains the detailed
function and algorithms of each category. Section 4
introduces the process of our proposed routing and MAC
protocol based on the node classification in section 3. Finally,
section 5 draws conclusions and shows the future works.
II. RELATED WORKS
Hierarchical (clustering) routing techniques can aid in
reducing energy consumption. The essential operation in
clustering is to divide the network into several virtual
clusters, firstly select a set of cluster heads, and then forming
clusters with the remaining nodes. Cluster heads are
responsible for coordination among the nodes within their
clusters and aggregation of received data, and
communication with each other and the base station. The
sensed data are aggregated, and then sent to the base station
by cluster heads. As shown in [1], the clustering can
obviously reduce the communication overhead for both
single-hop and multi-hop wireless sensor networks. LEACH
[1] is a classical clustering hierarchical protocol, which
incorporates randomized rotation of the high-energy cluster
head position among the sensors to avoid draining the energy
of any one sensor in the network. In this way, the energy
load of being a cluster head is evenly distributed among the
nodes. Being a cluster header is much more energy intensive
than being a non-cluster header node. LEACH incorporates
r
p
c
w
a
d
c
p
w
c
l
p
n
b
l
h
w
p
H
a
a
i
g

c
A
o
d
p
l
e
a
m
a
t
n
L
c
f
randomized r
position. One
consider the
whereas we d
algorithm. Se
dynamic netw
clusters, numb
priori, but in
with the dep
change every
large size of W
power to com
not realistic f
between CH a
The sub
literatures [1,2
head distribut
whole netwo
prolonged. Th
However, all
algorithm of
aggregated da
important, th
guaranteed.

F
Sources of
collisions, ov
Among them,
over one-half
during its life
protocols is to
listening, ove
efficient MAC
and contenti
multiple acce
achieved by d
then allocatin
node. Nodes t
LEACH proto
cluster. The t
figure 3(a). M
rotation of th
e shortcoming
energy factor
designed an mo
econdly, LEA
work. It is b
ber of nodes)
reality these
pletion of nod
round dynami
WSN. It is assu
mmunicate with
for a large sca
and BS is mult
bsequent LE
2,3,4] mostly
tion, so that
ork is reduce
he structure o
these hierarch
cluster format
ata transmissio
he reliability
Figure 1. Cluste
f energy waste
verhearing, ex
, energy waste
of the total en
etime. The ma
o reduce energ
erhearing and
C protocols i
ion-based pr
ss) is a typic
dividing the ra
ng unique tim
ake turns wak
ocol, it uses TD
time line of L
Most clustering
he high ener
g of LEACH
r when electin
ore efficient c
ACH doesnt
because param
are programm
parameters co
des. These pa
ically. LEACH
umed that eve
h others and B
ale WSN. So
ti-hop in our pr
EACH-optimiz
improve the c
the energy c
ed and the s
of clustering i
hical protocol
tion, give littl
on. Since the
y of transm
ering structure for
e in WSNs inc
xcessive mess
ed by idle list
nergy consume
ain objective o
gy waste cause
d excessive
include sched
rotocol. TDM
cal solution fo
adio frequency
me slots to ea
ke up and trans
DMA in data
LEACH proto
g WSN proto
rgy cluster h
is that it d
ng cluster he
cluster head el
work well i
meters (numb
med into the no
ontinuously c
arameters we
H is not scalab
ery node has en
BS directly, bu
the communi
rotocol.
zed protocol
capability of c
consumption o
system lifetim
is shown in F
s only consid
le considerati
aggregated d
mission shoul
r WSNs
clude idle liste
sage overhead
tening accoun
ed by a sensor
of most MAC
ed by collision
overhead. E
dule-based pro
MA (time-div
or the former
y into time slo
ach communi
smit/receive da
transmission w
ocol is as show
cols use TDM
header
oesnt
eaders,
ection
in the
ber of
odes a
change
used
ble for
nough
ut it is
cation
ls in
cluster
of the
me is
Fig. 1.
der the
on on
data is
d be
ening,
d, etc.
nts for
r node
C-layer
ns, idle
Energy
otocol
vision
. It is
ts and
cating
ata. In
within
wn in
MA as
the
clus
con
and
sca
no
pro
pop
is p
duty
avo
ene
In
bas
bot
pha
disj
met
floo
tag
bran
tree
is a
mes
algo
amo
con
data
A.
BS
eac
2 an
the
the
9 a
requ
nod
bran
11-
BS,
BN
exh
path
the
just
is
aba
nod

MAC protoc
ster.
Although T
nsumption obv
d the constrain
alability and fle
coordination a
ovide more sca
pular contentio
proposed to re
ty cycle opera
oidance, overh
ergy saving m
the same tim
sed MAC prot
th TDMA and
ase.
For multi-pat
joint paths h
thod. In H-S
oding protocol
their neighbo
anch that the n
e branch form
achieved witho
ssages. Most m
orithm in the
ounts of m
nsumption. We
a transmission
III. C
Branch Node
The definition
. According to
ch BN represen
nd 3 are BNs
neighbor with
discovery of
at first sets up
quest message
de 11, the resu
anch node 1. F
-9).
According to
, the node ha
N acts critical
hausted the wh
hs are correspo
BN doesnt j
t acts as a rout
below a limi
andons the role
de.
col to manage
TDMA prot
viously, it lim
nts of time a
exibility. Cont
among the no
alability and fl
on-based proto
educe the ener
ation. It also p
hearing avoida
may result high
me, collisions
tocols. In our
d content-base
th routing prot
has been cons
SPREAD [6]
l to find a set
ors as a child, s
node is located
m an alternative
out the need t
multi-path pro
phase of dat
message exch
e use hierarch
n.
CLASSIFICATIO
e (BN)
n of the BN is
o the branch
nts one branch
. If a node rec
h different BN
an alternative
p a primary p
from node 11
ults show that
Finally node 9
the simulation
as more burde
l role in the
hole branch is
ondingly faile
join the cluste
ter in the netw
itation value,
e of branch no
the access of
ocol can
mits the numbe
allocation so t
tention-based p
des accessing
lexibility. CSM
ocols. For WSN
rgy efficiency
provided schem
ance, synchron
h latency and
cant be avoi
proposed mec
ed MAC proto
tocol in WSN
sidered as th
, it uses th
of node-disjo
sibling or cou
d. Sensors foun
e disjoint path
to introduce a
otocols usually
ta transmissio
hanges and
hical structure
ON OF SENSOR
s the one-hop
aware floodin
h. As shown in
ceives a reque
N value, it can
path. As show
ath (2-4-9). T
1. After comp
node 11 come
finds an altern
n result in [6],
ens on data tr
network, be
separated and
ed. In order to c
er formation an
work. Furtherm
it should a
ode and transfo
f nodes within
reduce ener
er of joint nod
that its short
protocols requ
the channel a
MA series are t
Ns, S-MAC [1
y by using a l
mes for collisi
nization. But t
low throughp
ided in conte
chanism, we u
ocols in differ
Ns, building no
he most reliab
he branch-aw
oint paths. Nod
sin, based on t
nd on a differ
h. This algorith
any extra routi
y use flat routi
on, which cau
much ener
e in the phase
R NODES
neighbors of t
ng algorithm [
n figure 2, nod
est message fro
n be regarded
w in Fig. 2, no
Then it hears t
pare the BN w
es from differ
native path (1
the nearer to t
ransmission. T
ecause once i
d the downstre
conserve energ
nd data sense
more, if its ener
announce that
orms to a norm
n a
rgy
des
of
uire
and
the
11]
ow
ion
the
put.
ent-
use
ent
ode
ble
are
des
the
ent
hm
ing
ing
ses
rgy
of
the
[6],
de 1,
om
d as
ode
the
with
ent
-8-
the
The
its
am
gy,
. It
rgy
t it
mal
a

w
n
s
B
b
n
g
t
q
a
A
n
p

n
n
k
a
r
e
r
n
Fi
Node i wi
according to th
E
ETX is the
whereas ERC
number of me
satisfied, node
B. Cluster He
In our ap
based on the r
node. CH is
generation, d
transmission.
quicker than
algorithm is ne
In our pa
According to
neighbors of
probability:

(t P
i
Ei is the r
neighbors of
neighborhood
k is the optim
as about 5% o
range (0,1).
Compared
energy and th
randomized ro
After clust
node sets up a
igure 2. Branch
ill decide whe
he following p
(
B re
E E = > o
e energy con
is in data rec
essages. is t
e i mark itself
ead Node (CH
pproach, cluste
residual energy
in charge of
data process,
The energy
normal node
ecessary.
aper, CH is e
o the residua
each node, C
- = min ) k t
residual energ
node i, whi
table. N is the
mal number of
of N. is the
d with LEACH
he density of e
otation.
ters are forme
a TDMA sche
aware flooding a
ether it should
probability:
( * (
TX
E E N +
nsumption in
ception. N is
the coefficient
as the BN.
H)
er heads are
y and the num
f data receive
, data aggr
consumption
es. So an ev
elected in a d
al energy an
CH is elected
+ 1 (
E
E
to
i
o o
gy of node i.
ich can be r
e total number
clusters, whic
coefficient, w
H, we conside
each node, in
ed as in LEAC
edule and tran
algorithm
d be a branch
))
RC
E
data transmi
proportional t
t. If the condit
elected distri
mber of neighb
e, TDMA sch
regation and
n of CH is
venly CH el
distributed ma
nd the numb
d as the follo
(
1 , )
N
N
neig
o

Nneig is the su
retrieved from
r of all sensor n
ch can be calcu
whose value is
er both the re
nstead of the s
CH, the cluster
nsmits this sch

node,
(1)
ission,
to the
tion is
ibuted
bors of
hedule
data
much
ection
anner.
ber of
owing
(2)
um of
m the
nodes.
ulated
in the
esidual
simple
r head
hedule
to t
coll
C.
sub
auto
hea
and
hea
app
neig
tran
Thi
sys
the
run
wil
pro
sch
It r
sch
tim
tim
and
som
agg
this
very
diff
figu

F
D.
only
thes
effi
nod
obv
the nodes in
lisions among
Substitute Clu
Substitute Cl
bstitute of CH
omatically tak
ad is elected, n
d number of n
ad. The CH
points a node
ghbors. This
nsmitted corre
is could impro
tem.
During the pr
cluster head
nning out of en
l be lost. Its
ocess, all node
hedule except S
receives data a
hedule expires,
me slot for CH
meout, if SN re
d waits for th
me problems w
gregation and
s way, the TDM
ry reliable a
ference of our
ure 3.

igure 3. a) Time
Inter-medium
In the proces
y nodes in the
se nodes Inte
iciently reduce
des not joining
vious energy s
Setup
Pi(t) ADV JOIN
Setup
( )
the cluster. T
data message
uster Head No
luster Head N
H. When CH
kes over the w
nodes send its
neighbors in
compares th
as a substitu
strategy gu
ectly even if t
ove the reliabi
rocess of data
is in any ur
nergy, all the
s why we ele
s within the cl
SN. SN keeps
and proceeds d
, different wit
H to transmit
eceives RCV_
he next phase;
with CH, so SN
transmit the a
MA schedule
and flexible
r MAC protoc
e line of LEACH
m Node (IN)
ss of data tran
e delivery path
er-media Nod
e the idle liste
g the data tra
aving.

Frame
Round
TDMA
/SN
SN
TDMA
This ensures th
es.
ode (SN)
Node (SN) is
H cant work
works of CH.
information o
JOIN messag
hese paramet
ute who has m
uarantees the
the cluster he
ility and fault
transmission
rgent situation
sensed data w
ect a substitu
luster comply
awaken durin
data as the CH
th LEACH, th
a RCV_SUC
_SUC, it emp
; Otherwise it
N takes over t
aggregated dat
neednt be alt
fault toleran
col with LEA
b) Time line of
nsmission from
h need to be a
des (INs). By
ening and over
ansmission, so
Steadystate
Steadystate
hat there are
s defined as t
k normally, S
After the clus
f residual ener
ge to the clus
ters and fina
more energy a
data could
ead is exhaust
tolerance of t
within cluster
n, e.g., sudden
within the clus
ute node. In t
with the TDM
ng all the proce
H does. After t
here is a speci
C to SN. In t
ties the memo
t means there
the work of d
ta to the BS.
tered at all. It
nt scheme. T
ACH is shown
f proposed protoco
m CH/SN to B
awaken. We c
y this way,
rhearing of oth
o that to achie
Transmit
RCVSUC Transm
no
the
SN
ster
rgy
ster
ally
and
be
ted.
the
r, if
nly
ster
this
MA
ess.
the
ific
this
ory
e is
data
By
s a
The
n in
o
BS,
call
we
her
eve

mi
p
k
n
M
E
i
d
u
W
c
m
r
m
d
t

Figure 4. M
Before tran
phase beforeh
keep awaken
nodes go to s
MAC protoco
E. Other Nod
The remain
in charge of w
data transmiss
IV. THE
This proto
uses a hybrid
WSN is displ
categories of s

Figure
f.Phase on
This phas
message. Hop
rebroadcasts it
message is ma
decides wheth
total energy a
NEXTN
N
CH
IN1
INk
BN
Node
11
8
14
MAC protocol for r
nsmitting data
hand. After this
in the follow
leep until the
l for IN and ot
des
ned nodes are
works such as
sion (within clu
E PROPOSED R
ocol mainly con
MAC protoco
layed in figur
sensor nodes.
e 5. Node Classi
ne: Broadcasti
e starts from
p by hop, node
t once. The se
arked as the pa
her it is a Bran
and total numb
NEXTN
NEX
Routedisco
1
4
9
route discovery an
a, CH perform
s phase, INs m
wing transmiss
data transmis
ther nodes is s
e normal nodes
s data sensing
uster).
ROUTING AND M
nsists of three
ol for different
re 5, where w
ificationin propos
ing
m the BS broa
e receives broa
ending node o
arent. After thi
nch Node, store
mber of nodes
XTN
SUC
SUC
SU
Awake
overy
BS
2
6
10
13
16
nd data transmiss
ms a route disc
mark themselve
sion, whereas
ssion is over.
shown as figur
s. Normal nod
g, cluster form
MAC PROTOC
e phase. Our sc
t phases. The
we can see dif
sed mechnism
adcasting a re
adcast messag
f its firstly rec
is phase, every
es the paramet
for the next p
UC
SUC
Data
Sleep
Awake
Awake
Awake
Data
Transmission
- BN
3
5
7
12
- CH
- SN
- IN
sion
covery
es and
other
The
re 4.
des are
mation,
COL
cheme
entire
fferent
equest
ge and
ceived
y node
ters of
phase,
and
mu
adv
com
The
con
A.
abo
dec
bro
MA
it b
bac
pro
info
tran
sam
and
B.
pro
the
tran
wit
from
figu
tran
deli
betw
sam
our
algo
bot
n
d records the
ltiple paths in
vantage of
mmunication. E
e table of neigh
TABLE I
In this phase,
ntrol the collisi
Phase Two: C
After previou
out the total e
cides whether t
Once the nod
oadcasts an adv
AC protocol. A
belongs, it tra
ck to the cho
otocol. In JO
ormation of re
The cluster h
nsmits this sc
me time, accor
d neighbor num
Phase Three:
This phase m
opagation with
cluster hea
nsmission alon
In the data t
th the TDMA
m CH to BS
ure 4. Only CH
nsmission.
In the paper
ivery scheme
ween hierarch
me time, we co
There are som
r proposed pro
Firstly, we
orithm with m
th energy effic
neighborhood
nformation to t
the broadcas
Each node reb
hbor informat
I. TABLE OF
, the network
ion.
Cluster Forma
us phase, every
energy Eto and
to be a cluster
de has elected
vertisement m
After each nod
ansmits a join
osen cluster
OIN-REQ me
esidual energy
head node set
chedule to the
rding to the p
mber, CH selec
: Data Transm
mainly consist
hin a cluster, th
ad to the B
ng multi-paths
transmission w
A MAC proto
S, nodes use t
Hs, INs, BNs
V. CON
r, we propose
for WSNs, w
hical routing
onsider the MA
me original and
otocol:
integrated
multi-path rou
iency and relia
d information,
the BS. This m
st nature of
broadcasts onc
tion is as show
F NEIGHBOR INFOR
uses CSMA M
ation
y node has got
d sum of node
head in a distr
d itself to be
message (ADV
de has decided
n-request mess
head using a
essage, it als
and neighbor
ts up a TDM
e nodes in the
parameters of
cts a node as a
mission
ts of two step
hen the data p
BS, which is
.
within clusters
ocol. During t
the MAC pro
keep awake a
NCLUSIONS
ed a cluster-
which integrate
and multipath
AC protocol fo
d progressive
the hierarch
uting protocol
ability.
which provid
mechanism tak
f the wirel
ce and only on
wn in Table 1.
RMATION
MAC protocol
t the informati
es N. Each no
ributed manne
cluster head,
V) using a CSM
d to which clus
sage (Join-RE
a CSMA MA
so includes t
number.
MA schedule a
e cluster. In t
f residual ener
a SN.
ps: first the d
propagation fro
s a multi-ho
s, nodes comp
the transmissi
otocol shown
and join the d
based multip
es the advantag
h routing. In t
or this scheme
characteristics
hical clusteri
l. This provid
des
kes
ess
nce.

l to
ion
ode
er.
, it
MA
ster
EQ)
AC
the
and
the
rgy
data
om
ops
ply
ion
in
data
ath
ges
the
.
s in
ing
des
Secondly, we classify nodes into five types according to
their role in WSN. This strategy guarantees more fairness in
node energy consumption and prolongs the network lifetime.
Thirdly, both of the process of cluster formation and
multiple path discovery are in distributed manner, it
guarantees the scalability of the network.
Fourthly, mechanism of substitute Cluster Head Node
(SN) accounts for the situation of CH damaging. Moreover,
multipath discovery scheme can solve the problem when a
path is failure. These guarantee the robustness of the system.
Finally, we use two parameters of total energy and node
sum number (optimal cluster number) to improve the
performance of both cluster formation and routing discovery.
In our scheme, these two parameters are updated after every
round and calculated by the BS. Its an efficient way to
control the network quality.
Until now, this work is still in theoretic stage, so our
future works include compare this proposed mechanism with
several existing works, in aspects of energy consumption,
data reliability, network lifetime, latency, etc.

VI.REFERENCES

[1] W. Heinzelman, A. Chandrakasan, and H. Balakrishnan, An
Application-Specific Protocol Architecture for Wireless Microsensor
Networks, IEEE Trans. Wireless Comm., vol. 1, no. 4, pp. 660- 670,
Oct. 2002.
[2] O. Younis and S. Fahmy, HEED: A Hybrid, Energy-Efficient,
Distributed Clustering Approach for Ad Hoc Sensor Networks,
IEEE Transction on Mobile Computing, vol.3, no.4, pp. 366-379,
Octobor 2004.
[3] S. Bandyopadhyay and E. Coyle, An Energy-Efficient Hierarchical
Clustering Algorithm for Wireless Sensor Networks, Proc. IEEE
INFOCOM, Apr. 2003.
[4] M Mehrani, J. Shanbehzadeh, A. Sarrafzadeh, S. J. Mirabedini, and C.
Manford, FEED: Fault Tolerant, Energy Efficient, Distributed
Clustering for WSN, Proc. ICACT2010, Feb, 2010.
[5] D. Ganesan, R. Govindan, S. Shenker, and D. Estrin. "Highly-
resilient, energy-efficient multipath routing in wireless sensor
networks, SIGMOBILE Mob. Comput. Commun. Rev., 5(4):1125,
2001.
[6] W. Lou and Y. Kwon. H-SPREAD: a Hybrid Multipath Scheme for
Secure and Reliable Data Collection inWireless Sensor Networks,
IEEE Transactions on Vehicular Technology, 5(4):13201330, 2006.
[7] C. Intanagonwiwat, R. Govindan, and D. Estrin, Directed Diffusion:
A Scalable and Robust Communication Paradigm for Sensor
Networks, Proc. ACM/IEEE Intl Conf. Mobile Computing and
Networking (MOBICOM), 2000.
[8] A. Ouadjaout, Y. Challal, N. Lasla, M. Bagaa, SEIF: secure and
efficient intrusion-fault tolerant routing protocol for wireless sensor
networks, in: Proceedings of the Third International Conference on
Availability, Reliability and Security (ARES) 2008, March 47, 2008,
pp. 503508.
[9] Y. Jing, X. Mai, J. F. Xu, B. G. Xu, and H. Lu, A cluster-based
multipath delivery scheme for wireless sensor networks, IC-BNMT
'09, Beijing, pp. 286 291, Oct, 2009.
[10] M. S. Zaman, and G. MamaMurthy, Clustered and leveled disjoint
multipath routing algorithm for wireless sensor networks, AH-ICI
2009, Kathmandu, Nov, 2009.
[11] T.van Dam and K. Langendoen, An adaptive energy-efficient mac
protocol for wireless sensor networks, in ACM Sensys 03, 2003, pp.
171-180.

ISBN: 978-81-920575-7-6 :: doi: 10. 74005/ISBN_0768
A Data Mining Approach for Mobility Prediction Using Wireless Traces
J.Ananthi, R.Menaka
Research scholars, Department of ECE
Anna University of Technology
Chennai, India

Dr.V.Ranganathan
Prof., Department of VLSI
KCG College of Technology
Chennai, India

AbstractWith various wireless communication protocols
emerging for ubiquitous communication, mobility prediction
enables better management and efficient usage of resources to
provide improved services to mobile users. As most users are
mobile in the wireless network, it is necessary to include
mobility as an inherent part of the system. The users mobility
influences the network performance and the underlying
quality of service. Various studies have been conducted under
different scenarios to improve performance of the wireless
network under various mobility models. In this paper it is
proposed to apply data mining techniques to predict the user
based on wireless traces.
Keywords-Mobility Prediction, Instance Based Learner
(IBL), Alternating decision tree, Nave Bayes.
I. INTRODUCTION
With the drastic increase in mobile wireless users, networks
face bandwidth and resource shortages which may cause
network failure [1]. To provide guaranteed and QoS to
users, proactive methods are actively studied. As most users
are mobile, it is necessary to include mobility as an inherent
part of the system [2]. Mobility management handles the
methods of storing and updating the location information of
mobile users. Mobility prediction is also part of mobility
management as mobility of the user plays an important role
in efficient usage of resources. Mobility prediction may be
defined as determining the next access point (AP) that a user
will connect to in a network. The mobility is generally
studied by designing mathematical models that model
human mobility [3].

Timely handoffs should be seamless for uninterrupted active
sessions like file transfers, voice over calls, video/audio
streaming, etc. When a user is mobile, the connection is
constantly being switched from one AP to another. Thus
seamless handoffs are possible only if resources are
available at the next AP. Allocating resources to all
neighboring APs is not feasible as it leads to wastage of
resources and under utilization of the network. Efficient
allocations of resources improve resource utilization and
reduce latency in accessing the resources. Prior knowledge
about the movements of users is helpful as mobility
prediction is based on past movements. The mobility
prediction impacts handovers and, hence, improves the
network QoS. There are several problems and issues
involved in mobility prediction in a real environment. The
mobility pattern varies with different population; the
mobility pattern of a city user is different from that of user
in an educational campus. User mobility traces are used for
mobility prediction modeling.
The movement history of the user can be extracted from the
logs obtained from all the access points. When a mobile
device attaches itself to an AP, the unique number of the
mobile device is entered into the logs with a time stamp.
Using the logs obtained from all the APs, the frequently
used path by a particular wireless device can be traced and
this is called the user mobility pattern (UMP) [4]. Mobility
rules are formed using the UMPs. The users actual
movement is obtained,
1 2
, ,....,
i n
u ap ap ap =
where
i
ap is the AP accessed by the user in the movement
path of n access points in the defined time. The mobility
rules are in the form of
1 2
1 2 3 4
, ,
l l
l l l l

1 2 3
, ,....,
n
l l l l
the left part of the mobility rule is called head of rule and
the other is called tail of the rule. When rules are generated,
a confidence value is calculated for each rule. The rule with
the highest confidence is selected.

In this paper it is proposed to use data mining algorithms for
predicting user mobility using wireless trace available from
the Dartmouth College.

II. LITERATURE REVIEW
Considerable amount of research on mobility prediction is
in literature. Some of the related works are discussed in this
section.
Prasad, et al., [5] proposed a generic framework for mobility
prediction of user movement in wireless networks. The
primary goal of the proposed model was effective allocation
of the resources. The framework consists of a prediction
model based on Hidden Markov Model (HMM) to predict
users next location and a predictive mobility management
based on a control theoretic approach with a feedback
control system. Campus wireless trace dataset available
online was used for testing the model. The dataset provided
mobility history of the user such as user associations with
APs, duration of association and so on. On simulation the
effectiveness of the prediction model and the controller was
proven. The proposed framework can be incorporated into
existing wireless networks.
Akoush, et al., [6] proposed a hybrid Bayesian neural
network model for predicting location. The model used
probability theory with neural networks for movement
prediction of the user and to build a inference model. The
dataset contained 100 Nokia smart phone users was
collected from Reality Mining Project at MIT was used.
Trade off analysis was performed for location update and
paging for the wireless network. To enhance prediction
accuracy cells that the users do not use were grouped.
Results showed that the prediction accuracy of the proposed
model outperforms all other standard neural network
methods.
Kim, et al., [7] proposed a mobility method, constructed
using wireless user traces. The proposed method extracts
users mobility tracks and obtained information about
movement speed, pause times, waypoints between
destinations and destination transition probability. This
information was used to generate synthetic tracks and the
same was validated by comparing with the real tracks.
Experiments were conducted using traces of wireless users
from Dartmouth College. Results showed that the synthetic
tracks produced were similar to the real track.
Su, et al., [8] presented a mobility prediction method to
enhance unicast and multicast routing protocols. In the
proposed method, GPS position was piggybacked on data
packets during communication between two nodes and the
expiration time of the link was estimated. The routes were
reconfigured before the link was broken thus providing
seamless connection before connection breaks. The
simulation was implemented within GloMoSim library.
Simulation results indicated that the routes stayed connected
the longest were chosen by the use of mobility prediction.

III. METHODOLOGY
Dartmouth College provides mobility traces as a community
service to researchers. Mobility trace collected over a period
of three years within the Dartmouth College campus is
available, however we used one month syslog data in this
work. During the data collection period the college housed
over 5500 students and 1200 faculty. Initially 476 access
points were available and over the period of time this
increased to 566. All access points were programmed to
share the same SSID and hence mobile users were able to
seamlessly use the network across the campus.115 subnets
covered the 188 buildings and hence devices were forced to
obtain new IP addresses some times. The log recorded in a
syslog server included the timestamp to each message with
each message containing the AP name, MAC address of the
card and the type of message. The various messages used
are authenticated, associated, reassociated, roamed and
disassociated. Whenever a device chooses to use the
network it is first authenticated and this message is ignored
in this work. After authentication, the device must associate
with one of the access points enabling all traffic between
device and network. Reassociation occurs when another AP
with better signal strength is available. The trace of a single
day from users in the Dartmouth College is used in this
work. It is proposed to consider four attributes with three
attributes giving the previous location of the user and the
fourth attribute considering the time.

The Nave Bayes classifier is a statistical method
of classification using supervised learning methods. It is
based on Bayes theorem. The classifier can solve both
diagnostic and predictive problems involving both
categorical and continuous valued attributes. The Bayes
theorem is as follows:

( )
( )
( )
( )
. P B A P A
P AB
P B
|
| =

where
( )
P AB |
is posterior probability of A,
( ) P A
is prior
probability of A,
( )
P B|A
is posterior probability of B and
( ) P B
is prior probability of B.
For a set of tuples D, with n dimensional attribute vector;
( )
1 2
: , ,...,
n
X x x x where
i
x is the value of attribute
i
A .
There are m classes:
1 2
, ,....,
m
C C C . Bayesian classifier
predicts X belongs to a class if

( ) ( ) i j
P C X P C X | > | for 1 , j m j i s s =
It becomes computationally difficult to evaluate
( ) i
P X C |
if
number of attributes is high. In such a case Nave class
conditional independence is used for computation.

( )
1
n
k i
k
P x C
=
|
[

Nave Bayes though over simplified, it performs well in
many complex situations. When the dimensionality of the
data input is high Nave Bayes are very effective. The
parameters are estimated with very small amount of training
data.
Alternating Decision Trees (ADT) is generalization
of decision trees, voted decision tree and decision stumps
with boosting [9]. The classifiers generated by ADT are
easy to deduce and it provides a measure of confidence
called classification margin. In ADT the root node is a
prediction node and consists of a numeric score. The
prediction nodes are followed by decision nodes which are
similar to decision tree stumps. The decision layer is
f
p
p

I
t
b
t
g
t
t
o
c
t
l
d
b
l
a
i
n
a
d
N
I
c
t
c
o
o
followed by
prediction no
prediction lay
Fig
In an ADT th
the path reach
but on reachin
the children o
grows. The ru
the rules to m
the ADT is hi
on the predict
Instance-ba
compares new
training, inste
learns from ex
directly from
based learning
learning occur
a new query
instances is re
new query
algorithms is
distance meas
Nominal attrib
In this paper
class attribute
the Nave B
classification
obtained and
obtained resul
Algorithm
Nave Bayes
J48
IBL
prediction l
des and decis
yer and decisio
1: A simple a
he instances de
hes a decision
ng a predictor
of the node. Th
ules generate r
make real valu
igher as decis
ion.
ased learning
w problem in
ead of perform
xamples or ins
the training
g is a kind of
rs by simply s
instance is pr
etrieved from
instance. An
the nearest
sures are used
butes are hand
IV.
305 records w
es. A tenfold c
Bayes, J48, IB
accuracy an
compared. T
lts.
Classifi
accur
s 72.45
72.45
68.524
ayer, thus a
sion nodes. a
on layer is show
alternating dec
efine the path
n node it conti
r node the path
he base rules i
real numbers;
ued prediction
ion nodes hav
g or memo
nstances with
ming explicit g
stances and co
instances the
lazy learning
storing the pre
resented, a se
memory and
n example
neighbor alg
to handle the
dled using ove
RESULT
were used with
cross validatio
BL algorithm
nd relative ab
Table 1 and f

ication
racy a
59%
9 %
46 %
alternating be
a simple ADT
wn in figure 1
cision tree
hs in the tree.
inues with the
h continues w
increases as th
the boosting a
. The robustn
ve limited infl
ory-based lea
h instances se
generalization
onstructs hypot
emselves. Inst
. In IBL algor
esented data.
et of similar r
used to classi
of instance-
gorithm. Eucl
numeric attri
rlaps.
h 4 attributes
on was used to
ms. The resu
bsolute error
figure 2 show
Relative
absolute error
56.4997%
46.9009 %
47.1483 %
tween
T with
1 [9].

When
e child
with all
he tree
allows
ess of
luence
arning
een in
n. IBL
theses
tance-
rithms
When
related
ify the
-based
lidean
ibutes.
and 1
o train
ult for
were
ws the
Fig
TABLE 1: CLA
g 2: Classificat
In this paper
accuracy an
J48, IBL alg
traces from D
outperforms
research nee
mobility pre
vector mach
[1] P. S. Pra
of mobil
6
th
IEEE
Conferen
[2] D. Perkin
the per
Internati
vol. 4, pp
[3] M. Go
Underst
Nature, n
[4] Gohan Y
Manolop
predictio
Engineer
[5] Pratap S
Framewo
Utilizatio
[6] Sherif A
Moveme
Neural N
[7] Minkyon
a mobili
IEEE Inf
[8] W. Su,
wireless
0.00%
10.00%
20.00%
30.00%
40.00%
50.00%
60.00%
70.00%
80.00%
Na
Ba
ASSIFICATION A
ABSOLUT

tion rate.

V. CON
r it was propo
nd relative ab
gorithms for m
Dartmouth Co
the Nave
ed to be carrie
diction using
ine and Neura

REFER
sad, P. Agrawa
lity in hierarchi
E Consumer C
nce, January 200
ns, H. Hughes,
rformance of
ional Conferen
p. 20482052, 2
onzlez, C. H
tanding individ
no. 453, pp. 479
Yavas, Dimitrios
poulos, A dat
on in mobile en
ring 54 (2005) 1
S Prasad and
ork for Mobi
on in Wireless N
Akoush, Ahme
ent Prediction
Networks, IWCM
ng Kim, David K
ity model from
focom 2006.
S.-J. Lee, M.
networks, in: IE
ave
ayes
J48 I
ACCURACY AN
TE ERROR
NCLUSION
osed to study t
solute error i
mobility predic
ollege. Result
e Bayes alg
ed to improve
Radial basis fu
al network algo
RENCES
al, and K. M. Siv
ical mobile ad
Communications
09, pp. 15.
and C. Owen,
f ad hoc n
nce on Commu
2002.
Hidalgo, and
dual human m
9482, 2008.
s Katsaros, Ozg
a mining appr
nvironments. D
121146.
Prathima Agra
ility Prediction
Networks. IEE
ed Sameh, 200
Using Bayesi
MC'07, August,
Kotz, Songkuk
m real user tra
. Gerla, Mobi
EEE MILCOM,
BL
Clas
Acc
Rela
abs
ND RELATIVE
the classificati
in Nave Bay
ctions using d
obtained for J
gorithm. Futu
effectiveness
function, Supp
orithm.
valingam, Effe
hoc networks,
s and Network
Factors affect
networks, IE
unications (IC
A.-L. Barab
mobility pattern
gur Ulusoy, Yan
roach for locat
Data & Knowled
awal, A Gene
n and Resou
EE 2010.
07. Mobile U
ian Learning
, page 191-6.
Kim., Extract
aces. Proceedin
lity prediction
, October 2000.
ssification
uracy
ative
oluteerror

ion
yes,
data
J48
ure
of
port
ects
in
king
ting
EEE
CC),
bsi,
ns,
nnis
tion
dge
eric
urce
User
for
ting
ngs
in
.
[9] Y. Freund and L. Mason. The alternating decision tree
algorithm. In Proceedings of the 16
th
International
Conference on Machine Learning, pages 124133, 1999.

ISBN: 978-81-920575-7-6 :: doi: 10. 74012/ISBN_0768
Performance Analysis of Routing Protocols in Mobile Adhoc Netoworks
Th.Ibungngomacha Singh
Department of Computer Science & Engineering, Manipur Institute of Technology
Imphal,Manipur,India

AbstractA mobile ad hoc network (MANET) is a
decentralized, peer-to-peer wireless ad hoc network that is
capable of configuring itself. Since it is not a static
infrastructure network traditional routing protocols fail to
cope with this type of network topology that changes with
time. Recently researchers have proposed new MANET
routing protocols and comparing and improving existing
protocols before any standardization process. However
simulations results from different research community are
not consistent enough to claim standard MANET routing
protocols for a specific application. This is mainly because of
the lack of consistency in routing protocol models and
different application environments or scenarios including
networking and user profiles. Furthermore the performance
metrics used in comparative analyses differ among
researchers. Therefore it is premature to give a generalized
conclusion in regard to the performance of the existing
routing protocols. In this paper we present comparative
analyses of MANET routing protocols and many possible
performance metrics are considered in the study. The task is
developed using NS2 simulation software and results obtained
are presented to give a detail analyses of different types
MANET routing protocols.
Keywords- MANET, Routing Protocols AODV, DSR,
DSDV, ZRF.
I. INTRODUCTION
Mobile Ad Hoc Network (MANET) [1] is a collection
of communication devices or nodes that wish to
communicate without any fixed infrastructure and pre-
determined organization of available links. The nodes in
MANET themselves are responsible for dynamically
discovering other nodes to communicate. It is a self-
configuring network of mobile nodes connected by
wireless links the union of which forms an arbitrary
topology. The nodes are free to move randomly and
organize themselves arbitrarily; thus, the networks
wireless topology may change rapidly and unpredictably.
Routing is a core problem in networks for sending data
from one node to another. Wireless Ad Hoc networks are
also called Mobile Ad Hoc multi-hop wireless networks is
a collection of wireless mobile hosts forming a temporary
network without the aid of any established infrastructure or
centralized administration. Mobile Ad Hoc Networks
(MANETs) are characterized by a dynamic, multi-hop,
rapid changing topology. Such networks are aimed to
provide communication capabilities to areas where limited
or no communication infrastructures exist. MANETs can
also be deployed to allow the communication devices to
form a dynamic and temporary network among them. A
mobile Ad Hoc network (MANET) is receiving attention
due to many potential military and civilian applications.
MANETs have several salient characteristics: 1) Dynamic
topologies 2) Bandwidth-constrained 3) Energy
constrained operation 4) limited physical security.
Therefore the routing protocols for wired networks cannot
be directly used for wireless networks. Some examples of
the possible uses of ad hoc networking include students
using laptop computers to participate in an interactive
lecture, business associates sharing information during a
meeting, soldiers relaying information for situational
awareness on the battlefield and emergency disaster relief
personnel coordinating efforts after a hurricane or
earthquake. A MANET uses multi-hop routing instead of a
static network infrastructure to provide network
connectivity. Several routing protocols have been proposed
for mobile Ad Hoc networks. Recent studies on
comparative analyses of different MANET protocols can
be found in [8-11]. However the results presented in these
literatures are not consistent and only few performance
metrics are considered. In this paper we present a number
of ways of classification or categorization of these routing
protocols and performance comparison from classes are
presented.

The following sections in the paper are arranged as
follows: Different types of MANET routing protocols are
presented in Section II. In Section III, methodology used
and the performance metrics are defined. The simulation
results and discussion are given in section IV. Section V
concludes the paper.
II. MANET ROUTING PROTOCOLS
Based on when routing activities initiated, routing
protocols for mobile ad hoc networks may be broadly
classified into three basic categories as indicated in figure
1.
(a) Proactive or table-driven protocols
(b) Reactive or on-demand routing protocols
(c) Hybrid routing protocols

Figure 1. Classification of ad hoc routing protocols.
A. Proactive or table- driven protocols
Traditional distance-vector and link-state routing
protocols [1] are proactive in that they maintain routes to
all nodes, including nodes to which no packets are sent.
For that reason they require a periodic control message,
which leads to scarce resources such as power and link
bandwidth being used more frequently for control traffic as
mobility increases. Proactive routing protocols have their
routing tables updated at all times, thus the delay before
sending a packet is minimal. However, routing tables that
are always updated require periodic control messages that
are flooded through the whole network- an operation that
consumes a lot of time, bandwidth and energy. One
example of a proactive routing protocol is explained in the
next subsection.
1) The Destination Sequence Distance Vector (DSDV)
[3]
It is a proactive routing protocol. DSDV also called
table-driven routing protocol because each node maintains
routing table that contains sequence numbers and hop-by-
hop information. DSDV is based on Bellman-Ford routing
algorithm. Some major improvements have been made in
Bellman-Ford algorithm in order to make it suitable for
wireless environment and cope with count-to-infinity
problem. DSDV uses the sequence number to avoid count-
to-infinity problem and using this sequence number DSDV
distinguishes between stale and fresh routes. Nodes talk
with each others and update their routing tables. If change
in topology occurs, updates are transmitted. There are two
types of updates, time-driven (periodic updates) and table-
driven (updates because of significant change). In case of
any change, routing updates are transmitted to all other
nodes which may cause large overhead. In order to reduce
this overhead routing updates are sent in two ways: a full
dump way, where full routing table is sent to neighbors but
it happens only in case of complete topology change. An
incremental update: where only the entries change in the
route metric are sent [2, 3].
B. Reactive or on-demand routing protocols
Reactive routing protocols, on the other hand, operate
only when there is a need of communication between two
nodes. This approach allows the nodes to focus either on
routes that are being used or on routes that are in process of
being set up. Reactive routing protocols determine routes
between nodes only when they are explicitly needed to
route packets. However, whenever there is a need for
sending a packet, the mobile node must first find the route
if the route is not already known. This route discovery
process may result in considerable delay. An example of
reactive routing protocol is Ad hoc On- Demand Distance
Vector (AODV).

1) Ad Hoc On-Demand Distance Vector
Routing Protocol
It is reactive routing protocol. AODV also called
source-initiated routing algorithm, because AODV only
discovers the path to the destination when source wants to
send data. Established path between source and destination
remains as long as it is needed or becomes inaccessible.
Route discovery mechanism of AODV is based on route
request (RREQ), route reply (RREP), and route error
(RERR) messages. Since AODV is flooding in nature,
when there is need to discover path, source node floods
RREQ message to all neighboring nodes. This RREQ
message contains destination sequence number. This
sequence number helps in ensuring route validity and
prevents routing loops [2]. For example, a route with
greatest sequence number is always chosen by sending
node. After receiving RREQ message each neighboring
node checks the destination ID. When the path is found,
RREP message is sent back to requesting node. The path
followed by RREP message is used to send data packet. On
the other hand, when the path is not found, neighboring
nodes forward the RREQ to their neighbors. In case of link
breaks a RERR message is sent to source node informing
that link is no longer valid now. The route discovery
mechanism of AODV is similar to DSR and routing table
of AODV with destination sequence numbers is similar to
DSDV [3].
C. Hybrid routing protocols
Combining the proactive and reactive approaches result
in a hybrid routing protocol. A hybrid approach minimizes
the disadvantages, but also the advantages of the two
combined approaches. The Zone Routing Protocol (ZRP) is
such a hybrid reactive /proactive routing protocol. Each
mobile node proactively maintains routes within a local
region (referred to as the routing zone). Mobile nodes
residing outside the zone can be reached with reactive
routing. ZRP is an example of such a routing protocol.
1) Zone Routing Protocol
ZRP is designed to address the problems associated
with proactive and reactive routing. Excess bandwidth
consumption because of flooding of updates packets and
long delay in route discovery request are two main
problems of proactive and reactive routing respectively.
ZRP came with the concept of zones. In limited zone, route
maintenance is easier and because of zones, numbers of
routing updates are decreased. Nodes out of the zone can
communicate via reactive routing, for this purpose route
request is not flooded to entire network only the border
node is responsible to perform this task. ZRP combines the
feature of both proactive and reactive routing algorithms
[7]. The architecture of ZRP consists of four elements:
MAC-level functions, Intra-Zone Routing Protocol
(IARP), Inter-Zone Routing Protocol (IERP) and Border
cast Routing Protocol (BRP). The proactive routing is used
within limited specified zones and beyond the zones
reactive routing is used. MAC-level performs neighbor
discovery and maintenance functions. For instance, when a
node comes in range a notification of new neighbor is sent
to IARP similarly when node losses connectivity, lost
connectivity notification is sent to IARP. Within in a
specified zone, IARP protocol routes packets. IARP keeps
information about all nodes in the zone in its routing table.
On the other hand, if node wants to send packet to a node
outside the zone, in that case IERP protocol is used to find
best path. That means IERP is responsible to maintains
correct routes outside the zone. If IERP does not have any
route in its routing table, it sends route query to BRP. The
BRP is responsible to contact with nodes across Ad Hoc
networks and passes route queries. Important thing in
border casting mechanism of BRP is it avoids packets
flood in network. BRP always passes route query request
to border nodes only, since only border nodes transmit and
receive packets [5, 6, 7].

III. METHODOLOGY
The simulation is conducted in two different scenarios.
In the first scenario, the comparison of the two routing
protocols is done by keeping 20 connections fixed in
varying numbers of nodes-10, 20,30,40,50.
In the second scenario, the routing protocols are
evaluated in different number of connections while the
number of 20 nodes is fixed. The number of connections is
set to 5, 10,15,20,25.
The performance of the protocols depends on various
interrelating adhered metrics discussed in the next sub-
section.
A. PERFORMANCE METRICS
To analyze the relative performance of the protocols we
measured the following performance metrics from the trace
files.
1) Throughput:
Throughput is the amount of data transferred to the
destination through the network in a unit time expressed in
kilobits per second (Kbps).
kbps
time simulation
d transfere data of Amount
Throughput

2) Number of Packets sent:
It is the total number of packets sent to the
destinations generated by the sources.
3) Number of Packets received:
It is the total number of packets received by the
destinations sent from the sources.
4) Packet Delivery Ratio:
It is the ratio of the data packets delivered to the
destinations to those generated by the source.
% x100
packets sent Total
packets received Total
ratio delivery Packet

5) Routing Overhead:
It is number of routing packets transmitted per data
packet delivered at the destination.

% x100
packets ing transmitt Total
packets signalling routing Total
overhead Routing

6) Average end-to-end delay of data packets:
There are possible delays caused by buffering during
route discovery latency, queuing at the interface queue,
retransmission delays at the MAC, and propagation and
transfer times. The paper use Average end-to-end delay.
Average end-to-end delay is an average end-to-end delay
of data packets. It also caused by queuing for transmission
at the node and buffering data for detouring. This metric
describes the packet delivery time: the lower the end-to-
end delay the better the application performance.
n
1
n
1
time recv. TCP
time) recv. TCP - sent time (TCP
_End_Delay avg_End_to

The packet delivery ratio and the throughput are most
important for best effort traffic. And the routing overhead
evaluates the efficiency of the routing protocol.
IV. SIMULATION RESULTS
The study has been carried out using NS2 simulation
software under Linux operating system platform. In the
simulation, the performance of three routing protocols
namely, AODV, DSDV and ZRP are analyzed in two
scenarios. In the first scenario, numbers of nodes is varied
keeping 20 connections fixed and in the second scenario,
number of connections is varied while the number of
nodes, 20 is fixed.
A. Scenario 1:
In this first scenario, the comparison of the three
routing protocols is done by keeping 20 connections fixed
and varying numbers of nodes with some parameters with
a specific value as shown in table I.
TABLE I. SCENARIO 1 FOR SIMULATION OF AODV,DSDV AND
DSR
Parameter Value
Number of
nodes
10,20,30,40,50
Simulation
Time
100
Pause Time 2 s
Environment
Size
700 x 400
Packet Size 512 bytes
Maximum
Speed
10 m/s
Queue Length 50
Mobility
Model
Random Waypoint
Mobility

The performance metrics for the three routing protocols
are calculated and all relevant graphs are given in figure 2-
figure 3.
From figure 2(a) it is observed that in the beginning,
the average end-to-end delay of AODV leads ZRP and
then comes DSR. However, with the increase in number of
nodes, DSDV leads ZRP for a while but AODV continues
to lead them. ZRP starts to dominate over them with more
number of nodes. At last, ZRP dominates them and AODV
leads DSDV. Figure 2(b) gives the packet delivery ratio,
ZRP and AODV coincide. In the beginning, the packet
delivery ratio of ZRP and AODV is more than DSDV.
However, at the last, with increasing the number of nodes
DSDV dominates over them. Both ZRP and AODV
continue to coincide. In the case routing overhead AODV
leads over DSDV and ZRP. ZRP comes at the last from the
beginning to the end with increasing number of nodes as
presented in figure 3(a).From figure 3(b) the throughput of
the three protocols can be observed. In the beginning,
AODV leads over DSDV. ZRP start at the lowest among
them. As the number of nodes increases, DSDV starts to
lead AODV and ZRP. At the end, DSDV leads AODV
followed by ZRP.
B. Scenario 2:
In this second scenario, the comparison of the three
routing protocols is done by keeping 20 nodes fixed and
varying numbers of connections with some parameters
with a specific value are as shown in table II.
TABLE II. SCENARIO 2 FOR SIMULATION OF AODV
,DSDV AND ZRP
Parameter Value
Number of
connections
5,10,20,25
Simulation Time 100 s
Pause Time 2 s
Environment Size 700 x 400
Packet Size 512 bytes
Maximum Speed 10 m/s
Queue Length 50
Mobility Model Random Waypoint
Mobility

The performance metrics for the three routing protocols
are calculated and relevant graphs are shown in figure 4-
figure 5 respectively.
From figure 4(a) the average end-to-end delay of
AODV leads ZRP and ZRP leads DSDV with less no. of
nodes. With the increase in number of connections, at the
last, AODV continues to dominate over ZRP and DSDV
stays after ZRP. From figure 4(b) it can be observed that in
the beginning, the packet delivery ratio of AODV leads
DSDV and DSDV leads ZRP. With the increasing number
of connections, packet delivery ratio of AODV drops.
However, at the last with increasing the number of
connections, the packet delivery ratio of AODV rises and
leads ZRP but DSDV dominates them. In the beginning,
DSDV leads AODV and then comes ZRP. As the number
of connections increases, AODV leads DSDV but at the
last, DSDV dominates AODV and ZRP stays at the last as
shown in figure 5(a).In figure 5(b) DSDV leads over
AODV and AODV leads ZRP. However, with the increase
in number of connections, AODV leads DSDV but at the
last, DSDV dominates over AODV and ZRP remains at the
last.
V. CONCLUSION
In this paper we have presented a detail comparative
analyses of different routing protocols one each from three
different categories of MANET routing protocols under
different scenarios. Different performance metrics are
defined and used for the comparative analyses. The results
presented here are valid under the scenarios considered in
this paper. However there are many factors that may be
further investigated to study its impact on MANET routing
protocols. We believe that our work presented in this paper
may suitably be of help to the research community in this
challenging field.


(
[1] Comer D.E
and Archite
[2] Chowdhury
of three wir
streaming m
of inmic 20
2004.
[3] Perkins, C.
Sequenced
SIGCOMM
[4] C. E. Perk
Vector (AO
aodv-O~.txt
[5] Ahmed, M.
attributes,"

(a)
RE
E. Internetworkin
ectures,fourth edi
y, M.U.;Perera, D
reless multi hop a
mpeg4 traffic," Mu
04. 8th internatio
E. and Bhagwat,
Distance-Vector
M Comput. Commu
ins and E. M. R
ODV) Routing.
t, November 1998
; Yousef, S.; , "S
IEEE ISIE 2008 ,

Fig

EFERENCES
ng With TCP/IP:
ition, 2000.
D.; Pham,t.;"A pe
ad-hoc network r
ulti topic confere
onal , vol., no., pp
, P. 1994. Highly
routing (DSDV)
un. Rev. 24, 4 23
Royer. Ad Hoc
IETF lnternet D
8.
Self-configurable
,pp.2119-2124, Ju
(a)
gure 2. (a) Aver

: Principles, Prot
erformance comp
routing protocols
nce, 2004. procee
p. 516- 521, 24-26
y dynamic Destin
for mobile comp
4-244, Oct. 1994
On Demand Dis
Draft, draft-ietf-m
e zone routing pro
uly 2008.

rage end to delay

tocols,
arison
when
edings
6 Dec.
nation-
puters.
.
stance
manet-
otocol
[6]
[7]
[8]
[9]
[10
[11

vs no. of nodes (
(b)
] Pearlman, M
configuration
Selected Area
] Patel, B.; Sriv
protocols in M
2010 National
] A.Boomarani
Throughput an
Open Problem
] Abdul Hadi A
comparision o
Mobile Ad
Research,vol.3
0] S.A Ade,P
AODV,DSDV
Hoc Networ
Knowledgeme
2010.
1] T.K. Godder,
Hoc Routing P

(b) Packet deliver
)
M.R. and Hass,
for the zone r
s in Communicat
vastava, S.; , "Pe
Mobile Ad Hoc N
l Conference on, v
Malany, V.R.
nd Delay Compa
ms Cmpt. Math., v
Abd Rahman,Zuri
of AODV,DSDV
Hoc Networks
31,no.4,pp.566-57
P.A. Tijare,
V,OLSR and DS
rks,Int. J. o
ent Managemen
Performance Co
Protocols,CSEIJ

ry ratio vs no. of n
, Z.J., Determ
routing protocol
ions. v17 i8. 1395
erformance analy
Networks," Com
vol.,pp.1-5, 29-31
Sarma ,R.M.
arision of MANE
vol.2,no.3,septemb
iati Ahmed Zuka
V and I-DSDV R
s,European Jou
76,2009.
Performance
R Routing Proto
of Information
nt,vol.2,no.2 p
omparision and an
,vol1.no.1,april 2
(b)
nodes

mining the opti
. IEEE Journal
5-1414.
ysis of zone rout
mmunications (NC
1 Jan. 2010.
chandrasekaran,
ET Protocols, Int
ber 2009.
arnain, Performa
Routing Protocols
urnal of Scient
Comparision
ocols in Mobile
Technology
pp.545-548,decem
nalysis of Mobile
011.

mal
on
ting
CC),
,
t. J.
ance
s in
tific
of
Ad
and
mber
Ad

(a)

(a
(
(a) Rou

a) Average end to
(a)
(a) Routing o
uting overhead vs

o delay vs no. of c

overhead vs no. of
s no. of nodes(b)

connections(b) Pac

f connection (b)
Throughput vs no
(b)
cket delivery ratio

Throughput vs n
o. of notes
)
o vs no. of connec

no. of connections

ctions

(b)
s
ISBN: 978-81-920575-7-6 :: doi: 10. 74019/ISBN_0768
Literature Survey and Proposed Concept on Performance Issue of Multicasting
Routing Protocol in Wireless Sensor Network

Rahul Verma (Author)
M. Tech. Scholar (CSE)
BIST
Bhopal, India

Akhilesh A. Waoo (Author)
Asst. Prof. Dept. of M. Tech. (CSE)
BIST
Bhopal, India

P.S. Patheja (Author)
Asst. Prof. Dept. of M. Tech. (CSE)
BIST
Bhopal, India

AbstractNetworks are classified in to two main categories i.e.
wired network and wireless network. We have known that the
wireless network is important. To Study on wireless network
we have used Mobile Ad-hoc Networks (MANET) which is the
new research area. MANET is infrastructure less network
where routing is the way of transferring packet from one node
to another node in the network. Multicast routing is a
technique where a packet is send to a group of selected node
which is the member of that group, when multicast routing is
applied to a MANET overall performance of the network is
increased. MAODV and ODMRP routing protocol is the
example of multicasting routing. These protocols have
parameters like Packet delivery ratio, Control overhead,
Forwarding efficiency, end to end delivery and many others. In
This paper we are proposing a concept to solve some
outstanding problems like capacity of the network, power
efficiency of the network and most important thing optimal
designing of the network in MANETs. In this paper our
proposed concept will try to provide solution of the mentioned
problems. Here we will provide a design to network in such a
way where we can increase capacity, power efficiency and
other parameters which are predefine using insert new node
concept in existing network.
I. INTRODUCTION
A mobile ad hoc network (MANET) is a type of wireless
networks. This type depends on the mobile nodes and there
is no infrastructure in such type. There are no routers,
servers, access points or cables. Nodes (mobiles) can move
freely and in arbitrary ways, so it may change its location
from time to time. Each node may be a sender or a receiver,
and any node may work as a router and do all router
functions. This means that it can forward packets to other
nodes. Many applications of MANETs are implemented
and used until today like in meeting conferences; military
operations; search and rescue operations, all of them are
examples of MANET networks.
Multicasting in wireless ad-hoc network is a hot topic in
recent years. By multicasting, we mean the transmission of
packets from a source or a group of sources to a group of
one or more hosts that are identified by a single destination
address. Multicasting greatly reduces the transmission cost
when sending the same packet to multiple recipients.
Several multicast routing protocols have been recently
proposed for ad-hoc networks. These protocols can be
Tree-based or Mesh-based.
Dependent or Independent of the underlying
unicast routing protocol.
On-demand or Table-driven.

It can improve the usage of wireless links by sending
multiple copies of data packets using inherent broadcast
behavior of wireless transmission though reducing
transmission overhead and power consumption is a very
challenging part in multicasting. There are many
applications where one-to-many and many-to-many
transmissions are required.
Tree-based routing protocols have only single path from
source to destinations, so the broken links need to be
repaired. On the other hand, mesh-based routing protocol
maintains mesh of the connected components of the network
and therefore, has multiple paths from sources to multicast
destinations. This reduces repairing overhead due to
presence of alternate paths available in the network. Mesh-
based routing protocols lead to congestion under the
conditions of high traffic load which can result in low
packet delivery ratio. This paper summarizes the simulation
techniques and analysis of some of the multicast protocols
like MAODV [10] and ODMRP [11] in MANET
environment.
This paper is organized as follows: in section II we
present a brief survey on various research papers on
multicast routing protocol, section III include problem
finding in mobile ad-hoc network, section IV includes a
description of our proposed idea, section V including
expected outcome of proposed concept. Finally, section VI
provides concluding remarks and highlights our proposed
work.
II. LITERATURE SURVEY
In [1] paper, they have present a comparative
performance of two on-demand multicast routing protocols,
namely MAODV and ODMRP delivery ratio, control
overhead and forwarding efficiency in different multicast
effects such as increasing number of multicast sources and
receivers in the multicast group. This simulation composed
with 200 mobile nodes for this single active multicast group
in the network.
In Another paper [3] they have present simulation based
experiments to analyze the performance of On Demand
Multicast Routing Protocol by evaluating Packet Delivery
Ratio, End to End delay and average throughput. These
results are compared between ODMRP with AODV and
FSR routing protocols by varying number of nodes and
mobility. The comparison shows that ODMRP for ad-hoc
networks performs better as compared to AODV and FSR.
Here another developed technique which is define in
[12] have discuss, they have present a comparative
performance of three multicast protocols for Mobile Ad hoc
Networks ODMRP, AMRIS and MAODV focusing on the
effects of changes such as the increasing number of
receivers or sources and increasing the number of nodes.
Although some simulation results of MANET protocols
have been published before, these three protocols have not
been compared in isolation. In recent years, a number of
new multicast protocols have been proposed for ad hoc
networks. A systematic performance evaluation of these
protocols is done by performing certain simulations. The
applicability of multicast protocols to diverse situation are
also studied and discussed.
In [13], they have present a comparative performance
evaluation of three general-purpose on-demand multicast
protocols, namely ADMR, MAODV, and ODMRP,
focusing on the effects of changes such as increasing
number of multicast receivers or sources, application
sending pattern, and increasing number of nodes in the
network. They have use mobile networks composed of 100
or 200 nodes, with both a single active multicast group and
multiple active multicast groups in the network, in a wide
range of multicast scenarios. Although some simulation
results for these protocols have been published before, the
three protocols have not been compared, and prior studies
have focused on smaller networks using a small set of
simulation scenarios, many with only a single active
multicast group. They have focus here on the effects of the
protocols relative degree of on-demand behavior and their
performance in different multicast scenarios.
III. PROBLEM FORMULATION
If overhead in wireless network nodes decreases network
capacity, reducing the overhead will increase the capacity of
the network. It is postulated in this paper that the
introduction of an intelligently placed node can increase the
capacity of a network. To decrease the overhead
experienced within the network, nodes are required to have
the minimal number of neighbors possible. This can be
accomplished by lowering the transmission powers of
individual nodes to a level where the number of neighbors is
minimized but the network remains connected. The question
therefore arises, can the insertion of an additional node into
a MANET coupled with classical topology control
mechanisms lead to an increase in network capacity? If so,
where should this new node is added?

IV. PROPOSED WORK
Generally mobile ad-hoc networks (MANETs) are used
for military and civilian applications. There are still a lot of
interesting problems investigating the capacity and the
optimal design of these networks. In This paper we are
proposing a concept to solve some problems which is
finding in MANETs. Our main aim of proposed concept is
to increase the capacity and power efficiency of the
MANET using multicasting protocols like ODMRP,
MAODV. Here proposed concept providing solution for the
problems of increasing the capacity, power efficiency,
increasing packet delivery ratio, control overhead,
increasing forward efficiency and decreasing end-to-end
delay of a MANET. With the help of proposed concept we
will try to decrease the overhead of the network using
multicast routing protocol. Here we will try to insert new
node in the existing network. This new node will insert
when multicast routing will performed. With the help of this
new node we can reduce overhead, due to this we can
achieve the network capacity is maximized by minimizing
overhead. Basic principal of proposed concept is shown in
figure 1. here it is clear that overhead between two nodes is
reducing by adding new node. Due to this number of
messages can be easily transmitted between sources to
destination using alternate path.

Figure 1: Architecture of Proposed Concept

Figure 2 is showing flow chart of proposed concept. We
have start sending packets from source to sink and if all the
path already suffer from overhead problem then our recently
packet will also suffer these problem but proposed concept
is solving this problem using inserting new node in proper
location so we can divert or reduce overhead in network and
transfer packets form source to sink in easy way. We will
try to find proper location on the some observation like most
overhead path, less energy path, less distance and many
more.


Figure 2: Flow Chart of Proposed Concept

Advantages of multipath routing:-
Fault tolerance.
Load balancing.
Bandwidth utilization.
Decrease End-to-End delay.
V. EXPECTED OUTCOME
The multicasting routing protocol ODMRP will be
produce batter results as compare other multicasting routing
protocol. Our simulation will we concentrate on ODMRP
routing protocol. Following points for batter result
ODMRP used MESH topology from each source
ODMRP provides alternative paths and a link
failure need not trigger the re-computation of the
mesh, broken links will time out
ODMRP broadcasts the reply back to the source
ODMRP allows for multiple possible paths from
the multicast source back to the receiver
ODMRP does activate a multicast route
immediately
VI. CONCLUSION
A general conclusion is that, in a mobile scenario, mesh-
based protocols out performed tree-based protocols. The
availability of alternate routes provided robustness to
mobility. ODMRP is very effective and efficient in most of
our proposed parameter for simulation scenarios. However,
we did not cover every possible situation. While the
proposed concept of this paper will provide guidelines to
implementation of network, the final selection of a multicast
protocol should take into account other considerations
which cannot be valuated via simulation alone.
REFERENCES
[1] R. Pandi Selvam and V. Palanisamy A Performance Comparison of
MAODV and ODMRP Routing Protocols in Mobile Ad Hoc
Networks published in International Journal of Research and
Reviews in Ad hoc Networks (IJRRAN) Vol. 1, No. 3, September
2011, ISSN: 2046-5106.
[2] Abhijit Das Soumya Sankar Basu and Atal Chaudhuri A Novel
Security Scheme for Wireless Adhoc Network puvlished in Wireless
Communication, Vehicular Technology, Information Theory and
Aerospace & Electronic Systems Technology (Wireless VITAE),
2011 2nd IEEE International Conference 2011.
[3] Yudhvir Singh Yogesh Chaba Monika Jain and Prabha Rani
Performance Evaluation of On-Demand Multicasting Routing
Protocols in Mobile Adhoc Networks published in 2010
International Conference on Recent Trends in Information,
Telecommunication and Computing.
[4] Yudhvir Singh and Dr. Yogesh Chaba Security and Network
Performance Evaluation of KK' Cryptographic Technique in Mobile
Adhoc Networks published in IEEE International Advance
Computing Conference (IACC 2009) Patiala, India, 6-7 March 2009.
[5] E. M. Royer and C. E. Perkins, ``Multicast Operations of the Ad-Hoc
On-Demand Distance Vector Routing Protocol'', Proceedings of
ACM/IEEE MOBICOM'99, pp. 207-218, Seattle, WA, Aug. 1999.
[6] M. S. Corson and S. G. Batsell, `À Reservation-Based Multicast
(RBM) Routing Protocol for Mobile Networks: Initial Route
Construction Phase'', ACM/Baltzer Wireless Networks, vol. 1, no. 4,
pp. 427-450, Dec. 1995.
[7] L. Ji and M. S. Corson, `À Lightweight Adaptive Multicast
Algorithm'', Proceedings of IEEE GLOBECOM'98, pp. 1036-1042,
Sydney, Austral, Nov. 1998.
[8] E. Bommaiah, et al., `ÀMRoute: Adhoc Multicast Routing Protocol'',
Internet Draft, Aug. 1998, Work in progress.
[9] C. W. Wu, Y. C. Tay, and C. K. Toh, `Àd hoc Multicast Routing
Protocol utilizing Increasing id-numberS (AMRIS) Functional
Specification'', Internet Draft, Nov. 1998, Work in progress.
[10] Elizabeth M. Royer and Charles E. Perkins. Multicast Operation of
the Ad-hoc On-Demand Distance Vector Routing Protocol. In
Proceedings of the Fifth Annual ACM/IEEE International Conference
on Mobile Computing and Networking, Mobicom 99, pages 207
218, August 1999.
[11] S.-J. Lee, W. Su, and M. Gerla, On-Demand Multicast Routing
Protocol (ODMRP) for Ad Hoc Networks, Internet-Draft, draft-ietf-
manet-odmrp-01.txt, Jun. 1999.
[12] Aparna K Performance Comparison of MANET (Mobile Ad hoc
Network) Protocols (ODMRP with AMRIS and MAODV) published
in International Journal of Computer Applications (0975 8887)
Volume 1 No. 10 2010.
[13] Jorjeta G. Jetcheva and David B. Johnson A Performance
Comparison of On-Demand Multicast Routing Protocols for Ad Hoc
Networks published in December 15, 2004 CMU-CS-04-176.
ISBN: 978-81-920575-7-6 :: doi: 10. 74026/ISBN_0768
The Cohesive Countermeasure Schemes for Securing Wireless Sensor Networks
against Vulnerabilities and Attacks: Review

Ronnie D. Caytiles
Hannam University
Daejeon, Korea
Corresponding Author

Tai-hoon Kim
GVSA & University of Tasmania
Australia

Dominik lzak
University of Warsaw & Infobright
Warsaw, Poland

Abstract The term Wireless Sensor Network (WSN),
denotes a network of intelligent sensor nodes that monitors
physical and environmental conditions and can be deployed
anywhere and anytime. WSN provides many advantages, but it
is also coupled with new security threats and alters the
organizations overall information security risk profile.
Although implementation of technological solutions is the usual
respond to wireless security threats and vulnerabilities,
wireless security is primarily a management issue. Effective
management of the threats associated with wireless technology
requires a sound and thorough assessment of risk given the
environment and development of a plan to mitigate identified
threats. In this paper, we study the cohesive countermeasure
schemes for securing WSNs against identified vulnerabilities,
threats and attacks. We present an analysis to help network
managers understand and assess the various threats associated
with the use of wireless technology. A number of available
solutions for countering those threats are discussed.
Keywords- WSN, WSN Threats, WSN Security, sensor
nodes, encryption
I. INTRODUCTION
WSNs consist of multifunction sensor nodes that are
small in size and communicate wirelessly over short
distances. The sensor nodes integrate different properties for
sensing the environment, as well as data processing and
communication among other sensors. WSNs perform an
important role in many applications, such as battlefields
surveillance, patient health monitoring, home automation,
traffic control, environmental observation and building
intrusion surveillance.
The wireless networking has improved productivity
through increased accessibility to information resources and
easier, faster and less expensive network configuration.
WSNs provide convenience, cost efficiency, and ease of
integration with other networks and network components.
However, wireless technology also creates new threats. Since
communication for WSNs are through radio frequencies, the
risk of interception is greater than with wired networks.

If the message is not encrypted, or encrypted with a weak
algorithm, the attacker can read it, thereby compromising
confidentiality. Security objectives include: preserving
confidentiality, ensuring integrity, and maintaining
availability of the information and information systems.
WSNs present a handful of issues for network managers.
Unauthorized access points, unknown stations, captured
nodes, spoofed acknowledgements are just a few of the
problems addressed in WSN troubleshooting. Moreover,
onsite maintenance for remotely deployed sensor nodes is
infeasible, thus a thorough consideration of security solutions
and troubleshooting tools must be available.
In this paper, we identify the cohesive countermeasure
schemes that can address the risks offered by the threats and
attacks related to WSNs. Although these threats cannot be
totally eliminated, a desired level of security can be achieved
by adopting such countermeasure. The objective of this
paper is to assist managers in making decisions by providing
them with a basic understanding of the nature of the various
threats associated with wireless networking and available
countermeasures.
II. WSN SECURITY THREATS
This section covers the different attacks and threats that
relate to WSNs. Most of these attacks are similar to those
that apply to traditional networks. However a node capture
attack is a totally new and distinct phenomenon which does
not apply to traditional networks.
A. Node Capture Attack
One of the distinct attacks in WSNs is a node capture. In
this attack, an attacker gains full control over a sensor node
through a direct physical access. Then the attacker can easily
extract cryptographic primitives and obtain unlimited access
to the information stored on the memory chip of the captured
node through a reverse engineering process with the potential
to cause substantial damage to the entire WSN.
Three main factors that can aid the attackers during a
node capture attack:
If sensor nodes within the WSN share a key or keys
with neighboring nodes that is used to encrypt or
decrypt data.
A node capture attack has a great impact on the
structure or topology of a WSN.
The density of the WSN has a direct influence on
node capture attacks, having a similar affect to the
network structure.
B. Side Channel Attack
A side channel attack refers to any attack that is based on
information gathered from the physical implementation of a
cryptosystem, in contrast to vulnerabilities in the algorithm
itself [9]. For example the attacker monitors the power
consumption or the electromagnetic (EM) emanation from
such cryptographic devices, and then analyzes the collected
data to extract the associated crypto key.
Side channel attacks include:
Simple power analysis (SPA) is a technique that
involves directly interpreting power consumption
measurements collected during cryptographic
operations. No statistical analysis is required in such
an attack [5].
Simple electromagnetic analysis (SEMA) an
adversary is able to extract compromising
information from a single electromagnetic sample
[4].
In differential power analysis (DPA) an adversary
monitors the power consumed by cryptographic
devices, and then statistically analyzes the collected
data to extract a key in contrast to the simple power
analysis [5].
In differential electromagnetic analysis (DEMA),
instead of monitoring the power consumption, an
attacker monitors electromagnetic emanations from
cryptographic devices, and then the same statistical
analysis as that for the differential power analysis is
performed on the collected electromagnetic data to
extract secret parameters [4].
C. Denial of Services (DoS)
A Denial-of-Service attack (DoS) occurs when an
attacker continually bombards a targeted AP (Access Point)
or network with bogus requests, premature successful
connection messages, failure messages, and/or other
commands. DoS is an event that diminishes or eliminates a
networks capacity to perform its expected function through
hardware failures, software bugs, resource exhaustion,
malicious broadcasting of high energy signals,
environmental conditions, or any complicated interaction
between these factors [11].
Communication systems could be jammed completely if
such attacks succeed. Other denial of service attacks are also
possible, e.g., inhibiting communication by violating the
MAC protocol. These cause legitimate users to not be able to
get on the network and may even cause the network to crash.
These attacks rely on the abuse of protocols such as the
Extensible Authentication Protocol (EAP).
D. Software Attacks
In Software-based attacks on WSNs, an attacker may try
to modify the software code in memory or exploit known
vulnerabilities in the software code. A well-known example
of such an attack is a buffer overflow attack where a process
attempts to store data beyond the boundaries of a fixed
length buffer, thus, resulting in the extra data overwriting the
adjacent memory locations.
E. Routing Attacks
As described earlier, every node acts as a router in a
WSN. Routing and data forwarding are an important task for
sensor nodes. Routing protocols have to be energy and
memory efficient, but at the same time they have to be robust
against attacks and node failures. There have been many
power-efficient routing protocols proposed for WSNs.
However, most of them suffer from security vulnerabilities
of one sort or another. In the real world, a secure routing
protocol should guarantee the integrity, authenticity and
availability of messages in the existence of adversaries of
arbitrary power. Every authorized receiver should receive all
messages proposed for it and should be capable of proving
the integrity of every message and also the identity of the
sender.
Some of the routing protocol attacks are the following
[10]:
Black hole attacks or packet drop attack;
Spoofed, altered, or replayed attack;
Wormholes attack;
Selective forwarding attack;
Sinkhole attack;
HELLO flood attack;
Acknowledgement spoofing.
F. Traffic Analysis Attacks
All communication is WSNs is moving toward a base
station in many-to-one or many-to-few patterns. An attacker
is able to gather much information on the topology of the
network as well as the location of the base station and other
strategic nodes by observing traffic volumes and pattern.
There are two types of traffic analysis attacks in WSNs: a
rate monitoring attack and a time correlation attack. In a rate
monitoring attack an attacker monitors the packet sending
rate of nodes near the attacker and moves closer to the nodes
that have a higher packet sending rate. In a time correlation
attack an attacker observes the correlation in sending time
between a node and its neighbor node that is assumed to be
forwarding the same packet and deduces the path by
following the sound for each forwarding operation as the
packet propagates towards the base station [3].
G. Sybil Attacks
The Sybil attack is defined as a malicious device
illegitimately taking on multiple identities. For example, a
malicious node can claim false identities, or impersonate
other legitimate nodes in the network [1][2]. The Sybil attack
can affect a number of different protocols such as the
following:
Distributed storage protocols;
Routing protocols;
Data aggregation (used in query protocols);
Voting (used in many trust schemes);
Fair resource allocation protocols;
Misbehavior detection protocols.
To attack routing protocols a Sybil attack would rely on a
malicious node taking on the identity of multiple nodes, thus
routing multiple paths through a single malicious node.
However the Sybil attack can operate in different ways to
attack the protocols listed above.
H. Attacks on In-Network Processing
Data aggregation or in-network processing is very useful
in terms of reducing the communication overhead. However
there can be different types of attack on in-network
processing:
Compromise a node physically to affect aggregated
results [12];
Attack aggregator nodes using different attacks;
Send false information to affect the aggregation
results.
I. Attacks on Time Synchronization Protocols
Time synchronization protocols provide a mechanism for
synchronizing the local clocks of nodes in a sensor network.
Three of the most prominent protocols are the reference
broadcast synchronization (RBS) protocol, Timing-sync
Protocol for Sensor Networks (TPSN) and Flooding Time
Synchronization Protocol (FTSP).
Since most of the time synchronization protocols do not
consider security, an attacker can physically capture a
fraction of the nodes and injecting them with faulty time
synchronization message updates easily. This event can
make the nodes in the entire network out-of-sync with each
other.
J. Replication Attacks
There are two ways to launch replication attacks in
WSNs. First, an attacker can eavesdrop on communications
and resend old packets again multiple times in order to waste
its neighboring sensor nodes energy. Then, an attacker can
insert additional replicated hostile sensor nodes into the
WSN after obtaining some secret information from captured
sensor nodes or through infiltration.
III. WSN SECURITY THREATS COUNTERMEASURE
SCHEMES
A. Securing the Confidentiality of WSN Data Transmission
Several types of countermeasures exist for reducing the
risk of eavesdropping on wireless transmissions. The first
involves methods for making it more difficult to locate and
intercept the wireless signals. The second involves the use of
encryption to preserve confidentiality even if the wireless
signal is intercepted. The rest are attack specific
countermeasures.
1) Signal Hiding Techniques: In order to intercept
wireless transmissions, attackers first need to identify and
locate wireless networks. There are, however, a number of
steps that organizations can take to make it more difficult to
locate their wireless access points. The easiest and least
costly include the following: Turning off the service set
identifier (SSID) broadcasting by wireless access points,
Assign cryptic names to SSIDs, Reducing signal strength to
the lowest level that still provides requisite coverage or
Locating wireless access points in the interior of the
building, away from windows and exterior walls. More
effective, but also more costly methods for reducing or
hiding signals include: Using directional antennas to
constrain signal emanations within desired areas of coverage
or Using of signal emanation-shielding techniques,
sometimes referred to as TEMPEST, 1 to block emanation
of wireless signals.
2) The Use of Encryption: The best method for
protecting the confidentiality of information transmitted
over wireless networks is to encrypt or scramble all wireless
traffic and communications over the network. This is
especially important for organizations subject to regulations.
Two types of countermeasures can significantly reduce
the risk of such attacks: strong encryption and strong
authentication of both devices and users.
3) Countermeasures against Denial of Service Attacks:
Wireless communications are also vulnerable to denial-of-
service (DoS) attacks. Organizations can take several steps
to reduce the risk of such unintentional DoS attacks. Careful
site surveys can identify locations where signals from other
devices exist; the results of such surveys should be used
when deciding where to locate wireless access points.
Regular periodic audits of wireless networking activity and
performance can identify problem areas; appropriate
remedial actions may include removal of the offending
devices or measures to increase signal strength and coverage
within the problem area. The summary of countermeasures
for different DoS attacks are depicted in Table 1.
TABLE I. WSN LAYERS AND DOS DEFENSES
Sensor network layers and denial-of-services defense
Network
layer
Attacks Defenses
Physical Jamming

Tampering
Spread-spectrum,
priority
message, lower Duty
cycle,
region mapping,
mode change
Tamper-proofing,
hiding
Link Collision
Exhaustion
Unfairness
Error-correction
code
Rate limitation
Small frames
Network and
Routing
Neglect and
greed
Homing
Misdirection

Redundancy,
probing
Encryption
Egress filtering,
authentication,
Black Holes
monitoring
Authorization,
monitoring,
redundancy
Transport Flooding
De-
synchronization
Client puzzles
Authentication

4) Countermeasures against Software Attacks: The
following countermeasure can be considered to secure the
WSN software and be protected from being exploited by
malicious users:
Software authentication and validation, e.g., remote
software-based attestation for sensor networks;
Defining accurate trust boundaries for different
components and users;
Using a restricted environment such as the Java
Virtual Machine;
Dynamic run-time encryption/decryption for
software. This is similar to encryption/decryption of
data except that the code running on the device is
encrypted. This can prevents a malicious user from
exploiting the software;
Hardware attestation. The trusted computing group
platform and next generation secure computing base
provide this type of attestation. A similar model
could be used in sensor networks.
5) Countermeasures against Sybil Attacks: Proposed
solutions to the Sybil attack include the following:
radio resource testing which relies on the assumption
that each physical device has only one radio;
random key pre-distribution which associates the
identity of each node to the keys assigned to it and
validates the keys to establish whether the node is
really who it claims to be;
registration of the node identities at a central base
station; and
Position verification which makes the assumption
that the WSN topology is static.
B. Securing the WSN Access Point
Insecure, poorly configured wireless access points can
compromise confidentiality by allowing unauthorized access
to the network. Organizations can reduce the risk of
unauthorized access to wireless networks by taking these
three steps:
1. Eliminating rogue access points. The best method for
dealing with the threat of rogue access points is to
use 802.1x on the wired network to authenticate all
devices that are plugged into the network. Using
802.1x will prevent any unauthorized devices from
connecting to the network;
2. Properly configuring all authorized access points.
Organizations also need to ensure that all authorized
wireless access points are securely configured. It is
especially important to change all default settings
because they are well known and can be exploited by
attackers; and
3. Using 802.1x to authenticate all devices. Strong
authentication of all devices attempting to connect to
the network can prevent rogue access points and
other unauthorized devices from becoming insecure
backdoors. The 802.1x protocol discussed earlier
provides a means for strongly authenticating devices
prior to assigning them IP addresses.
C. Other Countermeasures
The following are several tips for countering attacks and
ensuring security for WSNs:
Use anti-virus and anti-spyware software, and a
firewall. Computers on a wireless network need the
same protections as any computer connected to the
Internet. Install anti-virus and anti-spyware software,
and keep them up-to-date. If your firewall was
shipped in the off mode, turn it on;
Turn off identifier broadcasting. Most wireless
routers have a mechanism called identifier
broadcasting. It sends out a signal to any device in
the vicinity announcing its presence. You dont need
to broadcast this information if the person using the
network already knows it is there. Hackers can use
identifier broadcasting to home in on vulnerable
wireless networks. Disable the identifier broadcasting
mechanism if your wireless router allows it;
Change the identifier on your router from the default.
The identifier for your router is likely to be a
standard, default ID assigned by the manufacturer to
all hardware of that model. Even if your router is not
broadcasting its identifier to the world, hackers know
the default IDs and can use them to try to access your
network. Change your identifier to something only
you know, and remember to configure the same
unique ID into your wireless router and your
computer so they can communicate. Use a password
thats at least 10 characters long: The longer your
password, the harder it is for hackers to break;
Change your routers pre-set password for
administration. The manufacturer of your wireless
router probably assigned it a standard default
password that allows you to set up and operate the
router. Hackers know these default passwords, so
change it to something only you know. The longer
the password, the tougher it is to crack;
Allow only specific computers to access your
wireless network. Every computer that is able to
communicate with a network is assigned its own
unique Media Access Control (MAC) address.
Wireless routers usually have a mechanism to allow
only devices with particular MAC addresses access
to the network. Some hackers have mimicked MAC
addresses, so dont rely on this step alone;
Turn off your wireless network when you know you
wont use it. Hackers cannot access a wireless router
when it is shut down. If you turn the router off when
youre not using it, you limit the amount of time that
it is susceptible to a hack;
Dont assume that public hot spots are secure.
Many cafs, hotels, airports, and other public
establishments offer wireless networks for their
customers use.
The above mentioned countermeasure tips are applicable
to wireless networks, yet, some of them may apply to WSNs
directly.
IV. CONCLUSION
WSNs popularity is due in part to their small size and
low cost of sensors, their operations and their networking
behavior, enabling them to provide significant advantages for
many applications that would not have been possible in the
past. They can provide a numerous opportunities to increase
productivity and minimize costs.
In this paper we have identified and described the
different vulnerabilities, threats and attacks that could
possibly put WSNs in a vital or critical situation. These
threats are analyzed and categorized to find and classify the
countermeasure schemes applicable to mitigate the risks for
each category. A reasonable and practical level of security is
possibly achieved by adopting a systematic approach to
assessing and managing risk regardless of all the threats and
attacks.
A regular study on the emerging threats should be
considered to ensure the security to be maintained at a
desired level. Many research challenges must be addressed in
order for WSNs to be realistically implemented and for them
to become part of everyday life.

REFERENCES
[1] T. Roosta, S. Shieh, S. Sastry: Taxonomy of Security Attacks in
Sensor Networks, 1st IEEE Int. Conference on System Integration
and Reliability Improvements 2006, Hanoi (2006) pp. 1315.
[2] A. Perrig, J. Newsome, E. Shi, D. Song: The Sybil Attack in Sensor
Networks: Analysis and Defences, 3rd Int. Symposium on
Information Processing in Sensor Networks 2004 (ACM Press, New
York, USA 2004) pp. 259268.
[3] J. Deng, R. Han, S. Mishra: Countermeasures Against Traffic
Analysis Attacks in Wireless Sensor Networks, First IEEE/Cerate Net
Conference on Security and Privacy in Communication Networks
(SecureComm) 2005, Athens (2005) pp. 113124.
[4] K. Gandol, C. Mourtel, F. Olivier, Electromagnetic Analysis:
Concrete Results, Published in C_ .K. Ko_c, D. Naccache, and C.
Paar, Eds., Cryptographic Hardware and Embedded Systems { CHES
2001, vol. 2162 of Lecture Notes in Computer Science, pp. 251{261,
Springer-Verlag, 2001.
[5] Power Analysis, http://en.wikipedia.org/wiki/Power_analysis,
Accessed: December, 2011.
[6] Wireless Sensor Networks, http://en.wikipedia.org/wiki/
Wireless_sensor_network, Accessed: December 2011.
[7] J.N. Al-Karaki, A.E. Kamal: Routing techniques in wireless sensor
networks: A survey, IEEE Wirel. Commun. 11(6), 628 (2004).
[8] F. Ye, H. Luo, J. Cheng, S. Lu, L. Zhang: A Two-Tier Data
Dissemination Model for Large-Scale Wireless Sensor Networks,
ACM/IEEE MOBICOM 2002 (2002) pp. 148159.
[9] T. Roosta, S. Shieh, S. Sastry: Taxonomy of Security Attacks in
Sensor Networks, 1st IEEE Int. Conference on System Integration
and Reliability Improvements 2006, Hanoi (2006) pp. 1315.
[10] Y.-C. Hu, A. Perrig, D.B. Johnson: Adriane: A Secure On-Demand
Routing Protocol for Ad Hoc Networks, Annual ACM Int.
Conference on Mobile Computing and Networking (MobiCom) 2002
(2002).
[11] Denial-of-Service Attack, http://en.wikipedia.org/wiki/Denial-of-
service_attack, Accessed: December, 2011.
[12] K. Kifayat, M. Merabti, Q. Shi, D. Llewellyn-Jones: Group-based
secure communication for large scale wireless sensor networks, J.
Information Assurance Security. 2(2), 139147 (2007).

ISBN: 978-81-920575-7-6 :: doi: 10. 74033/ISBN_0768
A Robust Design of RTU Sensor Nodes Communication for SCADA Systems

Sabah Mohammed
Department of Computer Science,
Lakehead University,
CANADA
Ronnie D. Caytiles
Department of Multimedia
Engineering, Hannam University
Daejeon, Korea
Tai-hoon Kim
GVSA and University of Tasmania
Australia

Abstract SCADA Communication plays a vital role for
Supervisory Control and Data Acquisition (SCADA)
Monitoring Systems. The early SCADA communication took
place over radio, modem, or dedicated serial lines. Today, it is
much more common for SCADA communications to travel
over LAN or WLAN. This process of communication over a
SCADA system involves several different SCADA system
components that include the sensors and control relays,
Remote Terminal Units (RTUs), SCADA master units, and the
overall communication network. Advanced RTUs where most
SCADA communication is gathered within the system, are
deployed for a more complete monitoring. The different values
from inputs and outputs, referred to as SCADA points, are
sent from individual sensors to the RTU which is responsible
for forwarding these SCADA communications to the master
station, or Human-Machine Interface (HMI). This paper
proposes an RFGSM Enabled SCADA RTU Communication
for Sensor Nodes that utilizes the available communication
networks such as cellular phone network and radio frequency
for exchanging data to minimize cost and human effort. The
approach makes use of the GSM network, a custom RF
solution, a low-cost microcontroller, and software embedded to
microcontrollers is developed to run the overall process.
Keywords- RF, GSM, RTU, SCADA systems, Sensor Nodes
I. INTRODUCTION
The traditional Supervisory Control and Data Acquisition
(SCADA) communication used the combination of radio and
direct serial or modem connections. Today, it is much more
common for SCADA communications to travel over LAN or
WLAN. This process of communication over a SCADA
system involves several different SCADA system
components that include the sensors and control relays,
Remote Terminal Units (RTUs), SCADA master units, and
the overall communication network. Advanced RTUs where
most SCADA communication is gathered within the system,
are deployed for a more complete monitoring. The different
values from inputs and outputs, referred to as SCADA
points, are sent from individual sensors to the RTU which is
responsible for forwarding these SCADA communications to
the master station, or Human-Machine Interface (HMI).

SCADA communications occur between Master Station
and the RTUs, and between the RTUs and the remotely
deployed sensor nodes. As SCADA communications
between the Master station and the RTUs nowadays utilizes
the LAN, WLAN, and the Internet, the link between sensor
nodes and RTUs cannot incorporate these technologies.
Sensor nodes contain SCADA communication data that are
sensitive and needs to be securely transmitted to RTUs. The
utilization of LAN for sensor nodes can be expensive to
install at remote sites, and it may not be feasible for an
organization to immediately incorporate all of their remote
sites into the LAN network.
This paper is concerned with the prototype design and
development of a communication scheme that would provide
remote access to the sensor nodes, through the combination
of Radio Frequency and Global System for Mobile
communications network. This proposed communication
scheme introduces the two different wireless communication
technologies to allow data communicate between the sensor
nodes and RTUs of SCADA systems. Understanding the
characteristics, properties, and behavior of GSM, RF and
different hardware components are vital to this concept. The
wireless processes will be controlled via a low-cost
microcontroller. The GSM and RF communication media are
fully utilized by introducing new network protocols.
The rest of this paper are organized as follows: Section II
discuss about RTU in SCADA systems; the proposed
scheme for communication among sensor nodes of RTU is
outlined in Section III; and the concluding remarks in
Section IV.
II. RTU IN SCADA SYSTEMS
A remote terminal unit (RTU) is a microprocessor-
controlled electronic device that interfaces objects in the
physical world to a distributed control system or SCADA
(supervisory control and data acquisition system) by
transmitting telemetry data to the system and/or altering the
state of connected objects based on control messages
received from the system. Another term that may be used for
RTU is remote telemetry unit; the common usage term varies
with the application area generally [9].
RTUs, IEDs, PLCs and DCS are increasingly beginning
to overlap in responsibilities, and many vendors sell RTUs
with PLC-like features and vice versa. The industry has
standardized on the IEC 61131-3 functional block language
for creating programs to run on RTUs and PLCs, although
nearly all vendors also offer proprietary alternatives and
associated development environments.
III. RF AND GSM ENABLED SENSOR NODES
COMMUNICATION
This section outlines the proposed scheme for
establishing a link between the remotely deployed sensor
nodes in geographical area and the RTUs. The
communication link utilizes the numerous capabilities of the
Radio Frequency (RF) and Global System for Mobile
Communications (GSM) to provide for data transmission for
SCADA systems. Fig. 1 illustrates the concept for the main
structure of the communication link.

Figure 1. RF-GSM Enabled Communication Link Structure for Sensor
Nodes and RTUs.
The communication link structure is composing of three
components that include:
Sensor Nodes
Mobile Network Controller (NC)
Central Server (CS) for RTUs
The Sensor Node is capable of performing some
processing, gathering sensory information and
communicating with other connected nodes in the network. It
produces a measurable response to sensory environment,
processes data for a variety of application, stores data
temporarily and transports the data to the Central Server for
RTUs if necessary through the Network Controller.
The Mobile Network Controller (NC) serves as router
device that is installed on high points like power poles. The
principal purpose of NC is data transfer and information
routing between Sensor Nodes and Central Server for RTUs.
The Central Server (CS) for RTUs manages the
collection of data from network devices. It also transfers the
data to the database for storage and retrieval.

Figure 2. A more detailed concept for the communication link between
RTUs and Sensor Nodes.
Furthermore, to illustrate the detail concept, Figure 2
shows how the data traverse from each of the devices.
Concisely, an active sensing device receives signals
proportional to the sensory environment referred to as
SCADA points. This SCADA points are periodically read,
accumulated and stored at a low cost but high-performance
microcontroller. The RF transmitter and receiver provide the
window for the sensor nodes to communicate with NC.
Microcontroller, RF Tx\Rx, and the sensing device form a
node. The node can include more components depending on
the additional functions added to the system. To enhance the
system reliability; a non-volatile EEPROM is used to store
the meter reading and other important data at a node or NC;
and reload the stored data at startup.
At the Mobile Network Controller, the utilitys orders to the
system are received as Short-Messaging-Service (SMS)
messages through a GSM modem. A microcontroller
analyzes the message and sends the proper commands to all
nodes in its area or to a specific node only. Any readings,
alerts or reports sent by the nodes are received in the NC,
reshaped to an SMS message form and sent again to the
central server of the RTU.
A. Microcontroller
A microcontroller is a small computer on a single
integrated circuit consisting of a relatively simple CPU
combined with support functions such as a crystal oscillator,
timers, watchdog timer, and serial and analog I/O etc.
The microcontroller will play as a processor in a Node
and NC, it controls and facilitates the data transmission
between the node and network controller.
There are two kinds of Microcontroller use in the design.
Since NC requires more USART for interfacing with GSM
and RF module, a microcontroller with two or more USART
is the choice. In this case the low-cost but powerful Arduino
Mega Development Board is the option, it satisfied the
requirement for the need of more USART. Arduino Mega is
an open-source microcontroller based in ATmega1280 [10].
The summary of the parameters for Arduino Mega is listed
on Table 1.
TABLE I. SUMMARY OF THE ARDUINO MEGA
Microcontroller ATmega 1280
Operating Voltage 5 V
Input Voltage
(Recommended)
7 12 V
Input Voltage (Limits) 6 20 V
Digital I/O Pins 54 (of which 14 provide
PWM output)
Analog Input Pins 16
DC Current per I/O Pin 40 mA
DC Current for 3.3V Pin 50 mA
Flash Memory 128 KB (of which 4KB
used by bootloader)
SRAM 8 KB
EEPROM 4 KB
Clock Spped 16 MHz
Serial USARTs 4

The Node requires only less number of pins for inputs and
outputs, one USART for interfacing with RF module, and
less memory size microcontroller. The Arduino Diecimila is
the appropriate microcontroller for this device, Arduino
Diecimila is a microcontroller board based on the
ATmega168, it is also an open source microcontroller which
the predecessor of Arduino Mega. Below is the summary of
the Diecimila shown in Table 2.
TABLE II. SUMMARY OF THE ARDUINO DIECIMILA
Microcontroller ATmega 328
Operating Voltage 5 V
Input Voltage
(Recommended)
7 12 V
Input Voltage (Limits) 6 20 V
Digital I/O Pins 14 (of which 6 provide
PWM output)
Analog Input Pins 8
DC Current per I/O Pin 40 mA
DC Current for 3.3V Pin 50 mA
Flash Memory 32 KB (of which 2KB
used by bootloader)
SRAM 2 KB
EEPROM 1 KB
Clock Spped 16 MHz
Serial USARTs 1

B. Global System for Mobile Communications
The basic idea in using GSM communications in NCs is
to receive order messages from the utility in SMS form
(Central Server for RTUs). The Microcontroller executes the
proper routine related to the message content and passes its
commands to the Nodes. Reversely, the system sends its
replies, whether they are readings, reports or alerts, to the
utilitys central server again in SMS form via the GSM
network. The Microcontroller highly benefits from the GSM
Module feature that it alerts for any action within the GSM
network, so the Microcontroller can always monitor what is
delivered to the GSM module from network.
In this proposed communication link scheme, for the
proof of concept an old Sony Ericsson T230 is used as
alternative to GSM module; it is much cheaper and available
to the local market. Figure 3 shows the block diagram of
GSM interfacing with Microcontroller in NC.

Figure 3. Block Diagram for GSM interfacing with the Microcontroller in
a Network Controller.
The operating functions of the GSM module (SE-T230)
are accessed through the use of AT commands; these AT
commands are conforming to GSM 03.04 standards. AT
commands are available via the serial interface of the T230
for function implementation. The AT standard is a line-
oriented command language. Each command is made up of
three elements: the prefix, the body, and the termination
character.
The prefix consists of the letters AT
The body is made up of individual characters; it
consists of a name and (if applicable) associated
values
The default termination character is <CR>(=
0x0D)
o Example: AT+CMGR="12"
This command reads the message with index 12 stored in
the default SIM card place. Commands for the SE-T230 can
be categorized in two categories:
Setup commands: used to set the different parameters
of the SE-T230 (e.g. Baud rate, response type)
Communication Commands: used to establish
communication using the SE-T230 (e.g. Read &
Send SMS)
C. RF Module
An important goal of the design is to make the system
cost effective; instead of putting GSM module on every node
in the system that wish to get readings, the RF modules to
link the NC with the surrounding Nodes. This will reduce
cost and will produce a nice challenge to stand for.
RF modules are specified by their operating frequency,
modulation type, maximum data rate and maximum distance
of operation (in both open fields and in buildings).
According to the communication theory, the antenna
length of the modules should be about 17 cm. No special
antennas where used to this design because the transmitter
and the receiver were too close to each other for the demo
purposes, but if they are too spaced and larger distance the

consideration of using the special antenna is highly
recommended.
For the proof of concept purposes, a low-cost 433 MHz
RF and the transmitter range is about 100 meters between
buildings(actually about 75 meters accuracy). The use of
sophisticated RF solution which has a range of around 2km
is well suggested in actual implementation.
D. Relay
The Electronic brick -5V Relay module uses HLS8L
relay module to control high-voltage electrical devices
(maximum 250V). This relay is controlled by
microcontroller with RF to enable remote connect and
disconnection. For demo purposes this design will use a
relay with maximum 250V and 7A.
E. LCD Module
The SerLCD v2.5 is a simple and cost effective solution
for interfacing to LCD based on the HD44780 controller.
The SerLCD module takes incoming 9600 bps TTL level
signal and displays those characters on the LCD screen.
The consideration of choosing SerLCD is the number of
pins necessary to interface to microcontroller. The typical
LCD used 11 pins to interface with microcontroller, while
SerLCD has only three wires - 5V, GND, and Signal are
needed to interface to the LCD.
The HD44780 dot-matrix liquid crystal display controller
and driver LSI displays alphanumeric. It can be configured to
drive a dot-matrix liquid crystal display under the control of
a 4- or 8-bit microprocessor. Since all the functions such as
display RAM, character generator, and liquid crystal driver,
required for driving a dot-matrix liquid crystal display are
internally provided on one chip, a minimal system can be
interfaced with this controller/driver.
F. Communication Link Software/Program
As mentioned above, an embedded software or program
will run the overall process in the Central Server such
sending command and receiving response from the Network
Controller. To develop this software a specific programming
language will be used considering the compatibility issues
for the RTU, microcontroller, and sensing devices.
The Central Server for RTUs is the main harvesting
component that collects data from all NCs (and Nodes) in the
system and manages process controlling too. The automation
of the system should mainly be implemented in the Central
Server. It should be connected to a database to store various
data and transactions from different nodes in the system.
IV. CONCLUSION
In SCADA systems, data communication plays a vital
role for monitoring and controlling the processes of the
entire systems. There are two main communication areas for
SCADA systems, that is, between SCADA Master Stations
and RTUs; and between the remotely deployed sensor nodes
and the RTUs. Advanced RTUs where most SCADA
communication is gathered within the system, are deployed
for a more complete monitoring. The different values from
inputs and outputs, referred to as SCADA points, are sent
from individual sensors to the RTU which is responsible for
forwarding these SCADA communications to the master
station, or Human-Machine Interface (HMI).
This paper have proposed an RFGSM Enabled SCADA
Communication link between the RTUs and Sensor Nodes
that utilizes the available communication networks such as
cellular phone network and radio frequency for exchanging
data to minimize cost and human effort. The approach makes
use of the GSM network, a custom RF solution, a low-cost
microcontroller, and embedded software for microcontrollers
is developed to run the overall process.

REFERENCES
[1] M. Choi, R. Robles, E. Cho, BJ Park, SS Kim, GC Park, and TH Kim,
A Proposed Architecture for SCADA System with Mobile Sensors,
The Journal of Korean Institute of Information Technology, http://ki-
it.or.kr/paper/paper_08_05/8%EA%B6%8C5%ED%98%B8/A%20
Proposed%20Architecture%20for%20SCADA%20System%20with%
20Mobile%20Sensors.pdf, Accessed: December, 2011
[2] R. Robles, M. Choi, TH Kim, The Function of SCADA in Critical
Infrastructure, The Journal of Korean Institute of Information
Technology, 2009.
[3] ISO/IEC 24730-1 Information technology -- Real-time locating
systems (RTLS) -- Part 1: Application program interface (API).
[4] Introduction to PLC and SCADA, http://tesla-
motors.connybrookus.com/2011/12/15/
introduction_to_plc_and_scada/ Accessed: December, 2011
[5] D. Bailey and E. Wright, "Practical SCADA for Industry", Elsevier,
2003.
[6] ISO/IEC 19762-5 Information technology - Automatic identification
and data capture (AIDC) techniques - Harmonized vocabulary - Part
5: Locating systems.
[7] GAO-04-628T. Critical infrastructure protection; Challenges and
efforts to secure control systems. Testimony Before the
Subcommittee on Technology Information Policy, Intergovernmental
Relations and the Census, House Committee on Government Reform.
March 30, 2004. http://www.gao.gov.new.items/d04628t.pdf,
Accessed: December, 2011.
[8] Wireless Sensor Network, http://en.wikipedia.org/wiki/
Wireless_sensor_network, Accessed: December, 2011.
[9] Remote Terminal Unit, http://en.wikipedia.org/wiki/
Remote_Terminal_Unit, Accessed: December, 2011.
[10] Arduino, http://en.wikipedia.org/wiki/Arduino, Accessed:
December, 2011.

ISBN: 978-81-920575-7-6 :: doi: 10. 74040/ISBN_0768
Secured SCADA System in Smart Grid Deployment

Yvette E. Gelogo Tai-hoon

Multimedia Engineering Department, Hannam University KimCulture Design Engineering Research Center,
133 Ojeong-dong, Daeduk-gu, Daejeon, Korea Hannam University,

Abstract-Supervisory Control and Data Acquisition
(SCADA)is a system that hasbeen a stalwart of electric utility
operations. It becoming increasingly complex as new
technologies arrives and new issues emerge on the road to a
modern electric grid.It is very important to keep tract the
security issues that arise together with its improvement. This
paper discusses the SCADA system vulnerabilities and
formulates the possible security measures to mitigate these
vulnerabilities as SCADA system becomes a big part of Smart
Grid.
Keywords:SCADA, Smart Grid, Cryptography, Security

I. INTRODUCTION
A smart grid is a digitally enabled electrical grid that
gathers, distributes, and acts on information about the
behavior of all participants (suppliers and consumers) in
order to improve the efficiency, importance, reliability,
economics, and sustainability of electricity services.
Power grid is managed through control centers,
electronic monitoring/control equipment at substations, and
an extensive communication network. Supervisory Control
and Data Acquisition (SCADA) systems are used
extensively to control and monitor the national power. Many
tests have been done and much vulnerability was discovered,
this vulnerabilities has not been publically disclosed, but
could be exploited by a knowledgeable attacker [1].
As a part of Smart grid, SCADA system play a vital
role in nations critical infrastructures and taking it to the
next level of enhance operations, SCADA system has been
integrated to IP-based network which make the SCADA
system take risk in terms of security. Though there have
been a lot of researches and studies made to mitigate these
vulnerabilities, this is not enough and the study is still going
on.

This paper proposes a methodology for securing SCADA
system in Smart Grid environment.The two focused is for
the network security to ensure that the packets sent
havereached for its intended receipts. And the other one is
the typical attacks that are being launch through malicious
software. These two ways of attacks uses different
mechanisms to launch an attack to host. It is better to study
both to have a more secure SCADA system.
For packet security, there is a lot of cipher scheme
being used to make sure the integrity of the message.The
attacks that can possibly launch with unsecured network is
the attacker can intercept the packet intended for another
SCADA components that is being modified by the attacker
and replace it malicious programs that can destroy the
intended recipients or host. Though this kind of attack will
cause a big damage, there is no concrete security
mechanism to combat such attack profoundly. Another is
when the attack is being launch through downloaded and
installed software which are malware, Trojan viruses and
other virus programs that the host is not aware of its
existence because also of the high cryptography used that
the IDS or NIISD , firewall cannot detect. This will cause a
lot of damage before it can be detected. This paper discusses
the mechanism to mitigate these attacks.

II. BACKGROUND
The following are the SCADA System Components:
1. Operator: Human operator who monitors the
SCADA system and performs supervisory control
functions for the remote plant operations.
2. Human machine interface (HMI): Presents data to
the operator and provides for control inputs in a
variety of formats, including graphics, schematics,
windows, pull down menus, touch-screens, and so
on.
3. Master terminal unit (MTU): The MTU presents
data to the operator through the HMI, gathers data
from the distant site, and transmits control signals

t
t
a
F
p
r
c
u
to th
betw
low
becau
interr
4. Comm
betw
Com
wirel
telep
5. Remo
in th
signa
from
MTU
betw
relati
close
micro
interf
distri
contr
trans
alteri
contr
Anot
Telem
with
moni
and t
Statio
Fig
SCADA
typical contro
terminal units
actuators, and
For the most
performed by
referred to a
consists of a
usually set to
he remote site
ween the MTU
and the contr
use of possi
ruptions.
munications m
ween the M
mmunication c
less or wired n
hone network
ote terminal u
he master/sla
als to the dev
m these device
U. An RTU
ween the RT
ively high an
ed loop. A Re
oprocessor-co
faces objects
ibuted control
rol and da
mitting telem
ing the state
rol messages
ther term that
metry Unit, t
the applicat
itors the field
transmits all th
on.
gure 1.Typical SC
systems are
ol system co
s (RTU) conne
d relaying info
t part, the br
y the Remot
as the RTU)
programmabl
o specific req
e. The transm
U and the remo
rol method is
ible time del
means: Comm
MTU and re
can be thro
networks, or t
k.
unit (RTU): F
ave architectu
vice under con
es, and transm
may be a PL
TU and con
nd the control
emote Termin
ontrolled elec
s in the phy
l system or SC
ata acquisiti
metry data to
of connected
s received f
may be used
the common
tion area ge
d digital and
he data to the
CADA System arc
primarily c
onsists of one
ected to a var
ormation to a
rains of a SC
te Terminal
. The Remo
le logic conve
quirements, ho
mission rate of
ote site is rela
usually open
lays or data
munication m
emote contro
ough the Int
the switched p
unctions as a
ure. Sends c
ntrol, acquires
mits the data t
LC. The data
ntrolled devic
method is u
nal Unit (RTU
tronic device
ysical world
CADA (superv
ion system)
the system a
d objects base
from the sy
for RTU is Re
usage term
enerally. An
analog param
Central Moni
chitecture
ontrol system
e or more r
riety of sensor
master statio
CADA system
Units (some
te Terminal
erter. The RT
owever, most
f data
atively
n loop
flow
method
ollers.
ternet,
public
slave
ontrol
s data
to the
a rate
ce is
usually
U) is a
e that
to a
visory
) by
and/or
ed on
ystem.
emote
varies
RTU
meters
toring
ms. A
emote
rs and
on [2].
m are
etimes
Units
TU are
RTU
allo
the
the
hum
usu
ofte
cha
con
hum
real
mo
per
ben
and
com
out
SCA
dig

its
tim
clos
gen
inst
to
tech
and
inte
pro
a bi
stan
num
netw
atta
sys
as
atta
netw
exp
that
imp
esp
wor
set
alig
ow human inte
RTU might
speed can b
man interventi
ually automati
en, a SCADA
anges to fun
nsidered closed
man interventi
l-time data on
re efficient
rsonnel safety
nefits are mad
d software in
mmunication
tside network
ADA architec
ital data transm
III.
Supervisory
communicati
me security wa
sed nature o
neral model o
talled in a priv
attack the s
hnology mode
d have started
erfaces and
otocols [6]. Be
igger issue.
The transitio
ndardized and
mber of conne
works and the
acks. Conseq
tems has com
extremely v
acks [7][8].
SCADA sys
work connec
posed SCADA
t accompany
plications res
pecially since m
rlds critical i
of network fa
gnment of netw
ervention, for
control the se
be changed or
ion. In additio
ically logged
A system wi
nction optim
d loop system
ion [2]. SCAD
n production,
control para
y, and reduce
de possible by
SCADA syste
protocols an
ks, including
ctures rely hea
mission.
SCADA VUL
Control and D
ion protocols
as not the pri
of the commu
of trusting the
vate network.
system becau
ernized, they
running more
have implem
ecause of this u
on from prop
d open solution
ections betwee
e Internet has m
quently, the
me into question
vulnerable to
stems are bei
ctivity. This
A networks to
todays TCP
sulting from
many of these
infrastructure.
acts can organ
work defenses
r instance, in a
etting of a con
r overridden
on, any chang
d for and/or d
ill monitor a
mally; SCADA
ms and run with
DA provides m
, operations,
adigms, impro
es costs of o
the use of sta
ems combined
nd increased
the Interne
avily on standa
LNERABILIT
Data Acquisit
designed de
mary concern
unications ne
e data on the
Outsiders wil
use of its cl
have become
modern servi
mented remot
upgrades, secu
prietary techno
ns together wi
en SCADA sy
made them mo
security of
n as they are i
cyberwarfar
ing modernize
development,
o the significa
P/IP networks
m this expos
e systems con
. Only with a
nizations fully
s and security
a factory settin
nveyer belt, a
at any time
ges or errors
displayed. M
and make slig
A systems a
h relatively lit
management w
implementatio
oves plant a
operation. The
andard hardwa
d with improv
connectivity
et. [4]. Mode
ard protocols a
IES
ion Systems a
ecades ago, th
n because of t
etworks and t
em since it w
ll find it diffic
lose nature.
e interconnect
ices such as W
te configurati
urity has becom
ologies to mo
ith the increas
ystems and off
ore vulnerable
SCADA-bas
increasingly se
re/cyberterrori
ed with TCP
, however, h
ant security ris
s. The potent
sure are hig
ntrol much of t
a comprehensi
demonstrate t
policy.
ng,
and
by
are
Most
ght
are
ttle
with
ons
and
ese
are
ved
to
ern
and
and
hat
the
the
was
cult
As
ted
Web
ion
me
ore
sed
fice
e to
sed
een
ism
/IP
has
sks
tial
gh,
the
ive
the

p
a
p
s
h
d
g
m
c
4

i
l
n
w
(
i
w
T
v
t
I
a
t
a
4

c
c
b
w
Q
d
IV
F
Typical
physically sec
and employin
password prot
security meas
hackers can
decode propr
gateways. Enc
methods to re
communicatio
4.1 Securing I
It is very
SCADA is no
in internet br
launch in TCP
need to secure
we propose
(CGA). CGA
identifier is g
way hash func
The binding b
verified by re
the hash with
IPv6 address
and auxiliary
the correspond
a certification
4.2 Key Mana
Authentic
SCADA syste
SCADA devi
command to
challenge the
be required to
who have eno
Quality system
difference in
V. SECURITY
Figure 2. Secure S
SCADA se
curing the ha
ng common c
tection and an
sures are hard
easily identif
rietary protoc
cryption and a
educe some o
ons.
IP Signaling
important to
ow run in IP-b
rought securit
P/IP layer. To
e the commun
a Cryptogra
Ais IPv6 add
generated by
ction from pub
between the pu
e-computing th
the interface i
can be protec
parameters a
ding private k
n authority or a
agement
cation is the p
em proves its i
ice receiving
perform cont
sending devic
o access the da
ough permiss
m administrat
security preve
Y METHODO
SCADA Commun
ecurity mea
ardware and tr
cyber security
nti-virus utiliti
der to enforce
fy confidentia
cols, and byp
authentication
f these cyber
secure IP sign
ased network.
ty vulnerabili
o combat these
nication in Lin
aphically Ge
dresses for w
computing a
blic key and a
ublic key and
he hash value
identifier. Me
cted by attach
and by signing
key. The protec
any security in
process by w
identity to ano
a critical m
trols or respo
ces identity.
ata and report
sion can acce
tion technique
ention. SCAD
OLOGY
nication
sures consis
ransmission m
y defenses su
ies. Communic
e since moder
al phone num
pass firewalls
are highly eff
threats to SC
naling since m
. Exposing SC
ities. Some a
e attacks, ther
nk layer. In thi
nerated Add
which the inte
cryptographic
auxiliary param
d the address c
e and by comp
ssages sent fro
hing the publi
g the message
ction works w
nfrastructure.
which one part
other. A
message, such
ond with data
Authenticatio
ts so that only
ss the inform
es can make a
DA web server
st of
media,
uch as
cation
rn day
mbers,
s and
fective
CADA
modern
CADA
attacks
re is a
is case,
dresses
erface
c one-
meters.
can be
paring
om an
ic key
e with
without
t of a
as a
a, can
n will
y users
mation.
all the
r must
alw
serv
from
the
req
sen
info
enc
mas
4.3

phy
Sec
pos
loca
cam
mo
was
US
soft
atta
from
SCA
kin

seri
SCA
Sm
way
SCA
full
thes
har
lead

[1]
[2] R
[3] 6
[4] T
[5]K
[6]C
[7] h
ways be secure
ver security so
m the custome
master server
quest will be
nd back the
ormation wil
cryption that is
ster and the re

Securing Har
It is very imp
ysically. Thes
curing hardwa
ssible inside j
ation security
mera and oth
nitor the area
sthe maliciou
B drive into
ftware became
acker was plan
m all of his c
ADA system
nd of attack.
It is very
iousness, and
ADA systems
mart Grid envir
ys.This paper
ADA system
ly secured SC
se aspect: Sec
rdware securit
d to more vuln
Study of Security
Security
http://www.inl.g
rent_issues.pdf
Ronald L. Krutz,
6DEPLOY IPv6 T
http://www.6de
nd%20Sensor%
Tai-hoon Kim,
Smart Grid
Applications, E
Korea Smart Grid
CNN Interactive,
1988,http://www
juvenile.hacker/in
http://www.tscm.c
e since the da
oftware can al
er or client wi
r. The client w
completed. T
requested inf
ll also be
s proposed to
emote assets.
rdware
portant to secu
se are the ta
are from any
job that may
y measure mu
her sensor de
a. One of the
us software th
o an exposed
e an instrume
nning a "mas
controlled ma
must be secu

V. CONC
important to
importance o
s, the challen
ronment in so
r examines th
as a part of
CADA system
curing IP Sign
ty. For neglec
nerable SCAD
REFERE
y Attributes of S
gov/scada/publica
"Securing SCAD
Training, Athens,
eploy.eu/workshop
%20Networks.pdf
Securing Commu
Environment In
Engineering & Dev
Institute, http://w
Teen Hacker F
w.cnn.com/TECH/
ndex.html [May 2
com/outsideplant
ata in it are ve
lso be added.
ill start with a
will be authenti
The SCADA m
formation to
encrypted u
be used betw
ure SCADA h
angible part
y disasters or
occur. In thi
ust be deploy
evices must
e recent SCAD
hat was deliv
d USB port.
ent for botne
ssive" denial o
achines. This
ured physical
CLUSION
understand
of these newly
nges associate
ome different a
he security r
smart grid. I
we must con
naling, Key M
cting this sec
DA system.
ENCES
Smart Grid System
ations/d/securing_
DA Systems" Wile
22/06/201,
ps/20100621_ath
unication of SCA
nternational Jou
velopment, Issue
www.smartgrid.or
Faces Federal C
/computing/9803/
2011]
t.html
ery critical. W
Communicati
an http request
icated before t
master will th
the client. T
sing the sam
ween the SCAD
hardware devic
of the syste
r calamities a
is case, physi
ed. Surveillan
be installed
DA attacks [1
vered through
This malicio
et attackand t
of service atta
only shows th
lly to avoid t
the complexi
y interconnect
ed with the n
and more urg
requirements
In order to ha
nsider security
Management a
curity areas w
ms Current Cy
Issu
_the_smart_grid_
ey Publishing, Inc
hens/8%20IPv6%2
ADA Components
urnal of System
2, Volume 5, 201
r.kr/10eng6-1.php
Charges, March
/18/
Web
ion
t to
the
hen
The
me
DA
ces
em.
and
ical
nce
to
10]
h a
ous
the
ack
hat
this
ity,
ted
new
ent
for
ave
y in
and
will
yber
ues,
_cur
c.
20a
s in
ms
11.
p
18,
[8] Washington Post. Dissertation could be security threat.
http://www.washingtonpost.com/ac2/wp-dyn/ A23689-2003Jul7
[9] MetinOzturk, Philip Aubin, "SCADA Security: Challenges and
Solutions" June 2011 / White paper
[10] Smart Grid news, Smarter Grid.. Struggling SCADA?
http://www.smartgridnews.com/artman/publish/Technologies_SCAD
A/Smarter-Grid-Struggling-SCADA-1561.html
ISBN: 978-81-920575-7-6 :: doi: 10. 74047/ISBN_0768
Tele-Operative Ophthalmoscope

A.N.Nithyaa

R.Premkumar
Department of Biomedical Engineering Department of Biomedical Engineering
Rajalakshmi Engineering College Rajalakshmi Engineering College
Thandalam,Chennai,India Thandalam, Chennai,India.

A.R.Anuprasan

Bharathkumar.V
Department of Biomedical Engineering Department of Biomedical Engineering
Rajalakshmi Engineering College Rajalakshmi Engineering College
Thandalam,Chennai,India Thandalam,Chennai,India

Abstract - The slit lamp is an instrument consisting of a
high-intensity of light source that can be focused to shine a
thin sheet of light into the eye. In most developing and
many areas of developed
countries, access to quality eye care are inadequate which
leads to blindness. In todays modern heath technology the
ophthalmologist needs digital photography for clinical
examination and for diagnosing tumours in the eye. The
price of a slit lamp increases with its quality for accurate
identification of eye tumours. Tele-Operative
Ophthalmoscope offers great hope to overcome public
health problem and can be used for teaching medical
professionals with low cost. Research activities can be
carried out using this system. This system provides a real
time video of the eye to the students without any
discomfort and inconvenience to the patients. While
teaching the students the concern staff has to save the
images of the eye to provide a clear idea about that image.
In case of direct visualisation the ophthalmologist has to
do the adjustments for each and every student, since the
position of the visualization will change if there is any
disturbance. To reduce the changes and to provide direct
visualization, a coupler with camera is fitted to the viewing
system and is connected to a pc. The paper envisages
designing and testing a new Tele-Operative
Ophthalmoscope, which gives the direct visualization of
the eye on a monitor or on a screen, so that all the students
can view the image at the same time in a class room. For
the purpose of experimentation Webcam is used.

KEY WORDS
Ophthalmoscope, Telemetry

I. INTRODUCTION

Thin slit light beam enables an optical section for in
vivo inspection of different layers of the eye. Since its
invention in 1911, the slit-lamp has become an
indispensable tool for eye examination. [1]
Ophthalmology is ideal for telemedicine. Most eye
pathology is visible to inspection. While blindness is
devastating, many eye diseases develop slowly. Thus, if
the disease is detected early in its course and is
appropriately treated, we can save many people from
needless suffering from visual disability. In most
developing and many areas of developed countries,
access to quality eye care is inadequate leading to the
preventable blindness. Passive tele-ophthalmology
where patients clinical data (such as fundus
photographs) are collected, transmitted, and later
interpreted by ophthalmologists are well developed.

[2-3] However, due to its dynamic nature and
requirement of special skills mentioned above, slit-lamp
examination requires active tele-ophthalmology
solutions. To address this pressing public health
problem, we have developed a Tele-Operative
Ophthalmoscope, which gives the direct visualization of
the eye on a monitor or on a screen, so that skilled
ophthalmologist can teach the students without
disturbing the patient and can provide lively the eye
images. Producing lively in the classroom is a
wonderful way to spark the students imaginations and
draw out their creativity.

II. TELEOPERATIVE SLITLAMP
MICROSCOPE
For ophthalmologic examination, a slit lamp
microscope is generally used as a fundamental
diagnostic device, as shown in Fig. 1. With the slit lamp
microscope, the eye specialist's basic tasks include
adjustment of the microscope position using the joystick
to find a focal position and to switch slit types using the
knobs on the slit lamp unit to set an appropriate
diagnostic condition. The eye specialist might use a
fronting lens and a blue filter together with the
microscope to conduct additional
examinations. The organization of these special
devices requires great skill and experience for
microscope control in addition to knowledge of clinical
medicine.

Fig 1. Examination of a Slit Lamp

In previous works, a specialist - an expert remote
presenter-controls a slit lamp microscope at the patient
site. The captured high-quality still or moving images of
eyes are sent to a classroom.
Ophthalmologists play a relatively peripheral role in
teaching the medical student. In contrast, our approach
introduces a tele-operative slit lamp microscope can
provide a real time technology to capture high-quality
still or moving images through the network, and
conduct a classes and diagnosis in a timely manner.
This setting might require an assistant at the patient site
with limited knowledge related to the ophthalmic
peripherals, hardware and software. The assistant might
be a general doctor who is not an ophthalmologist, or
perhaps a nurse who would support the
ophthalmologist's slit lamp operation.

Fig 2. Focused and slightly blurred area
Human eye has a spherical surface. Only features
that are located in the plane defined by the perfect
object distance appear perfectly focused. All other
features are blurred. In addition, attaching the camera at
one side of the slit lamp microscopy can cause the
narrow vision image, see Fig.2.

III. GENERAL BLOCK DIAGRAM

Fig 3.1 General Block Diagram
3.1 Description of Block Diagram
3.1.1.1 Slit Lamp
Today the slit lamp is the ophthalmologists most
frequently used and most universally applicable
instrument. Biomicroscopy of the living eye is a routine
ophthalmologic examination. The slit lamp enables the
user to inspect individual eye segments in quick
succession to obtain a general impression of t h e
eye and make a diagnosis.
I n a slit lamp, the mo s t important type of
illumination is the optical section. Al l other
techniques are variations. For survey examination of
the anterior segment the slit is adjusted to full aperture.
This results in a circular, very bright and evenly
illuminated field t h a t is slightly smaller than the
microscope's field of view. By placing a ground glass
into the optical path the entire field of view is
illuminated. I t is well known t h a t the structure of
transparent objects such as the cornea, anterior
chamber, eye lens, and body can only be seen poorly
in transmitted or reflected light, as the relative
amplitude modulation o f light is too weak a n d
the phase modulation is not perceived by the eye.
However, such objects can generally be observed well
in scattered or fluorescent light.

Fig 3.2 The overall system
A slit lamp exam may be done:

As part of a routine eye exam along
with other procedures to evaluate the eye, such
as ophthalmoscopy, vision testing, or tonometry
(to measure pressure in the eye).
To look at structures in the back of the
eye, such as the optic nerve or retina.
To help detect disorders in the
structures in the front of the eye, such as
infection or injury to the cornea, cataracts,
conjunctivitis, or iritis.
To help detect and monitor glaucoma
or macular degeneration.
To check for a foreign body, such as a
metal fragment, on or in the eye.
To detect eye problems that may be
caused by other diseases, such as diabetes or
rheumatoid arthritis. Routine slit lamp exams are
important to detect eye problems at an early
stage and to guide treatment if eye problems
develop.
To monitor complications such as
bleeding after an eye injury.
To monitor complications such as
cataract formation that occurs because of
chemotherapy, radiation treatment, or after a
bone marrow transplant.

3.1.1.2 Coupler
A coupler is a device that connects two ends with
out any loss of energy. The coupler is two piece Teflon
coupling. The camera is fitted inside the coupler at the
correct length and the two pieces are tightened. The
modified slit lamp supports video camera installation
for viewing images of the patients eye on screen. The
image coming from one end of the slit lamp is focused
on to the webcam for high fidelity. The webcam used
picks up the image directly and delivers to the PC. The
coupler is made up of PVC, in order to have good
focusing. This duplication of the natures high fidelity
system is achieved after many trial and errors.

3.1.1.3 Coupler Design

Fig 3.3Top View of First Part

Fig 3.4 Top View of Second Part

Fig 3.5Top View of Third Part

Fig 3.6Side View of First Part

Fig 3.7 Side View Of Second Part

Fig 3.8 Side View Of Third Part

Fig 3.9Total Coupler

Fig 3.10 Standard Photograph from the eye

IV. CONCLUSIONS

For the purpose of experimentation webcam is
used, which gives the direct visualization of the eye on a
monitor or on a screen, so that all the students can view
the eye at the same time in a class room. Since the eye is
very sensitive and importance to human life, the quality of
the image that used for doctors' diagnose plays an
important role in examining the patient's illness.
Furthermore, slit lamp examination can suffer the patient
because of high luminance, so less used time would be
better. Tele-Operative Ophthalmoscope system has
specificity approaching those of the conventional slit-
lamp. The remaining challenges are to miniaturize the
prototype, to develop a low cost mass production of
the
units, and to optimize the utilization of internet
bandwidth.

V. ACKNOWLEDGEMENTS
The authors wish to thank the Chairman and
Chairperson of Rajalakshmi Engineering College,
Chennai for providing all the facilities required for
carrying out this work. They also thank the Principal,
Rajalakshmi Engineering College, Chennai for his
constant encouragement and guidance.

REFERENCES

[1]. J.K. Ledford, The Slit Lamp Primer, Slack Incorporated,
Thorofare, N.J., 2006.
[2]. S. Kumar, M.L. Tay-Kearney, I.J. Constable, K.Yogesan., Internet
based ophthalmology service:impact assessment Br J Ophthalmol
89: 1382-83, 2005.
[3]. J.G. Camara, RE Rodriguez., Real-time
telementoring in Ophthalmology Telemed J 4: 375-377, 1998
[4]. Richard j. Jamara, Od, faao, frans van de velde, md,
and eli peli, msc, od, faao Scanning Eye
Movements in Homonymous Hemianopia
Documented by Scanning Laser Ophthalmoscope
Retinal Perimetry 1040-5488/03/8007-0495/0 VOL.
80, NO. 7, PP. 495504 Optometry Vision Science
Copyright 2003 AmericanAcademy of Optometry
[5]. J.G. Camara, RE Rodriguez., Real-time telementoring in
Ophthalmology Telemed J 4: 375-377,1998.
[6]. Austin Roorda, David Merino, Kaccie Y. Li, and Yuhua Zhang
Miniaturization of Adaptive Optics Scanning Laser
Ophthalmoscope in Frontiers in Optics, OSA Technical Digest
(CD) (Optical Society of America, 2009), paper FWX1.
[7]. Perednia D. A., Allen, A., "Telemedicine technology and
clinicaapplications" JAMA, 53, 273, 1995.
[8]. H. Li, "Telemedicine and ophthalmology," Survey of
Ophthalmology, vol. 44, 1999, p. 61-72.
[9]. M. Moore, "The evolution of telemedicine," Future Generation
Computer Systems, vol. 15, 1999, pp. 245-254.


ISBN: 978-81-920575-7-6 :: doi: 10. 73942/ISBN_0768

Abstract- Routing protocols for mobile ad hoc networks
(MANETs) have been explored extensively in last few
years. Much of this work is targeted at finding a feasible
route from a source to a destination without considering
current network traffic or application requirements.
Routing may let a congestion happen which is detected by
congestion control, but dealing with congestion in reactive
manner results in longer delay, and unnecessary packet loss
and requires significant overhead if a new route is needed.
Routing should not be aware of, but also be adaptive to,
network congestion. Adaptation to the congestion helps to
increase both the effectiveness and efficiency of routing.
These problems are solved by the congestion-aware routing
protocols in certain degree. These protocols which are
adaptive to congestion status of mobile ad-hoc network can
greatly improve the network performance. In this paper,
we present the survey of congestion adaptive routing
protocols for mobile ad-hoc network. Finally, the future
direction of congestion-aware routing protocols is
described.

Keywords: Ad hoc networks, congestion aware routing,
Congestion metric, congestion adaptability

I. INTRODUCTION
Wireless ad-hoc network is usually defined as a set
of wireless mobile nodes dynamically self-organizing a
temporary network without any central administration or
existing network infrastructure. The node in the wireless
ad-hoc network can serve as routers and hosts. So, they
can forward packets for other nodes if they are on route
from source to destination. Routing is important problem
in wireless ad-hoc network. Traditional working
protocols cannot work well in wireless ad-hoc network
because of the characteristics of the wireless ad-hoc
networks. Since, mobile nodes have limited transmission
capacity they mostly intercommunicate by multihop
relay. Multihop routing is challenged by limited wireless
bandwidth, low device power, dynamically changing
network topology, high vulnerability to failure. To
answer these challenges, many routing algorithms in
MANETs were proposed. There are different dimensions
to categorize them: proactive routing Vs reactive routing
or single path routing Vs multipath routing. In proactive
protocols, route between every two nodes are established
in advance even though no transmission is in demand. In
reactive protocols, route is discovered when needed
transmission and released when transmission no longer
takes place. Congestion is one of the most important
restrictions of wireless ad-hoc network. It may
deteriorate the performance of whole network. In the
current design routing is not congestion-adaptive.
Routing may let the

congestion happen which is detected by congestion
control. But dealing with congestion in reactive manner
results in longer delay and an unnecessary packet loss
and requires significant overhead if the new route is
needed. But, now there is another dimension for
categorizing for routing protocols: congestion adaptive
Vs. congestion un-adaptive routing. Our motivation is
that congestion is dominant cause for packet loss, long
delay, and high overhead in MANETs.
These problems become visible in large scale
transmission of traffic intensive data such as multimedia
data where congestion is more probable and negative
impact of packet loss on the service quality is of more
significance. In this paper we studied congestion routing
protocols like CRP(Congestion Adaptive Routing
Protocol)[7],ECARP (Efficient Congestion Adaptive
Routing Protocol )[11],CARP(Congestion Aware
Routing Protocol),CADV(Congestion Aware Distance
Vector)[12],CARA(Congestion Aware Routing plus rate
Adaptation)[12],CARM(Congestion Aware Routing
Protocol for Mobile Ad-hoc Network)[12].
The remaining part of the paper is organized as
follows: In section II we provide the studied congestion
aware routing protocols. In section III comparison
between these algorithms is presented. In section IV we
concluded the paper.
ALGORITHMS
There are many routing algorithms in mobile ad-hoc
networks for routing and congestion free networks.
Some of them are explained below:
A. Congestion Adaptive Routing Protocol (CRP):
Congestion Adaptive Routing is a congestion
adaptive unicast routing protocol for mobile ad-hoc
network. CRP protocol tries to prevent congestion from
occurring in the first place. In CRP, every node
appearing on a route warns its previous node when prone
to be congested. So, CRP uses the additional paths called
as bypass for bypassing the potential congestion area
to the first non-congested node on the primary route. It
reduces packet delay. But, at the same time CRP tries to
minimize bypass to reduce protocol overhead. Hence,
the traffic is split over bypass and primary and
adaptively to network congestion. Hence, 1) power
consumption is efficient.2) Congestion is resolved
Protocols A Study of Congestion Aware
Adaptive Routing in MANET

Mr.S.A.Jain Mr.S.R.Kokate
Research Scholar Research Associates
MAE, Alandi (D) MAE, Alandi (D)
Pune University, India Pune University, India


beforehand and at the same time there is small packet
loss rate.

RP is on-demand and consists of the following
components.
a. Congestion Monitoring: When no. of packets
coming to the node exceeds its carrying capacity, node
becomes congested and its starts losing packets. Various
metrics are used for node to monitor congestion status.
Main parameters are percentage of all packets discarded
for lack of buffer space, the average queue length, the
no. of the packets timed out and retransmitted, average
packet delay. In all these parameters, rising number
indicates growing congestion.
2) Primary Route Discovery: Sender discovers the
route to the receiver by broadcasting the REQ packet
toward receiver. The receiver responds REQ by sending
the REP packet on same path that the REQ previously
followed. This is called primary route and nodes on this
are called primary nodes. To reduce traffic due to the
primary route discovery and better deal with
Congestion in the network, 2 strategies are adopted
1) REQ is dropped if arriving at a node which is having
congestion status as red 2) REQ is dropped I arriving
at node already having a route to destination .
b. Bypass Discovery: A primary node periodically
broadcasts a UDT i.e. update packet. This packet
contains the nodes congestion status and set of tuples
[destination D, next green node G, distance to green
node, n] for each node appearing as a destination in
primary table. For this reason is when node P receives an
update packet from next primary node P
next,
about the
destination D, P will be aware of congestion status of
next. This causes the congestion to know about the next
green node of P which is n hops away from primary
route. But if the next hop is yellow or red, congestion
will be there if data packets continue to be forwarded on
PP
next
. But, CRP tries to keep congestion from
occurring in the first place, P node starts to select bypass
route toward G-the next green node of P known from the
UDT packet. This bypass search is similar to primary
route search, except that 1)the bypass request packets
TTL is set to 2*m and 2)bypass request is dropped if

arriving at node already present on primary route. It
can be also possible that no bypass is found. So, in such
situation packets are delivered to destination by
following primary route.
C. Traffic Splitting and Congestion Adaptability:
When the bypass at a node is found, data packets coming
to this node are not necessarily spread over bypass and

primary route. To avoid the bypass from being
congested no packet is forwarded on bypass unless any
primary node is red i.e. congested. The basic idea behind
traffic splitting is that when primary link consists of less
congested node, traffic on primary link should be
increased, otherwise it should be reduced. Bypass and
primary routes cannot include more than 2 common
nodes, but different bypass paths can share common
node. This increases chance to discover a bypass. But,
because of this bypass node may become congested if it
has to carry large loads of bypass traffic. But, this can be
solved, by splitting probability adjustment for congestion
adaptation. The probability adjustment is as shown in
TABLE I.
b. Multipath Minimization: To reduce the protocol
overhead, CRP tries to minimize using multiple paths. If
the probability p to forward data on a primary link
approaches 1.0, this means the next primary node is far
from congested or the bypass route is highly congested.
In this case, the bypass at the current node is removed.
Similarly, if the next primary node is very congested (p
approaches 0), the primary link is disconnected and the
bypass route becomes primary. To make
the protocol more lightweight, CRP does not allow a
node to have more than one bypass. The protocol
overhead due to using bypass is also reduced partly
because of short bypass lengths. Each bypass connects to
the first non-congested node after the congestion spot,
which should be just a few hops downstream.

c.Failure Recovery: CRP is able to quickly resume
connectivity after a link breakage by using bypass routes
currently available. There are 3 min cases of failure
Primary link failure: When one of link on primary
route fails, the initial node sends a DISC packet towards
sender along route. This DISC goes on recording nodes
and it stops at node having bypass. This node if finds
that its bypass destination is there in DISC, that bypass is
not used and DISC is forwarded upstream towards
sender till it finds a node with bypass and not having
failed node as its destination. If both these cases are not
Congestion Bypass
status=green
Bypass status=yellow Bypass status=red
Next primary node is green P:=p+(1-p)/4 P:=p+(1-p)/3 P:=p+(1-p)/2
Next primary node is yellow P unchanged P unchanged P:=p+(1-p)/4
Next primary node is red P:=p-(1-p)/2 P:=p-(1-p)/4 Find another bypass

TABLE I
SPLITTING PROBABILITY ADJUSTMENT


there DISC is sent to the sender and it will find new
primary route.
d.Bypass link or node fails: In this case bypass node
which finds this failure sends a BPS_DISC packet
through bypass route to primary node and that bypass is
removed.
e.Primary node fails: If node on the primary route
fails, its previous node sends DISC packet along primary
route. If the bypass node detects some failure, it will also
send BPS_DISC packet along bypass until reaching a
primary node. When primary node received both these
packet, it removes bypass and DISC packet is forwarded
along primary route. Then this is handled same as first
case. If BPS_DISC packet doesnt arrive at the primary
node on time that bypass is used as primary route. But, if
it comes late, it is ignored. But, route remains broken but
it will recover soon because another DISC packet will be
sent back.
To evaluate the performance of the CRP following
parameters are used:
a.Packet delivery ratio: Percentage of data packets
received at the destination out of the number of data
packets generated by the CBR traffic sources.
b.End-to-End delay: It is the accumulative delay in
data packet due to buffering of packets, new route
discoveries, queuing delay, MAC-layer re-transmission,
and transmission and propagation delays.
c.Routing Overhead: The ration of the amount in
bytes of control packets transmitted to the amount in
bytes of data received.
d.Normalized power consumption: The ratio of the
amount in bytes of both control and data packets
transmitted to the amount in bytes of data received.

B. An Efficient Congestion Adaptive Routing
Protocol for Mobile Ad Hoc Networks (ECARP):

An efficient congestion adaptive routing protocol is
better than every other routing protocol during heavy
traffic loads. The ECARP, routing protocol ensures high
availability of alternative routes and reduce the rate of
stale routes. ECARP is having mainly AODV as its base.
This can be achieved by increasing the parameters of
routing protocols (especially in AODV) that normally
take more time for link recovery. These parameters are
active_route_time-out, route_reply_wait_time,
reverse_route_life, TTL_start, TTL_increment,
TTL_threshold and delete_period.
ECARP Congestion Control Algorithm
This algorithm provides solution to improve routing
protocols due to constrained environment.
Step 1: Check the occupancy of link layer buffer of
node periodically. Let N
c
be the congestion status
estimated.
Step 2: Compute N
c
=Number of packet buffered in
buffer/Buffer Size
Step 3: Set the status for congestion. It can be
indicated by three statuses Go, careful, and Stop.
[ Go indicates there is no congestion with N
c
],
carefull indicates the status likely to be congested with
N
c
and Stop indicates the status already
congested, N
c
1.]
Step 4: Invoke congestion control routine when link
failed event has occurred in data transfer with using
active route or N
c
1.
Step 5: Assume that neighboring will have alternate
route or non-congested route to the destination.
Step 6: Make Query to non-congested neighbors for
route to destination
Step 7: after obtaining the routes from the neighbors,
select route with minimum hops.
Step 8: Once route is finalized start sending the data
packets through non-congested route.
Step 9: If there is no alternative route to destination
then start splitting the traffic to the less congested route.
Step 10: Traffic splitting effectively reduces the
congestion status at the next main node.

d. Congestion aware routing plus Rate Adaptation
(CARA):

The base use of CARA protocol is DSR. The route
discovery mechanism of DSR is modified. This protocol
mainly aims to find the bypass route for congested zones
or nodes. This can be achieved by combining the average
MAC utilization and the instantaneous transmission
queue length to indicate the congestion level of nodes in
the network. When source wants to transmit data to the
destination node, it broadcasts RREQ packets. When
intermediate code receives RREQ, it checks its
congestion level. If the congestion level is higher then it
discards the RREQ. When RREQ arrives at the
destination node, though destination node is congested or
not it handles the RREQ and replies RREP. So, route
without congested node is established.
CARA uses two metrics to measure congestion
information first is average MAC layer utilization. The
instantaneous MAC layer utilization is considered as 0
only when the medium around the node is available at
the beginning of a transmission and as 1 when the node
is not idle. (e.g. detecting physical carrier or detecting or
back off due to virtual carrier sensing.) As, the
instantaneous MAC layer utilization is either 1 or 0 the
average value with in the period indicates the use of
wireless medium around the node.
Second metric used is instantaneous transmission
queue length. If the node has many packets waiting in
the queue, it causes long packet latency or even dropping
of packets. So we can say that node is congested now.
The above mentioned metric can veraciously reflect
the congestion conditions around the node. This protocol
tries to minimize the congestion in two ways: 1) It
forbids the RREQ packets to propagate in the congested
area. 2) It guides the route around the congested area or
nodes instead of across them. As a result of this no
conditional transmission burden generate in these areas.
e.Congestion Aware Routing protocol for Mobile ad
hoc networks (CARM):

A congestion aware routing protocol for mobile ad
hoc networks uses a metric incorporating data rates,
MAC overhead and buffer delay to control the
congestion. The CARM protocol introduces a new
parameter called WCD (Weighted Channel Delay) to
measure congestion level and adopts a route
ELDC(Effective link Data-rate Category) to avoid the
MDRR(Mismatched data-rate route) problem. The
MDRR problem is shown in following fig.3

Fig. 3 An example of MDDR problem

The data rate of route shown by dashed (A-B-D-G) is
limited by teaming fast link (B-D) with slow link (A-B
and D-G).
As mentioned earlier, the CARM protocol introduces
a new parameter called WCD (weighted channel delay)
to measure congestion and it is given as :

WCD=a Q +
(1+b)T
MACALL
+T
data

where Q is the number of buffered packets for this
link. T
data
=L
data
/ R is the data transmission time, L
data
is
the length of data in bytes or bits and R is the data rate of
the link. T
MACALL
is total time spent at the MAC layer.
The constants a and b are parameters with values
between 0 and 1 which are used to weight T
MACALL
. By
weighting T
MACALL
can avoid misjudgment of congestion
as shown in fig.2

Fig. 2 Two scenarios with the same overall delay but different
MAC and transmission delay due to different data-rates and congestion
levels.
In CARM, source node broadcasts RREQ packets
with ELDC and WCD information when it attempts to
transmit data to the destination. Intermediate nodes
compare source ID, source sequence, and ELDC of the
RREQ packets they receive from neighbors, and drop the
RREQ packets whose source ID and source sequence
number are the same with that of other RREQ packets
received earlier and ELDC is lower than the earlier
RREQ packets. Only the destination node can responds
to the RREQ packets by sending RREP packets back to
the source along the route from which they came. The
route is established when the first RREP arrives at the
source. The subsequent RREP packets are cached for the
spare routes. The utilization of the congestion metric,
WCD, is very special in CARM protocol. Because the
priority of route packets is higher than data packets, the
route packets can be forwarded without queuing. That is,
the congestion level information inherent in queuing
delays is lost. The author proposed a RREQ-delay
scheme. An RREQ is forwarded with a delay of the
WCD that is calculated according to the WCD
information in the RREQ at the intermediate nodes. The
lower the congestion level of link is, the smaller the
delay of RREQ packets are, the earlier the RREQ
packets arrive at the destinations. This scheme ensures
that the RREQ packets of routes with lower congestion
level arrive at the destination first and congested links
are eliminated in the routes. This all causes high
overhead. So, overhead in case of CARM is very high.

E. Congestion-Aware Distance Vector (CADV):

The CADV protocol is based on proactive protocol,
DSDV. In a distance vector routing protocol, every host
maintains a routing table contains a distances from itself
to possible destinations. A mobile host in ad -hoc
network acts like a single server queuing system. Delay
in sending packet is related with congestion. In CADV,
each entry is related with delay expected. This helps to
measure congestion at the next hop. The expected delay
is computed follows:

(1)
Where n is the number of sent packets & L is the
length of MAC layer packet queue. E [D] estimates the
time. A newly arrived packet has to wait before it is send
out.In CADV, routing decision is made based on
distance to the destination as well as the expected delay
at the next hop showed in (1) CADV gives the routes
with low expected delay, higher priority. CADV tries to
avoid congestion and tries to balance traffic by giving
priority to a route having low expected delay.
CADV routing protocol consist of three components:
1) Traffic Monitor: It monitors traffic going out
through the link layer. Currently it keeps track of
average delay for sending one data packet in receipt

period of time. Time period is specified by route
maintainance component.
2) Traffic Control: It determines which packet is the
next to send or drop. It reschedules packets if needed. It
supports a drop tail FIFO queue and provides
functionality to queue packets.
3) Route maintainance: It is the main component. It
performs the work of exchanging information with
neighbors, evaluation and maintaining routes. It manages
the traffic monitor and traffic control component. CADV
better support for QoS. The real time performance of
CADV is good, and end to end delay was short. The
overhead of CADV is unacceptable when the network is
large. Through put also decreases the performance of
CADV is may be well in the small & steady wireless ad-
hoc network.

F. Congestion Aware routing Protocol (CARP)

CARP is an on-demand routing protocol. It uses
information gathered from MAC layer to discover
congestion free routes. CARP uses combined weight
matrix in its standard cost function to check for the
congestion level. The multiple paths are computed
during the route discovery. Calculate node weight matrix
NM which assign a cost to each link in the network and
select maximum throughput paths.
NM = (L
q
* D
rate
)/(OH
mac
* D
avg
)
1) Route request: Consider the route
S-P1-P2-P3-D
To initiate congestion-aware routing discovery, the
source node S sends a RREQ. When the intermediate
node P1 receives the RREQ packet, it first estimates all
the node weight metrics.
The node P1 then calculates its node weight NMP1
RREQP1P2
P2 calculates NMP2 and forward the RREQ packet
RREQP2P3
Finally the RREQ reaches the destination node D
with the sum of node weights
RREQP3D
2) Route Reply: The destination node D
sends the route reply packet RREP along with total
node weight to the immediate upstream node P3
RREQDP3
Now P3 calculates its cost C based on the
information from RREP as
CP3= (NM
p1
+NM
p2
+NM
p3
)-(NM
p1
+NM
p2
)
By proceeding in the same way, all the intermediate
hosts calculate its cost .On receiving the RREP from all
the routes, the source selects the route with minimum
cost value.

II. COMPARISONS

Congestion is a dominant reason for packet drops in
ad hoc networks.CRP sends packets on both bypass
paths and primary routes simultaneously. So, incoming
traffic is distributed on primary and bypass route
depending on current congestion status of network.
Congestion is subsequently better resolved .In ECARP
some parameters of AODV such as TTL_start,
TTL_increment are increased. So, it ensures the high
availability of alternative routes and reduces the rate of
broken rut removal process. CADV is not congestion
adaptive. It offers no remedy when the existing route
becomes heavily congested. So, CADV improves AODV
in delivery ratio only. The real time performance of the
CADV is good and the End-to-End delay is short. The
disadvantage of the CADV is that since, each node
maintains all the routes to the nodes in the network and
changes the route information periodically, the overhead
for maintaining the routing tables is huge. The overhead
of the CADV is unacceptable when the network is large
or the topology changes frequently. The throughput
decreases sharply at the same time. So, CADV may
perform well in the small, steady wireless ad -hoc
network. By studying the algorithms of CARM, CARA
and CADV it is conclude that overhead of the CARM
and CADV are higher than CARA, the delay of CADV
is shorter than the other two.
III. CONCLUSION
It is clear from algorithms available for having
adaptive solution for congestion in the network as due to
vast pay load on networks, which may be due to flooding
of packets or may be due to repeat requests on the basis
of error correction techniques. Congestion metrics still
remains a great challenge for the future work. It is quite
important to obtain an optimal approach that combines
related parameters collected from physical layer, MAC
layer to measure congestion. Finally we can conclude
that congestion is the problem associated with the
network and has to be countered by having compromised
solution rather than elimination.

REFERENCES

[1] D. B. Johnson, D. A. Maltz, Dynamic source routing in ad hoc
wireless networks, Mobile Computing, pp. 153181, 1996.
[2] Josh Broch, David A. Maltz, David B. Johnson, Yih-Chun Hu,
Jorjeta Jetcheva. A Performance Comparison of Multi-Hop Wireless
Ad Hoc Network Routing Protocols, The Fourth Annual ACM/IEEE
International Conference on Mobile Computing and Networking, pp. 1-
13, 1998
[3]Chen, X., H. M. Jones, A .D .S. Jayalath, Congestion-Aware
Routing Protocol for Mobile Ad Hoc Networks, Vehicular
Technology Conference 2007, pp. 21-25, 2007
[4] S. Ramanathan, M. E. Steenstrup, A survey of routing techniques
for mobile communications networks, mobile networks and
applications, Vol. 1, pp. 98104, 1996.
[5] C. E. Perkins, Ad hoc on demand distance vector (AODV)
routing, IEFT Internet Draft, November 1998,
[6] H. Raghavendra and D.A. Tran, Congestion Adaptive Routing in
Ad Hoc Networks (Short Version), Proc. ACM Intl Conf. Mobile
Computing and Networking (MOBICOM), Oct. 2004.
[7] H. Raghavendra and D.A. Tran, Congestion Adaptive Routing in
Ad Hoc Networks IEEE Transactions on Parallel and Distributed
Systems, Vol. 17, No. 11, November 2006.
[8] D.A. Tran and H. Raghavendra, Routing with Congestion
Awareness and Adaptivity in Mobile Ad Hoc Networks, Proc. IEEE
Wireless Comm. and Networking Conf. (WCNC), Mar. 2005.

[9] Xiaoqin Chen, Haley M. Jones and Jayalath, "Congestion-Aware
Routing Protocol for Mobile Ad Hoc Networks" IEEE 66
th
conference
on Vehicle Technology, pp.21-25, October 2005.
[10] S. J. Lee and M. Gerla, Dynamic load-aware routing in ad hoc
networks, Proceedings IEEE International Conference on
Communications, pp. 32063210, June 2001.
[11] T.G.Basavraju, S.K.Sarkar, C.Puttamadappa, M.A.Gautham,
ECARP: An Efficient Congestion Adaptive Routing Protocol for
Mobile Ad-hoc Networks, ITS Telecommunication Proccedings,
pp.715-718, 2006.
[12] Wu Wei, Zhang Zhongzhao, Sha Xuejun, Qin Danyang A Study
of Congestion-Aware Routing Protocols For Wireless Ad-Hoc
Network.

ISBN: 978-81-920575-7-6 :: doi: 10. 73935/ISBN_0768
Exploiting the Bit Error Diversity for Efficient Error Detection in Multihop
Wireless Network
Yanthungbeni Humtsoe
PG Student ME Computer Department
Pune Institute of Computer Technology
Pune, India.

Abstract Errors are inherently present in wireless network,
exploiting the bit error diversity and correcting the error for
packet transmission is necessary for improving the
throughput. In multihop wireless network when transmission
goes over multiple hops, error increases which degrades the
performance of the network. Various methods already exist
for controlling the error to improve the throughput. Scattered
Random network coding exploit the usefulness of random
network coding and classify the block into good (low error
probability) and bad block (high error probability). In this
paper we proposed a new approach for controlling the error.
Proposed work uses the concept of forward error correction
for correcting and controlling the error on the bad bits,
16QAM is used for modulation symbol and shortest
remaining algorithm is used for giving priority to messages.

Keywords- Multihop, Throughput, Error control, bit error,
Error detection.

I. INTRODUCTION
In multi-hop wireless networks, communication
between sender and receiver is carried out through a series
of intermediate nodes whose function is to relay
information from one node to another. In unreliable
wireless channel, errors are naturally present. Multihop
wireless network suffer from low throughput as error
increases over multiple hops. It is important to design an
efficient transmission and error control schemes to
improve the throughput in multihop wireless network.
Network coding is introduced in [1] to improve the
throughput performance. In network system, the
intermediate node simply replicate and forward the
received information to the destined node ,while in
network coding ,the intermediate node combine the
information received from different sources and forward to
the output node. In [2] a network coding approach to
cooperative diversity is proposed, a scenario is considered
in which two partners cooperate to transmit information to
a single destination. Both the partner transmit their locally
generated information and the partner relay information
which lead to an encoding scheme in which the algebraic
superimposition of its local and relay information is
transmitted by each partner. At the destination decoding is
done by repeating the process between the codeword from
the two partners. In a communication network, various
kind of error exists during the packet transmission. Errors
include packet error, random error and so on. Network
error correction coding (NEC) is introduced in [3] to deal
with these errors. Earlier work on network coding mainly
focus on multicast, in [4] Network coding has been applied
in multihop wireless network to take advantage of the
multiple routes from sender to receiver in unicast flows.
The 2N-QAM (Quadrature Amplitude Modulation)
modulated symbol consists of N-bits. For the 2N-QAM
with N larger than 2(eg.8QAM (PSK), 16QAM, 64QAM,
256QAM) the reliabilities of the bits Gray-mapped onto
the modulated symbol is different.16QAM contains 4bits
and 64QAM contains 6bits.

II. RELATED WORK
Random network coding [5] is a technique for
scattering information in a communication network, it is
prone to errors. The attachment of even a single corrupted
packet will affect all the information receives; therefore
error control is of great interest in random network coding.
As many problem occur for error control in random
network coding, in [6] the problem of error control is
considered and the expression of the error control problem
is given in terms of rank metric codes. Efficient Network
coding based reliable multicast is introduced in [7]. In [8]
the author examines the error control problem in
noncoherent random network coding. In network coding,
error control is an important problem since they are highly
impressionable to packet error. In [9] an error control code
in Random Network coding has been introduced. The
Random network coding has a property called rateless in
which the packet are encoded as a linear combination of
the original packet and all the independent packet are
useful. A receiver collect every data (correct block) that it
receive without describing which block is from which
source, once it receive all the blocks, the original data
blocks can be correctly reconstructed. In this sense,
random network coding can be considered as a rateless
code. When random network coding
is applied to multihop wireless networks, the sender
encodes the data and transmits the coded blocks to the
relay, which will forward the data to the end receiver.
In[10]Drizzle is proposed to see whether network coding
will be still used in multichannel wireless networks and it
is a new solution to maximize the throughput in the
presence of errors. In [11] a signal constellation
rearrangement for HARQ is presented where for
retransmission symbol mapping is used and at the receiver
before demodulation signal combination is implemented
and then the combined signal is demodulated. In [12] a
novel algorithm for searching HARQ with constellation
rearrangement is proposed. In wireless communication due
to unreliable wireless channels retransmission is done to
remove the errors. Scattered Random Network coding (S-
RNC) is introduced in [13] to improve the throughput, it is
designed to examine the benefit provided by the random
network coding.

III. PROPOSED WORK
[13] Introduces bit error diversity and proposes
Scattered Random Network Coding (S-RNC) to exploit
this diversity to improve the throughput for multihop
wireless communication. However, S-RNC relies on
random network coding and block differentiation (each
block must have a priority). S-RNC can be extended and
generalized to remove these two restrictions. We propose
to do so under two different heads-
1. Stronger Error Correction for `bad' bits.
2. Deciding which bits go to the `good' bits.
To improve the throughput we consider a new approach
for controlling the errors. In this approach, Strong error
correction will be provided to the bad bits as they have a
high probability of error. This will ensure not only error
resilience, but also reduced coding overhead. FEC
(forward error correction) is applied on the bad bit for
controlling the errors. A scheduling algorithm is proposed
that schedules the block so as to maximize the throughput
of the system. We choose the shortest remaining message
(SRM) algorithm so the message with the shortest length
will have the highest priority.The message with shortest
length will be put on the good block and longest length are
put on bad block .In order to avoid indefinite waiting for
long messages SRM with ageing is adopted as the packet
grows older their priority increases and for this an age a
is appended to all the messages (age is used only by
scheduler and is never transmitted).Initially age are set to 1,
so (M = {(m
1
, a
1
= 1), (m
2
, a
2
= 1),(m
3
, a
3
= 1), .}) and it
is incremented every time when the priority function is
called. Also, Shortest Remaining message algorithm is
used which gives priority to the message with the shortest
length with the hope that message can be sent to the
receiver quickly with less number of errors.
M is a set of messages that is
M = {m
i
| m
i
M, i}.
Each message (m
i
) consists of one or more bits. We can
represent m
i
as a regular expression as,
m
i
= (0|1)for all i.
Each message is composed of several blocks. If there
are n
i
blocks in message m
i
then
m
i
= [b
i1
b
i2
bi3 bin].
Place these blocks in block queue (BQ).Thus,
BQ=BQ b
ij
i, 1 j n
ij
Forward error correction (FEC) is used to control error
on bad bits. Since we are using 16QAM, the replication
FEC
for 16QAM is, ([b
i
.b
j
] converted to [b
i
..b
2j
]).
In 16QAM each block is composed of 4bits,
b
k
=[ b
k1
b
k2
b
k3
b
k4
],
the code word will be-

C=
11 12 13 14
21 22 23 24
c c c c
c c c c
(
(

=
1 2 1 2
3 4 3 4
k k k k
k k k k
b b b b
b b b b
(
(

To place the good or bad bits in their appropriate
position
Good matrix (GM) and bad matrix (BM)are used. The
good bits and bad bits are mixed together for transmission.
As
!6QAM is used 4bits per symbol can be send at a time
and the good and bad bits are mixed together in these
4bits.The code word will be-
C= (b
g
) GM + ( b
b
) BM
The algorithm begins when it receives a set of
messages M that are to be transmitted from the network
layer. Each message in M is partitioned into 4-bit blocks
since 16QAM is used as modulation symbol. These blocks
are appended to BQ. m
g
(lower value indicate higher
priority) and m
b
(higher value indicate lower priority) are
selected from among the messages in M . It is possible that
there is only one message in M or all messages have the
same priority (length). In order to exploit bit error
diversity to increase throughput, we need at least two
message with different priorities .Hence, if m
g
= m
b
, we
transmit the remaining blocks without any further
processing. In the other case, we move to the next step. In
the next step, we generate the sets GB (good block) and
BB (bad block).FEC is a flag that indicated whether the
Replication FEC is to be applied. We initialize it to `true'.
In preparation for placing the good/bad blocks in the
good/bad position, we select the first block from GB and
denote it by b
g
. The next step is the Replication FEC
(forward error correction). We select the first block from
BB (denoted as b
b
). Finally, we are ready to place the
blocks in the right position. As b
g
has been transmitted, we
can now remove it from BQ. After removal of any block
from BQ, we must check whether BQ is empty (stopping
condition) or if there is only one block in BQ (trivial
case).We also remove b
g
from GB. As GB contains all the
blocks of the message m
g
, an empty GB indicates that the
entire message has been transmitted. Hence, we remove m
g

from M. As BQ is not empty, but GB is empty, we need to
fill the next message in m
g
. We do this in the next two sub
steps. As before, we check if there is only one message or
the priority of all the messages are the same. Now we
proceed to b
b.
If FEC is true we should not perform FEC in
the next iteration (instead we should use b
2
). If FEC is false,
we have transmitted both b
1
and b
2
. Hence, we remove b
b

from BQ and BB. The reasoning is same as that used for
removing b
g
from BQ and GB. The process will continue
until the entire messages are being transmitted.

Algorithm: Input: Set of messages (M)
Output: Transmission of message.
1. Partition the message in M into blocks.
2. Append each block to BQ
3.(a)Select m
g
such that priority(m
g
) =
min{priority(m
i
)} i
(b)Select m
b
such that priority (m
b
) = max {priority
(m
i
)}
i
4. if (m
g
== m
b
) {transmit(BQ) and return}
5. (a) GB = {b | (b BQ) and (b m
g
)}
(b) BB = {b | (b BQ) and (b m
b
)}
6. FEC = true
7. Select first block in GB: b
g
(say)
8. if (FEC == true)
(a) Select first block in BB: b
b
(say)
(b) Define b
1
and b2 as follows-

1
2
b
b
| |
|
\
=
1 2 1 2
3 4 3 4
k k k k
k k k k
b b b b
b b b b
(
(

9. C = b
g
GM + b1 BM
10. Transmit C
11. Remove b
g
from BQ
12. .(a)if (length(BQ==0){return}
(b)else if (length(BQ)==1){Transmit(BQ) and
return}
13. Remove b
g
from GB
14. if (length(GB==0)
(a) Remove m
g
from M
(b) Select m
g
g
) =
min{priority(m
i
)}
i
(c) GB={b |(b BQ) and (b m
g
)}
(d) If (m
g
== m
b
) {transmit (BQ) and return}
15. if (FEC == true)
(a) FEC = false
(b) b
1
= b
2

16. else
(a) FEC = true
(b) Remove b
b
from BQ
(c) If (length (BQ) == 0) {return}
(d)Else if(length(BQ)==1){transmit (BQ) and
return}
(e) Remove b
b
from BB
(f) If (length(BB)== 0)
i. remove m
b
from M
ii. Select m
b
b
) = max
{priority
(m
i
)} i
iii. BB = {b | (b BQ) and (b m
b
)}
17. go to 7

IV . CONCLUSION
In this paper, we have provided a review for the reasons
of poor throughput in multihop wireless network. Error
control scheme has been introduced for random network
coding. Scattered random network coding relies on random
network coding and they put the bits of odd number blocks
in good bit position and even number blocks in bad bit
position. When transmission goes over multiple hops the
odd number blocks are protected from getting error while
the even number blocks will get dirtier. We can improve the
performance on the existing system by implementing a
strong error control scheme. To improve the throughput,
We have proposed a new approach by providing strong
error correction on the bad bits as they have a high
probability of error.FEC (forward error correction) is
applied on the bad bits for error correction and shortest
remaining message (SRM) algorithm is also used to give
priority to the message with the shortest length with the
hope that message will be delivered to the receiver quickly
and to avoid indefinite waiting time SRM with ageing is
also adopted.

REFERENCES

[1] R. Ahlswede, N. Cai, S. R. Li, and R. W. Yeung, Network
information flow, IEEE Trans. Inf. Theory, vol. 46, no. 4, pp.
12041216,Jul. 2000.
[2] L. Xiao, T. Fuja, J. Kliewer, and D. Costello, A network coding
approach to cooperative diversity, IEEE Trans. Inf. Theory, vol. 53,
no. 10,pp. 37143722, Oct. 2007.
[3] Z.Zhang,Theory and application of network error correction
coding, proceeding of the IEEE, vol.99, pp.406-420, 2011.
[4] S. Katti, H. Rahul, W. Hu, D. Katabi, M. Medard, and J. Crowcroft,
XORs in the air: Practical wireless network coding, in Proc. ACM
SIGCOMM, 2006.
[5] T. Ho, M. Medard, R. Koetter, D. Karger, M. Effros, J. Shi, and B.
Leong,A random linear network coding approach to multicast,
IEEE Trans. Inf.Theory, vol. 52, no. 10, pp. 44134430, Oct. 2006.
[6] D. Silva, F. Kschischang, and R. Koetter, A rank-metric approach
to error control in random network coding, IEEE Trans. Inf.
Theory, vol. 54,no. 9, pp. 39513967,Sep 2008.
[7] K.Chi, X.Jiang, B.Ye, H.Y.Shwe, and S.Horiguchi, Efficient
Network coding-Based End-to-End Reliable Multicast in Multi-hop
Wireless Network,Proceeding of the 15
th
Asia-Pacific Conference
on communication(APCC2009), 2009
[8] R. Koetter, F. R. Kschischang,Coding for errors and erasures in
Random Network Coding, IEEE Trans. Inf. Theory, vol. 54,
no.8,pp. 35793591, Aug. 2008.
[9] R.Ahlswede and H.Aydinian,On error control codes for Random
Network coding, Workshop on Network Coding, theory and
Applications, 2009.
[10] R.Y.Kim,J.Jin and B.Li,Drizzle:Cooperative symbol-level Network
Coding in WiMAX,IEEE Trans.veh.Technol.,vol.59,
no.3,pp.1415- 1432,Mar.2010.
[11] Z. Zhao, L. Zhou, X. Zheng, and S.-E. Park, Enhanced
constellation rearrangement for harq with symbol combining, in
Proc. IEEE MWS, Jul. 2009, pp. 1115.
[12] M.L. Jia, J.M. Kuang, and Z.W. He, Enhanced HARQ employing
LDPC coded constellation rearrangement with 64QAM, in
Proc.Int.Conf.Commun.Technol.,2006, pp. 14.
[13] R.Y. Kim, J. Jin, B. Li, Scattered Random Network Coding for
Efficient Transmission in MultihopWireless Network,IEEE
Trans.on Vehicular Tech., vol.60, 2011.


Part III
ICCA 12
Volume 4

Establishing the Relationship between Customer Relationship
Management and E- Business

Manu Chopra
Research Scholar
Singhania University, Rajasthan

Abstract- Customer relationship management (CRM) is
also known as relationship marketing or customer
management. The goal of CRM is to aid organizations
in better understanding each customers value to the
company. CRM is a very important business initiative
because during times of economic dislocation, customer
retention repeatable revenue is a matter of survival.
CRM is the method that businesses have to deal with
the newly minted customer economy.
Now-a-days, CRM is emerging as CRM e-Business. E-
Business is a term that is basically used to describe a
business that runs on the internet and that makes use of
the existing internet technologies to increase the
profitability and efficiency of the business. The need for
CRM e-Business solution has been fuelled as products
have become more commoditized and the number of
competitors has soared with pricing differences getting
narrower. All this has placed immense pressure on the
company and organizations are fast realizing that they
need to achieve customer satisfaction if they want their
profits to rise.
In this paper we will try to establish a relationship
between the CRM and the E-platform. CRM, despite
being one of the most profitable customer strategies,
still it has some failures. The most important aspect of
CRM problems is its excellent ability to achieve
customer retention but its failure to do so. This is
indirectly responsible for CRM collapse. What actually
is CRM failure and why does it occur? Generally one of
the reasons for this failure is because most
organizations that actually employ CRM, experience a
lot of confusion about its attributes and what it really is.
Some would define it as a business strategy while others
view it as something to do with technology.

Keywords: CRM, E-Commerce, E-platform.

I. INTRODUCTION
As we all know this is the world where every
industries boundary are not confined to the boundary
of the industries. These are extended deep in the
customer. This let to the rising of the concept called
Customer Relationship Management; Basically the

Volume 1. Copyright 2012 Techno Forum Group,
India.
ISBN: 978-81-920575-7-6:: doi: 10.
73074/ISBN_0768
need to keep in touch with the customer increases
the significant of the CRM software. Now a day there
is no companies which like to create a great
relationship with the customer. The need of the
company for such type of relationship increases the
customer satisfaction that industry or company is
thinking for them.

II. CUSTOMER RELATIONSHIP
MANAGEMENT
The so called typical customer no longer exists,
and companies have been learning this lesson the
hard way. Until very recently, business was more
concerned about the whats then about the
whos. In other words, companies were found as
selling as many products and services as possible,
without regard to who was buying them. Most
corporations cling to this product centric view even
today, basing their organizational structure and
compensation plans on the products they sell, not the
customer who buy.
Nowadays the competition is just a mouse click
away. Embattled companies are slouching towards
the realization that without customers, products dont
sell and services dont materialize. They have been
forced to become smarter about selling, and this
means becoming smarter about whos buying,
companies are reading the competitive writing on the
wall looking the technology for a leg up. This,
combined with the oft quoted factoid that it costs a
company six times more to sell a product to a new
customer then it does to sell to an existing one the
old bird in the hand thus coming to roost has
motivated business to try to maximize existing
customer relationship and the main way to squeeze
every drop of value from existing customer is to
know who the best customers are motivate them to
stay that way. Indeed a good starter definition of
customer relationship management is:
The infrastructure that enables the delineation of
and increase in customer value, and the correct means
by which to motivate valuable customers to remain
loyal-indeed, to buy again.
CRM is the seamless co-ordination between sales
customer service, marketing field support and other

customer touching functions. It integrates people,
process and technology to maximize relationship
with all customers and suppliers. CRM results in a
number of benefits to an organization like increased
margins improved customer satisfaction ratings and
decreased administrative costs. The goal of CRM is
to aid organizations in better understanding each
customers value to the company, while improving
the efficiency and effectiveness of communication.
CRM captures, analysis, and distributes all relevant
data from customers and prospect interactions to
everyone in the organizations. CRM emphasized
customer acquisition and is recognized as one of the
most viable tools used to future a companys success
in the highly competitive business world.
There are three major areas that focus on
customer satisfaction, sales, marketing and services.
The functionality of and between these three fields is
essential to successfully connecting a companys
front and back offices to facilitate effective,
enterprise wide coordination. The professional sales
force predicts and proposes the real time analysis of
information and distributes this information to the
company and business partners. Marketing
concentrates on personalizing customer preferences
and offering them satisfying experiences service is
associated with the companies call centers and
coordinate interaction between web email and other
communication Medias. These fields are developed
further with the help of CRM automation.

III. CRM CHALLENGES
CRM despite being one of the most profitable
customer strategies, still it has some failures. The
most important aspect of CRM problems is its
excellent ability to achieve customer retention but its
failure to do so.
Various problems with CRM are:
a. Exorbitant costs
b. Inadequate focus on objectives
c. Inefficient Resources
d. Inappropriate matrices
e. Complex Systems.
f. No customer focus
g. Slow returns
CRM does fail in some instances. That cannot be
disputed. CRM reviews shoe that the failure rates of
CRM could reach a high of an amazing 80% in a few
years time. Thats entirely in contravention of all the
highs that have been predicted right? It just evidences
the fact that if CRM is to be employed it has to be
done with the utmost caution exercised and with full
attention to implementation.
Some software companies also fails in
implementing CRM, on this front the chief mistake
could be the way they handle SFA as a component.
What happens is that despite the fact that the
endeavor to help sales persons actually gain ground
in their sales efforts they wind up doing the opposite.
Since the systems are devised chiefly by people who
are not at the fore front and who know very little
about sales activities personally, this has an impact
on its user friendly attributes. Sales personnel
therefore find them quite difficult to handle and more
often then not non productive as well.
CRM packages tend to fail also on account of
their inflexibility or shall we say their rigidity. A
CRM solution tends view the customer as yet another
transaction. But customers are far from that. Seeing
them as figures and numbers is one of the loop holes
in the CRM industry. The human element is lost in
the entire spectrum of activity. Decisions are required
to be made on a daily basis that can not be
incorporated in the CRM package itself. It has
everything to do with spontaneous, decisive choices
being made by the employees who are face to face to
customers.
One reason CRM system often fail is because the
employees themselves are reluctant to use them or
unable to use them in the appropriate manner. Armed
with insufficient knowledge and misconceptions they
are hesitant and ill equipped to deal with the CRM
efforts. Disappointed with the initial implementation
results employs start to grumble. Reluctant to give it
a chance they stop giving off their best efforts only to
aid CRM failure. The hype about CRM is often to
blame leaving employs believing in the magic
wand sweeping away all their difficulties when the
actual pitfalls are felt, they stop working towards
CRM goals. Result CRM failures.
CRM also fails because businesses seldom bother
to make the right choice of CRM software and often
do so after little or no deliberation. Ensuring that the
chosen solution can and will support all their
business processes, is essential and must be carried
out, but it is something businesses seldom do
resulting in CRM collapse.
So Whos Responsible For CRM Failure?
Actually its a combination of both
Organizations employing CRM as well as those
selling them. Organizations havent learnt how to
employ its business processes in the right manner so
as to get the most out of its CRM implementation
while CRM sellers need to make sure that their
systems are user friendly and easily adaptable and
most importantly take into account the various factors
that have been so far overlooked. Companies need to
learn the golden rule. Merely installing CRM
packages does not ensure success. It is by no means a
surety of positive results. Understanding the fact that
the strategy does have its drawbacks, working on
these drawbacks and doing their best to get the most

out of their CRM implementation alone will
obliterate CRM failures and yield success. What is
`highly required is the ability to focus on the business
needs, choose a CRM package that works towards it,
employ the right resources and assume the right
metrics. Adopting these measures would go a long
way in alleviating CRM problems.

IV. CUSTOMER RELATIONSHIP
MANAGEMENT AND E-BUSINESS
The phenomenon of building a relationship with
customers via internet is known as electronic CRM
(e-CRM). The objective of CRM and e-CRM are the
same the difference is the medium used for
providing services to the customer. E-CRM is a
multifaceted strategy that helps companies
understand, anticipate and manage customer needs. A
major thrust of it involves segmenting customers and
offering appropriate and differentiated services for
each of these levels. It mainly uses the electronic
media to integrate and simplify customer- related
business processes, drastically reducing costs of
customer- facing operations while achieving CRMs
primary goal to enhance the customer experience. E-
Business does have its own difficulties though. It is a
lot more complex than that what people think it is. In
order to succeed at e-Business an organization needs
to basically figure out just hoe it wants top do its
business, what products it wants to sell, what are its
key objectives and who are its targeted customers.
E-Business transactions have crossed the billion
marks and are scheduled to hit $1 trillion in the year
2006. With such astounding statistics, it is but clear
that the advantages that it offers have far surpassed
company expectations and have roped in users
rapidly. Ignoring the phenomenal growth of e-
Business is futile. Companies preferring to stick to
traditional methods and processes will without a
doubt are left behind in the rat race. Since the internet
explosion globally, consumers prefer e- Business as it
smacks of ease of usage. E- Business is a term that is
basically used to describe a business that runs on the
internet and that makes use of the existing internet
technologies to increase the profitability and
efficiency of the business.

V. REASONS FOR THE GROWTH OF CRM
E-BUSINESS
5.1. The reasons for CRM e-Business growing to
such enormous heights is internet technology,
benefits of electronic technology and the easily
available solutions that exist. Companies find that
availing of CRM e- Business facilities boosts the
efficiency to a great extent. The reason e- Business
has succeeded is on account of its ability to provide
cost effective solutions that require very little time to
implement.

5.2. Traditional business methods have proved
tiresome, costly and very time consuming. As a result
businesses have been looking around for a better
solution. CRM e- Business solutions proved to be just
the strategy that was required. It catered to existing
business problems and proved that it had the potential
to deal with potential difficulties as well. Flexibility
and the ability to adapt to a changing environment is
something that CRM e-Business finds easy to do.
5.3. CRM e Business includes companies that
use the internet to acquire products or supplies for in
house production. This is known as e-procurement. It
ahs the potential to actually cut down cost
considerably making it a good bet for companies
looking at achieving cost effectiveness.
5.4. It also has the added advantage of being able
to track purchases and manage the purchasing efforts
efficiently. It helps in the buying and selling of
products. It also helps the organization to use
electronic chat as a means of technical support and
customer support. This helps any company to save
time and effort and at the same time provide
opportunities that the traditional system could not.
Whats advantageous here is that a virtual computer
system can be used and with the download of a
program, all the relevant information about hardware
and software specifications can be adequately sent to
the support team directly.
5.5. In addition it manages to maintain a central
server or email list as a method of distributing
information, rather than having costly and time
consuming individual ones. In the pats few years
almost all companies have become to some level or
the other an e- Business. CRM e- Business is right
now increasing the chance of building sales and
increasing the sales revenue by increasing the area of
operation, reducing operating costs, increasing
productivity and thereby hiking the efficiency of the
supply chain and the CRM application as well.
5.6.CRM e- Business solutions give companies a
well planned and easily integrated e- Business
strategy that caters to both the customer needs as well
as the corporate needs. Both these need to be
adequately catered to in order that company
objectives de fulfilled. The net result of
implementing CRM e Business strategies is
satisfied customers and overall productivity.

VI. CONCLUSION
Customer relationship management also known as
relationship marketing or customer management,
plays a vital role in todays competitive world and if
implemented properly, can yield very positive results

in retaining the existing customers and attracting the
new ones. CRMs goal is to aid organizations in
better understanding each customers value to the
company.
CRM inspite of being so fruitful, still faces some
challenges or fails sometimes due to the improper
implantation of it or may be due to the organization
those selling it. To overcome these challenges what is
highly required is the ability to focus on business
needs, choose a CRM package that works towards it,
employ the right resources and assume the right
metrics. This all can help organization to achieve
high profits and in retaining customers.
In the same manner e-CRM is the right tool for
the present scenario as it saves time and resources of
an organization and provides good enough outcome
in terms of customer.

ACKNOWLEDGEMENT
I would like to thank my official supervisors and
Assistant Professor Dr. Ashish N. Jani for all their
help, advice and encouragement.

REFERENCES
[1]. Jill Dych, The CRM Handbook, A business guide to
customer relationship management; 7
th
Edn.2007: 28-42.
[2]. Prof. V.Sudhakar, Indian Journal Of Marketing, e-CRM; Dec
2009; 33-39.
[3]. Jill Dych, The CRM Handbook, A business guide to customer
relationship management; 7
th
Edn.2007: 128-141.

Success and Failure Factors of e-Government project implementation in Yemen:
A systematic Literature review

Mahdi Abdullah Alsebaei
International IT Policy Program
Technology Management, Economics and Policy Program
College of Engineering, Seoul National University
Seoul, South Korea

Abstract - The aim of this paper is to explore the current
situation of e-government projects implementation in
Yemen and understand the success and failure factors from the
perception of the Yemeni e-government project team. The
methodology is exploratory method which is to investigate the
academic literature about e-government implementation in
developing countries focusing on the success and failure factors
of implementation. Also the structured telephone interviews
were done with the top decision makers of the e-government in
Yemen. The situation of Yemen in the area of communication
and technologies will be presented. The findings show that
Yemen started to establish a good plan for e-government
projects and the implementation concept started to be well
organized by the e-government projects team. The researcher
as a member of this team has identified many activities in
order to determine the most significant factors that help the
government for a proper implementation for the system of
e-government. However, Yemens government still need to
work hard regarding the Internet quality and the level of
education for citizens in terms of computer technologies and IT
skills. Yemen also needs to overcome the issue of the gap
between genders as well as between urban and rural areas.
The study will address important and significant factors
preventing the implementation of e-government and
establishing of e-services in Yemen. Finally; the requirements
for successes E-government implementation will be elaborated.
And the study ends up with recommendations as solutions for
better E-government implementation in the future for the
Republic of Yemen.

Keywords: E-government, Success factors, failure factors,
E-government in Yemen, Yemen.
I. INTRODUCTION
Electronic government or e-government suggests the use of
information technology to enable and improve the efficiency
with which government services are provided to citizens,
employees, businesses and agencies; it increases the
convenience and accessibility of government services and
information to citizens, (Carter and Belanger, 2005).
E-government comprises the use of ICT in order to deliver

Proc. of the Intl. Conf. on Computer Applications Volume
1. Copyright 2012 Techno Forum Group, India.
ISBN: 978-81-920575-7-6 :: doi: 10. 73081/ISBN_0768
public services to citizens and businesses. It entails the
transformation of public services available to citizens using
new organizational processes as well as new technological
trends (Gunter, 2006).
E-government is designed to facilitate a more integrated
mode of governance. It encapsulates the relationships
between governments, their citizens as well as their suppliers
by the use of electronic means (Means and Schneider, 2000).
In brief, the goal of e-government is to achieve in making
government services more accessible, more citizen focused,
more relevant to citizens as well as more responsive to their
needs and expectations.
In this perspective, e-government is the use of
technology, especially web-based Internet applications in
order to enhance access to and efficiently deliver government
information and services (Brown and Brudney, 2001).
(Heeks, 2003) asserts that the use of information and
communication technologies as a means to improve the
activities of public sector organizations is the definition of
e-government. According to the World Bank (2004), e-
government refers to the use of information technologies by
government agencies that have the ability to transform
relations with citizens, businesses, and other arms of
government. Last but not least, according to the Centre for
Democracy and Technology (2002), e-
government is the use of information and communications
technologies in order to transform government by making it
more accessible, effective and accountable.
The researcher conducted this study to review the
situation of e-government in Yemen and analysis its related
concepts which is among the early stage of E-government
project. The study will analyze several factors that contribute
to the success implementation of E-government project in
local government of Sanaa city- the capital of Yemen.

II. E-GOVERNMENT CONCEPTS
A. Main objectives of e-government
Make it easier to find information about
government services.
Improving access to government services by
making them online 24/7.
Helping to improve the quality of services.
Helping to interact more effectively with
government organizations and related agencies so that
more services can be provided.
Helping to build trust between government
Challenging the existing ways of working.
B. E-government Applications
Government-to-Government (G2G):
G2G is the interconnection between the government
agencies. This function or application serves both internal
processes and activities (between public organizations
themselves) and external ones also (between government
organizations, citizens and businesses). The ultimate aim of
the (G2G) function is to enhance inter-government
organizations' processes by streamlining collaboration and
coordination.
In order to realize a single access point, collaboration and
cooperation among different governmental departments and
agencies is compulsory. Online communication and
cooperation allows government agencies and departments to
share databases, resources, pool skills and capabilities,
enhancing the efficiency and effectively of processes
(Valantina, 2004).
Government-to-Citizens (G2C):
G2C is the interconnection between the government
agencies and the public citizens. With this function or
application, the public organizations publish information and
contact details, and offer regular services online. The
ultimate aim of this application is to give users different
options and communication channels for government
transactions.
G2C allows customers to access government information
and services instantly, conveniently, from everywhere, by
use of multiple channels (PC, Web TV, mobile phone or
wireless device). It also enables and reinforces their
participation in local community life send an email or
contribute to an online discussion forum (Valantina, 2004).
Government-to-Businesses (G2B):
G2C is the government communication with the private
business sector. G2B allows businesses as well as individuals
to have transactions with the government, examples being:
renewing registrations, lodging taxes, downloading tenders
information, and many others. The government-to-business
(G2B) application is as useful as the G2C system, enhancing
the efficiency and quality of communication and transactions
with business. Therefore, this should be considered by the
governments.
Companies everywhere are conducting business-to-
business e-commerce in order to lower their costs and
improve inventory control. The opportunity to conduct
online transactions with government reduces red tape and
simplifies regulatory processes, therefore helping businesses
to become more competitive. The delivery of integrated,
single-source public services creates opportunities for
businesses and government to partner together for
establishing a web presence faster and cheaper (Valantina,
2004).
Government-to-Employees (G2E):
G2E is the interconnection between the government
agencies and the employees from different sectors (public
and private G2E is perhaps the least adopted application of e-
government. Scholars and countries around the world usually
focus on the first three applications only.
G2E is an effective way to provide e-learning, bring
employees together and to promote knowledge sharing
among them. It gives employees the possibility of accessing
relevant information regarding: compensation and benefit
policies, training and learning opportunities, civil rights laws,
etc (Valantina, 2004).
However, Yemen is still in the first function G2G which
is the interconnection between the government agencies.
There is no full interaction with citizens because they can
only brows the information about government services from
the different government agencies through the government
official portal which is announced in 2009. Moreover, there
is no electronic interaction with the Yemeni government
with the private business sector yet as well as the employees
from both public and private sectors, these maybe targeted to
be established after completing all the stages required to be
implemented for the e-government system in Yemen.
C. Benefits of E-government
OECD (2003) thoroughly examined e-government
initiatives in its members countries (Australia, Austria,
Belgium, Canada, Czech Republic, Denmark, Finland,
France, Germany, Greece, Hungary, Iceland, Ireland, Italy,
Japan, Korea, Luxemburg, Mexico, Netherlands, New
Zealand, Norway, Poland, Portugal, Slovak Republic, Spain,
Sweden, Switzerland, Turkey, UK, USA), and listed the
advantages of e-government as follows:
- Improves efficiency in processing large quantities of
data
- Improves services through better understanding of
users requirements, thus aiming for seamless online
services.
- Helps achieve specific policy outcomes by enabling
stakeholders to share information and ideas.
- Assists a governments economic policy objectives by
promoting productivity gains inherent in ICT and
e-commerce.
- Contributes to governments reform by improving
transparency, facilitating information sharing and
highlighting internal inconsistencies.
- Helps in building trust between governments and their
citizens, an essential factor in good governance by using
internet-based strategies to involve citizens in the policy
process, illustrating government transparency and
accountability.

III. PURPOSE OF THE STUDY
THE MAIN PURPOSES OF THIS STUDY ARE AS FOLLOWS:
To understand the current level of e-government
project implementation in Yemen.
To determine the critical success and failure factors
of e-government project in Yemen as being perceived by top
decision makers and projects team.
To come up with some recommendations that
should bring into attention for the e-government project
implementation in Sanaa city the capital of Yemen.

IV. E-GOVERNMENT MODELS
Several e-government models have been proposed by
researchers and International organizations such as (UN
2008, Layne & Lee 2001) and some others. We briefly
discuss these models in the following subsections:
A. UN e-government maturity model 2008: A Five
stages model:
United Nations has identified stages of e-government
from different perspective. UN developed a five stage e-
government model which builds upon the sophistication of
online presence. As a country migrates upwards through the
various stages as shown in the figure, it is ranked higher in
the web measure index (UN, 2008).

Figure 1: UN E-government Maturity Model
Source: UN E-government Index 2008.
Stage I - Emerging: A governments online presence is
mainly comprised of a web page and/or an official website;
links to ministries or departments of education, health, social
welfare, labor and finance may/may not exist.
Stage II - Enhanced: Governments provide more
information on public policy and governance. They have
created links to archived information that is easily accessible
to citizens, as for instance, documents, forms, reports, laws
and regulations, and newsletters.
Stage III - Interactive: Governments deliver online
services such as downloadable forms for tax payments and
applications for license renewals.
Stage IV - Transactional: Governments begin to
transform themselves by introducing two-way interactions
between citizen and government. It includes options for
paying taxes, applying for ID cards, birth certificates,
passports and license renewals, as well as other similar G to
C interactions, and allows the citizen to access these services
online24/7. All transactions are conducted online.
Stage V - Connected: Governments transform themselves
into a connected entity that responds to the needs of its
citizens by developing an integrated back office
infrastructure. This is the most sophisticated level of online
e-government initiatives and is characterized by:
1. Horizontal connections (among government agencies)
2.Vertical connections (central and local
government agencies)
3. Infrastructure connections (interoperability issues)
4. Connections between governments and citizens
5. Connections among stakeholders (government, private
sector, academic institutions, NGOs and civil society)
B. Developing fully functionally E-
government: A four stage model By Layne & Lee 2001:
Based on technical and managerial feasibility, Layne and
Lee (2001) proposed a four-stage e-government model.
These four stages are:
Stage 1- Cataloguing: government online
presence
Stage 2- Transaction: citizens transact
with government electronically.
Stage 3- Vertical integration: Local
government connected to different functions
or service of the government.
Stage 4- Horizontal Integration:
Integration across different functions and services of
government.
Figure 2: Layne and Lee (2001), Egovernment Stages Model
Source: Developing fully functionally E-government:
A four stage model by Layne & Lee 2001.

According to the model, the first stage is the cataloguing
of information on a Web site. This involves the initial efforts
C: Connected
Transact: Transaction
-Horizontal
Integration
-Systems
integrated across
different
functions
--Real one stop
shopping for
citizens
-Vertical
Integration
-Local Systems
linked to higher
level systms
-within
similar
functionalitie
Transaction
-Services and
Forms are
online
-Working D.B
supporting
online Catalogue
-Online presence
-Catalogue
presentation
-Downloadable
Forms

of government to be present online. In this stage, the
information is not a transactional operation. Online presence
of the government saves its time and money because that
will reduce the workload on frontline offices. The second
stage is, Transaction, aims to extend the capability of
catalogue stage and enables to perform a simple online
transaction. The Vertical Integration stage transforms
government services and integrating them at different levels.
The horizontal Integration stage is the maturity stage with
unified and seamless services.
The two models differ because they are developed from
different perspectives. UN (2008) five-stage model focus on
a web-based public service. Layne & Lees (2001) four-stage
model is developed based on a general or an integrated
perspective combining technical, organizational, and
managerial feasibility. However, the model proposed by
Layne & Lee (2001), four-stage model misses out on the
political participation component and does not address the
possible changes in the way decisions are made in
government. Yemen government is most likely deal with the
different stages of e-government by following both of these
models. (E-government project team Documentation, 2008).

V. ICT AND E-GOVERNMENT IN YEMEN
In this part the researcher is trying to present the current
situation of Yemens ICT and e-government. Therefore this
part will be explained in five subsections as follows:
A. Yemen overview:
Yemen is one of the developing countries which located
in the southern part of the Arabian Peninsula. It is bounded
on the north by Saudi Arabia and by the Arab Sea. Oman lies
in the west of the Republic of Yemen and the Red Sea lies in
the west of the Republic of Yemen. The total area of the
Republic of Yemen is about 555,000 square Kilometer, and
the population of Yemen is 23 million (NIC, 2010). The
official language of the country is Arabic and Islam is the
official religion of the country. The Yemeni Rial (YR) is the
official currency unit. The Republic of Yemen has three
national independence days: September 26, 1962 when the
king of North the Republic of Yemen, at that time, was
overthrown and making the country a republic instead of a
kingdom, November 30, 1967 when South the Republic of
Yemen, at that time, had become independent from United
Kingdom. The Unification Day on May 22, 1990, when the
Republic of Yemen was established by the merger of South
and North of Yemen.

Figure 3: Map of the Republic of Yemen.

B. ICT Sector in Yemen
Information and Communication Technology (ICT) is
considered nowadays an efficient tool serving the progress
and development of countries (ESCWA, 2007). Like other
less developed countries, the republic of Yemen is seeking to
develop this vital sector to carry on its development process
(ESCWA, 2007). Yemen is in lowest level of ICT
infrastructure in world, this level is characterized by the
following: (a) low penetration rates of fixed and mobile
telephone lines; (b) lack of an environment conducive to
widespread use of telecommunication services by businesses
and individuals; and (c) insufficient national bandwidth,
inadequate backbone for voice and data telecommunication
and insufficient number of Internet players in the market
(ESCWA, 2009).
The following table 1- shows that Yemen country index
in term of using computers and Internet, telephone usage.

Table 1: Telecommunication Infrastructure index and its components
Country
Index
value
Estimated
Internet users
per 100
inhabitants
Main fixed
telephone
lines per 100
inhabitants
Mobile
subscribers
per 100
inhabitants
Personal
computers
per 100
inhabitants
Total fixed
broadband
per 100
inhabitants
Yemen 0.0298 1.44 4.48 13.76 2.77

0.00

Source: United Nations E-government Survey 2010.

According to (ESCWA, 2007), Yemens effort towards
building an Information Society can be summarized as
follows:
The National Information Center (NIC) was established
during the second half of 1995.
The ministry of Telecommunication and Information
Technology (MTIT) in 2003, it was entrusted with IT
management in addition to development and organization
of communication policies. It has many campuses deal
with the IT such as:
1. Telecommunication and Information Technology city
2. Electronic Library
3. General Telecom Council
4. Data communication
5. Yemen Electronic Government Portal.
The central statistics Department, affiliated to the
Ministry of Planning and International Cooperation,
adopted a national strategy on statistical work.
C. TelephoneCapacity
According to (ESCWA, 2007), Yemen is still among the
countries that benefit little from communication services.
The main challenges and obstacles are the countrys
difficult topography, scattered population groups and low
density, weak infrastructure services and low level of
distribution networks in the cities as well as in rural areas.
The following table shows the majority indicators for
Yemen telephone communications.

TABLE2: TELEPHONE COMMUNICATION INDICATORS (URBAN AND RURAL)
DURING 2003-2005
INDICATOR/YEAR 2003 2004 2005 2006 2007
URBAN COMMUNICATION
TOTAL CAPACITY OF
COMMUNICATION
NETWORK
1,505,192 1,564,349 1575,169 1,580,966 1,573,476
CAPACITY FITTED(LINE) 1,161,041 1,243,838 1,278,315 1,300,042 1,326,125
Population (People) 19,131,977 19,721,631 20,305,269 20,900,532 21,538,995
Telephone density
(telephone/ per 100
inhabitances)
3.5 4.1 4.4 4.63 4.74
Lines Vacancies (line) 476,157 445,702 376,930 331,714 304,137
Number of working
telephone lines (line)
684,884 798,136 901,385 968,328 1,021,988
The increase in the
working telephone lines
142,680 113,252 103,249 66,943 53,660
The number of main
exchanges (number)
47 52 54 49 49
The number of sub
exchanges (number)
208 222 223 229 229
Rural Communication**
Rural Communication
Capacities fitted (Line) 101,344 128,385 240,245 190,575 193,578
Lines operating (line) 62,593 96,027 124,921 156,190 168,718
Percentage of lines
operating to equipped
capacities (%)
62 75 52 82 87

Source: NATIONAL PROFILE OF THE INFORMATION SOCIETY IN YEMEN, ESCWA 2009.
D. Internet and mobile services in Yemen
According to (Al-Eryani, 2009), Internet services in Yemen
started in 1996 with one single provider. Second provider
has been started working in April 2002. Wireless access to
network was lunched at the end of December 2006.
Between2004-2006, Yemen improved its mobile
communication services as the number of mobile telephone
subscribers jumped from 1.483 million in 2004 to 3.201
million 2006 (ESCWA, 2007). The following table shows
the majority indicators for Yemen internet and mobile
services.
Table 3: Internet and Mobile Phone Indicators in Yemen
Source: NATIONAL PROFILE OF THE INFORMATION SOCIETY IN YEMEN,
ESCWA 2007

E. E-government development in Yemen

The Yemen government, like most other developing
countries, is still trying to implement IT in government
organization, and try establishing some projects as initial
step to make e-government such as the National Program for
Information Technology, that known as Yemen
e-government project 2003. The following Table-4 shows
that Yemen is still one of the lowest countries in terms of
E-government development between 2008-and 2010.
Table 4: E-government developments in western Asia (including Yemen)
between 2008 and 2010.
Country
E-government development
index value
World e-government
development ranking
2010 2008 2010 2008
Bahrain 0.7363 0.5723 13 42
Cyprus 0.5705 0.6019 42 35
United Arab Emirates 0.5349 0.6301 49 32
Kuwait 0.5290 0.5202 50 57
Jordan 0.5278 0.5480 51 20
Saudi Arabia 0.5142 0.4935 58 70
Qatar 0.4928 0.5314 62 53
Turkey 0.4780 0.4834 69 76
Oman 0.4576 0.4691 82 84
Azerbaijan 0.4571 0.4609 83 89
Lebanon 0.4388 0.4840 93 74
Georgia 0.4248 0.4598 100 90
Armenia 0.4025 0.4182 110 103
Syrian Arab Republic 0.3103 0.3614 133 119
Iraq 0.2996 0.2690 136 151
Yemen 0.2154 0.2142 164 164
Source: United Nations E-Government Survey 2010.

VI. E-GOVERNMENT PROJECTS IN YEMEN
Information Communication Technology (ICT) is
considered nowadays an efficient tool serving the progress
and development of countries (ESCWA, 2007). The Yemen
government, like most other developing countries, is trying
to implement ICT in government organization, and try
establishing some projects as initial step to make
e-government system such as the National Program for
Information Technology, that known as Yemen e-
government project (MCIT, 2003), and the official
government portal which has been announced in 2009. As
the previous Table 4 compare Yemen to other countries
around the world and show that Yemen is still one of the
lowest countries in term of E-government development
between 2008-2010.
A. First E-government project in Yemen (2003):
The first attempt to build e-government in Yemen was
announced in 2003 and its websites was launched at the
same year but with failure. However, the project was
derailed in the absence of change-inducing environment
(ESCWA, 2007). Moreover, most of the available
information systems lack of technical compatibility and
updated information (NIC, 2005). This first e-government
project faced some problems which lead to the failure of the
project; these problems can be summarized as follows:
1. Absence of clear vision and strategy for the purpose of
the e-government project.
2. Absence of coordination and integration between the
e-government project team and different stockholders
from the government ministries and related agencies.
3. Absence of political leadership.
4. Conflicting priorities about the responsibilities among
involved organizations toward the e-government
project.
5. Carefulness from some organization to diffusion their
information.
6. Lack of technical knowledge and common technical
culture.
7. Lack of sense and social awareness of the importance
of e-government system.
8. Lack of clear methodologies and models.
9. Lack of e-readiness among the government
organizations employees including the top
management and leaders.
10. The gap between the project vision and the real
situation.

Because of the above factors, the project was failed. The
failure of the first e-government project was seen from other
top management point of view as adjournment not failure
due to the government prior projects such as improving
education sectors and health sectors.
B. Second E-government project in Yemen (2008):
This project was established in September 2008.The
main aim was to implement and completely design the
Yemen official electronic portal. The first stage was to
design the electronic information content of all government
organizations, taking into account the daily updating. In this
stage the citizen will benefit from e-government website in
two different ways. First way is finding information from
the portal via the internet. Second way, is inquiring about
the government services via the emails and contacts
available on the government portal.
From the new e-government plan, there will be projects
working in parallel with the main project aim to develop the
technical culture. Other projects will aim to diffusion the
technical skills by doing many technical skills by doing
many technical courses in many locations in Yemen such as
summer camps (Al-Eryani, 2009).
The second stage in the new e-government plan will be
considering the concept of interaction between e-
government and citizens in one direction (semi-interaction).
The third stage of this plan will be focusing on e-payment.
This stage will aim for full interaction in two directions.
However, the Yemen e-government portal at
www.yemen.gov.ye is the first step toward the full
implementation of the e-government project. The main aim
of this government portal is to make a unified container for
all government ministries and agencies in one site. This
electronic portal must be the official source of all the
government information. This portal can be considered as
the first track to the e-services and the first step toward the
full e-government.
According to (Al-Eryani, 2009) this portal has many
benefits such as: reduce cost of searching information,
reduce time and effort of searching information, and prepare
the society for the next step which toward the semi
interaction with the government websites.
One of the main reasons behind the success of designing
and launching the government portal with all ministries of
Yemen government is the combination of selecting the
project team which consists of ten members five of them are
from the ministry of telecommunication and information
technology for the technical and hosting issues and the other
five are from the Information center of the Secretary of the
cabinet at the Prime minister office for the administration
and policy issues. This leads to get a good response from the
different ministries and related agencies to upload their
information as required from the project team to the official
portal in a particular time and each of them take the
responsibility of upgrading their information. This could not
be happen without the interfere of the secretary of the
cabinet in the Prime minister office which gives directions
Indicator/Years 2003 2005 2006
Overall internet subscribers (subscriber) 74615 109127 158476
Overall internet subscribers (per 100 inhabitants) 0.38 0.54 0.76
Internet users (user) 221000 330000 450000
Overall number of mobile phone subscribers 1483233 2277553 3201796
Mobile phone subscribers per 100 inhabitants 75 11.2 15.3
Number of computers in the country (computer) 210000 270000 330000
Computer dissemination density
(one device per 100 inhabitants)
1.1 1.3 1.6
and instruction to the different ministries and agencies to be
on time to do their responsibilities with the respect to the
scheduling time specified by the project team.

VII. CONCLUSION AND RECOMMENDATIONS
The OECD (2003) defines e-government as the use of
information and communication technologies, and
particularly the internet, as a tool to achieve better
government. In this description, the internet is defined as a
requirement and a possible medium for e-government. Thus,
the first step, countries need to define their clear vision and
strategy for e-government system. This broad vision of e-
government should be shared with the all different
stakeholders (government agencies, Citizens, Businesses,
and others). From the e-government projects team
information and phone call interviews and also researcher
point of view as one member of the team, it has been clear
that there are failure and success factors of implementing e-
government system in the metropolitan city Sanaa-the
capital city of the republic of Yemen. These factors can be
summarized as follows:
Success factors:
1. Make the government employees e-ready. That
means train them for the required ICT skills and knowledge
as well as make them awareness of the benefits and
importance of the e-government system.
2. Make the citizens e-ready. That means to reduce
the gap between the citizens in the technology culture and
skills.
Failure factors:
1. The ambiguous vision of the conceptual of the
e-government.
2. The absence of the readiness and awareness among
the employees and the citizens.
3. The digital divide between the communities.
Recommendations:
From the researcher point of view, some
recommendation that Yemeni government should strongly
consider towards e-government implementation which could
be no technological factors as well. These recommendations
can be summarized in two stages:
Stage1: E-readiness Infrastructure: This has to include
the following:
1. Education: improve the quality of
education and offer education for all.
2. Gender: reduce the gap between males
and females and offer them equal education and
chances.
3. Income: since the GDP per capita is
around $1000 USD, the government should reduce the
cost of computers, training, and internet access and
make them available for the all layers of citizens with
cheaper prices.
4. Rural and Urban: reduce the gap between
the citizens in the rural and urban areas by offering
them.
Stage2: E-readiness Diffusion: This has to include the
following:
1. Diffusion of computers and internet culture
2. Clarify the thought that Computer is only for
specialist.
3. Clarify the misunderstanding of dangers of the
internet websites.
4. Clarify the conceptual of e-government benefits
and importance for citizens and employees.
Finally, it can be said that Yemen published its official
portal in 2009 towards preparing for the whole e-
government project implementation. The vision is now clear
and the processes are somehow identifies and established.
One of the main reasons behind the success of designing
and launching the government portal with all ministries of
Yemen government is the combination of selecting the
project team which consists of ten members, five of them
are from the ministry of telecommunication and information
technology for the technical and hosting issues, and the
other five are from the Information center of the Secretary
of the cabinet at the Prime minister office for the
administration and policy issues. This leads to get a good
response from the different ministries and related agencies
to upload their information as required from the project
team to the official portal in a respected time, therefore each
of them take the responsibility of upgrading their own
information. This could not be happen without the interfere
of the secretary of the cabinet in the Prime minister office
which gives directions and instruction to different ministries
and agencies to be on time to do their responsibilities with
the respect to the scheduling time specified by the project
team.
ACKNOWLEDGMENT
This work was supported by the Ministry of Knowledge
Economy (MKE), South Korea, under the International IT
Policy Program (ITPP) at Seoul National University.
REFERENCES
[1] United Nations (2008). UN E-Government Survey 2008: From e-
government to Connected Governance, Department of Economic and
Social Affairs, Division for Public Administration and Development
Management, available at:
http://unpan1.un.org/intradoc/groups/public/documents/un/unpan0286
07.pdf
[2] Layne, K. and Lee, J.W. (2001).Developing fully functional E-
government: A four stage Model, Government Information Quarterly
18(2), 122-136.
[3] Carter, L. and Belanger, F. (2005). The utilization of e-
government services: citizen trust, innovation and acceptance factors,
Information system J 15, 5-25.
[4] Valentina (2004). E government for Developing Countrie;
Opportinuties and Challenges. The Electronic Journal on Information
Systems in Developing Countries, EJISDC (2004) 18, 1, pp. 1-24.
[5] Heeks, R. (2003) E-government for Development Projects Fail: How
can Risks be Reduce? iGovernment working paper series, Paper no.
14. Available at http://
http://unpan1.un.org/intradoc/groups/public/documents/NISPAcee/U
NPAN015488.pdf
[6] OECD,2003 OECD e-Government Studies: The e-Government
Imperative, OECD Publications, France, 2003
[7] ESCWA (Economic and Social Commission for western Asia),
(2007) National Profile of the information society in Yemen, United
Nation available at:
http://www.escwa.un.org/wsis/reports/docs/Yemen_2005-E.pdf
[8] The Yemeni e-government portal project documentation, (2008,
2009).
[9] Dada, D. (2006). The Failure of E-government in Developing
Couturiers: A Literature Review. The Electronic Journal on
Information Systems in Developing Countries, EJISDC (2006) 26, 7,
1-10.
[10] United Nations (2010). UN E-Government Survey 2010: Leveraging
e-government at a time of financial and economic crisis, United
Nations, New York, 2010. Available at:
http://unpan1.un.org/intradoc/groups/public/documents/un/unpan0388
51.pdf
[11] World Bank (2009). World Development Report, available at
http://econ.worldbank.org/ accessed on September 15, 2010.
[12] NIC (National Information Center) (2005), E-government Yemens
particularities and methodology for achieving ambition. Paper
submitted to Yemens Shoura council meeting.
[13] Basu, S. (2004), E-Government and Developing Countries: An
Overview, International Review of Law Computers & Technology,
Vol 18, No. 1, pp. 109132.
[14] Sharifi, H. and Zarei B, (2004). An adaptive approach for
implementing e-government in I.R. Iran, Journal of Government
Information 30 (2004) 600619, doi:10.1016/j.jgi.2004.10.005


A Design Model of Banking Website in Perspective of Behavioral Feasibility Study

Mr. Saravanan.C Mr. Vinay R
Assistant Professor, Department of MCA, Student, 3
rd
Semester, Department of Mechanical Engineering
R.V.College of Engineering R.V.College of Engineering
Bangalore, India Bangalore, India

AbstractIn Todays network world, security is the major
concern all over the world. When new security measures have
developed to solve the existing threat, a new attack is produced
by the malicious users. This paper deals with problems related
to middle aged group people and aged people while using bank
websites for transactions and other works. Phishing is one of
the reason where people avoid using e-banking. In this paper
we are giving suggestions to enhance the feature of banking
websites based on behavioral feasibility study by using
simulation

Keywords-Security,Phishing,e-banking,behavioral feasibility.
I. INTRODUCTION
Phishing is the word which is toping the list of cyber
crimes as this is the starting crime which leads to all other
cyber crimes. Phishing is the act of acquiring private or
sensitive data from personal computers for use in fraudulent
activities. Today Phishing websites are increasing at an
alarming rate day by day. According to Indian Computer
Emergency Response team [CERT- In] in the survey
conducted by them, out of 109 security incidents reported in
the month of September 2011, 28% incidents were related to
Phishing [1].
Many people avoid e-banking stating Phishing as one of
the reason. Aged people also find difficulties because of their
vision problem that increases with their age. Decreasing
ability to focus on near tasks related to computer screen,
Lack of confidence and trust worthy in using banking
website make people to ignore using online banking
transactions. In terms of security, people are also worried
about their net bank login details to be hacked by
unauthorised malicious users. Customers are certainly
concerned of giving their bank account number online.
Numerous studies have identified lack of trust as one of the
main reasons that consumers are still reluctant to conduct
their financial transactions online
In case of emergency, users dont get appropriate responses
from the server as the form of Account statements and other
online transaction details. Most of this occurs due to
technical problems. Users also feel that internet banking
involves in a paperless business environment. They fear that

Proc. of the Intl. Conf. on Computer Applications Volume
1. Copyright 2012 Techno Forum Group, India.
ISBN: 978-81-920575-7-6 :: doi: 10. 73088/ISBN_0768

if they do not have written proof of transactions, they will
lose the ability to prove anything. They want the control a
piece of paper seems to give them. Educated users are also
still using the bank passbook for their banking transactions
and not using the facilities of the net banking for the
transactions. The issue of customer unfamiliarity with the
internet, which is prominent among senior citizens, has
recently caught some attention, because these customers
believe that they are left at a disadvantageous position and
become very reluctant in doing business online. Many
Procedures are involved in using net banking as compared to
ATMs. Banking through online has emerged as an important
resource for controlling the transactions, reducing the cost by
replacing paper works which in turn leads to efficient
productivity and profitability [2].
Fig 1 represents the statistics of phishing rate for the year
2011, from January till September. In the month of April
2011, there was high rate of phishing incidents.Fig 2
represents the genuine website of banking website and Fig 3
represents phishing website which differs only in the URL.

Fig 1. Phishing Statistics in the year 2011
(January to Septembet)

Fig 2. Genuine Banking Website

Fig 3. Phishing Website which differs in URL
II. EXISTING PROBLEMS WITH BANKING WEBSITES
Text color and Font size issue
The texts at the site are typed with a very small font and
the font color is practically identical to the background color.
It prevents the visitors from studying the information at the
site as such fonts are very tiring for the eyes. The
background color of some of the websites is too bright and
motley. The text gets lost on such background and it is hard
for visitors to read anything.
Slow Server Response Times
Some bank websites contains more graphics than the
desired level which results in delay in loading the page.
Sometimes building a page also involves connections to
database servers, slowing down the process even further. A
user does not bother why response times are slow. All they
know is that the site does not offer good service. The level of
trust gradually reduces due to low response times
Annoying objects in the websites
Aged Users find Flashing and blinking in the website to
be irritating for them .They also find pages with irrelevant
material such as advertisements distracting to them.
Actions
Scrolling and other mouse activities are a learned
behaviour and become more difficult with age. Using pull-
down menus, scrolling lists and scrolling pages is difficult to
understand, accessing the items what is clickable is
problematic for some elderly users.
Security issues
Security and data confidentiality issues have been a
major barrier in the banking industry in the current period
[3]. Some customers avoid online banking as they perceive it
as being too vulnerable to fraud. The security measures
employed by most banks can never be completely safe but it
becomes less secure if users are careless or computer
illiterate. An increasingly popular criminal practice to gain
access to a user's finances is Phishing, whereby the user is in
some way persuaded to hand over their password(s) to a
fraudster.

Table 1 represents the general problems faced by elder users
in using banking website online. Elder users suggest 3 major
factors like security, simple design of website and training in
deciding the usability of banking related websites.
Banking organizations can concentrate more on these factors
so as to increase the number of elderly online users.Fig 4
represents the top priority factors suggested by them to
enhance the banking website features.

Table 1. Percentage division of problems faced by elder Users

Sl.No Problems faced by elder users
Percentage

1 Formatting
22 %

2 Slow Server Response time
24%

3 Annoying objects
18 %

4 Security issues
22 %

5 Lack of Computer Awareness
14 %

Fig 4. Major factors suggested by elderly users
III. DESIGN MODEL OF THE WEBSITE
Banks can take the following measures to increase the
feasibility of the elderly users
Design User Friendly website

The bank website should be formatted in a user friendly
manner. Computer jargon and many technical terms used on
sites that confuse the users should be avoided .The website
should be designed in a such a way that elder customers
should be familiar to operate.
Visual Design for website
Experienced older users can scan pages as well as
younger users, but older users with vision deficits need to be
accommodated with suitable contrast along with headings to
help them narrow their visual search Aged users generally
prefer larger text.
Website usability
Users should be able to achieve specified goals with
effectiveness, efficiency and satisfaction. Good conventions,
Clear description of the utilities and concise content in the
website should be maintained

Interactive Webpage
Design convention such as underlined links should be
followed Bank Websites should have more help options and
should be more interactive with GUI like ATM.

IV. GENERAL STEPS TO BE CONSIDERED BY BANKS
Banks should provide suggestions and necessary advices
on the importance of security precautions to its elder
customers [4].Rural area resident people are not aware of
online banking and other features of banking web site
[5].Banks and financial organizations can take the following
steps to increase number of elderly users to use online
banking.
Banks should educate their customers and manage
general consumers perception of the risks in
transacting on the internet.
In order to maintain security and privacy of the
elderly customers, when using online banking,
Banks can organize seminar and workshops for
them.
Some Elder customers prevent online banking due to
lack of computer skills. They can be educated on
Some Basic computer skills needed to conduct
online banking.
Banks should insist on the convenience that online
banking can provide to elder people, such as
avoiding long queue, in order to motivate them to
use it.
Banks can also motivate the elderly users on cost
saving that online can provide, such as reduction in
transaction cost by use of online banking
Banks should increase awareness among elder
customers by conducting awareness programs and
make them feel online banking is also secure for
conducting financial transactions like traditional
banking.
V. CONCLUSION
Banking through online has a very bright and promising
future in the eyes of Elder customers. If Banks provide them
with necessary support and suggestions, majority of the
elderly customers will try to experience online banking.
Elder customers have strong confidence in their banks but
their confidence level in the technology is at low level.
Banks should increase the level of awareness They are ready
to adopt online banking if banks can take necessary actions.
Based on the suggestions from the elderly users, banks can
modify the design pattern and user interface of the website.

REFERENCES
[1] CERT-in Monthly Security Bulletin, Indian Computer Emergency
Response team, Department of Information Technology, Ministry of
Communications and Information Technology
[2] Pooja malhotra, Balwinder singh, The Impact of Internet Banking on
Bank Performance and Risk: The Indian Experience, Eurasian
Journal of Business and Economics 2009, 2 (4), pp 43-62.
[3] Neha Dixit, Dr. Saroj K. Datta Acceptance of E-banking among
Adult Customers: An Empirical Investigation in India Journal of
Internet Banking and Commerce, August 2010, vol. 15, pp.2.
[4] Andrea Schaechter, Issues in Electronic Banking: An overview,
IMF Policy Discussion Paper, International Monetary Fund 2002.
[5] Jiaqin Yang, Mike Whitefield and Katja Boehme ' New issues and
challenges facing e-banking in rural areas: an empirical study' Int. J.
Electronic Finance, Vol. 1, No. 3, 2007 pp. 336 - 354

Semantic Web Technique Apply E-Museum Domain Using the Protg Tool for
Ontology Based Semantic Search

K.Palaniammal,
Research Scholar, IT Department ,
Thiagarajar College of Engineering,
Madurai-625015, Tamilnadu, India.

Abstract : The Semantic Web is an extension of the current
web in which relevant information is given to the required
people by taking into account the meaning of the words.
This paper presents semantic web technique apply E-
museum domain using the Protg tool for ontology based
information retrieval. It involves XML, RDF and Ontology
while searching the information. The ontology is being
created with Protg which is the tool used as an ontology
editor. Results show that more relevant information is
retrieved using the proposed approach.

Keywords: Ontology, RDF, Semantic Web, Information
Retrieval, E-museum, Protg Tool.

I. INTRODUCTION
Most of the search engines search for keywords to
answer the queries from users. The search engines
usually search web pages for the required information.
However they filter the pages from searching
unnecessary pages by using advanced algorithms. These
search engines can answer topic wise queries efficiently
and effectively by developing state-of art algorithms.
However they are vulnerable in answering intelligent
queries from the user due to the dependence of their
results on information available in web pages. The main
focus of these search engines is solving these queries
with close to accurate results in small time using much
researched algorithms. However, it shows that such
search engines are vulnerable in answering intelligent
queries using this approach. They either show inaccurate
results with this approach or show accurate but (could
be) unreliable results with the keywords based searches
they usually provide results from blogs (if available) or
other discussion boards. The user cannot have a
satisfaction with these results due to lack of trusts on
blogs etc. To get the trusted results search engines
require searching for pages that maintain such
information at some place. This requires including
domain knowledge in the web pages to help search
engines in answering intelligent queries. The layered
model of Semantic Web provides solution to this
problem by providing tools and technologies to enable
machine readable semantics in current web contents.

Volume 1. Copyright 2012 Techno Forum Group,
India.
ISBN: 978-81-920575-7-6 :: doi: 10.
73095/ISBN_0768
Web search is a key technology of the Web, which is
essentially based on a combination of textual keyword
search with an importance ranking of the documents
depending on the link structure of the Web. For this
reason, it has many limitations, and there are a plethora
of research activities towards more intelligent Web
search, called semantic search on the Web, or also
Semantic Web search, which is currently one of the
hottest research topics in both the Semantic Web and
Web search (see [8] and [7], respectively).
The semantic web is a meaningful web. It is a mesh
of information linked up in such a way as to be easily
processable by machines, on a global scale [2]. The
semantic web backbone technology is ontology.
Ontology is about the exact description of things and
their relationships. For the web, ontology is about the
exact description of web information and the relationship
between web information [1] [3]. The semantic web is
based on a language such as XML, RDF, OWL and tool
that make use of these languages. The semantic web is
generally built on syntaxes which use URIs to represent
data, usually in triples (subject, predicate, and object)
based structures: i.e. many triples of URI data that can be
held in database or interchanged on the World Wide Web
using a set of particular syntaxes developed especially
for the task. These syntaxes are called Resource
Description Framework syntaxes.
In this paper, we propose the semantic web based
search engine for museum domain with annotations
created for every collections. We apply the power of
XML meta-tags deployed on the web page to search the
queried information. The XML page will be consisted of
built-in and user defined tags. The metadata information
is extracted from this XML into RDF. The RDF triples
are populated by inputting through XForms [16]. These
tags will help the system in getting answer from reliable
sources. In this work we focused on the museum
domain. SQL is used for retrieving results for our sample
queries. Semantic interoperability is achieved by using
ontologies while the use of XML/RDF ensures machine
understandability.
The implementation of the semantic web depends on
technologies such as XML, RDF, and ONTOLOGY
which are all W3C recommendations. In this paper we
described in the next section literature survey, section 3
presents what is museum, museums and the semantic
web, section 4 presents implementations, section 5 for
conclusions and finally shows references.

II. LITERATURE SURVEY
The SCULPTEUR [14] project is an ontological
based solution for navigating, searching and retrieving
digital cultural heritage information from multiple
distributed digital museums. The nature of the work from
this view point is global ontology based semantic
integration for multiple heterogeneous databases [15].
Hakia [10] is a general purpose semantic search
engine that search structured text like Wikipedia. Hakia
calls itself a meaning-based (semantic) search engine
[11]. Theyre trying to provide search results based on
meaning match, rather than by the popularity of search
terms. The presented news, Blogs, Credible, and galleries
are processed by hakia's proprietary core semantic
technology called QDEXing [10]. It can process any kind
of digital artefact by its SemanticRank technology using
third party API feeds [12]. A single query by the user
brings results from any repository including Web, News,
Blogs, Video, Images, Hakia Galleries and also from
Credible Sources [12]. For short queries the site displays
results in categories, instead of a standard list as shown
in current search engines. For longer queries, Hakia
highlights relevant phrases or sentences. The results are
somehow relevant and reliable but Hakia does not reveal
its inside technology. Hakia take the searched query and
find the results in many categories for example from
galleries, videos etc. so it took more time than the usual
search engines in the retrieval of results [10].
DeepDyve [13] is a powerful, professional research
engine that lets users access expert content from the
Deep Web the part of the internet that is not indexed
by traditional search engines. It indexes every word in a
document, but also computes the factorial combination of
words and phrases in the document and uses some
industrial strength statistical techniques to assess the
informational impact of these combinations [11]. The
presentation of search results is very complex. It presents
the users with many advanced options for refining,
sorting or saving the search. The search results are
however relatively easy to navigate. The results
presented are only for the paid customers, they are not
available for the general public.

III. MUSEUMS AND THE SEMANTIC WEB
A. Museum
A museum is a non-profit making, permanent
institution in the service of society and of its
development, and open to the public, which acquires,
conserves, researches, communicates and exhibits, for
purposes of study, education and enjoyment, material
evidence of people and their environment.
A special characteristic of cultural collection contents
is semantic richness. Collection Items have a history and
are related in many ways to our environment, to the
society, and to other collection items 4]. This semantic
network is not limited to a single collection, but spans
over other related collections in other museums. The
network of semantic associations can be extended to
contents of other types in other organization, as well.
Much of the semantic web content will be
published using semantic portals. Such portals
typically provide the end-user with two basic
services: a search engine based on the semantics of
the content and dynamic linking between pages
based on the semantic relations in the underlying
knowledge base. Semantic web technology enables
new possibilities when web searching museum
collections on the semantic web

B. Using semantic web technologies for E-
Museums
Museums possess large amounts of digitized data and
Meta data in their collection database. A special
characteristic of cultural collection contents is
semantically richness with an interconnected nature.
Collection items have a history and are related in many
ways to our environment, to the society, and to other
collection items. Semantic web Meta data and ontology
standards and tools, such as RDF and RDF schema can
be used for managing versatility of concepts and
association in a novel way and for making it explicit for
the museum visitors and researchers [6]. The web makes
it possible to consolidated data from distributed
heterogeneous collection and publishes it for anybody
who has an Internet access.
To realize these ideas, we have developed a
demonstrational prototype of a semantic web portal
called Annotation for Museum collections. This system
contains collection of cultural and artifacts. The project
has nine modules as following
Archaeology
Art
Anthropology
Numismatics
Botany
Geology
Zoology
Chemical conservation
Childrens Museums
The collections include many more sub contents. The
contents come from the collection of the National
Museum, Tanjore Museum, Chennai Museum, Madurai
museum etc...
C. Objectives
The main goals of developing the system are the
following

Global view to distributed collections
It is possible to use the heterogeneous distributed
collections of the museums participating in the system as
if the collections were in a single uniform repository.
Museums information retrieval
The system supports intelligent information retrieval
based on ontological concepts, not on simple keyword
string matching as is customary with current search
engines [9].
Semantically interoperability
A most interesting aspect of the collection items to
the end-user is the implicit semantic relations that relate
collection data with their context and to each other. In
museum collections, such associations are exposed
dynamically to the end-user by defining them in terms of
logical predicate rules that make use of the underlying
ontologies and collection metadata.
Easy local content publication
The portal should provide the museums with a
cost-effective publication channel. The success of the
www is the easy of content publication.

IV. IMPLEMENTATIONS
A. Ontology for Document type and Contents:
Ontology comprises the vocabulary. It is made up of
two parts namely, document and document type.
Document is created for Museum collections and
document type consists of contents for it. Ontology is
created for RDF. RDF has five fields namely, class,
subclass, properties, domain and range.
Class: Classes are abstract groups, sets, or collections
of objects. It may contain individuals, other classes, or a
combination of both.
Relationships (subclass): An important use of
attributes is to describe the relationships (also known as
relations) between objects in the ontology. Typically a
relation is an attribute whose value is another object in
the ontology.
Attributes: Objects in the ontology can be described
by assigning attributes to them. Each attribute has at least
a name and a value, and is used to store information that
is specific to the object it is attached to. In this document,
Museum collection is implemented. The user can view
the concepts from contents and also from the properties
of contents as a day-to-day process through this museum
collection. This property contains the Archaeology, Art,
Anthropology, Numismatics, Zoology, Geology,
Childrens museums, Chemical conservation, and
Botany. The user can view more, using the properties of
links in the form of step-by-step process through these
museum collections. The user can update any related
concepts at anytime using the properties of links.
B. Building Ontology
Ontologies are the core technology for the
Semantic Web. Ontology defines the terms used to
present a domain of knowledge that is shared by people,
databases, and applications. The Semantic Web thus
provides explicit meaning to the information and services
available on the Web for automated processing and
information integration based on the underlying
ontology. Ontology is an explicit and formal
specification of a conceptualization. The Ontologies are
useful for improving the accuracy of Web searches.Thus
search engines can look for pages that refer to a precise
concept in ontology. Web searches can exploit
generalization/ specialization information .Thus, if a
query fails to find any relevant documents, the search
engine may suggest to the user a more general query and
if too many answers are retrieved, the search engine may
suggest to the user some specializations. Museum
ontology is developed to share common understanding of
the structure of descriptive information among people,
among software agents and between people and software
and to enable reuse of domain knowledge which avoids
re-inventing the wheel and to introduce standards to
allow interoperability. The reasons for creating an
ontology is to make domain assumptions explicit i.e.,
easier to change domain assumptions and easier to
understand and update legacy data and to separate
domain knowledge from the operational knowledge.
In this paper use a number of standard markup
languages, such as RDF, to formally model information
represented in web resources. The Semantic Web is a
web of machine processable meanings underpinned by
shared and formally defined ontologies. In order to
coordinate different semantic web activities, museum
ontology is explicitly defined to share a contextual
conceptualization of the museum domain. This allows
the users to make their resources more machines
processable by collaboratively constructing an enriched
layer of the semantic web that links museum artifacts
with formal semantics to support other semantic
activities such as semantic query, aggregation and
information retrieval.

Fig. 1 OWL Classes
The Fig. 1 shows the ontology for museum services.
The Ontology is created by having the root node as
thing then follows the various categories like
Archaeology, Art, Anthropology, Numismatics, Botany,
Geology, Zoology, chemical conservation, and childrens
museum. Each sub category has many other nodes
related to it. It consisting OWL classes, relationships of
OWL classes and individuals for museum ontology. The
museum ontology is created using protg 3.4.7.
C. Ontology for Various Modules Associated With
Contents
The content was annotated using RDF and
reference to various ontologies. The dataset is metadata
of over 30000 collections of objects, e.g. Archaeology,
Art, Anthropology, Numismatics, Botany, Geology,
Zoology, chemical conservation, and childrens museum
in RDF triples format results as below.

Fig.1 Semantic web search engines

Fig.2 RDF Triples for museums collections

D. Graph of the Data Model
Here the data model for home page and
numismatics are created. The language used for creating
data model is XML and RDF.

Fig.3 Data Model
4.5 Framework for the semantic web museums
RDF can be used as a general framework for data
exchange on the web. The RDF data model distinguishes
between resources, which are object identifiers
represented by URIs, and literals, which are just strings.
The subject and the predicate of a statement are always
resources, while the object can be a resource or a literal.
In RDF results, resources are always drawn as ovals, and
literals are drawn as boxes.

Fig.4 Image Annotations
RDF was used in annotating historical cultural contents.

V. CONCLUSIONS
In this paper presented the design and implementation
of semantic web technique to improve web search in
museum collections. Semantic Web search engine was
created, Ontology and RDF were developed. This work
presented the notion of semantic original meta-data
model allowing ontology based named-entity, indexing,
and retrieval. Number of issues related to the
representation and the usage of the semantic annotation
were addressed. The fields discussed include the
semantic web, information extraction, e-museums,
ontology related topics, conceptual modeling, RDF
framework modeling and web services. The system may
also extend to support other domains like education,
tourism, and any information retrieval system

REFERENCES

[1]. Publishing and Using Ontologies as Mash-Up Services: Kim
Viljanen, Jouni Tuominen, and Eero Hyv onen Semantic
Computing Research Group (SeCo) Helsinki University of
Technology and University of Helsinki.
[2]. Introduction to Semantic Web Technologies Ivan Herman,
2010 Semantic Technology Conference in San Francisco,
CA, USA.
[3]. knowledge representation with ontology: C.brewster, the
present and future, intelligent systems , IEEE vol19(1)-2004
[4]. Personalized Context-aware recommendations in
SMARTMUSEUM: combining Semantics with
Statistics.Innar Liiv, Tanel Tammet, Tuukka Ruotsalo and
Alar Kuusik: Proceedings of the The Third International
Conference on Advances in Semantic Processing
(SEMAPRO 2009), IEEE Computer Society, October, 2009
[5]. MuseumFinland - Finnish Museums on the Semantic Web.
Eero Hyvnen, Eetu Mkel, Mirva Salminen, Arttu Valo,
Kim Viljanen,: Journal of Web Semantics, vol. 3, no. 2, pp.
25, 2005.
[6]. SMARTMUSEUM Knowledge Exchange platform for
Cross-European Cultural Content Integration and Mobile
Publication: tomi kauppinen et al-2009.
[7]. The dawn of semantic search. Computer, J. Hendler. Web
3.0: 43(1):7780, 2010.
[8]. Next generation Web search.
R. A. Baeza-Yates and P. Raghavan. In S. Ceri and M.
Brambilla, editors, Search Computing, LNCS 5950, pp. 11
23. Springer, 2010
[9]. Intelligent Semantic Web Search Engines: International
journal of Web & Semantic Technology (IJWesT) Vol.2,
No.1, January 2011 DOI: 10.5121/ijwest.2011.2103 34 A
Brief Survey G.Madhu1 and Dr.A.Govardhan2
Dr.T.V.Rajinikanth3
[10]. An Empirical Evaluation on Semantic Search
Performance of Keyword-Based and Semantic Search
Engines:D. Tmer, M. A. Shah, and Y. Bitirim, Google,
Yahoo, Msn and Hakia, 2009 4th International Conference
on Internet Monitoring and Protection (ICIMP 09) 2009.
[11]. "Top 5 Semantic Search
Engines".http://www.pandia.com/
[12]. GoWeb: a semantic search engine for the life science
web. BMC bioinformatics,H. Dietze and M. Schroeder, Vol.
10, No. Suppl 10, pp. S7, 2009.
[13]. "DeepDyve: The largest online rental service for
scientific, technical and medical research
http://www.deepdyve.com/
[14]. New way to search, navigate and use multimedia
museum collections over the web, museums and the web,
Canada-2005. Mathew addis, kirk martinez et al.
[15]. Global Ontology construction for heterogeneous digital
museums: hong-Zhe Liu,IEEE-2007

ISBN: 978-81-920575-7-6:: doi: 10. 74411 /ISBN_0768
Customer Relationship Management In Banks

DR.P.ANBUOLI
1
, T.R.THIRUVEN KAT RAJ
2

Faculty, Department of Management Studies, AssistantProfessor,
Anna University of Technology Madurai, Bharath Niketan Engineering College,
Madurai, Tamil Nadu Theni, Tamilnadu

Abstract - In todays scenario, companies hunt for
customers to please them in every respect and offer them the
best deal. Customer is everything, who is the new
realization of every profit-making company because they
are the revenue and profits. The liberalization and
globalization of Indian economy has taken place almost a
decade ago, the focus point in any service organization is
customer service. It is more so in the banking industry.
Customer service is the base for business expansion because
of the stiff competition prevalent in the banking industry.
The survival of banking business is dependent on customer
services.
Keyword:CRM,BANK,CUSTOMER

I. INTRODUCTION
Customer Relationship Management (CRM) in the
Indian banking system is fundamental to building a
customer-centric organization. CRM systems link
customer data into a single and logical customer
repository. CRM in banking is a key element that allows a
bank to develop its customer base and sales capacity. The
goal of CRM is to manage all aspects of customer
interactions in a manner that enables banks to maximize
profitability from every customer. Increasing competition,
deregulation, and the internet have all contributed to the
increase in customer power.
By entering into your premises, the customer is
giving an opportunity to serve him, but you are not doing
any favour by serving
In literature, many definitions were given to describe
CRM. The main difference among these definitions is
technological and relationship aspects of CRM. Some
authors from marketing background emphasize
technological side of CRM while the others consider IT
perspective of CRM. From marketing aspect, CRM is
defined by [Couldwell 1998] as a combination of
business process and technology that seeks to understand
a companys customers from the perspective of who they
are, what they do, and what they are like. Technological
definition of CRM was given as the market place of the
future is undergoing a technology-driven metamorphosis
[Peppers and Rogers 1995]. Consequently, IT and
marketing departments must work closely to implement
CRM efficiently. Meanwhile, implementation of CRM in
banking sector was considered by [Mihelis et al. 2001].
They focused on the revaluation of the critical satisfaction
dimensions and the determination of customer groups
with distinctive preferences and expectations in the
private bank sector. The methodological approach is
based on the principles of multi-criteria modeling and
preference disaggregation modeling used for data analysis
and interpretation. [Yli Renko et al. 2001] have focused
on the management of the exchange relationships and the
implications of such management for the performance and
development of technology-based firms and their
customers. Spesifically the customer relationships of new
technology-based firms has been studied. [Cook and
Hababou, 2001] was interested in total sales activities,
both volume-related and non-volume related. They also
developed a modification of the standard data envelope
analysis (DEA) structure using goal programming
concepts that yields both a sales and service measures.
[Beckett-Camarata et al. 1998] have noted that managing
relationships with their customers (especially with
employees, channel partners and strategic alliance
partners) was critical to the firms long-term success. It
was also emphasized that customer relationship
management based on social exchange and equity
significantly assists the firm in developing collaborative,
cooperative and profitable long-term relationships.
Importance of Customer Satisfaction
The importance of customer service for
capturing business in banks has become a focus point in
all stages of marketing of banking services, since almost
all banks are offering more or less the same products with
little changes in nomenclature. Here, the concept of
enhancing the customer satisfaction is a pivotal point to
attract more and more people to banking by a particular
bank. In the present day situation, customers becoming
more and more demanding, bankers need to evolve new
approaches, new and innovative products to keep pace
with growing expectations. With the opening up of the
economy, customer satisfactions has acquired new
meaning and different dimensions as banks are
discovering more and more innovative products to meet
increasing competition among themselves. In order to
service the customer more efficiently to the highest level
of satisfaction, technology up gradation is an important
aspect to speed up the servicing in all spheres of banking
and to fulfill the customer needs.

Technology
Technology is helpful in upgrading the services to
meet the competition, it will become a fallout if not
operate effectively and efficiently. Technology-driven
products such as debit cards, credit cards, tele-banking,
mobile phone banking, internet banking, fax, e-mail, help
desk services, etc. have given more access to customers to
communicate their needs instantly and transact their
business promptly
Customer Expectations
Customer service is required to be upgraded to satisfy
the customer expectations. To render efficient customer
service, the bankers have to first understand, what really
the customer wants. Therefore, the bank managements
have to continuously innovate and make strategies and
evolve new products to attract the customers. In order to
achieve this task, banks have to prepare projects, keeping
in view, the priorities of each segment of customers
classified into middle value customers and high net worth
customers. This will enable banks to innovate products to
customer needs
Communication
Interaction with customers is an important aspect in
providing better service to the customers to their fullest
satisfaction. Communication with customers at front
office level will help bankers to really identify the needs
and priorities of customers. Therefore, the role of Human
Resources Department of the bank is to provide efficient
staff at all levels, wherever there is an interaction with
customer either directly or indirectly. Proper
communication skills are required for the staff to deal
with the customers. It is one of the top priorities of banks
to provide good service.
Relationship Management
Relationship managers role in providing a better
Service to the customers is pivotal to the expansion of
banking business. This RMs maintains rapport,
particularly with high net worth customers and provides
them with all the services they need. This will help in not
only retaining the existing customer base, but also to
capture the new clientele. Relationships are more
important to provide good customer services.
Alternatively, good service will help the bankers to build
up relationships. Unless the service is provided to the
fullest satisfaction of the customers, taking further
references from the existing customers to establish new
relationships is difficult. Again this requires the service to
the existing customers and their patronage to bank.
Mapping the Customer Needs
In the present day banking, thought the banking
services are technology-driven and major role is played
by technology in all decisions on customer services, the
rested methods like customer surveys continue to help the
managements in identifying the customer needs.
Identifying the customer needs is important before
providing the service, so that the actual requirement of the
particular segment of customer is met with. This will help
managements to have an insight into customer needs and
attitudes. How to present the service
Banking is a service industry and bankers have to be
identified by the customer for whom the service is
extended. It is the starting point, where the banker comes
into contact with the customer. Therefore, banker should
have good attitude, appeal and good appearance. Bankers
have to cultivate a good relationship with the customer.
These factors have a great impact on customer service. In
such situations, customers will become loyal to the
services provided by banker and patronise the banker.
Banker should understand the customer needs and
identify the common areas of interest. Relationship
building is an integral part of banking business.
Customer Grievances
Receiving customer complaints, listening to the
grievances patiently and solving the problems is one of
the important areas of customer service. For this, a
separate mechanism is provided by banks, such as
customer grievance cells in individual banks and
ombudsman at each state. Customers are free to contact
the grievance cells in banks to make complaints, about the
service, problems etc. Banks will immediately attend to
such grievances and settle the issues.
Change of Organizational Structure for Better
Servicing
The financial sector reforms have brought about a
need for restructuring of banks in providing good
customer service. The number of innovative financial
products offered by banks is increasing steadily. The
public sector banks are still continuing with old patterns
in their structure. They are overstaffed selling different
schemes and all types of products. This structure is
hampering customer service. In these days of super
specialization, public sector banks have to segregate their
structure by classifying the banking as rural banking and
commercial banking. In such case, they can compete
effectively with new breed of private sector banks.

II. CONCLUSION
Customer service should be the main motto for
improvement of business. Technology implementation
should be speeded up, such as computerization, electronic
banking. The banking industry is vibrant and active in
India, due to technological revolution, Banks have to
utilize this opportunity to become more and more strong
organizations providing essential services.
REFERENCES
[1]. Alis, O.F., Karakurt, E., & Melli, P., 2000, Data Mining for
Database Marketing at Garanti Bank, Proceeding of the
International Seminar "Data Mining 2000", WIT
Publications.
[2]. Beckett-Camarata, E.J., Camarata, M.R., Barker, R.T., 1998,
Integrating Internal and External Customer Relationships
Through Relationship Management: A Strategic Response to
a Changing Global Environment, Journal of Business
Research, 41, 71-81.
[3]. Berry, M.J.A. & Linoff G.S., 2000, Mastering Data Mining:
The Art and Science of Customer Relationship Management,
John Wiley & Sons, Inc.
[4]. Cabena, P., Choi H.H., Kim I.S., Otsuka S., Reinschmidt J. &
Saarenvirta G., 1999, Intelligent Miner for Data Applications
Guide, IBM Redbooks, SG24-5252-00.
[5]. 5.Cook, W.D., & Hababou, M., 2001, Sales Performance
Measurement in Bank Branches, Omega, 29, 299 307.
[6]. Couldwell, C., 1998, A Data Day Battle, Computing, 21
May, 6466.
[7]. Hosking, J.R.M., Pednault, E. P. D. & Sudan, M., 1997, A
statistical perspective on data mining, Future Generation
Computer Systems, 13, 17-134.
[8]. Mihelis, G., Grigoroudis, E., Siskos, Y., Politis, Y., &
Malandrakis, Y., 2001,
[9]. Customer Satisfaction Measurement in the Private Bank
Sector, European Journal of Operational Research, 347-360.
[10]. Peppard, J., 2000, Customer Relationship Management
(CRM) in Financial Services,
[11]. European Management Journal, Vol. 18, No. 3, pp. 312327,
[12]. Peppers,D., & Rogers,M., 1995, A New Marketing Paradigm,
Planning Review, 23(2), 1418.
[13]. Ryals, L., & Knox, S., 2001, Cross-Functional Issues in the
Implementation of Relationship Marketing Through
Customer Relationship Management, European Management
Journal, Vol. 19, No. 5, pp. 534542.
[14]. Yli-Renko, H., Sapienza, H.J., Hay, M., 2001, The Role of
Contractual Governance Flexibility in Realizing the
Outcomes of Key Customer Relationships, Journal of
Business Venturing, 16, 529555.
[15]. 13.Yuan, S.T., & Chang, W.L., 2001, Mixed-Initiative
Synthesized Learning Approach
[16]. For Web-Based CRM, Expert Systems with
Applications,20,187-200.

ISBN: 978-81-920575-7-6:: doi: 10. 74418 /ISBN_0768
ANALYSIS OF FOREIGN DIRECT INVESTMENT IN SERVICE SECTOR
: A STUDY WITH REFERENCE TO INDIA

Dr. K. Maran,
Director - Sairam Institute of
Management Studies, Chennai
India

R. Anitha
Research Scholar, Bharathiar
University, Coimbatore India
R. Udhayasankar
Faculty, Management Studies,
Sairam Inst. of Tech. Chennai -
India

Abstract - During 1990, there was a great setback in
the Indian economy due to the economic crisis. The
Indian economy suffered from low per capita income,
chronic unemployment, low capital formation,
increased trade deficit, low standard of living, low
growth in GDP, low infrastructural development etc. In
view of the economic crisis and the shortage of capital,
the Government of India realized the importance of
foreign capital for the development of the country.
From the period 1991 onwards there has been
continuous effort taken by the Government in
attracting foreign capital and hence this period is
termed as the post liberalization period. The initiatives
taken by the Government include increasing the
investment cap, providing concessions in taxes and
offering tax holidays. India was then attracted to
foreign investors; as a result there was a massive
increase of foreign investment flow into the country.
The contribution of FDI inflow into the country was US
$ 129 million in 1991-92 to US $ 37,182 in 2009-10.
During 2006, the share of FDI in this sector was
increased drastically from 30.72 percent in 2005 to
61.94 percent in 2006. From 2006 onwards, an upward
trend of FDI in this sector was observed. There is a
significant change from manufacturing to service sector
from 2006 onwards.

Keywords: FDI, Economy, Service Sector, Estimated,
Pre and Post Liberalization etc.,

I. INTRODUCTION
The Indian economy has made remarkable
progress during the pre-liberalization period. The
agricultural sector during the period has progressed
through green, yellow, white and blue revolutions.
The industrial sector has grown annually by 6 to 7%
and the tertiary sector has contributed to the growth
of the Indian economy. During 1990, there was a
great setback in the Indian economy due to the
economic crisis. The Indian economy suffered from
low per capita income, chronic unemployment, low
capital formation, increased trade deficit, low
standard of living, low growth in GDP, low
infrastructural development etc. In view of the
economic crisis and the shortage of capital, the
Government of India realized the importance of
foreign capital for the development of the country.
From the period 1991 onwards there has been
continuous effort taken by the Government in
attracting foreign capital and hence this period is
termed as the post liberalization period. The
initiatives taken by the Government include
increasing the investment cap, providing concessions
in taxes and offering tax holidays.
For the first four decades after independence in
1947, the economic policies of the Indian
government were characterized by planning, control
and regulation. Until the 1980s, Indias development
strategy was focused on self-reliance and import
substitution. There were periodic attempts at market-
oriented reform, usually following balance of
payments pressures, which induced policy responses
that combined exchange rate depreciation and an
easing of restrictions on foreign capital inflows.
However, these controls were relatively narrow in
scope and had little impact on actual inflows, which
remained small. The situation changed dramatically
with the onset of reform programmes introduced in
the early 1990s in the aftermath of the balance of
payments crisis of 1991.
Broadly speaking, Indias approach towards
external capital flows can be divided into three main
phases. In the first phase, starting at the time of
independence and spanning up to the early 1980s,
Indias reliance on external flows was mainly
restricted to multilateral and bilateral concessional
finance. Subsequently, in the context of a widening
current account deficit during the 1980s, India
supplemented this traditional external source of
financing with recourse to external commercial loans,
including short-term borrowings and deposits from
non-resident Indians (NRIs). As a result, the
proportion of short-term debt in Indias total external
debt had increased significantly by the late 1980s.
The third phase was marked by the balance of
payments crisis of 1991 and the initiation of the
reform process.
The broad approach to reform in the external
sector was made in the Report of the High Level
Committee on Balance of Payments based on the
recommendations made by C. Rangarajan, Chairman,
Economic Advisory Council in 1991. The objectives
of reform in the external sector were conditioned by
the need to correct the deficiencies that had led to
payment imbalances in 1991. Recognizing an
inappropriate exchange rate regime, unsustainable
current account deficit and a rise in short-term debt in
relation to official reserves were amongst the key
contributing factors to the crisis. A series of reform
measures were put in place. They included a swift
transition to a market-determined exchange rate
regime, dismantling of trade restrictions, a move
towards current account convertibility and a gradual
opening-up of the capital account. While liberalizing
private capital inflows, the Committee recommended,
inter-alia: a compositional shift away from debt to
non-debt-creating flows; strict regulation of external
commercial borrowings, especially short-term debt;
discouragement of the volatile element of flows from
NRIs; and a gradual liberalization of outflows.
Service sector was the second attracted sector for
foreign investors during the period from 1991 to
2005. But from 2006 onwards, this sector became the
most preferred sector for foreign investors when it is
compared to manufacturing and primary sector.
61.94 percent in 2006. From 2006 onwards, an
upward trend of FDI in this sector was observed.
Table 4.8 and Chart 4.5 elaborates the fact that the
sectoral composition of FDI has undergone a
significant change from manufacturing to service
sector from 2006 onwards. The major shift in the
composition of FDI is due to the policy liberalization
and the changes in sectoral policies of FDI from time
to time. Earlier, many sectors which were not open to
FDI were thrown open and limits of many others
have been raised in the post liberalization period.
During 1991-2005, only 43 industries were opened to
foreign direct investment but from 2006 onwards,
due to further liberalization of FDI policies, the
number of industries has increased to 64.
Further, the phenomenal shift of FDI in India
from manufacturing to service sector has caused the
need to compare sectoral composition of FDI flow in
India with that of the developed and developing
countries in the world.
II. STATEMENT OF THE PROBLEM
During 1990, India was facing with a serious
economic and Balance of Payment crisis. It was the
time when Government felt the need for foreign
capital to supplement domestic capital to overcome
the crisis and to stimulate economic growth. In 1991,
the Government liberalized the economy by opening
the doors to foreign investors so as to bring India in
tune with the global economy. India was then
attracted to foreign investors; as a result there was a
massive increase of foreign investment flow into the
country. The foreign investment inclusive of both
FDI and FPI which was US$133 million in 1991-92
increased to US $ 69,557 million in 2009-10. The
contribution of FDI inflow into the country was US $
129 million in 1991-92 to US $ 37,182 in 2009-10.
61.94 percent in 2006. From 2006 onwards, an
upward trend of FDI in this sector was observed.
There is a significant change from manufacturing to
service sector from 2006 onwards.
OBJECTIVES OF THE STUDY
To analyze the forecasting of FDI
Inflows into India
To identify the trends and status of FDI
Inflows into Service Sector
III. REVIEW OF LITERATURE
Balasundaram Maniam and Amitiava Chatterjee
(1998) analysed on the determinants of US foreign
investment in India; tracing the growth of US FDI in
India and the changing attitude of the Indian
Government towards it as a part of the liberalization
program. They find that only the relatively weak
exchange rate appears to be a significant factor and
that the US FDI has been increasing in dollar
amounts and relative percentage growth. They call
for an improvement in infrastructure and reductions
in red tape and protectionism to encourage further
growth.
Nagesh Kumar (2000) has made an exploratory
attempt to examine the patterns of MNC related
mergers and acquisitions in India during the nineties
with the help of an exclusive data base. He finds that
the liberalization of policy framework since the early
nineties has led the MNCs to increasingly use the
Merger and Acquisition route to enter and strengthen
their presence in the country. In the recent years, two
fifths of all FDI inflows took the form of M&A s
compared to virtually all of FDI inflows coming from
Greenfield ventures earlier. The deals relating to
MNCs are predominantly horizontal rather than
vertical in nature. In terms of development
implications, he finds that FDI inflows in the form of
M&A s are of an inferior quality compared to
Greenfield investments. These findings, therefore,
emphasize the need for adopting a comprehensive
competition policy framework in India.
Constantina Kottaridi and Thanasis Stengos
(2010) make a contribution to the existing literature
on the foreign direct investment and economic
growth nexus by contrasting past empirical evidence
and conventional wisdom and arriving at some
interesting new results. By applying non-parametric
methods and taking into account non-linear effects of
initial income and human capital on economic
growth. They explore the FDI effect on growth in
much greater detail than previous studies. Their
findings not only confirm the non-linear effect of
human capital in the presence of FDI inflows but also
suggest that FDI inflows are growth enhancing in the
middle-income countries while there is a two-
regime FDI effect for high-income countries. This
new finding appears to be independent of OECD
country membership.
Elizabeth Asiedu and Donald Lien (2011), in
their empirical studies examine the impact of
democracy on foreign direct investment assume that
the relationship between democracy and FDI is the
same for resource exporting and non-resource
exporting countries. This paper examines whether
natural resources in host countries alter this
relationship. We estimate a linear dynamic panel-data
model using data from 112 developing countries over
the period 19822007. We find that democracy
promotes FDI if and only if the value of the share of
minerals and oil in total exports is less than some
critical value. We identify 90 countries where an
expansion of democracy may enhance FDI and 22
countries where an increase in democratization may
reduce FDI. We also find that the effect of
democracy on FDI depends on the size and not the
type of natural resources.
IV. METHODOLOGY
Research Design
The design adopted for the study is Descriptive
Research Design, as the study is concerned with
describing the characteristics and specific predictions
about Foreign Direct Investments in India during a
particular period.
V. PERIOD OF STUDY
The main purpose of the study is to analyze the
magnitude, causes and effects of Foreign Direct
Investment inflow into the country. Hence FDI
inflows into the country is analyzed for a period of 30
years from 1980-81 to 2009-10. For better analysis,
the period is classified into two phases: the first phase
is the period of Pre-Liberalization i.e., period from
1980-81 to 1990-91 and the second phase is the Post
Liberalization period from 1991-92 to 2009-10. The
rest of the analysis, that is, the country wise source,
sectoral analysis, relationship between FDI and FPI,
factors which determine FDI flow and its impact on
the economy are analyzed for a period of 19 years
from 1991-92 to 2009-10.
The period under study is important for a variety
of reasons. First of all, it was during July 1991 that
India opened its doors to private sector and to foreign
investors by liberalizing and globalizing the
economy. Secondly, the experiences of South-East
Asian countries by liberalizing their economies in
1980s became the stars of economic growth and
development in early 1990s. Thirdly, Indias
experience with its first generation economic reforms
and the countrys economic growth performance was
considered as safe havens for FDI which led to
second generation of economic reforms in India in
first decade of this century. Fourthly, there is a
considerable change in the attitude of both the
developing and developed countries towards FDI.
They both consider FDI as the most suitable form of
external finance. Fifthly, there is an increase in
competition for FDI inflows particularly among the
developing nations.
VI. SOURCES OF DATA
The study is based on secondary data which is
published sources of data collected from various
sources. The data was extracted from the following
sources; Handbook of Statistics on the Indian
economy, RBI, various issues, UNCTAD, WIR
series, various issues, Economic Survey, Government
of India, various issues, World Development
Indicators, World Bank, Department of Industrial
Policy and Promotion (DIPP), Secretariat of
Industrial Assistance (SIA), Central Statistical
Organization (CSO).
HYPOTHESIS
The hypotheses for the study are to test
whether FDI inflow shows a positive trend over the
study period.
VII. RESULT AND DISCUSSIONS
Trends in FDI Inflows during 2010-11 to 2014-
15
After analyzing the volume of FDI Inflows
in India, during the Pre and Post Liberalization period
the trend in FDI inflows for next five years i.e., from
2010-11 to 2014-15 is being projected. The
projection of FDI inflow to the country is estimated
by using Auto Regression Moving Average
(ARIMA) model. The results of the model are
summarized in table 1.1 and the estimated FDI
inflows from 2010-11 to 2014-15 is tabulated in table
1.2.
Table 1.1: Result of ARIMA model of FDI Inflows
Mod
el
Coeff
icient
Std.
Error of
Coeffici
ent
t
value
p
value
AR1 1.449 0.18
9
7.
649
0.00
0**
AR2 -.652 0.21
8
2.
988
0.00
9**
Year 8409.
651
251
7.263
3.
340
0.00
4**
Con - 503 3. 0.00
stant 1678024
8.439
7067.20 331 4**

** denotes significance at 1% level
Log Likelihood : -
206.973
Alkalines Information Criterion (AIC):
421.946
Schwarzs Bayesian Criterion (SBC) :
425.724
From the above result, the p-value of the constant,
first and second lag is very less. This indicates that
the fit is good. Further, the Alkalines Information
Criterion (AIC) and Schwarzs Bayesian Criterion
(SBC) both measure goodness of fit and account for
model complexity. The ARIMA model equation is
fitted as ,(FDI)
t
= -16780248.439 + 1.449 (FDI)
t-1
-
0.652 (FDI)
t-2
+
t

Where, (FDI)
t
= first order difference in FDI =
(FDI)
t
- (FDI)
t-1
Table 1.2: Estimated FDI Inflows in India during 2010-11
to 2014-15
(Amount Rs. In Cr.)
Ye
ar
Esti
mated
Amount
% of
Growth
in FDI
over
Previous
year
95% Confidence
Limit
Low
er CL
Uppe
r CL
201
0-11
178,
110 1
146,
101
210,1
20
201
1-12
172,
800 -3
111,
528
234,0
71
201
2-13
165,
622 -4
79,2
79
251,9
66
201
3-14
160,
380 -3
55,1
66
265,5
95
201
4-15
159,
161 -1
41,1
80
277,1
42

Fig. 1.1 : Projection of FDI Inflows in India from 2010-11
to 2014-15

Table 1.2 shows a marginal increase of one
percent in FDI inflow as when compared to the
period 2009-10. Then from the period 2011-12 to
2014-15 a downtrend of FDI inflow is observed. FDI
inflow into India is estimated to be Rs.172800 crore
in 2011-12 as against Rs.178110 crore in 2010-11,
that is, the growth trend of FDI inflow into the
country shows a decline of 3 percent in 2011-12 as
when compared to 2010-11. This fall in FDI inflow
into the country is expected to further decline during
2012-13, that is, a 4 per cent decline will be
anticipated as against the previous year. During
2013-14 and 2014-15, the fall in growth rate of FDI
inflow is found to be reduced to 3 percent and one
per cent as against the previous years. The projection
of FDI inflow in India for the next five year period
from 2010-11 to 2014-15, shows that FDI into the
country will have a downward trend from 2011-12
onwards. Therefore, Government should take suitable
steps to overcome this situation.
Trends in Service Sector
First of all we observe that the FDI inflow in
services has been phenomenal in the last two
decades. Since the onset of liberalization, the country
received a huge amount of FDI in service sector
because of the tremendous growth potential that it
possesses. This sector has ranked among the top ten
sectors attracting FDI since 1991. Services which
comprise of financial and non financial services,
banking services and insurance, outsourcing and
research and development services recorded third
place during 1991-99 and 2000-05 by attracting 9.71
and 9.61 percent of total FDI inflows during the two
periods. However, it moved to rank one by adding
22.96 percent share to the cumulative FDI for the
period from 2006 to 2010.
Among the sub sectors of services sectors,
financial services attract 9.13 percent of total FDI
inflows. It is followed by banking services 2.11
percent, insurance 1.52 percent, non-financial
services 2.09 percent, outsourcing 0.72 percent,
research and development 2.33 percent, courier 0.32
percent and other services 3.38 percent respectively.
Mauritius was placed the top of the chart by investing
43.09 percent in service sector followed by Singapore
13.88 percent, UK by 11.21 percent and USA by 8.52
percent. FDI inflows in service sector are heavily
concentrated around two major cities in India that is,
Mumbai by 62.90 percent and Delhi 12.48 percent.
Further using Auto Regression Moving Average
(ARIMA) model FDI inflow in Service Sector is
projected for next five years. The results of the model
are summarized below -


Table 1.3: Result of ARIMA model of FDI Inflows to Service
Sector
A.
M
odel
Coeffic
ient
Std.
Error of
Coefficie
nt
t
valu
e
p
value
AR1
0.303

0.320
0
.946
0.372
Yea
r

28751.483

8984.229
3
.200
0.013
*
Con
stant
575165
37.647
18004
421.558
3
.195
0.013
*
** denotes significance at 1% level
Log Likelihood : -
137.405
Alkalines Information Criterion (AIC) :
280.810
Schwarzs Bayesian Criterion (SBC) :
282.004
The above result shows that the p-value of the
constant and first lag is very less. It indicates that the
test is fit. Further, the Alkalines nformation Criterion
(AIC) andSchwarzs Bayesian Criterion (SBC) both
measure oodness of fit and account for model
complexity. The ARIMA model equation is fitted as,
(FDI)
t
= -57516537.647 + 0.303 (FDI)
t-1
+
t

where, (FDI)
t
= first order difference in FDI =
(FDI)
t
- (FDI)
t-1

B. Table 1.4 : Actual and estimated values of
FDI Inflows to Service Sector
(Amount Rs. in Million)
C
ear
Act
ual
Values
Est
imated
Values
95%
Confidence
Limit
%
of
Gro
wth
in
FDI
over
Previ
ous
Year
D.
owe
r CL
U
pper
CL
199
1-99
40,
376.82
-
42,321
.49
-
260,020
.14
1
75,3
77.1
6 -
200
0
1,8
61.50
11,
501.48
-
172,435
.83
1
95,4
38.7
9 -
200
1
8,2
02.24
19,
859.82
-
160,571
.20
2
00,2
90.8
3
3
41
200
2
15,
431.39
41,
817.08
-
136,212
.71
2
19,8
46.8
7
8
8
200
3
13,
903.59
64,
043.68
-
112,735
.00
2
40,8
22.3
6
-
10
200
4
11,
455.83
83,
615.45
-
93,086.
66
2
60,3
17.5
6
-
18
200
5
31,
455.14
102
,908.3
2
-
74,893.
27
2
80,7
09.9
1
1
75
200 178 129 - 3 4
6 ,584.7
4
,006.4
3
51,049.
17
09,0
62.0
2
68
200
7 145
,099.5
2
193
,646.3
7
10,2
24.82
3
77,0
67.9
2
-
19
200
8 339
,475.1
2
203
,529.6
8
15,6
89.98
3
91,3
69.3
8
1
34
200
9 279
,182.5
4
282
,493.1
0
89,2
55.22
4
75,7
30.9
8
-
18
201
0 284
,249.2
7
84,7
12.70
4
83,7
85.8
4 2
201
1

305
,820.2
9
73,6
98.55
5
37,9
42.0
3 8
201
2

332
,394.8
9
81,6
25.91
5
83,1
63.8
7 9
201
3

360
,486.4
1
93,9
95.60
6
26,9
77.2
2 8
201
4

389
,037.8
2
107,
171.27
6
70,9
04.3
6 8
201
5

417
,728.6
4
120,
149.10
7
15,3
08.1
8 7
Note: Actual values compiled from SIA, Newsletter,2009.

Fig. 1.2: Trends in Service Sector

Table 1.4 and Fig.1.2 shows the trend in FDI
Inflow in Service Sector and the projection of inflow
into this sector is estimated for a period of six years
from 2010 to 2015 using ARIMA model. It is
observed that during this period FDI in this sector
shows an upward trend. The share of FDI in this
sector was found to have a 2 percent increase in
2010. Later during 2011 and 2012, foreign
investments are expected to have a further increase of
8 percent and 9 percent. But from 2013 onwards, the
share of foreign investment in this sector is expected
to gradually reduce to 8 percent in 2013 and 2014
and to 7 percent in 2015.
VIII. CONCLUSION
After liberalizing the economy in 1991, Foreign
Direct Investment played a major role in the growth
of the country. The findings of the study reveal that
there has been increase in sectoral growth of the
economy, increase in the level of Personal
Disposable Income, Exports, Imports and rise in the
level of employment. A rise in all the above factors
leads to an overall growth of the country creating a
positive impact on the economy. Although there has
been increased flow of FDI into the country during
the post liberalization period, it is found that the
global share of FDI in India is very less when it is
compared to other developing countries. Further, it
was found that the realization of approved FDI into
actual is very less over the years. Therefore, to
overcome this situation, the Government of India
should take serious steps to simplify the procedures
and remove red-tapism so that actual FDI can be
raised. Government should also revise the sectoral
cap and bring more sectors under the automatic route.
Further, agreement of Double Taxation treaties
should be signed with other countries in order to
increase bilateral trade between the countries.
Therefore, there is an urgent need to adopt innovative
policies and good corporate governance practices on
par with international standards, by the Government
of India, to attract more and more foreign capital in
service sector to enhance its performance and also to
contribute more in Indian economy.
REFERENCES
[1]. Aditya KR Bajaj and Swastik Nigam Globalization in
the Indian Pharmaceutical Industry Foreign Direct Investment
spillovers and implications on Domestic Productivity: 1991-2007,
is a research project done under IIM Ahmedabad, 2007.
[2]. Annual Survey of Industries, Ministry of Statisitics and
Programme Implementation, Government of India, various issues
[3]. Avik Chakrabarti (2003), A Theory of the spatial
distribution of Foreign Direct Investment, International Review of
Economics & Finance, Volume 12, Issue 2, 2003, pp. 149-169.
[4]. Balasubramanyam.V.N and David Spasford, Does
India need a lot more FDI? Economic and Politically Weekly,
April 28, 2007.
[5]. Balasundaram Maniam and Amitiava Chatterjee. The
Determinants of US Foreign Investments in India: Implications and
Policy Issues, Managerial Finance, Vol. 24, No. 7, 1998, pp. 55-
62.
[6]. Constantina Kottaridi, Thanasis Stengos (2010),
Foreign Direct Investment, Human Capital and non-linearities in
Economic Growth, Journal of Macroeconomics, Vol. 32, No. 3,
September 2010, pp. 858-871.
[7]. Cooper Donald R. and Schindler Pamela S. (2000),
Business Research Methods, Tata McGraw Hill Publishing
Company Ltd., New Delhi.
[8]. Dharmendra Dhakal, Raja Nag, Gyan Pradhan, Kamal
P.Upadhyaya, Exchange Rate Volatility and Foreign Direct
Investment: Evidence from East Asian Countries, International
Business & Economic Research Journal, Vol.9, No.7, July 2010.
[9]. Elizabeth Asiedu, Donald Lien (2011), Democracy
Foreign Direct Investment and Natural Resources, Journal of
International Economics, Vol. 84, No. 1, May 2011, pp. 99-111.
[10]. Ganesh.S, Who is afraid of Foreign firms? Current
Trends of Foreign Direct Investment in India, Economic and
Political Weekly, Vol. 32, No.22, May 31-June 6, 1997, pp. 1265-
1274.
[11]. Jayachandran.G and A.Seilan, A Casual relationship
between Trade, Foreign Direct Investment and Economic Growth
for India, International Research Journal of Finance and
Economics, Issue 42, 2010.
[12]. Jayashree Bose, Foreign Direct Investment Inflows in
India and China A Sectoral Experiences, ICFAI University
Press, Hyderabad, 2007.
[13]. Johnson Andreas (2004), The Effects of FDI Inflows
on Host Country EconomicGrowth,
http://www.infra.kth.se\cesis\research\publications\ working
[14]. Nagesh Kumar.N, Mergers and Acquisitions by MNCs
Patterns and Implications, Economic and Political Weekly,
August 5, 2000.
[15]. Sufian Eltayeb Mohamed and Moise G.Sidiropoulos,
Another Look at the Determinants of Foreign Direct Investment
in MENA Countries: An Empirical Investigation, Journal Of
Economic Development, Volume 35, No.2, June 2010.


ICCA Volume 4

Hochgeladen von

Dokumentinformationen

Copyright

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

ICCA Volume 4

Hochgeladen von

Copyright:

I nt er nat i onal Conf er ence on Comput er Appl i cat i ons 2012

Ubiquitous, embedded computing, e.g., in domestic

skin color and splines for contour descriptions. In

Fig. 2. Screenshot of the Camera

Figure 3 shows an interaction between

Gives complete spanning

Function 4 F forwards the query message to one of its

Das könnte Ihnen auch gefallen