Ncaessn Part 4

National Conference on Advancements in Embedded Systems and Sensor Networks (AESSN - 2015)
HARDWARE AND SOFTWARE CO-DESIGN ISSUES IN H.264

DECODER
T.Chandrasekhar , R.Ravi Kumar
M.Tech ,Kluniversity
Associate Professor
Department of electronics and computers engineering
Department of electronics and communication engineering
chandrasekhartupakula@yahoo.com , rayalaravi2013@gmail.com
Abstract- In this paper we talk about mixed architectures of a H.264/AVC video decoder. Here software part of decoder was
implemented in NIOS II processor on a FPGA prototyping board (Stratix III). Software and hardware architectures was
proposed to increase the decoder output performance. Based upon the time execution parameters, data dependency
constraints, the decoder partitioning is applied. Here the inverse 4x4 Intra process is implemented with hardware accelerator.
It consists of inverse 4x4 Intra prediction, inverse transformation and inverse quantization. By implementing inter prediction
as a hardware module the decoder through can be increased
Keywords
H.264 decoder, software-hardware architectures , inverse 4x4 intra prediction, inverse transform, inverse quantization ,inter
prediction
INTRODUCTION H.264 DECODING PROCESS
H.264/ AVC codec is a multimedia application produces high-

quality video contents with low binary rate. In H.264 encoder
video quality performances is obtained by video coding tools
and techniques. Many algorithms were proposed for
improving in term of quality and bitrates. For providing a
perfect high quality image H.264/AVC should process the
video in reduced time. The design realizes a multimedia
system on the chip can be partitioned into three types pure
software designing, pure hardware designing and mixed
software and hardware implementation. Software architectures
implementation in [2], are flexible and of high design level. It
is mainly based on processor like RISC based CPU or a
Digital Signal Processor. However, the hardware
implementation, [3] and [4], is high efficient because it
Fig 1: H.264 decoder block diagram
provides multiple task executions in parallel. But design time After decoding the Network Abstraction Layer parameters, the
is more. With respect to software design, it is less flexible. data is entropy decoded by two ways: Context-based
Examples of hardware prototyping boards are FPGA and Adaptive Variable Length Decoding and Exp-Golomb.
ASIC. Due to the software and hardware implementation [5-6- CAVLD produces quantized coefficient array but it is more
7], fast development and high performance throughput can be time consuming. Exp-Golomb is used for others syntax
achieved. elements It is used to reconstruct and reorder the data Inverse
In this paper combined architecture is proposed to implement quantization produces a set of coefficients .inverse
H.264 decoder on FPGA. NIOS II is the processor. The quantization reconstructs data. On input according to Qp
inverse 4x4 intra prediction and inter prediction is hardware value. Qp is the quantization parameter fixed by the encoder.
accelerator. In previous video coding standard, Inverse Discrete Cosine
Transform was used. H.264/AVC use a separable Inverse
Integer Cosine Transform. It has 4x4 inverse Discrete Cosine
Transform. Inverse transform is applied on each 4x4 block.
DC coefficients of each 4x4 block are assembled in a matrix
and applied inverse Hadamard transform .The 4x4 intra
prediction modes are suitable for significant information
ISSN: 2348 - 8387 www.internationaljournalssrg.org Page 283

within a frame. Each 4x4 block is predicted independent from Inverse quantization process inputs include CAVLC outputs
spatially neighboring coefficients. One of nine prediction coefficients of 16 blocks. First coefficients are organized in a
modes are applied to each Macro block. In inter prediction 16 coefficients buffer. Then they are sending to inverse
motion vector is extracted from bit stream. Then, motion quantization component with a control signal. After 16
compensation module is applied. It consists of adding motion transfers of 16 coefficients each time, all inputs coefficients of
vector coordinates to a corresponding block in reference inverse quantization are ready to start execution. For IP
frame. Result is a reconstructed block. Block size can changes output, a buffer is used to put coefficients by set of 16. This
from one motion vector to other. The de-blocking filter
Memory model is used for input and output to minimize

Perform an in-loop filtering to reduce blocking artifacts memory size. Instead of putting all 256 coefficients in
generated by image partitioning and quantization. After memory wait bus transfer, only 16 coefficients are in wait
inverse quantization and inverse transform, the de-blocking state. In addition to this option doesnt affect execution time or
filter compare the edge values of each 4x4 block with adjacent transfer time.
block to select the level of filtering.
SOFTWARE IMPLEMENTATION
Software implementation is necessary before going to the

mixed implementation to evaluate time executions of each
modules in decoder. This profiling can give an idea about
Module complexity. De-blocking filter consumes 30% of

overall decoder time execution. Then inverse quantization and
inverse transform have second place by 21%. The aim of our
research is to define software and hardware solution of H.264
decoder. All decoder modules will be designed as hardware IP
and for only control operations will be done on the software
part of architecture. To attain this purpose, de-blocking filter
was first implemented as hardware IP in previous work [18]
using ESL tools. The second IP is subject of this proposed
work. In order to have a better partitioning between software
and hardware, most demanding modules in term of time Fig 2: Intra prediction hardware architecture
execution need to be transformed on hardware accelerator.
Despite importance of time execution, it is not only constraint Motion Estimation finds the best matching candidate block
to be used to define block that need to be accelerated. Data between a current macroblock and its reference frames in a
dependency and possibility of parallelization in algorithm are search window. It can efficiently reduce the temporal
also a major factors. In fact, inverse quantization, inverse redundancies. With motion estimation, we can promote the bit
transform and inverse prediction presents a good tradeoff rate effectively by transferring the motion vectors of a MB.
between the time execution and algorithmic constraints Thus, motion estimation is one of the most important
functions in many video coding standards. In H.264/AVC,
The standard supports multiple reference frames in order to
find the objects which appear suddenly. Besides, it also
HARDWARE ARCHITECTURE OF INVERSE 4X4
INTRA PROCESS AND INTER PREDICTION supports variable block sizes coding in matching blocks.
Although these two functions indeed increase the coding gain,
both of them will greatly increase computational complexity
Inverse 4x4 intra prediction, inverse quantization, inverse with the comparison of previous standards which only support
transform and inter prediction are decoder modules. Enclosing one or two reference frames and only one fixed
them on one IP is a way to minimize data transfer between MB size.
hardware and software architecture part. Architecture of The selected frames for prediction are indicated by a reference
inverse 4x4 Intra prediction and Inter prediction is given by index which is associated with the frame index in the buffer.
Figure 7. Accelerator input are inverse quantization inputs, The process of encoding which is based on the values
prediction modes and macroblock position. Output is only predicted to form the best matching candidate blocks is called
reconstruct macroblock. To calculate the equations of Motion compensation. When the best matching candidate
prediction mode, the inverse 4x4 intra prediction module blocks is found, it will be used to form a reference frame to be
needs 16 prediction modes and the macroblock position within pasted on the current frame. Then, a vector called motion
a frame (MBX, MBY). vector can be drawn from the reference frame. If an encoder
wants to transfer a block data in Frame t, there are two ways to

get it. Let the encoder transfer all pixels of this block to CONCLUSION
decoder. Since there is not much difference between Frame t
and Frame t-1, and the data of Frame t-1 had already In this paper, mixed architecture of H.264 decoder was
reconstructed by a decoder and is stored in the buffers, we can developed. Hardware accelerator is inter chain and 4x4 intra
take Frame t-1 as a reference and exploit the data of Frame t-1 chain. It contains a inter prediction, 4x4 intra prediction,
in the decoder buffer. Thus, we just need some information, inverse quantization and inverse transform. Hardware design
is optimized to overcome tradeoff between video time
constrain and hardware on chip area. As results, accelerator
the motion vector, to construct the Frame t there are many speed is greater than 20% of software speed
criteria used to judge the best matching block. One of the most
Widely used criteria is sum of absolute differences (SAD). The REFERENCES
function of SAD is
[1] Wiegand (Ed.), T, Draft ITU-T Recommendation H.264/AVC
and Draft ISO/IEC 14496-10 AVC, Joint Video Team of ISO/IEC
JTC1/SC29/WG11 &ITU-T SG16/Q.6 Doc. JVT-G050, Mar.
Where (m, n) is the distance of the candidate block [2] Werda. I, Dammak. T, Grandpierre. T, Ben Ayed MA, Masmoudi
corresponding to coding block C at position (i , j), and N is the N, Real-time H.264/AVC baseline decoder implementation on
macroblock size. H.264/AVC defines the maximum of (m, n) TMS320C6416, Springer-Verlag 2010, J Real-Time Image
Processing.
as search range for different levels. The motion vector MV is
defined as: [3] Warsaw .T, Lukowiak. M, Architecture design of an H.264/AVC
decoder for real-time FPGA implementation, Application-specific
Systems, Architectures and Processors (ASAP'06), 2006 IEEE.
[4] Fan C.-P. and Cheng Y.-L., FPGA implementations of low
latency and high throughput 4x4 block texture coding processor for
H.264/AVC,Journal of the Chinese Institute of Engineers, vol. 32,
no. 1, pp. 3344, 2009.
[5] Kun Y, Chun Z, Guoze DU, Jiangxiang XIE, Zhihua W, A
Hardware-Software Co-design for H.264/AVG Decoder, Solid-State
Circuits Conference, 2006. (ASSCC). IEEE Asian pages : 119-122.
[6] WANG S-H, PENG W-H, A Software-Hardware Co-
Implementation of MPEG-4 Advanced Video Coding (AVC)
Decoder with Block Level Pipelining, vol. 41, no. 1, pp. 93-110,
2005.
[7] Ben Atitallah A., Kadionik P., Masmoudi N., Lvi H. Design
FPGA implementation of a HW/SW Platform for Multimedia
Embedded Systems, Automation for Embedded Systems Journal,
Pages 1-19, Springer 2008.
[8] Babionitakis K., Doumenis G., Georgakarakos G., Lentaris G.,
Nakos K., Reisis D., Sifnaios I., Vlassopoulos N., A real-time
Fig 3 : Inter prediction hardware architecture.
H.264/AVC VLSI encoder architecture, Journal of Real-time Image
Processing , vol. 3, no. 1-2, pp. 43-59, 2008.
PROPOSED ARCHITECTURE
[9] Tran X-T., Tran V-H, An Efficient Architecture of Forward
Transforms and Quantization for H.264/AVC Codecs, REV Journal
on Electronics and Communications, Vol. 1, No. 2, April June,
2011.
[10] Kthiri M., Kadionik P., Le Gal B., Lvi H., Ben Atitallah A.,
Performances analysis and evaluation of Xenomai with a
H.264/AVC decoder, ICM 2011.
[11] Damak T., Werda I., Samet A., Masmoudi N., DSP CAVLC
implementation and Optimization for H.264/AVC baseline encoder,
15th IEEE International Conference on Electronics, Circuits, and
Systems, ICECS 2008, MALTA.
[12] Richardson, I.E.G ,H.264/AVC and MPEG4 video
Intra and inter compression. Video Coding for Next Generatiopn Multimedia,
prediction processor
Wiley editor, 2003.

[13] Damak T., Werda I., Ben Ayad M-A, Masmoudi N, An

Efficient Zero Length Prefix Algorithm for H.264 CAVLC Decoder
on TMS320C64, 2010 International Conference on Design &
Technology of Integrated Systems in Nanoscale Era (DTIS).
[14] Malavar, H. Hallapuro, A. Karczewicz, M. Kerofsky, L. Low
complexity transform and quantization in h.264/AVC, IEEE
Transactions on circuit and system for video technology, Vol. 13, No.
7, pp. 598-603, 2003.
[15] Soon-kak Kwon, A. Tamhankar, K.R. Rao, Overview of
H.264/AVC / MPEG-4 Part 10, Journal of Visual Communication
and Image Representation, Vol. 17, No 2 , pp 186216, Apr. 2006.
[16] Kessentini A., Kaaniche B., Werda I., Samet A., Masmoudi N.,
Low complexity intra 16x16 prediction for H.264/AVC
International Conference on Embedded Systems & Critical
Applications ICESCA,2008, Tunisia.
[17] Werda I., Chaouch H, Samet A, Ben Ayed M-A, Masmoudi N.,
Optimal DSP Based Integer Motion Estimation Implementation for
H.264/AVC Baseline Encoder, The International Arab Journal of
Information Technology, Vol. 7, No. 1, January 2010.
[18] Damak T., Werda I., Masmoudi N., Bilavarn S., Fast
prototyping H.264 deblocking filter using ESL Tools, 2011 8th
International Multi-Conference on Systems, Signals & Devices,SSD.
[19] Stratix III device, http://www.altera.com/
[20] Artisan Components. TSMC 0.18m 1.8-Volt SAGE-XTM
Standard Cell Library Databook, 2001.
[21] Lindroth T, Avessta N, Teuhola J, Seceleanu T, Complexity
Analysis of H.264 Decoder for FPGA Design, ICME 2006.

A NOVEL SYSTEM TO PROVIDE CAB SAFETY FOR WOMEN
AND IMPLEMENTING WITH INTERNET OF THINGS
KONA GOWTHAMI K.SRIPATH ROY

Konagowthami93@gmail.com koganti_sripathroy@kluniversity.in
M.Tech student Asst Professor
K.L.UNIVERSITY K.L.UNIVERSITY
ABSTRACT: Constructing an intelligent monitoring device is embedded with our

traffic monitoring system which depends on application which senses the heart beat of a
automatic identification of vehicles. Presently, person driving the vehicle and if there are any
automatic identification technology abnormalities in the heart beat, then our
supported image and vehicle tracking falls application performs a dual role. One of the
within the lure due to its low recognition rate. applications uses a GPS to track the location
Thus it is necessary to use new technologies information of the user and send this location
supporting Internet of Things give a new information via SMS, email simultaneously, an
approach for it. During this paper we tend to emergency signal is sent to Microcontroller. The
explore this issue and propose a possible in-vehicle device design works using Global
theme. Firstly RFID technology is used. Positioning System (GPS) and Global system for
Secondly, we tend to obtain positioning mobile communication / General Packet Radio
information of vehicles by victimization GPS Service (GSM/GPRS) technology. The device is
technology. Thirdly, as GPRS provides high- embedded inside a vehicle whose position is
speed wireless IP services for mobile users, determined and tracked in real-time. A
totally supporting the TCP/IP, we choose microcontroller is useful here to control the GPS
wireless GPRS theme to transmit knowledge and GSM/GPRS modules. The vehicle tracking
of mobile objects. The automatic detection system uses the GPS module to get geographic
and transmission of knowledge provided a coordinates at regular time intervals. The GPRS
basic guarantee for constructing an intelligent module is used to transmit and update the
vehicle tracking system. vehicle location to a database. A smart phone
application is also developed for continuously
Index terms--intelligent vehicle tracking, monitoring the vehicle location. To provide
Internet of Things, RFID, GSM/GPRS, GPS security for the information provided in internet
we use an RFID-based electronic identity
security platform of the internet of things to
I. INTRODUCTION realize the efficient security administration of
personal network identity.
Intelligent vehicle tracking system plays an
important role especially for the safety of II. MATERIALS AND METHODS
women. An efficient vehicle tracking system is
2.1 INTERNET OF THINGS:
designed and implemented for tracking the
movement of any equipped vehicle from any Internet of Things (IOT) semantically suggests
location at any time. This application is targeted that a worldwide network of interconnected
to a person who in emergency. Heart rate objects unambiguously addressable, supporting

common place communication protocols, network formation of knowledge transport layer

which may be a paradigm that is quickly gaining contains wired network and wireless network, its
ground in the state of affairs of contemporary supporting technology mainly includes GPS,
wireless telecommunications. The basic plan of GPRS, the net so on. The top layer of the net of
this idea is that the pervasive presence around us Things is processing and data exchange layer,
of a range of things or objects like (RFID) whose task is to complete information exchange
tags, sensors, actuators, mobile vehicles, etc. and processing, information calculation,
which, through unique addressing schemes, area information storage and alternative functions.
unit ready to act with every other and join forces Actualization of the IoT thought into the real
with their neighbours to achieve common goals. time is possible through the integration of many
Potentialities offered by the IoT change the useful technologies, like EPC, RFID, GPS,
development of an enormous variety of GPRS, Internet, WSN etc. the subsequent is that
applications, of which can be sorted into the the more introductions for the principles of those
subsequent domains: 1.Traffic transportation and key technologies.
provision domain 2. Healthcare domain.3.
Sensible atmosphere (home, office, plant) 2.2 RFID and EPC
domain. 4. Personal and social domain. The Radio Frequency Identification (RFID) systems
three-tier design of the Internet of Things is as give direct object identity sensing. They use a
shown in Figure1. device (RFID tag) to receive and send remote
commands. RFID systems contain tags, readers,
hosts and antenna. There is a small inexpensive
tag in every RFID object that has every product
a novel identity. Once Associate in Nursing
RFID reader sends missive of invitation signal,
the RFID tag responds to the readers reading
and writing request. RFID offers wireless
communication between the tags and readers
with non-line-of-sight readability, which
eliminates manual information assortment and
introduces the potential for automatic
identification method.
Fig 1. Three-tier architecture of internet of 2.3 GPRS:

things
The general packet radio service
The lowest layer is associate object-object (GPRS), a data extension of the mobile
network, namely a network that takes use of a telephone normal GSM, is emerging because the
range of sensors, RFID to create object 1st true packet-switched design to allow mobile
identification and information reading and subscribers to profit from high-speed
writing between objects. This layer is transmission rates and run information
information acquisition layer whose supporting applications from their mobile terminals. it's a
technologies area unit principally EPC, RFID, GSM-based wireless packet switching
etc. When data pools along, it wants technology, providing finish to finish and wide-
transmission, wherever a second layer is forms, area wireless science property, whose purpose is
that is termed information transport layer. The to supply packet-based sort of information

services for GSM users. GPRS provides high-

speed wireless science services for mobile users,
fully supports the TCP/IP, dynamically allocates
science addresses for the mobile sites and
achieves mobile net functions. Any kind of
business within the fastened net will be able to
be achieved through GPRS mobile networks.
F
ig 3. Diagram of GPS system
III. RESULTS
The objective of intelligent vehicle tracking

system is to actualize automatic watching for
traffic vehicles. In order to succeed in this
objective, we should always first solve
problems. The primary one is the way to
establish and find these mobile objects. Future
one is the way to acquire the positions of them.
At last, detected information of those mobile
objects should be transmitted from outside to
Fig 2. Vehicle tracking using GPRS watching center for processing and calculation,
C. GPS and once mobile vehicles area unit far from
cities on the far side the vary coated with web,
GPS works as shown in Figure 3.Ground how to carry the information for them? To solve
management station consists of master these issues, we've designed a unique network
management station, observance station and architecture for intelligent traffic watching as
injecting station. Master management station shown in Figure 4.
performs calculation of satellite annual and
correction parameters of satellite clock, and
injection of those information into the satellites.
Furthermore it controls the satellite and issue a
directive to that, and it conjointly has constant
perform as a monitoring station. observance
station performs receiving satellite signals;
observance the operating standing of satellites;
injecting station performs injecting satellite
annual and correction parameters of satellite
clock.
Fig 4. Network architecture

Transportation Research Part B 40 (2006) 857

871
IV. CONCLUSIONS
[3] Chin-Ling Chen, Yong-YuanDeng, RFID
From the higher than style and analysis, we will system with mutual authentication and privacy
acquire the following conclusions. Using RFID, protection, 22 (2009) 12841291
GPS, GPRS and network technologies Internet
of Things is found to construct associate [4] J.W.-K. Hong, S.-S. Kwon, J.-Y. Kim,
intelligent traffic monitoring system, that makes WebTrafMon: Internet network traffic
the latter as a section of the former. Secondly, monitoring and analysis system based on web
intelligent vehicle tracking system based on
internet of Things. Thirdly, the technologies of [5] Ensia European Network and Information
internet of Things makes it potential that a Security Agency-A Roadmap for a pan-
complete automation in observation system from European eIDM Framework by 2010, 2006.
information observe to information
[6] Ensia European Network and Information
transmission, and to intelligent decision-making, Security Agency. Report on the state of pan-
from vehicle management to main road European eID initiatives, 2009.
congestion management.
[7] Wei Yuan, Automatic Tracking System of
REFERENCES: Vehicles Based on GPS and GSM, vol.4, no.4,
[1] Jianhe Du, Lisa Aultman-Hall, Increasing pp.28-29
the accuracy travelling information from [8] Shiqing Liu, Integration and Application
passive multi-day GPS travel datasets Design of GPS and GSM System, vol.23,
[2] Hanif D. Sherali, Jitamitra Desai, Hesham no.12, pp.85.
Rakh, A discrete approach for locating
Automatic Vehicle Identification readers ,

DEVELOPMENT OF SMALL WIRELESS SENSOR NETWORK
FOR WATER MANAGEMENT USING XBee BASEDTECHNOLOGY

S. Anusha [1] DR. K. Srinivasa Ravi [2]
Sulamanusha481@gmail.com ravi.kavuluri@kluiversity.com
MTECH STUDENT (ES) PROFESSOR
Department of electronics and computer science Engineering
K L University,Guntur,A.P.,India
Abstract: Now a days the development of quantity across the wireless sensor networks to
contribute to unbalance of natural ecological generate a data pertaining to utilization of water and
system which leads to changes of climate condition raising alerts in terms of alarm/messages/mails
&natural resources such as water quality following any destruction in the safety norms for the
&quantity. For this we use a wireless sensor drinking water quality and usage of amount of water.
network (WSN) to monitoring the quality and The estimated outcome shall be a new model for
quality of water. Wireless sensor networks have online, real time water monitoring system that should
been achieving widespread applicability in water be used to improve efficiency of the water supply
quality monitoring. The purpose of water quality distribution network. The determination of water
like PH level, temperature, salinity, conductivity, quality such as PH level, temperature, salinity,
fluoride, turbidity, dissolved oxygen etc is conductivity, fluoride, turbidity, dissolved oxygen etc
measured and sends the data to the base station or is measured and sends the data to the base station or
control /monitoring room. The main objective of control /monitoring room. The main objective of this
this paper is the development of the suitable paper is the development of the suitable model for
model for online monitoring the both quality & online monitoring the both quality & quantity of
quantity of water at different water resources water at different water resources such as rivers,
such as rivers, lakes, canal etc. In this paper, the lakes, canal etc.
essential design and realization of wireless sensor
A currently becoming popular and widely used
network featuring a low power consumption of the
technology based on wireless sensor network is
XBee primary based technology is used for
extensively used in this project as it is able to provide
transmission & reception. XBee primary based
flexibility, low cost implementation and reliability. A
communication technology is the most recent
high power transmission with a relatively low power
version for zigbee. Transmission range of Zigbee
consumption XBee PRO based wireless sensor
technology using XBee to be improved by using
network technology is applied in this work. XBee
XBee-PRO and 9XTend OEM RF Module. XBee
PRO is a communication standard for use in the
used the zigbee protocol the only variation is
wireless sensor network defined by the XBee PRO
lower cost, high power and easy to use.
that adopting the Digi international standard for its
Index Term wireless sensor network, water reliable communication. It is preferred due to its
quality &quantity monitoring, XBee technology features that fulfill the requirement for a inexpensive,
easy to use, low power consumption and reliable data
I.INTRODUCTION communication between sensor nodes. The
GPS/GPRS are used for sending the
To develop an local, intelligent support system for messages/alerts/mails to the base stations.
online monitoring of the water flow , water quality&

The development of graphical user interface and automatic online monitoring of temperature,
(GUI) for the monitoring purpose at the base turbidity, water level, and PH of lake water.
monitoring station is another main component in the
project. The GUI should be able to display the III.SYSTEM FLOW:
various parameters being monitored continuously in
Master design:
real time. Several measurement and performance
analysis to evaluate the reliability, feasibility and
PC
effectiveness of the proposed monitoring system are slave 1 sub master1 a RS232 based
also presented. XBee Remote
module terminal
II. LITERATUE SURVEY:
Recent advances, through in wireless sensor

networks have provided researchers with an easily
deployable, scalable, flexible and relatively
inexpensive for real time distributed data monitoring. Here we designed a network; under that network we
Adhoc wireless sensor networks are self organized
connected two slaves. Two slaves are used for
networks composed of a large number of sensor
nodes that interact with their environment and different base stations. These Slaves are implemented
communicate in a wireless fashion. IN the previous using XBee PRO and ARM 7 combination. It will
paper A high power transmission with a relatively work as wireless sensor networks. Various Sensors
high power consumption Zigbee based wireless are connected to Arm 7 via inbuilt ADC and output
sensor network technology is applied in this work. of sensor is nothing but the measured parameters
Zigbee is a communication standard for use in the from respective slaves. Both slaves placed inside the
wireless sensor network defined by the Zigbee
Alliance. In this paper, the essential design and water to acquire the different parameters.
realization of wireless sensor network featuring a low
power consumption of the XBee based technology is The arm 7 is connected to the XBee PRO
used for transmission & reception. XBee based module through RS 232.The XBee PRO module have
communication technology is the most recent version its range 30 meter from the slaves. By selecting the
for zigbee. different XBee PRO module we can increase the
distance between Master and the Slaves.
WIRELESS SENSOR NETWORK:
Entire system contains four sensors to
Wireless sensor network is the wide range platform measure four parameters of water. Temperature,
due to its low cost, small size, low power turbidity, salinity & water level .These sensors are
consumption, flexibility, portability, scalability installed at each node in target area. Measured
features. WSNs (wireless sensor network)with parameters are analog in nature we converted it in
advantages arrangements, collection of variety of digital form by using inbuilt ADC. A temperature
parameters high detection accuracy and high sensor LM35 was used to measure the temperature.
accountability of the monitoring network etc. Measured Temperature is transmitted to the XBee
A wireless ad hoc sensor network consists of a terminal using two Xbee modules. A ARM
number of sensors spread across a geographical area. microcontroller handles coordination of various
Each sensor has wireless communication capability modules in the circuit. It can perform control in either
and some level of intelligence for signal processing open loop or closed-loop fashion. In open-loop
and networking of the data. A WSNs is a Ad-hoc control, one can adjust the frequency, the duty cycle,
network composed of great number of tiny low cost and the polarity (for turning purposes) of the
and low power consumption sensing nodes which are actuation pulses remotely on a PC. In closed-loop
capable of sensing, calculating and communicating control, the actuation voltage will be computed. The
data .This paper develops water quality and quantity microcontroller also interfaces with the sensor.
monitoring system based on wireless sensor network Through the Xbee PRO module, the ARM
which is applied to artificial lake to realize remote

microcontroller establishes communication with the

PC.
2.4 XBee PRO UNIT:
Slave Design:
The XBee and XBee-PRO RF Modules were
Temperature engineered to meet IEEE 802.15.4 standards and
support the unique needs of inexpensive, low-power
wireless sensor networks. The modules require less
Turbidity power and provide reliable delivery of data between
Inbulit devices. In this paper, the fundamental design and
ADC ARM implementation of wireless sensor network featuring
Salinity
a high power consumption of the XBee based
7 technology is used for transmission & reception.
XBee based communication technology is the latest
Water level version for zigbee. XBee used the zigbee protocol the
only difference is lower cost, high power and easy to
use. Here different sensors are used for the sensing
XBeePRO different parameters like temperature, PH, turbidity.
RS232
Tx/Rx
The XBee-PRO900 RF module is ideally
suited for low-latency point-to-multipoint networking
applications. Capable of point-to-point, peer-to-peer
and point-to-multipoint networking, the XBee-900 is
best solutions where RF saturation and absolute
transmission distance are dominant to the application.
2.3 SENSOR UNIT:
A sensor unit is basically consists of several

sensors used to detect the predetermined parameters
that indicate the quality of water. In this work, three
types of sensor; pH sensor that senses the acidity of
basicity of the water, temperature sensor and
turbidity sensor based on phototransistor are used.
All the sensors use battery for its operation. The
information being sensed by the sensors are then
converted into electrical signal and go through the
signal conditioning circuit that functions to make sure
the voltage or current produced by the sensors is
proportional to the actual values of parameters being
sensed. Then it is passed to a ARM 7 microcontroller
Fig: XBee-PRO 900 RF module
or microprocessor that processes it to the value
understandable by human.
The sensor node consists of various sensors The features of the XBee PRO 900 RF
to measure the water quantity parameters like flow, module is: The frequency band is 900 MHz The
velocity, pressure. These parameters which are interference immunity is used the frequency hop
crucial to determine the quantity & quality of potable spread spectrum (FHSS). The data rate is up to 250
water are logged in a central server through a Kbps. The distance is Up to 1 mile (1600 m), up to
gateway node. 2500 ft (750 m) international variant. The voltage
level is up to 3.6 V DC.

IV.SYSTEM CONFIGURATION water quality and quantity while taking the different
inputs from sensors nodes.
The examined system is a wireless sensor
network deployed for monitoring of a water V.CONCLUSION
distribution and supply system. The sensor network
consists of 520 plus sensor nodes monitoring flow Overall, the proposed in this paper, implementation
and pressure of a nodes system. These nodes are of high power XBee PRO based WSN for water
distributed with in roughly 50sq.km. The majority of
quality and quantity monitoring system offering low
the sensor nodes are deployed water lakes, rivers,
seas etc .All nodes including their sensors, radios and power consumption with high reliability is presented.
loggers are commercially manufactured and are The utilization of high power WSN is suitable for
designed to be water proof. Electronics within these activities in industries involving large area
were custom made with the possibility for minor monitoring such as built-up, construct, mining etc.
imperfections. Each sensor node is equipped with a Another important fact of this system is the simple
GSM modem capable of GPRS data connectivity. installation of the system where the base station can
Data collected by individual sensor nodes are relayed
be placed at the local residence close to the target
via a public GSM/GPRS network. Sensor nodes are
powered using a battery pack with an estimated area and the monitoring task can be done by any
lifetime of approximately twenty four months. person with minimal training at the beginning of the
GPS is a positioning service, while GPRS is system installation. Performance modeling is
a data service used in mobile phones. GPS is used to completely different environment is one important
identify your location on earth, while GPRS is used aspect to be studied in the future as different kind of
to access emails, and to browse the internet.GPS monitoring application requires different
communicates with a collection of satellites that orbit
the earth, while GPRS communicates with a configuration during system installation and
terrestrial tower.GPS requires three or more stations techniques.
to work, while GPRS requires just one.
REFERENCES
2.6 SOFTWARE DESIGN:
[1].Water Quality Monitoring System Using Zigbee Based
Wireless Sensor Network 1Zulhani Rasin and 2Mohd Rizal
GUI (graphical user interface) software is Abdullah Faculty of Electrical Engineering Universiti Teknikal
employed to get the stable remote access to Malaysia Melaka (UTeM) Melaka, Malaysia
observance the real time control and monitoring the 1zulhani@utem.edu.my, 2rizal_dem@yahoo.com.
variable rate irrigation controller.GUI is the most
developed software is a kind of interface that enables [2]. Monitoring water distribution systems: understanding and
managing sensor networks . D. Ediriweera and I. W.Marshall
to the user to interact with electronic device through
Lancaster Environment Centre, Lancaster University, Lancaster,
graphical icons and visual indicators. Its accustomed LA1 4YQ, UK Received: 5 February 2010 Published in Drink.
develop to interface the hardware and software Water Eng. Sci. Discuss.: 12 April 2010 Revised: 23 August 2010
modules to the wireless sensor nodes. Its associate to Accepted: 2 September 2010 Published: 27 September 2010.
interface for logging data from the different sensor
[3]. Automatic monitoring & Reporting of water quality by using
nodes. WSN Technology and different routing methods A.C.Khetre1,
Prof.S.G.Hate2 Department of E&TC, Raisoni College of
The actions in a GUI are usually performed Engineering, Ahmednagar.
through direct manipulation of the graphical parts.
[4]. Design of a Water Environment Monitoring System Based on
additionally to computers, GUIs are often found in
Wireless Sensor Networks Peng Jiang 1,*, Hongbo Xia 1, Zhiye
hand-held devices like MP3 players, media players, He 1 and Zheming Wang 21 Institute of Information and Control,
gaming devices office and industry devices. The term Hangzhou Dianzi University, 310018, Zhejiang Province, China;
"GUI" tends to not be alternative low-resolution E-mails: sink@stu.hdu.edu.cn (H.X.), endyriver@tom.com (Z.H.)
2 Environmental Science Research & Design Institute of Zhejiang
kinds of interfaces with slow resolutions, such as
Province, 310007, China; E-Mail: wzhem@sohu.com (Z.W.) *
video games (where HUD is preferred), it is with in Author to whom correspondence should be addressed; E-Mail:
the tradition of the pc science research at the PARC pjiang@hdu.edu.
(Palo Alto Research Center). Here, for monitoring the

[6]. Sensor networks for monitoring and control of water

distribution systems A. J. Whittle 1, M. Allen2, A. Preis3 and M.
Iqbal3 1 Department of Civil and Environmental Engineering,
Massachusetts Institute of Technology, Cambridge, MA, USA.
[7]. Implementation of Wireless Data Transmission in Monitoring

and Control R.Nagarajan, Member, IEEE and R.Dhanasekaran,
Senior Member, IEEE
[8]. Chae, M.J., Yoo, H.S., Kim, J.R., and Cho, M.Y., Bridge
Condition Monitoring System Using Wireless Network (CDMA
and Zigbee), 23rd International Symposium on Automation and
Robotics in Construction ISARC 2006, Tokyo, Japan, 3 5 Oct
2006. Sangmi Shim, Seungwoo Park and Seunghong Hong,
Parking Management System Using Zigbee, IJCSNS
International Journal of Computer Science and Network Security,
Vol. 6 No. 9B, September 2006.
[9]. http://www.digi.com/products/wireless-wired-embedded-
solutions/zigbee-rf-modules/point-multipoint-rfmodules/xbee-pro-
xsc for the XBee PRO information.
[10]. http://www.zigbee.org/.
[11]. http://www.digi.com/lp/xbee/

HANDLING LOCK HOLDER PREEMPTION FOR

COEXISTING GPOS&RTOS ON EMBEDDED
VIRTUALIZATION LAYER
ABSTRACT put of GPOS that supports shared memory multi
processors.
The real time resource management with in the
INTRODUCTION
Linux kernel is dramatically improving as a result of
Real-time embedded systems like smart phones
the real time Linux community. The development of
become highly efficient along with the enhancements
the commercial products involves the reusing of
of CPUs target in their market. But their worthless
existing real time applications in embedded systems
features introduced substantial engineering cost. The
without increasing their cost because the real time
key difficulty in the expansion of such devices comes
applications run on real time OSes whos OS API is
from the incompatible requirement of them: low
completely different from the POSIX interface. A
latency and high throughput must be established in
VMM that executes multiple operating systems at the
one system. This requirement is tough to satisfy with
same time is the promising solution, but presently
existing OSes, because all of them are considered as
existing virtual machine monitors such as Xen and
either Real-Time Operating System (RTOS) or
KVM are firm to be used for embedded systems due
General Purpose Operating System (GPOS).
to their through put oriented design and complexities.
RTOSes, like eCosor TOPPERS1, are designed and
vlk, which is a light weight processor abstraction
developed for executing real-time tasks such as
layer it provides the virtual CPU`s for the guest
processing wireless communication protocols. In a
operating systems and schedules them according to
distinctive case, these tasks run periodically for short
their priorities. Sometimes vlk schedules Linux with
time. The feature of executing such deadline sensitive
a low priority and RTOS with a high priority. The
tasks depends on on the limitation to RTOSes. For
performance degradation problem caused by LHP is
example, most RTOSes cannot change the number of
common every virtualization layer. Delayed
tasks vigorously. On the other hand GPOSes, like
preemption mechanism is used to resolve LHP. DPM
Linux, are designed and developed for executing
mostly based on the Para-virtualization technology
tasks which contain of major amount of computation.
because it does not waste the CPU time and also
Some of them in desktop computers are latency
implement with less effort. Well at all introduced the
sensitive is offering the comfortable experience to
hardware based solution called spin detection buffer
users, but missing deadlines is not deadly for them.
for detecting the meaningless spin of vCPUs
The impact from the real time Linux community has
produced by LHP. Some earlier proposals can avoid
significantly improved the real-time resource
this problem, but none of them cares about the real-
management capability of Linux. However, there is
time responsiveness of guest OSes. So the
always a compromise between satisfy in real-time
approaches are not suitable for embedded systems.
constraints and achieving maximum throughput .In
We have developed a new method for avoiding the
order to develop such a modern real-time embedded
LHP problem. The approach can ensure both the real
system which needs to fulfill conflicting
time responsiveness of RTOS and the high through
ISSN: 2348 8387 www.internationaljournalssrg.org Page 296

requirements, combining multiple OSes on a virtual fatal performance degradation of guest OSes when
machine monitor can be an effective approach. RTOS and SMP GPOS share the same physical CPU.
Virtual machine monitors, e.g. KVM , Xen and This performance problem comes from the
VMware, are traditionally used in the area of data phenomenon called Lock Holder Preemption (LHP).
center or desktop computing for executing multiple It is a general phenomenon of virtualization layers,
OS instances in one physical machine. Their ability hence a solution for this problem was already
of executing multiple OSes is also smart for proposed. However these existing solutions only
embedded systems because they make it possible to focus on the throughput of guest OSes, therefore the
implement the system which has multiple OS virtualization layers that execute RTOSes cannot
behaviors. If there is a virtualization layer which has adopt these solutions. To the best of our knowledge,
a capability of executing GPOS and RTOS in one there is no virtualization layer that can execute RTOS
physical machine, development of real time and GPOS on a multicore processor without
embedded systems can be simpler. In [2], Armand performance degradation caused by LHP, and is
and Gien presented several requirements for a distributed as open source software. Our laboratory is
virtualization layer to be suitable for embedded developing an open source virtualization layer for
systems: combining RTOS and Linux on embedded systems
1. It should execute a current operating system and its that adopt multicore processors, named vlk (vCPU
maintained applications in a virtualized environment, Layer in Kernel), a forked project from our original
such that modifications required to the operating project named SPUMONE. During the development
system are minimized, and performance overhead is of this virtualization layer, we faced many difficulties
as low as possible. specific to embedded systems. They come from the
2. It should be straightforward to move from one limitation of hardware resources, the requirement of
version of an operating system to another one; this is engineering cost, or scheduling RTOS and SMP
especially important to keep up with frequent Linux GPOS on the same CPU. Because of these
evolutions. difficulties, we believe that virtualization layers for
3. It should reuse built-in device drivers from their real-time embedded systems should be developed as
existing execution environments with no open source software for incorporating various
modification. insights from a wide range of community.
4. It should support existing legacy often real-time Basic Architecture
operating systems and their applications while 3.1 User-Level Guest OS vs. Kernel Level Guest OS:
guaranteeing their deterministic real-time behavior. There are many traditional approaches to execute
There is no open source virtualization layer that has multiple operating systems on a single processor in
a capability to satisfy above all order to comprise multiple functionalities.
requirements.VirtualLogix2 VLX is a virtualization Microkernels implement guest OS kernels at the user
layer designed for combining RTOS and GPOS, but level. When using microkernels, various remote
it is proprietary software. OKL4 micro visor is a instructions, traps and interrupts in the OS kernel
microkernel based virtualization technology for need to be virtualized by transferring their code. In
embedded systems, but performs poorly as the nature addition, since OS kernels are to be executed as user
of microkernels. In addition, we found that there is level tasks, The OS kernel communicated with

application tasks via inter-process communication. vCPUs, vCPU0 and vCPU1. vCPU0 is executed by
Therefore, many parts of the OS need to be modified. pCPU0 and vCPU1 is executed by pCPU1. RTOS
VMMs are another approach to execute multiple has one vCPU, vCPU2. This is executed by pCPU1.
OSes. If a processor offers a hardware virtualization So both vCPU1 and vCPU2 are executed on pCPU1.
support, all instructions that need to be virtualized Unlike typical microkernels or VMMs, vlk itself and
trigger traps to VMM. This makes it possible to use guest OS kernels are executed in the privileged level
any OSes without any modification. But if the .Since vlk provides an interface slightly different
hardware virtualization support is incomplete, certain from the one of the underlying processor, we simply
instructions still need to be complemented by modify the source code of guest OS kernels, a
replacing some code to virtualize them. Most of the method known as Para-virtualization. This means that
processors used for the embedded systems only have some privileged instructions should be replaced to
two protection levels. So when kernels are situated in hypervisor calls, function calls to invoke vlk API, but
the privileged level, they are tough to isolate. On the the number of replacements is very small. Thus, it is
other hand, if the kernels are located in the user level, very easy to port a new guest OS or to upgrade the
the kernels need to be modified significantly. Most of version of a guest OS on vlk. vlk does not virtualize
embedded system industries prefer not to modify a peripheral devices because traditional approaches
large amount of the source code of their OSes, so it is incur significant overhead that most of embedded
desirable to put them in the privileged level. Also, the systems could not tolerate. In vlk, since device
virtualization of MMU introduces significant drivers are implemented in the kernel level, they do
overhead if the virtualization is implemented by not need to be modified when the device is not shared
software. Therefore, we need reorder mechanisms to by multiple OSes.
reduce the engineering cost, to ensure the reliability
of the kernels and to exploit some advanced
characteristics of multicore processors.
vlk: A Multicore Processor based Virtualization
Layer for Embedded Systems
vlk is a thin software layer for multiplexing a single
physical CPU(pCPU) core into multiple virtual
CPU(vCPU) cores. The current target processor of
vlk is the SH4a architecture, which is very similar to
the MIPS architecture, and is adopted in various
Japanese embedded system products. Also, standard LITERATURE SURVEY
Linux and various RTOSes support this processor. SPUMONE
The latest version of vlk runs on a single and The method we are trying to establish is for a virtual
multicore SH4a chip. Currently, SMP Linux, machine monitor called SPUMONE [1]. This is a
TOPPERS, and the L4 are running on vlk as a guest Para-virtualization technology that works as a thin
OS. The basic abstraction of vlk is vCPU as depicted abstraction layer between hardware and OSes.
in Figure 1. In the example of this figure, vlk hosts Virtualization technology categorized into Para-
two guest OSes, Linux and RTOS. Linux has two virtualization requires modification of guest OSes,

but overhead of the virtualization layer is relatively SPUMONE requires modifying the source code of
low. In SPUMONE, each OS runs on a vCPU guest OSes. The amount of modification is however
provided and scheduled by SPUMONE. The vCPU very small: only the entry of interrupt service routine
scheduler is activated by interrupt and through the and functions for issuing sleep instruction. There is
sleep instruction of guest OSes. When the guest OSes several similar works focusing on combining a RTOS
issue a sleep instruction, SPUMONE selects the next and a GPOS together. RTLinux [3] treats Linux as an
runnable vCPU assigned to its physical CPU (pCPU). idle task of the real-time OS. Linux is therefore
If there is no runnable vCPU, a special vCPU allowed to run when there is no real-time task to run.
representing an idle state is executed just like the idle Adeos [2], targeting the x86 architecture, enables
task of traditional OSes. And when interrupt is raised, running multi OSes on one system without
SPUMONE delivers it to the vCPU which the modification of guest OSes. It only requires inserting
interrupt number is assigned to. SPUMONE kernel module, but highly depends on the unused
leverages the interrupt priority mechanism of privilege levels of the x86 Architecture.
hardware. Thanks to this mechanism, it is possible to
mask interrupts partially. The interrupt controller
judges which interrupts should be masked based on
their priority. For example, if we assume that there
are two devices with different priorities, one being
the timer device which rises an interrupt periodically
and the other being the Ethernet controller which
rises an interrupt when it receives a frame, if a high
priority is assigned to the timer device and a low one
to the Ethernet controller, the interrupt service
routine of the OS processing the interrupt of the timer
device can mask the interrupt of the Ethernet device.
This feature is key to implementing SPUMONE. In III. PROBLEM STATEMENT: MULTICORE
the example described above, if the RTOS is assigned SPECIFIC

the timer device and GPOS the Ethernet device, PERFORMANCE PROBLEM
SPUMONE delivers timer interrupts to the vCPU of SPUMONE is a successful technology on single core
the RTOS, triggering its execution. If the Ethernet environment. But on multicore systems, depicted in
controller receives a frame and tries to raise an Fig.1, a critical performance degradation of the
interrupt during the execution of the RTOS, the GPOS is observed. Fig.2 is the result of running the
pCPU will not be interrupted by the Ethernet backbench benchmark. The leftmost bar, labeled 4
controller because the priority of the timer device is cores, is the score of Linux running on 4 dedicated
higher than the one of the Ethernet controller. This physical cores. The rightmost bar, labeled 3 cores, is
interrupt priority mechanism is a common feature of the score of Linux running on 3 dedicated physical
CPU targeting embedded systems. Our test cores. The bars in the middle, labeled with percent-
environment, SH-4A, provides it, as well as ARM- ages, are the scores of Linux running on top of 4
based processors. As the description above implies, physical cores while sharing one core with the RTOS.

The percentage describes the CPU time consumed by that occurrence of blocking by IPC and overhead of
the RTOS. As this graph describes, when the CPU context switch is reduced. Today, co-scheduling is
time consumption of the RTOS is larger than 70%, used for scheduling virtual machines to avoid LHP.
the scores of hack bench is lower than when Linux is For example, VMWare [7] employs a customized
running on 3 cores only. This performance version of this method. However, co-scheduling still
degradation occurs because of the Lock Holder wastes a non-negligible amount of CPU resource, so
Preemption (LHP) [4] phenomenon. In general, LHP some projects explored other ways to solve this
is caused by preemption of the vCPU executing a problem. Uhlig, et al proposed a method to avoid
thread holding a busy-wait mutex (e.g. spinlock). LHP in both Para-virtualization and full virtualization
LHP causes a significant waste of time because when in [4]. The method for Para-virtualization is
other vCPUs try to acquire the lock acquired by indicating the holding of a lock by a guest OSes with
preempted vCPU, the vCPU holding the lock cannot flags of the VMM. If the flags indicate the thread of a
continue to execute the critical section protected by guest OS is holding a lock, the VMM never preempts
the lock because it is preempted. Therefore, other the guest OS but sets another flag indicating
vCPUs trying to acquire the lock will spin in vain delayed preemption. When the thread of the guest OS
until the blocked vCPU is scheduled again and finally releases the lock, it checks the delayed preemption
releases the lock, provoking the observed flag, which, if turned on, yields the CPU to another
performance degradation. There is also another VM. This method is efficient for virtualization
problem related to LHP. OSes running on a multicore environments containing GPOSes only, but cannot be
processor use Inter Core Interrupt (ICI) for some applied to RTOSes because it causes delay in
types of synchronization. For example, ICI is used interrupt delivery. Wells et al. proposed another
for TLB shootdown. In a virtualized environment, method to avoid LHP using extended hardware in [5].
this can cause significant performance degradation or The extended hardware component, called Spin
deadlock if the destination vCPU is preempted. Detection Buffer (SDB), features eight content-
addressable memory entries that can hold unique
stores and loads instruction. During a given period it
records stores and loads. If the entries are not full,
SDB indicates that a thread of a guest OS is holding a
lock. All these work introduces an extra delay to the
real-time responsiveness of guest OSes, so these
techniques cannot be applied in a satisfactory manner
to embedded systems.
CONCLUSION
References
The traditional way to solve the LHP problem is co-
[1] Hitoshi Mitake, Tsung-Han Lin, Hiromasa Shimada, Yuki
scheduling [6]. Co-scheduling was originally Kinebuchi,Ning Li, and Tatsuo Nakajima. Towards co-existing of
designed for multi process programs frequently Linux and realtime
interacting with each other via IPC. A scheduler OSes. In Proceedings of Ottawa Linux Symposium, 2011.
[2] Armand, Francois and Gien, Michel. A Practical Look at
implementing this method tries to execute processes
Micro-Kernels and Virtual Machine Monitors. In Proceedings of
communicating with each other in same time slice, so the 6th IEEE Conference

on Consumer Communications and Networking Conference, 2009.

[3] RedBend software. Mobile Virtualization: How it Works.
[4] Kinebuchi, Y. and Morita, T. and Makijima, K. and Sugaya, M.
and Nakajima, T. Constructing a Multi-OS Platform with Minimal
Engineering Cost. In Analysis, Architectures and Modelling of
Embedded Systems, 2009
[4] Volkmar Uhlig, Joshua LeVasseur, Espen Skoglund, and Uwe
Dannowski. Towards Scalable Multiprocessor Virtual Machines.
In VM'04: Proceedings of the 3 rd conference on Virtual Machine
Research And Technology Symposium

EMBEDDEDDEVICE AS A SERVICE:
NEW PARADIGM IN CLOUD COMPUTING
Cloud computing is an Internet based
computing wherever resources unit area
ABSTRACT
shared and maintained at datacenters that
Deploying an embedded system to act as a might be geographically located across
controller for electronics is not new. Now-a- globe. One will rent a virtual server, needed
days these types of systems are all around us memory, OS, applications or load his own
and are used for a multiple of purposes. In softwares on that and access it from any
distinction, cloud computing may be part of the world as per his uses. Data will
comparatively new approach for Computing be stored and secured in these data-centers
as a whole. This thesis project discussed which may be accessed by authorized users.
these two technologies so as to make a Its modified the total worlds scenario from
bridge between these two widely completely owning private computers and servers in a
different platforms. Therefore a bridge private network to on-demand access
should enable new ways of exposing network of shared computing re-sources
features and doing maintenance on with minimum management burden to
embedded devices. This could save prospective users, therefore cloud is an
companies not only time and money while merger of a number of most effective
dealing with maintenance tasks for computing technologies and processes for
embedded systems, However this cloud reliable, resilient and secure applications.
additionally avoid the required to host this Embedded Cloud computing is the new
maintenance software systems on dedicated dimension in cloud computing arising from
servers rather these tasks could use cloud the merging of the Embedded computing
resources only if required. This paper with it. Most of the tiny embedded systems
explores such type of bridge and presents will be joined (wired/wireless) which may
techniques suitable for joining these two communicate and perform the job. However
computing paradigms together. those systems might have an upper hand on
the cloud, then most of the processing might
be shifted to the cloud server and various
data centers that will be reducing the load of

devices. Embedded device as a service is the for Internet. Cloud computing is described
proposed model in this paper. as a pay as you go model, which means
1 INTRODUCTION the system is often right-sized. You pay only
1.1 CLOUD COMPUTING for what you use.
Cloud computing has over the previous few 1.2 EMBEDDED COMPUTING
years become a serious platform for

companies due to its ability to reduce costs Embedded systems are deployed in various
and because this model leads to a managed situations to act as controllers. Such systems
IT infrastructure that may be used to are quite accepted today. These systems are
dynamically provision and dimensional completely different in designs, capabilities,
services. The cloud consists of both and usage. While connecting these systems
hardware and software which is provided by to the Internet, it has been done by varying
a data center for which a customer pays only degrees, however in various cases these
for the resources that they use. Cloud systems have only been connected to
computing make use of virtual machines internal networks. Enabling these systems to
(VMs) running on clusters of computers. secure the function which is used as Internet
Computing has seen changes over decades. enabled devices that require consideration of
Cloud computing will be able to use the embedded systems. Now-a-days an
applications through the internet, which may increasing fraction of these embedded
be reached using a Web browser, where as systems are being connected to the Internet
the business pertaining softwares and data and form an Internet of things. An
are stored on servers which is a totally embedded system is a computer system that
remote location. From mainframe is created for specific controlling functions
computing to client-server model, it has which is installed in a larger system with
cleverly picked up some of the similarities real-time computing constraints. It is
from autonomic computing, grid computing, embedded as part of a device that includes
utility computing and many more which hardware and mechanical parts. By contrast,
have been the major contributors of this a general-purpose computer, such as a
master mind. In fact, the cloud is a metaphor personal computer (PC), is designed to be
flexible. At present, embedded systems
control various number of devices in protocols, signal processing, user interface

common use today. Embedded computing and so on.
seeks to design architectures that can 1.LITERATURE REVIEW
execute particular applications functions like
1.3 CLOUD COMPUTING
ARCHITECTURES
Apart from these architectural layers, the
The Architecture of cloud computing is cloud is also deployed in various models
depends on Front end and Back end. The such as Public, Private and Hybrid Cloud.
front end is the one which is able to notice Public cloud is already out on internet and
the clients and the back end is the cloud Private cloud is private to a particular
itself made up of storage devices, servers organization. Public cloud is mostly
etc. The cloud-based solution which allows managed by organizations which deal
you to access any of the remote devices, multiple users at a time with the help of
anywhere, anytime, from any devices. virtualization. Data is secure and multiple
Accessing your devices through desktop, requests are dealt at a common platform of
someone elses computer or a Smartphone cloud. It is much larger than Private cloud.
allows for device and location Private cloud is mostly for single client with
independence. The cloud that offers a way total control on security, data and service.
for companies to pursue opportunities Hybrid cloud is basically the combination of
quickly and cost effectively. Before cloud both public and private clouds. Private cloud
services, software developers had to develop is enhanced with resources of a Public cloud
or buy, configure, and maintain their own and helps to maintain a better reliability and
servers and software applications. The cloud broader functioning of the cloud. Though its
provides the ability to deploy a cloud based a complex structure because it should be
solution quickly and easily with little to no judged, the distribution of the application to
capital expenditure accelerating time to run on which model, whether Public or
market. The important feature which we Private cloud. It is really advantageous when
look in a cloud solution is Hardware large resources need small computing on
Agnostic looks for vendors who offer a Public cloud.
hardware agnostic platform.
1.4 CURRENT CLOUD SERVICES or pay-for-use model. It helps an enterprise

1.4.1 IaaS to have a virtual infrastructure and services.
Customers can handle their infrastructural
Infrastructure as a service (IaaS) In IaaS, the needs sitting in any part of the world with
user utilizes "fundamental computing the help of just one console. A group of IT
resources" such as processing power, experts is a necessity to provide
storage, networking components, or infrastructure services with resources like
middleware. The consumer can control the server, storage and network. These resources
operating system, storage, and deployed are the upper layer of infrastructure
applications. IaaS customers are often managed by the user.
companies with extensive IT expertise who
desire access to computing power but dont
want to be responsible for installing or
maintaining the hardware. Cloud computing
derives this feature from well known Utility
Computing. In utility computing, computing
resources such as storage and services are
charged on the amount of time you use. Like
the meters in houses to measure the unit of Figure 2.1: An example topology of a
electricity, water, cooking gas etc. This IaaS provider.
model has advantages like you need not
1.4.2 PaaS
have to own resources and maintain them,
instead rent them. There would not be in any
Platform as a service (PaaS) PaaS is a cloud-
trouble for licensing of softwares,
based platform that companies can use to
hardwares and other network issues. IaaS
develop their custom applications or write
requires the benefits from this existing
software that integrates with existing
architecture. This is the foundation of cloud
applications. The user controls the
services. It needs high technical competency
applications running within the environment
and involves IT architects who can very well
(and possibly has some control over the
provide Infrastructure as a Service to
hosting environment), but does not control
customers. It is based on On-Demand model
the operating system, hardware or network

infrastructure that they are running on. PaaS
is currently the smallest segment of the
cloud computing market and is often used
by established companies looking to
outsource a piece of their infrastructure.
Platform provided to the developers is the
core area to design softwares. It provides a
sort of framework for applications and hence Figure 2.3: An example topology of a
its not unwise to call PaaS a kind of PaaS provider.
Framework Computing. Applications
designated for definite platforms cannot run 1.4.3 SaaS
on other platforms. There is no need to
worry about the underlying architecture, one Software as a service (SaaS) SaaS is the
can create applications with necessary tools largest and most mature part of the delivery
and requirements, and put to use in disposal model within the cloud and is an application,
environment. This facilitates a rapid or suite of applications, that reside in the
development and disposal of applications. cloud instead of on a users hard drive or
Development environments include device. Google Maps, Salesforce.Com, and
IDEs which are configured or integrated to Shutter fly are examples of commonly used
PaaS for designing, building and validating SaaS applications .It offers specific software
of applications. Net beans, Microsoft Visual based solutions running in the cloud.
Studio and others as well are connected to Customers can choose from applications
components of PaaS and helps in such as email and other collaboration tools
development of applications on the platform. to be used by thin clients and/or end users. A
simple example is the Gmail email service
offered by Google and used by users via
their web browsers.

software partitioning, fabrication of

hardware-software development, code
design and simulation etc. Hence an easy
embedded prototyping board was chosen for
the purpose of interfacing. Embedded
prototyping board is a series of
microcontrollers development boards
designed for fast, flexible and low-risk and
Figure 2.2: An example topology of a professional rapid prototyping.
SaaS provider.
2 PROPOSED CLOUD
1.5 EMBEDDED SYSTEM DESIGN ARCHITECTURE
Embedded System Design is unique because

it is a hardware-software co-design problem.
Both, software and hardware must be
designed to together to make sure that the
implementation functions properly and is
reliable and cost effective Wayne Wolf
describes four major tasks that are:
1. Partitioning function to be implemented
in interacting and smaller pieces.
2. Allocating function is implemented
directly in hardware and software running
Figure 6: Proposed architecture for
on microprocessor
cloud services
3. Scheduling function is times at which
functions are executed, which is important IaaS has the feature of lending OS, memory,
when single hardware unit is shared. network and other physical resources.
There are various issues in designing of Devices can also be considered a separate
embedded systems such as hardware physical entity. But in PaaS, you deploy

your code and APIs associated will help to mobile phones uses cloud services to extend
program the device. Hence, EDaS has a their resources, such as iCloud on Apple
domain overlapping the both services. Using devices and Sky Drive on Microsoft devices.
IaaS is deployed in the project but using Embedded Devices can thus be monitored
PaaS is yet to be explored. and managed from any location and
collected data shared with multiple
4 CONCLUSIONS AND FUTURE applications which also reside in the cloud-
WORK a key consideration for industrial automation

projects.
In this paper embedded devices interfacing

3 REFERENCES
with cloud computing by using hardware-
[1]. Sun Microsystems. Introduction to cloud computing, white
software partitioning functions and paper. 2009.
embedded prototyping board, then the

[2]. Peter Mell and Tim Grance. The NIST Definition of Cloud
proposed model in this paper is Embedded Computing. csrc.nist.gov/publications/nistpubs/800-145/SP800-
Device as a Service (EDaS). This paper also 145.pdf, 2009, National Institute of Standards and Technology,
Information Technology Laboratory.
presented and evaluated the deployment of a
solution to bridge use of embedded systems [3]. Joe Bungo, Embedded programming in the cloud: a novel
approach to academia, January/February 2011, IEEE.
by users via a cloud deployed service. This
creates an environment in which the end [4]. Wolf, Wayne H. Hardware-Software Co-Design of Embedded
Systems, Vol 82, no.7, pp. 968-969, July 1994, IEEE.
users will be able to interact with embedded
systems via a fault tolerant solution creating [5]. Mbed Hardware Specification,
http://mbed.org/handbook/mbed- Microcontrollers.
new capabilities for interaction. . An Tutorial.UG682 (v1.0) April 27, 2009
embedded cloud design is presented that

[6]. Cloud Enterprise Cloud Installation guide,
consists of distributable Process Description https://help.ubuntu.com/community/UEC/CDInstall.
Language (PDL), Distributed Middleware

[7]. Embedded Cloud Computing Services Joseph Zaloker,
(DiMiWa), and an infrastructure. As a Director Technical Marketing, Arrow Electronics.
result, PDL can execute distributed

processes and share resources as services
over heterogeneous IoT devices with help of
DiMiWa and the infrastructure. Modern


A Security to vehicles using Smartphone Networks

R. Pavani Mrs P.S.G Arunasri
pavanirontala@gmail.com arunasri_2012@kluniversity.in
M .TECH ASSOC. PROFFESOR
Department of electronics and computer science engineering
KL University, Guntur, A.P., India.
Abstract:
Modern vehicles area unit progressively being interconnectedwith pc systems

that collect information from conveyance sources and net services. Badly,
this creates a non-negligible attack surface, that extends oncevehicles area
unit partially operated via smartphones. During this letter, a hierarchically
distributed system design that integratesa smartphone with classical embedded systems
is bestowed,and Associate in ad-hoc, end-to-end security layer is intended to demonstratehow
a smartphone will move firmly with a contemporaryvehicle while not requiring modifications
to the present in-vehicle network. Experimental results demonstrate the effectiveness ofthe
approach.
1.Introduction
The current trend in automotive product and services is to improve the accessibility of the vehicles
through novel services, that need an affiliation to some Internet-based source. This can be used to
gather information on the external environment (e.g., traffic conditions, weather forecasts, vehicle
position and orientation, usually integrated inside the on-board vehicle management systems), and to
supply infotainment services. In doing thus, the new devices that act with the vehicle (e.g., modern
motion-picture show systems, GSM, and Bluetooth connections) lead to associate degree accumulated
attack surface, which can alter associate degree adversary to interrupt into the vehicle itself, inflicting
severe safety hazards. Recently, many researchers highlighted this facet and with success attacks
against totally different vehicles every of those works showed that it absolutely was doable to take
management of sure functionalities of the vehicle, and interfere with safety-critical or sensitive
elements. These vulnerabilities hamper novel solutions (e.g., smartphones to unlock the vehicle door
or to start out the engine), because of the risk of flourishing attacks. Adding security mechanisms to
vehicles could be a difficult task, because the connected embedded architectures are ordinarily
designed with safety needs rather than security ones in mind. Additionally, the obtainablecomputing
resources area unit generally tailored to suit tightly to thecontrol systems desires, so as to limit the
prices. This clearlyrestricts the choices on the market for adding a security layerin later design phases.
Recently, however, securityrequirements area unit gaining additional and additional priority,
particularly inthe communication between embedded computers with ad-hocwireless networks .In this
letter, we tend to take into account a smartphone-in-the-loop vehiclearchitecture within the use case of
a simple machine vehicle.We propose a security answer that protect against attacks byaddressing the
challenges raised higher than, meeting eachperformanceand period constraints. what is more, we tend

to express the capabilities of the target design under consideration (i.e.,no input capabilities on the
vehicle aspect, restricted output capabilities,and lack of a sure execution surroundings on the
mobiledevice). Our projected answer permits a smartphone to determinea secure session layer over
associate insecure radio affiliation,which provides further security guarantees notwithstanding
thesecurity mechanisms already enforced within the physical layer(if any). As a result, the complete
application layer is transparentlysecured. The relevance and also the effectiveness of the
projectedsolution is supported by experimental results.
2.System Architecture
The system design into account directly relatesto the vehicle management logic.
The management logiccan be split into two main stages: the high-level stage takecares of the
vehicle motion and energy management, whereas the lowlevelstage takes care of
information acquisition and exploit. Thislayout is of common
in complicated automotive management systems,as they're typically characterised by cascade
structures that exploitthe frequency-separation paradigm so as to
decouplenested management loops. Consequently, we have a tendency to translate the logical division
between the two management loops into a technological separation:The high-level and therefore
the low-level management routines runon completely different devices, so resulting in a hierarchically
distributedcontrol system. The border between these two levels of abstractionmay be derived in step
with completely different policies likecomputational concerns, questions of safety, technological
constraints.Clearly, the two subsystems should share information.As a consequence, a channel that
guaranteesthe study and interconnection between the two subsystemsis needed. The
system designcomprises two main components. The
primary component isthe entrance electronic management unit (ECU), that is physicallymounted on
the vehicle, that runs the low-level management logiccommunicates with sensors associate degrade
actuators via an in-vehicle network (e.g., the will bus). The entry is provided with a radio interface
that enables wireless communication between the in-vehicle network and external devices. The
second part is an external device that works closely with the vehicle ECUs via the radio interface. In
our situation, the external device could be a mobile device that runs the high-level management
routines and acts as a driver-to-vehicle interface. This paradigm is extremely appealing and is gaining
increasing interest among vehicle makers, as drivers area unit probably to be already aware of mobile
apps, and because this methodology facilitates each package updates and the integration with different
web-based services . We with success enforcement , the same system design. Specifically, we have a
tendency to enforce associate degree intelligence vary extender for light-weight electrical vehicles,
with the goal of optimizing the energy consumption by actively modifying the vehicle dynamic
behavior,. This task is accomplishedwith a two-layer structure. A high-level controller keepstrack of a
reference profile, for the battery state of charge (SoC). The profile is generated by taking into
consideration the route length and its elevation profile. The mobile device implements the SoC
controller among associate degree ad-hoc app that we have a tendency to develop , that includes
navigation options that leverage on Internet-based services (e.g., Google Maps API). Furthermore, the
low-level management loops enforce speed and acceleration constraints ( and in, which permit to
satisfy the specified energy consumption profile. The low-level controllers act on the gas handle gap
to ensure that the resurgent behaviour of the vehicle (i.e., speed and acceleration ) is unbroken within
the prescribed limits. The entry and executes the low-level management loops on a 16-bits PIC
microcontroller with a central processing unit speed of twenty million instructions per second , and
communicates with sensors and actuators via bus. The mobile device communicate via a Bluetooth
layer. They exchange both format and time period management information. Format data is packed

into a forty eight bytes frame and therefore the communication is unidirectionalfrom the mobile
device to the entry one. On the contrary, the time period communication is bidirectional: The entry
one sends a 64-bytes payload each zero.2 s (5 Hz), whereas the mobile device communicates a 6-
bytes control-data packet whenever the vehicle travels fifty m. Simulation results and experimental
information collected on a image lightweight two-wheeled electric vehicle prove the effectiveness and
therefore the hardness of the projected approach. The vehicle equipped with the SoC controller saves
about 200thof the energy equipped by the battery, with relation to a nominal driving behaviour.
3. Issues
In the same situation, the mobile device and also the Gateway EU (Electronic Unit) exchange
sensitive knowledge. If this knowledge is compromised by Associate in Nursing someone, then the
practicality of the management system, and so the vehicle driveability could also be severely
affected: Depending on the attackers skills, the driving force even may loose the management of the
vehicle. Our focus is on the Bluetooth layer. The protocol encompasses a two-phase session setup:
when the pairing method, that permits the peers urge to grasp every other and got wind of the network
properties, the particular communication is enabled. Reckoning on the protocol version, different
security features are out there. However, the first Bluetooth standard and its successors, with the
introduction of the secure simple pairing (SSP) protocol , suffer from varied security vulnerabilities
thanks to weak cryptographical primitives. The protection of most Bluetooth applications (e.g., in
embedded scenarios) depends on a static PIN solely, with no thanks to modify it.
4. Secure Layer Services
Given the applying state of affairs and also the same security issues, it's necessary to plan an
application-level security mechanism that mitigates the vulnerabilities that exist the wireless link.
Such security layer should be free from the underlying wireless layer and should permit secure
communication between the mobile device and also the vehicle. In our attack model the mortal is
aware of the radio protocol in use, and is readyto transmit and receive discrete information packets on
the radio interface.The objective of the wrongdoer is to get access to the information changed between

the vehicle and also the mobile device, and ultimately manipulate the EU execution flow. We
concentrate on the applying layer. Therefore, attacks against the physical layer (e.g., jamming) or
attacks that need physical, even temporary, access to the vehicle (e.g., forceful shutdown) fall outside
the scope of our security layer.
A . Secure Analysis
We derive the necessities of our security layer through the evaluation of the appliance situation by
means of that trust domains and trust relationships between human action parties (or entities). A party
is taken into account a trustworthy domain if we have a tendency to trust its correct processing and
execution of the package implementation, and therefore its integrity. Otherwise, we have a tendency
to contemplate the party as associate untrusted domain. Betting on the characteristics and also the
security properties of the communication between entities, we can define trustworthy relationships (or
accepted dependence) between entities. Our answer is intended to account for security flaws (e.g.,
unknown knowledge leaks) or dependencies of the appliance layer on proprietary elements of an
entry.
B . Secure Session Layer
Our security layer follows the two-stage protocol .The primary stage sets up associate end-to-
end relationship between each application layers (i.e., on the mobile device and
on the ECU). Owing to the constraints of the state of affairs (e.g., distribution
of the mobile application through app stores, property
capabilities of the ECU), we have a tendency to don't assume any precomputed,
static credentials or scientific discipline keys on the mobile device,
nor use a public-key infrastructure on the ECU: solely the
vehicles owner is ready to initiate the primary stage by sanctioning the
one-off authorization procedure on the vehicles facet. For example,
this procedure may be enabled by pushing a buttononly
reachable mistreatment the vehicle key. A classic PIN-based procedure
is not continuously possible, owing to the restricted input capabilities on
the EU facet (e.g., absence of keypads). At intervals a brief time
span the EU accepts a mobile devices identity and therefore the user
receives the identity information of the EU .The
second stage ensures that the period of time communication necessities
are met. To the current finish, it implements a radially symmetrical scientific discipline
scheme that establishes a secure communication session.
The radially symmetrical session secret is derived from the semipermanent
secret changed throughout the primary stage, and some random knowledge
generated on the mobile device.
We enforced our two-stage approach on the entry
ECUs microcontroller. We have a tendency to use a ECDH key-establishment
scheme (asymmetric cryptography) on a homogenous curve
(NIST P-192). For every authentication method, the mobile
device computes a replacement random key set and transmits the corresponding
public key to the EU. In distinction to the key set of
the mobile device, the EU possesses a static semipermanent key set
for the key institution theme . For the
session cryptography, we have a tendency to enforce AES during a chaining block cipher

(CBC) mode with a 128-bit key. The key-derivation

function is enforced per the quality and provides
a recent 128-bit radially symmetrical key for every session. For the
implementation of our security layer on the mobile device, we
choose the OpenSSL library. Besides these two scientific discipline
schemes, we have a tendency to enforce the SHA-1 hash operate and outlined
a protocol structure for the mixing during a communication protocol
stack.
C. Secure evaluation
Our answer will mitigate the protection threats beneath the human model . The goal of our security
layer is to forestall attacks through the radio interface. As our answer implements cryptographic
session layer, that removes the dependency from proprietary implementations, dramatically reducing
the risk of exploitation. Though the offender obtains access to the ECU via the radio interface, the
appliance information is encrypted with the session key. The offender has less possibilities of getting
the cryptanalytic, long-run secret, than in a very regular Bluetooth pairing. Specifically, a man-in-the-
middle attack is troublesome to conduct: the offender would want to be among the communication
range: 1) throughout the heritage Bluetooth pairing process; and 2) throughout the primary stage of
our security protocol (i.e., the exchange of the general public keys). Solely the vehicle owner will
alterthe authorization method for a mobile device (e.g., in his owngarage) and, a lot of significantly,
among a predefined and short time span. Rather than compromising the ECUs security layer, an
attacker could perform an infatuated attack against the mobile device (e.g., mobile malware). Our
security framework addresses this type of security threat by providing the cryptanalytic mechanisms
under the developers authority and is versatile with respect to future updates to the mobile device or
OS. In fact, we tend to square measure ready to amendment any cryptanalytic primitive or protocol so
as to shield from actual or future vulnerabilities. We assume that the protection of the mobile
applicationsand so of our security layeris supported the integrity of operatingsystem and its
services.
5. Results
In this section, we tend to gift the experimental results and justifyour planned answer. In our
experimentalsetting, the entry is put in on a light-weight, electric simple machine vehicle present in
production and made by Piaggio. The entryECU implements intelligent, range-extending algorithms.
Our security protocol has the two main operating modes: pairing and payload exchange. Pairing is
active once the mobiledevice is paired with the vehicle, when the standard Bluetooth pairing
mechanism has taken place. In pairing mode, we tend to measure the performance of the uneven
cryptography each on the mobile device and on the entry, and also the performance of the key-
generation routine (on the mobile device). The payload exchange mode activates once the AES key
are literally changed, and encrypted or secret writing takes place. During this mode, we analyze the
performance of the secret writing (on the mobile device) and also the performance of the secret
writing.The payload consists of 64bytes of information, which has an artifact theme for supporting
discretionary payload size. The pairing could be a one-shot task, whereas the system ordinarily works
in payload-exchange mode. At runtime, the payload exchangemust satisfy period constraints. Here,
the bottleneck lies within the Bluetooth stack, as a result of the AES encryption-decryption of the 64-
bytes payload is dead whenever one in every of the peers transmits or receives a message via
Bluetooth (i.e., every 200 ms). To check its performance, we tend to collected runtime information
each on a machine and on a true implementation on the take a look at vehicle, to ensure a correct

characterization. Each at runtime and pairing time, the execution time could be an important
performance indicator. The results support the practicability of the projected approach in sensible
applications. Of course, the bottleneck of the key exchange is that the entry EU attributable to its
lower computer hardware speed: The execution time is just about an 130ms20 times larger than the
common time recorded on the mobile device.
6. Future Work
In this paper it's same that the present system is of going solely in the case of two wheelers and
myproposed one is that for four wheelers to underneath go sure changes like vehicle
detection,broadcasting ,temparature identification, wheather statement,trafficking, signalling,
bluetooth identification in vehicles whereas in motion and jointly describes the identification of the
gap between one vehicle to the opposite one. This provides U.S.A. clarity regarding what the longer
term work on the vehicles by exploitation smartphone technologies and networks. By this the
upgrading of phones takes place and improvement in code takes place. By this the whole networking
gets improved and jointly provides clarity however managing of those sources takes place i.e via
bluetooth or wi-fi.....GSM etc...vehicular resources that area unit stopped by the road facet also are
determined exploitation smartphone to smartphone association i.e by end-to-end adhoc association .
7. Conclusion
Finally we tend to associate during this paper that it's an fashionable connection to smartphone bound
technological version of the realm in networking section with an automotive embedded networking
background. We tend to design enforce and evaluate the protection layer over thesmartphone
association from bluetooth to entry and thus extend it to a lot of security issues connected like whether
or not prediction,approximately distance between vehicles,accidental sensations, busy roads,vehicular
sources stopped by road facet etc...Hence this will be done in such simplest way.
8. References
[1] I. Rouf, R. Miller, H. Mustafa, T. Taylor, S. Oh, W. Xu, M. Gruteser, W. Trappe, and I. Seskar,
Security and privacy vulnerabilities of incar wireless networks: A tire pressure monitoring system
case study, in Proc. 19th USENIX Conf. Security, Berkeley, CA, USA, 2010, pp. 2121.

[2] S. Checkoway, D. McCoy, B. Kantor, D. Anderson, H. Shacham, S. Savage, K. Koscher, A.

Czeskis, F. Roesner, and T. Kohno, Comprehensive experimental analyses of automotive attack
surfaces, in Proc. 20th USENIX Conf. Security, Berkeley, CA, USA, 2011, pp. 66.
[3] F. Stajano, Security for Ubiquitous Computing. Hoboken, NJ, USA: Wile, 2002.
[4] A. Dardanelli, M. Tanelli, B. Picasso, S. Savaresi, O. di Tanna, and M. Santucci, A smartphone-

in-the-loop active state-of-charge manager for electric vehicles, IEEE ASME Trans. Mechatron., vol.
17, no. 3, pp. 454463, 2012.
[5] C. Spelta, V. Manzoni, A. Corti, A. Goggi, and S. M. Savaresi, Smartphone-based vehicle-to-

driver/environment interaction system for motorcycles, IEEE Embed. Systems Lett., vol. 2, no. 2, pp.
3942, Jun. 2010.
[6] A. Dardanelli, M. Tanelli, and S. M. Savaresi, Active energy management of electric vehicles
with cartographic data, presented at the 2012 IEEE Int. Electr. Veh. Conf., 2012.
[7] Microchip Technology Inc., 16-bit dsPIC Digital Signal Controllers.
[8] NIST Special Publication 800-121 Revision 1, Guide to Bluetooth Security: Recommendations of
the National Institue of Standards and Technology 2012.

ROBUST AND ECONOMICAL TECHNIQUE TO EXTRACT ROADS

FROM SATELLITE PICTURES
Department of Electronics and Computers, K L University, Guntur, AP.
G.S.K.Sharma Ch. Lakshmi Manoja
Professor, M.Tech Student,
Sarma_gollapudi@kluniversity.in manoja.ch24@gmail.com
ABSTRACT: Image classification is an

important research area in computer vision.
Organizing images into semantic categories
can be extremely useful for searching and
browsing through large collections of images.
Road information has a fundamental role in
modern society. In todays world of growing
population, the need for urban planning is
very high. Road extraction from optical
satellite images is an economic and efficient
way to obtain and update a transportation
database is explained, its extraction can be of Fig: Image Taken from urban area
great help. This paper presents an integrated
method to extract urban main-road from
satellite optical images. The other applications
INTRODUCTION
of road extraction are: Identification of The Road extraction from digital pictures has
isolated buildings that need to be detected and drawn a special attention within the previous
updating of GIS database according to the couple of decades. Various ways has been
requirements of the human expertise. In this developed which has semi automatic and
method, roads are extracted solely based on automatic road extraction. Image classification
their color. The steps in the algorithm are is a very important analysis space within the
easy to follow and implement. It is also less field of image process. pictures are classified in
time consuming and an automatic method. line with the visual and discourse information
gift in it. it's a difficult task in varied
application domains, as well as satellite image
Key Words: Multi spectral Image, Median classification, syntactic pattern recognition,
Filtering, Morphological Operations, Edge diagnosis, biometry, video surveillance, vehicle
Detection. navigation, industrial visual scrutiny, vehicle
navigation etc. Image classification becomes
terribly troublesome if pictures contains bleary
and noise. Another drawback in image

classification is that several classification ways The basic steps concerned during this
depend upon the scene complex ness .i.e, if technique are given. The essential steps
large amount of objects are in image, then it's concerned are: the conversion of a vector
difficult to categorise. image to a scalar image using spectral angle,
the scalar image is filtered employing a median
Road extraction plays one among the key roles filter to get rid of noises that affects satellite
in vehicle navigation system, urban coming up pictures, filtered image is regenerate into a
with, disaster management system and traffic binary image, the binary image still contains
management system relating to the betterment some noise which is removed using
of human lives. Thus, the necessity for road morphological operations. The edges of the
extraction employing a sturdy and economical extracted road will be determined and finally
methodology is additionally high. Currently, the extracted road is overlaid onto the initial
there are some ways to extract roads manually image.
and mechanically. . Road extraction explained
during this paper depends solely on the colour
of the road. The advantage of this methodology
is that road pictures from any style of satellite
may be used provided it's more than 0.5m
resolution. Here, the images considered are
multispectral images. Multispectral images are
those pictures that accommodates 3 or
additional spectral bands. Any type of roads Fig:- Basic Steps Envolved
may be extracted supported their color. The
algorithmic rule is implemented using
MATLAB. The main disadvantage of the Road Extraction algorithm
various higher than given ways is that the issue
to supply the most effective parameters for a The various steps within the extraction
algorithm is explained below
particular given image. The remaining part of
the paper is organized as follows. The planned All multispectral images are thought-about as a
steps and extraction algorithms are explained. vector since it contains 3 or additional spectral
bands. Processing of an image as a vector is
kind of troublesome. so as to convert a
PROPOSED ALGORITHM multispectral image into a scalar there are
several strategies however here we have a
The first step during this technique is that the tendency to use spectral angle to convert the
creation of a database. Multi Spectral image is multispectral image to scalar. The spectral
taken as the input database. The input image angle of a picture is given clearly in. The
should contain multi spectral road images spectral angle of a picture is determined
whose road intensity values are in a particular victimization the spectral band, r is that the
range. By considering totally different intensity reference component and p is that the current
ranges for any type of roads can be extracted. component of a selected spectral band. Figure

shows the spectral angle image of a picture operator is employed as a result of the
from the info. The 2 vital benefits of perimeters is extracted with larger accuracy.
victimization spectral angle for conversion are: The edges of the roads are shown in Figure.
(i) the formula rely solely on the spectral angle The ultimate step is to overlay the extracted
and not on the quantity of spectral bands gift road onto the scalar image of the first image.
within the image, therefore this technique is Overlaying of the result helps as an instance
applied to multispectral pictures with any range the accuracy of the road extraction. In the final
of spectral bands (ii) supported the reference image, the skinny lines indicate the ways of
component elite, any variety of roads is roads within the image. the Final image is
extracted. The roads to be extracted are either evolved.
tarred or dirt tracks. The output for the spectral
angle is the image will be in darker color. On
the spectral angle image median filtering is EXPERIMENT AND RESULT
finished to get rid of the noise that affects the
satellite image. Once considering differing The database for road extraction will be created
kinds of filters, median filter is that the most based on the colour of the roads. the
apt one to cut back noise in satellite image. photographs will purchased from companies
Figure shows the image when median filtering. selling satellite pictures or the apt free pictures
The filtered image is then born-again into a from net will be downloaded. MATLAB 7.10
binary image for simple and quick process. The software package platform is employed to
edge for binary conversion is 0.07. The image perform the road extraction. The scale pictures
when changing it into a binary image. The used is 512* 512. From the Figure below it's
binary image still contains several unwanted clear that some of the objects apart from roads
pixels. Morphological operations are applied are detected. This is often as a result of those
on the image to remove the unwanted signals. objects are having the colour among the
Morphological operations removes unwanted particular range of roads. These objects might
pixels supported the foreground and therefore be little components of barren land and parking
the background of a picture. The importance of tons.
morphological operations is additional clearly
understood by relating. Since the operations are
done on the binary image. The MATLAB
operator used for binary image is bwmorph.
The image obtained when applying
morphological operations is given in Figure.
When applying the morphological operations
we get the clean roads however it's vital to get
the edges of those roads for clear identification
of the roads. Gradient filter is employed for the
edge detection and therefore the variety of
operator used for the detection is Sobel. Sobel

CONCLUSION
The roads play an important role in urban
designing. The formula introduced is automatic
one. It needs solely little interaction from the
users. The formula was enforced to notice
roadways from satellite pictures with resolution
Fig: Original Image
greater than 0.5m. The vital and key parameter
of this formula is that the color of the roads
within the data base. Different types of roads
will be extracted supported this formula. Since
extraction is alone supported color, some of the
barren lands and little areas of parking lots also
are being extracted. This is often as a result of
the locations even has the same constituent
intensity values as that of roads. Totally
different techniques like usage of Digital
Fig: Grey Image Elevation Models (DEM), active contours and
computer science ways may be enclosed to get
rid of the unwanted objects that are being
extracted. The formula enforced is quick,
strong and simple to know and implement.
REFERENCES
[1] J. Mena, State of the art on automatic road extraction for
GIS update: a novel classification, Pattern Recognition Letters,
vol. 24, pp. 3037
3058, 2003.
[2] I. Laptev, H. Mayer, T. Lindeberg, W. Eckstein, C. Steger,
and A. Baumgartner, Automatic extraction of roads from aerial
images based
Fig: Binary Image on scale space and snakes, Machine Vision and Applications,
vol. 12, pp. 2331, 2000.
[3] M. Rochery, I. Jermyn, and J. Zerubia, Higher order active
contours, International Journal of Computer Vision, vol. 69, pp.
2742, Aug.
2006.
[4] M. Barzohar and D. B. Cooper, Automatic finding of main
roads in aerial images by using geometric-stochastic models and
estimation,
IEEE Transactions on Pattern Analysis and Machine
Intelligence, vol. 18, pp. 707721, July 1996.
[5] M. Bicego, S. Dalfini, G. Vernazza, and V. Murino, Automatic
road extraction from aerial images by probabilistic contour
tracking, in
IEEE International Conference on Image Processing, ICIP03,
Fig: Final colour Image evolved by edge detection vol. 3, pp. 585588, 2003.

COMPRESSIVE OBJECT TRACKING A REVIEW AND ANALYSIS

BASED ON HUMAN COMPUTER INTERACTION
SASIDHARAN.R1, MALATHI.K2
1
(Department of Software Engineering/Indian Institute of Information Technology/India)
2
(Assistant Proffessor , DatabaseEngineering/ Indian Institute of Information Technology/India)
Abstract: The objective of this article is to audit the vision and it has been studied for several decades. In this
tracking strategies, characterize them into distinctive survey paper, the studies that are related to our work are
classifications, further more distinguish new patterns. To give summarized. Tracking the non-stationary appearance of
an improvement to the solution of drift problem in online objects with different pose, illumination variations and
tracking a separate section called compression tracking has occlusions still remains a challenging task for the researchers'
been chosen. In this survey, we have taken various community.
compressive tracking techniques along with their working, The general object tracking methods are classified into
merits and demerits. Most of the methods include object three categories, namely, point tracking, kernel tracking and
segmentation using background subtraction. The following
Silhouette tracking. In point tracking, the objects detected in
methods use diverse strategies like Mean-shift, Kalman filter,
consecutive frames are described by points, and therefore
Particle filter etc. This paper presents a survey on the association of the points relies on the previous object
compressive object tracking using the state of art models used. state which might embrace object position and motion. This
The designed models which are successfully applied using approach needs an external mechanism to observe the objects
compressive sensing concepts reduces the number of pixels in each frame. In Kernel tracking, kernel refers to the
and these methods are efficient in feature extraction and object form and look, for instance, the kernel may be an
dimensionality reduction with high accuracy.. oblong guide ortemplate elliptical form with an associated hist
ogram. Objects are half-tracked by computing the motion of
Keywords:compressive tracking; background subtraction; the kernel in consecutive frames. This motion is sometimes
particle filter; visual tracking
within the sort of a constant quantity transformation like
quantity transformation like translation, rotation, and affine.
In Silhouette tracking, tracking is performed by estimating the
I. INTRODUCTION object region in every frame. Silhouette following ways use
An object tracking system is used generally for the them knowledge encoded within the object region. This
observation of persons or objects on the move and supplying a information is often within the kind of look density and form
timely ordered sequence of respective location data to a models that are sometimes within the kind of edge maps.
model. Object tracking, generally finds its use in videos. It is
the process of locating a moving object (or multiple objects)
over time using a camera. It has a variety of uses, some of II. COMPRESSIVE OBJECT TRACKING
which are: human-computer interaction, security and TECHNIQUES
surveillance, video communication and compression, 1. Compressive Target Tracking using Particle Filtering
augmented reality, traffic control, medical imaging and video Method:
editing. It additionally estimates the movement of an object Tracking multiple targets in video is a classic problem in
in a picture plane because it moves around a scene. There has many areas. It is challenging for a variety of reasons,
been an incredible propagation of video surveillance including camera movement, illumination variability,
camera in public locations like stores, ATMs, highways, occlusions and not the least, the large dimensionality of the
traffic signals, schools, buses, subway stations, and airports so observed signal. It then becomes feasible to apply a number of
as to sight and track the moving objects. Detection estimation methods in order to find the target locations. They
and tracking of moving objects is extremely necessary to address sequential estimation, i.e. the filtering problem.
watch public transportation, and important assets. In the If the dynamics and/or the observations are nonlinear but
environments under surveillance monitoring, background gets the noise remains Gaussian, then the unscented Kalman filter
changing frequently for this reason, it is necessary to update (UKF) proposed by E.Wan et al [2] is used currently as one of
the value in the background images. First of all, in the real the best solutions. But still it leads to large errors in the true
time video tracking the object has to get separated from the posterior mean and covariance of the transformed Gaussian
background and then the tracking has to be made. For this random variable (GRV), which may lead to sub-optimal
kind of segmentation various approaches are available and the performance and sometimes divergence of the filter.
discussions about the various approaches are explained in this
paper. Tracking is an important research topic in computer

Object not necessary to capture individuals identity to perform this

trackin task. In fact, it could be sufficient to segment the crowd in
order to find the average person dimension, and to track the
Point Kernel Silhouett motion of people as a whole regardless of individuals.
trackin trackin e tracking Moreover, Rachlin and Baron [8] proposed the secrecy of
compressive sensing measurements. They asserted that
Dete Proba Mult Tem Cont Sha acquiring, transmitting and storing the video sequence in the
rmini bilisti i- plate our pe projection domain is computationally secure, it means that the
Mat
random projections enable decoding and any kind of further
Clas
processing. In fact, even though CS cannot achieve a perfect
Vie Stat Direct
w sifier e Mini security level. But Cevher et al. [9] had shown that CS can be
effectively used to perform background subtraction in the
projection domain, provided that the background estimator is
Vari Heu linear.
atio risti
nal c Cossalter et al [10] proposed a new coding scheme suitable
for video surveillance applications that allows tracking of
Fig 1: General object tracking methods [1] video objects without the need to reconstruct the sequence,
One for each target, still to improve the performance thus enabling privacy protection. At the decoder, they exploit
without undue complexity, they look at compressive sensing the sparsity that characterizes background subtracted images
[3], which allows to the exact reconstruction of sparse signals in order to recover the location of the foreground object. In
from a small number of incoherent measurements, typically by addition, by leveraging compressive sensing, they achieve
solving an L1 minimization problem. It is assumed that compression, encoding a limited number of random
compressive sensing is only a low-dimensional subspace of projections, as well as secrecy.
the state changes so that re-estimation is carried out whenever
3. 3. Compressive Sensing Used for Radar Tracking
the prediction residual increases beyond a threshold. If they
Compressive sensing and processing of radar waveforms
consider using a standard Kalman filter in the reduced state
enables high-resolution tracking while using low sampling
subspace [4], the compressive measurements are used for
rates and inexpensive processing. Compressive sensing
multi-view tracking. But still it has problem, where
enables the analog to information conversion of signals that
compressive sensing inversion needs to be performed at every
sample them well below the Nyquist rate. The processing is
time step.
performed using the compressive samples which are much less
Recently E.Wang et al [5] proposed a system which is in number than the Nyquist rate samples. Compressive
made effective by applying compressive sensing ideas in a processing, however, results in an increase in estimation error
multi-particle-filter frame-work, it is possible to preserve as compared to processing the recovered signal or the signal
tracking performance while achieving considerable sampled at Nyquist as per Davenport et al. [11]. In a delay-
dimensionality reduction, avoiding costly feature extraction Doppler estimation problem, the estimation error is due to an
procedures. Additionally, the target locations are estimated increase in the sidelobes of the AF associated with the
directly, without the need to reconstruct each image. This can transmitted waveform. This AF structure deterioration will in
be done using linear measurements which, under certain turn increase tracking error. To overcome the above
conditions, preserve crucial observability properties. mentioned drawbacks, Ioannis Kyriakides [12] proposed a
compressive sensing and processing is applied to single target
tracking. Moreover, estimation using compressively sampled
2. Compressive Sensing Used for Object Tracking in and processed Bjorck CAZAC sequences is shown to be
Video Sequences improved over estimation using linear frequency modulated
Now a days video surveillance systems are waveforms sampled at the Nyquist rate. This shows that low-
widely used in many places like airports, banks, shops, traffic rate acquisition and processing maintains reliable tracking
monitoring and within the premises of private houses so as to performance at high resolution, while simplifying the receiver
track an object or entities. However, this form of tracking is and reducing computational expense using adaptive scheme.
useful for the security purposes, but still it affects the privacy Compressive sampling is performed in a single step after
of an individual. Dufaux et al. [6] proposed an efficient having considered all available information. As per the above
privacy enabling technology for Motion JPEG 2000 videos information it is found that Ioannis Kyriakides [13] had
which consists of scrambling the transform-domain proposed an effective adaptive compressive sensing and
coefficients of the regions of interest in a video sequence. processing scheme, applied to the radar tracking problem. The
Moreover, since there are some problems in recovering the adaptive scheme naturally incorporates sequentially updated
full scene once the encryption key is made available at the information on target state that is readily available from a
decoder. In the same way, Chan et al. [7] observed that it is

particle filter based tracker. The proposed method is shown to combination, most algorithms typically operate on the premise
improve tracking performance as compared to a non-adaptive that the model of the target object does not change drastically
scheme, while maintaining a low sampling rate and a over time. The main challenge of visual tracking can be
computationally inexpensive operation. attributed to the difficulty in handling the appearance
variability of a target object. Intrinsic appearance variability
4. Improved Adaptive Compressive Sensing and includes pose variation and shape deformation, whereas
Processing (ACSP) Method extrinsic illumination change, camera motion, camera
viewpoint, and occlusions inevitably cause large appearance
Compressive sensing is able to preserve information in a variation. Due to the nature of the tracking problem, it is
waveform while sampling it at a sub-Nyquist rate. imperative for a robust algorithm to model such appearance
Compressive processing is then performed on the low- variation.
dimensional compressively sensed measurements, instead of
using reconstruction [14] that is followed by higher To enhance the robustness of such object trackers,
dimensional processing. CSP, however, although simple to Realizing the limitations Isard et al [17] introduced particle
implement, increases estimation error as compared to Nyquist filters to visual tracking and presented the Condensation
sensing and processing (NSP) according to Davenport et al algorithm for contour tracking in which multiple plausible
[11] and Kyriakides [12]. Tracking performance when using interpretations are propagated over time. This probabilistic
CSP can be improved by utilizing available tracking approach has demonstrated success in tracking the outline of
information such as information coming from the targets target objects. However, the representation scheme employed
motion model and past measurements. The proposal of (curves or splines) ignores the internal appearance of the
Kyriakides [12,15] an adaptive compressive acquisition target, and is not updated to account for variations in its
method was proposed, utilizing information available from the appearance, due to pose or illumination change.
estimated probability distribution of the target state. Supervised discriminative methods used for classification
This implementation of ACSP identifies a set of delay- and regression have also been exploited to solve visual
Doppler shifts that are likely to appear in the radar return tracking problems. Avidan [18] developed a tracking
waveform with nonzero probability. Although this ACSP algorithm that employs the support vector machine (SVM)
method was shown to reduce tracking error versus non- classifier within a optic flow framework. Although this
adaptive CSP, it roughly approximated the distribution of the algorithm has demonstrated success in tracking specific
target state to be uniformly distributed with fixed parameters objects, e.g., cars from a mounted camera in a moving vehicle,
throughout the duration of the tracking scenario. This resulted significant effort is required in training a SVM. Later,
in a fixed sized dictionary of delay-Doppler shifts. Williams et al [19] developed a method in which an SVM-
based regressor was used for tracking. As a result of training
To improve estimation performance an adaptive the regressor on in-plane image motion, this method is not
compressive sensing and processing (ACSP) method is effective in tracking objects with out-of-plane movements. A
proposed by Kyriakides [16] that creates a dictionary more elaborate mixture model fit via an online EM algorithm
containing delay-Doppler shifts likely to be found in the radar was recently proposed by Jepson et al. [20], in which three
return waveform according to tracking information. This components were used to model the responses of wavelet
improved ACSP method improves tracking performance by filters, and thereby account for appearance variation during
adapting the size and content of the dictionary compared to tracking. However, their appearance model treats pixels within
non-adaptive CSP and a fixed sized dictionary ACSP method. the target region independently (ignoring their covariance) and
The method demonstrates the improvement in tracking thus does not have notion of the thing being tracked. This
performance when adjusting the size of the delay-Doppler can result in modelling background rather than the foreground,
dictionary based on tracking information when using a fixed thereby failing to track the target object.
dictionary size.
Later very recently David et al [21] proposed a method
5. Compressive Sensing Method Used For Visual that, during visual tracking, which employs two method for
Tracking correctly updating the sample mean, and a forgetting factor to
ensure less modeling power is expended fitting older
Visual tracking essentially deals with non-stationary data,
observations. Both of these features contribute measurably to
both the target object and the background, that change over
improving overall tracking performance. Recently T.Bai et al
time. Most existing algorithms are able to track objects, either
[22] proposed a structured compressive sensing based tracking
previously viewed or not, in short durations and in well
algorithm for intelligent optical sensing, which exploits the
controlled environments. However, these algorithms usually
random feature reduction and the structured sparse
fail to observe the object motion or have significant drift after
representation of the target visual appearances. The efficiency
some period of time, due to drastic change in the objects
of the tracker is improved by a random feature reduction
appearance or large lighting variation in its surroundings.
together with the Block Orthogonal Matching Pursuit (BOMP)
Although such situations can be ameliorated with recourse to
algorithm. This method can achieve a more efficient tracking
richer representations, effective prediction schemes or

without losing the robustness compared to the reference the high dimension feature space by using compressive
trackers. sensing. The extraction of features reduces the computation
complexity and guarantees the real time of tracking. These
6. Visual tracking via Sparse Representation algorithms will provide the accuracy, robustness, and speed.
Compressive sensing or sparse representation has played a
fundamental role in many research areas. The problem is to 7. 7. Background Subtraction In Compressive Sensing
exploit the compressibility and sparsity of the true signal and Background subtraction is fundamental step in
use a lower sampling frequency than the Shannon-Nyquist automatically detecting and tracking moving objects with
rate. Sparsity then leads to efficient estimation, compression, applications in surveillance, teleconferencing and even 3D
dimensionality reduction, and modelling. Roughly speaking, modeling. In all applications that require background
compressive sensing is a technique for reconstructing a signal subtraction, the background and the test images are fully
(e.g., an image) using the prior knowledge that the sampled using a conventional camera. After the foreground
reconstruction is sparse or compressible le. With this estimation, the remaining background images are either
technique, the signal is represented by a sparse set of basis discarded or embedded back into the background model as
functions. That is, all of the coefficients corresponding to the part of a learning scheme. A motion detection algorithm
basis functions vanish except for a few. The challenges in begins with the segmentation part where foreground or
designing a robust visual tracking algorithm are caused by the moving objects are segmented from the background. The
presence of noise, occlusion, varying viewpoints, background simplest way to implement this is to take an image as
clutter, and illumination changes. Williams et al [23] extended background and take the frames obtained at the time t, denoted
the use of statistical learning algorithms for object by I(t) to compare with the background image [28].
localization. This approach is demonstrated in real-time
Lamarre et al [29] proposed background subtraction
tracking systems where the sparsity of the RVM means that
algorithm that uses multiple competing hidden-Markov
only a fraction of CPU time is required to track at frame rate.
models (HMMs) over small neighbourhoods to maintain a
However problem of automatically recognizing human valid background model in all situations. They used the DCT
faces from frontal views with varying expression and coefficients of JPEG encoded images directly to minimize
illumination, as well as occlusion and disguise. These computation and to use local information in a principled way.
problems are solved by Wright et al [24]. This new framework Later Piccardi [30] presented a review about various
provides two crucial issues in face recognition: feature background subtraction methods. Many video sources,
extraction and robustness to occlusion. The theory of sparse however, are in a compressed format before processing.
representation helps predict how much occlusion the W.Wang et al [31] proposed an approach to construct
recognition algorithm can handle and how to choose the background models directly from compressed video. The
training images to maximize robustness to occlusion. The proposed approach utilizes the information from DCT
problem is that the full potential of sparsity in robust object coefficients at block level to construct accurate background
detection and recognition together is yet to be uncovered. models at pixel level. This method is accurate and has lowest
computational cost.
In the same way, Xue Mei et al [25] proposed a robust
visual tracking method by casting tracking as a sparse Aggarwal et al [32] presented a fast and robust method for
approximation problem in a particle filter framework. moving object tracking directly in the compressed domain
Specifically, to find the tracking target in a new frame, each using features available in MPEG videos. But several issues
target candidate is sparsely represented in the space spanned are addressed like handling of full occlusions, fast camera
by target templates and trivial templates. The sparsity is motion, multiple object tracking and unsupervised tracking of
achieved by solving an l1-regularized least-squares problem. objects. In the same way Uttam et al [33] considered the
The proposed approach demonstrates excellent performance in background subtraction from adaptive compressive
comparison with previously proposed trackers. Still there are measurements, with the assumption that the background-
some problem in simultaneous tracking and recognition. subtracted images lie in a low-dimensional subspace. While
this assumption is acceptable when image tiling is performed,
Recently, T. Zhang et al [26] formulate object tracking in
background-subtracted images are sparse in an appropriate
a particle filter framework as a multi-task sparse learning
domain, spanning a union of low-dimensional subspaces
problem, which is denoted as Multi-Task Tracking (MTT).
rather than a single subspace.
Since the model particles as linear combinations of dictionary
templates that are updated dynamically, learning the Recently Garrett et al [34] proposed a method that
representation of each particle is considered a single task in adaptively chooses the number of compressive measurements
MTT. Compared with regular exhaustive search-based collected based on cross validation (CV) theory in CS. Their
methods, the main advantage of Monte Carlo sampling contribution extended the use of CV techniques to the
methods is the reduction of sampling patches. Lan Wang et al estimation of time varying signals and provided a practical
[27] proposed an effective method for Markov Chain Monte algorithm for adaptively changing the data rate of the system
Carlo (MCMC) sampling on compressive sensing on visual in response to scene activity.
tracking discriminative Haar-like features are extracted from

8. Kernel Based Tracking using Compression 9.Compressive Tracking in Real Time Scenario
One of the more general-purpose image processing The objective of real time tracking is to associate target
operations is to convolve an image with a kernel. A kernel is objects in consecutive video frames. The complexity of the
essentially a template describing a local neighbourhood, where problem is when the tracked object changes orientation over
the elements of the kernel specify the local weightings given time. Real time tracking is the process of locating
to each neighbouring pixel, and each pixel of the input image a moving object (or multiple objects) over time using a
is replaced by the convolution of its neighbourhood with the camera. Adding further to the complexity is the possible need
kernel. A general purpose convolution allows for the to use object recognition techniques for tracking. For these
application of Gaussian smoothing, local averaging, edge situations video tracking systems usually employ a motion
detection, Laplacian functions, and many other so-called model which describes how the image of the target might
kernel operations. change for different possible motions of the object.
Comaniciu et al [35] introduced a new framework for Various researches have done effective real time video
efficient tracking of non-rigid objects. A kernel-based tracking compression. Dorin et al [42] proposed a new method for real-
technique introduced, uses the basin of attraction of the time tracking based on the mean shift iterations and found the
similarity function. They employed a metric derived from the most probable target position in the current frame. The
Bhattacharyya coefficient as similarity measure, and use the dissimilarity between the target model and the target
mean shift procedure to perform the optimization. Generally candidates is expressed by a metric derived from the
semi-parametric kernel density estimation techniques are Bhattacharyya coefficient. But still there was some drifting
adopted. This technique fails under cases of occlusion. Vinay problem with these approaches. Later Grabner et al [43]
et al [36] proposed an improved kernel based object tracking employed ideas from semi-supervised learning and on-line
by performing the localization using a generalized boosting for feature selection. Thus so trained on-line
(bidirectional) mean shift based optimization. This makes the classifier was used in a tracking framework in order to
method resilient to occlusions. However, it is difficult to discriminate the object from the background. The knowledge
update this target model which is explained by Shen et al [37] from labelled data can be used to build a on-line classifier and
presented the target representations fragility by breaking they had demonstrated successful tracking of different objects
these trackers over a long image sequence. in real-time on various challenging sequences.
Shen et al [38] proposed a novel approach to kernel based Babenko et al [44] addressed the problem of tracking an
visual tracking, which performed better than conventional object in a video given its location in the first frame and no
single-view kernel trackers [37]. They proposed a support other information. They used a tracking by detection
vector machine (SVM) and the generalized mean shift (MS) technique where these methods train a discriminative classifier
tracker and then implemented by maximizing the classification in an online manner to separate the object from the
score. Compared with the plain MS tracker, it is much easier background. And also proposed a novel online MIL algorithm
to incorporate on-line template adaptation to cope with for object tracking that achieved superior results with real-
inherent changes during the course of tracking. Sam Hare et al time performance. However Li at el [45] addressed the
[39] applied a method that used a kernelized structured output problems in l1 tracker to provide the solution to those problems
support vector machine (SVM), which was trained online to by proposing real-time compressive sensing tracking (RTCST)
provide adaptive tracking and also introduced a budgeting by exploiting the signal recovery power of compressive
mechanism which prevented the unbounded growth in the sensing (CS). Dimensionality reduction and a customized
number of support vectors which would otherwise occur orthogonal matching pursuit (OMP) algorithm were adopted to
during tracking. To reduce visual drift problem which was accelerate the CS tracking. RTCST still produces competitive
encountered in object tracking, a two-stage sparse tracking accuracy compared to the l1 tracker.
representation method was proposed by Yan et al[40] to
Wu et al [46] proposed an effective and efficient tracking
improve the performance of the classifier and robustness of
algorithm integrating motion estimation appearance model
the algorithm, the kernel function was applied on the sparse
based tracking in the compressed domain . The idea is that the
representation. Moreover, the dimension of the target was
features are extracted from the multi-scale image feature space
reduced via compressive sensing.
based on compressive sensing theories. The motion
In order to have more accuracy using kernel concept we information has been integrated into appearance model based
can consider the concepts proposed by Dehshibi et al [41] a tracking by introducing motion estimator, i.e, particle filter. In
method for kernel-based object tracking in order to deal with the year 2014, Zhang et al [47] and made more effective and
partial occlusion. They used particle filter to estimate target efficient compressive tracking. A very sparse measurement
position accurately. The incremental Bhattacharyya matrix is constructed to efficiently extract the features for the
Dissimilarity (IBD) based stage was designed to consistently appearance model, where they compressed the sample images
distinguish the particles located in the object region from the of the foreground target and the background using the same
others placed in the background. The performance of this sparse measurement matrix. The tracking task is formulated as
method was evaluated for real world scenarios. a binary classification via a Naive Bayes classifier with online
update in the compressed domain. The proposed compressive

tracking algorithm runs in real-time and performed Tracking in general, may be a difficult downside. Troubles in
challenging sequences in terms of efficiency, accuracy and tracking objects can emerge because of camera movement,
robustness. illumination variability, occlusions non rigid object structures,
pose variation and shape deformation. In this paper we where
discussing many tracking methods and the techniques which
III.DISCUSSION AND CONCLUSION filter based tracker in case of radar tracking.
Paper Methods used Merit Demerit Success Rate (SR) / Location Error Rate
Accuracy Rate (LER)/ Average
(AR) Tracking Error (ATE)
Eric Wang et al [5] Compressive Particle Reduce the number of pixels - - -
Filtering (CPF) and accurate in a multiple-
target setting
T.Bai et al [22] Structured More efficient implementation - AR >68% LER <20%
Compressive Sensing with the random projection
based Tracking and
richer descriptive capabilities
Xue Mei et al [25] Simultaneous tracking Target templates are tracker will lock - -
and recognition, l1 dynamically on the occluding
minimization updated and non negative object
constrains are filtered
Lan Wang et al [27] Two-stage compressive This method is crucial for used to alleviate - -
tracking MCMC object tracking compared with drift problem
other tracking
Shen et al [38] generalized kernel- Improves tracking - - ATE = 4.2 % (appx)
based tracking accuracy and better than mean
shift method
Yan et al[40] Dimensionality Effective Compression - SR is between 82% - ATE < 12 %
reduction and kernel and robust object tracking and 99%
sparse representation solve drift problems
Li at el [45] Real-Time Consistently highest accuracy more accurate and - ATE < 30%
Compressive Sensing and robustness among all the robust results have
Tracking And compared tracking algorithms. to achieved.
Orthogonal Matching
Pursuit (OMP)
Algorithm
Zhang et al [47] Fast Compressive Track the right objects - SR = 99% Centre LER < 10%
Tracking accurately
TABLE I: Comparison of Compressive Object Tracking Based On Accuracy And Error Rate
Specifically, in visual tracking a structured compressive based tracking, Radar tracking . We also discussed detailed
sensing based tracking algorithm for intelligent optical about the background subtraction which is very much need in
sensing, this exploits the random feature reduction and the tracking scenarios for subtracting the target objects from the
structured sparse representation of the target visual background. We have pointed out various Troubles in tracking
appearances in the real time scenarios. However, there is a objects that are emerging because of camera movement,
need for more efficient solutions can be made by using very illumination variability, occlusions ,non rigid object
sparse measurement matrix that is constructed to efficiently structures, pose variation and shape deformation .In this
extract the features for the appearance model used in survey, we provided a detailed summary on different methods
compressive tracking applications. Overall, we believe that (like particle filtering , Kalman filter etc.) in various scenarios
additional sources of information, in particular prior and to overcome the troubles that we mentioned above .
contextual information, should be exploited whenever p
At the end of this paper, a discussion is made to point the
ossible to attune the tracker to the particular scenario in future work needed to improve the tracking algorithm for
which it is used. object detection and recognition under high occlusion.
Furthermore, we will investigate proficient recognition
In conclusion, we show a broad overview of item
modules for persistent tracking (where items vanish and return
following systems furthermore give a concise audit of related
after a drawn out stretch of time).We expect that this survey
points. We isolate the compressive tracking method into
on compressive tracking in video with rich theoretical details
various categories based on the use of this method in various
of the tracking methods along with bibliography contents will
applications like visual tracking, real time tracking , kernel

give valuable contribution to research works on object [21] David A. Ross Jongwoo Lim Ruei -Sung Lin Ming -Hsuan Yang,
Incremental Learning for Robust Visual Tracking, Springer, Int J Comput
tracking and encourage new research.
Vis (2008) 77: 125141
[22] Tianxiang Bai, Youfu Li, and Jianyang Liu, Structured Compressive
Sensing for Robust and Fast Visual Tracking , Sensors, 2012 IEEE.
IV. REFERENCES [23] O. Williams, A. Blake, and R. Cipolla, Sparse Bayesian Learning for
Efficient Visual Tracking, IEEE Trans. Pattern Analysis and Machine
Intelligence, vol. 27, no. 8, pp. 1292-1304, Aug. 2005.
[1] Yilmaz, O. Javed, and M. Shah, Object tracking: A survey, ACM [24] J. Wright, A.Y. Yang, A. Ganesh, S.S. Sastry, and Y. Ma, Robust Face
Computing Surveys, vol. 38, no. 4, 2006. Recognition via Sparse Representation, IEEE Trans. Pattern Analysis
[2] E. Wan and R. van der Merwe, The unscented Kalman filter for and Machine Intelligence, vol. 31, no. 2, pp. 210-227, Feb. 2009.
nonlinear estimation, in Proceedings of the IEEE Adaptive Systems for [25] X. Mei and H. Ling, Robust visual tracking and vehicle classification
Signal Processing, Communications and Control Symposium (AS-SPCC), via sparse representation, IEEE Transactions on Pattern Analysis and
2000. Machine Intelligence, vol. 33, no. 11, pp. 22592272, 2011.
[3] D. Donoho, Compressed sensing, IEEE Transactions on Information [26] T. Zhang, B. Ghanem, S. Liu, and N. Ahuja, Robust visual tracking via
Theory, vol. 52, no. 4, pp. 12891306, 2006. multi-task sparse learning, in Proceedings of IEEE Conference on
[4] D. Reddy, A.C. Sankaranarayanan, V. Cevher, and R. Chellappa, Computer Vision and Pattern Recognition, pp. 20422049, 2012.
Compressed sensing for multi-view tracking and 3-D voxel [27] Lan Wang, Pingyang ,Visual Tracking Based On Compressive Sensing
reconstruction, in Proceedings of the IEEE International Conference on MCMC Sampling, Systems, Man, and Cybernetics (SMC), 2013 IEEE
Image Processing (ICIP), 2008. International Conference on 13-16 Oct. 2013.
[5] Eric Wang, Jorge Silva and Lawrence Carin, Compressive Particle [28] B. Tamersoy (September 29, 2009). "Background Subtraction Lecture
Filtering For Target Tracking IEEE/SP 15th Workshop on Statistical Notes". University of Texas at Austin
Signal Processing, 2009. [29] Lamarre,M., Clark, J.J.: Background subtraction using competing
[6] F. Dufaux, M. Ouaret, Y. Abdeljaoued, A. Navarro, F. Vergnenegre, and models in the block-DCT domain. In: ICPR. (2002).
T. Ebrahimi. Privacy enabling technology for video surveillance. In [30] Piccardi, M.: Background subtraction techniques: a review. In: IEEE
Proceedings of SPIE, volume 6250, pages 205216, 2006. International Conference on Systems, Man and Cybernetics. Volume 4.
[7] A. Chan, Z. Liang, and N. Vasconcelos. Privacy preserving crowd (2004)
monitoring: Counting people without people models or tracking. In IEEE [31] Wang, W., Chen, D., Gao, W., Yang, J.: Modeling background from
Conference on Computer Vision and Pattern Recognition, Anchorage, compressed video. In: IEEE Int. Workshop on VSPE of TS. (2005) 161
Alaska, 2008. 168
[8] Y. Rachlin and D. Baron. The secrecy of compressed sensing [32] Aggarwal, A., Biswas, S., Singh, S., Sural, S., Majumdar, A.K.: Object
measurements. In Proc. Allerton Conference on Communication, Control, Tracking Using Background Subtraction and Motion Estimation in
and Computing, Monticello, IL, USA, September 2008. MPEG Videos. In: ACCV, Springer (2006) 121130.
[9] V. Cevher, A. Sankaranarayanan, M. Duarte, D. Reddy, R. Baraniuk, and [33] Uttam, S., Goodman, N.A., Neifeld, M.A.: Direct reconstruction of
R. Chellappa. Compressive sensing for background subtraction. In Proc. difference images from optimal spatial-domain projections. In: Proc.
European Conf. On Computer Vision, Marseille, France, October 2008. SPIE. Volume 7096., San Diego, CA (Aug.2008)
[10]Cossalter, M., Marco Tagliasacchi, and Giuseppe Valenzise. "Privacy- [34] Garrett Warnell, Dikpal Reddy, and Rama Chellappa : Adaptive Rate
enabled object tracking in video sequences using compressive Compressive Sensing For Background Subtraction, Acoustics, Speech
sensing." Advanced Video and Signal Based Surveillance, 2009. and Signal Processing (ICASSP), 2012 IEEE International Conference on
AVSS'09. Sixth IEEE International Conference on. IEEE, 2009. March 2012.
[11]Davenport, Mark A., et al. "Signal processing with compressive [35] D. Comaniciu, V. Ramesh, and P. Meer, Kernel-based object tracking,
measurements." Selected Topics in Signal Processing, IEEE Journal IEEE Trans. Pattern Anal. Mach. Intell., vol. 25, pp. 564577, 2003.
of 4.2 (2010): 445-460. [36] Vinay P. Namboodiri, Amit Ghorawat, and Subhasis Chaudhuri ,
[12]Kyriakides, Ioannis. "Radar tracking performance when sensing and Improved Kernel-Based Object Tracking Under Occluded Scenarios,
processing compressive measurements." Information Fusion (FUSION), Springer-Verlag Berlin Heidelberg 2006.
2010 13th Conference on. IEEE, 2010 [37] C. Shen, M. J. Brooks, and A. van den Hengel, Fast global kernel
[13]Ioannis Kyriakides ,Adaptive Compressive Sensing And Processing For density mode seeking: applications to localization and tracking, IEEE
Radar Tracking, Acoustics, Speech and Signal Processing (ICASSP), Trans. Image Proce., vol. 16, no. 5, pp. 14571469, 2007.
IEEE International Conference on 2011. [38] C. Shen, J. Kim, and H. Wang, Generalized kernel-based visual
[14]J. A. Tropp and A. C. Gilbert, Signal recovery from random tracking, IEEE Transactions on Circuits and Systems for Video
measurements via orthogonal matching pursuit, IEEE Trans. on Technology, vol. 20, no. 1, pp. 119130, 2010.
Information Theory, vol. 53, no.12, pp. 46554666, 2007. [39] S. Hare, A. Saffari, and P. Torr, Struck: Structured output tracking with
[15]I. Kyriakides, Adaptive compressive sensing and processing of delay- kernels, in Proceedings of the IEEE International Conference on
doppler radar waveforms, IEEE Transactions on Signal Processing, vol. Computer Vision, pp. 263270, 2011.
60, no. 2, Oct 2011. [40] Qingsen Yan, Linsheng Li , Kernel sparse tracking with compressive
[16] I. Kyriakides, Configurable compressive sensing and processing using sensing, Computer Vision, IET (Volume:8 , Issue: 4 ), 2013.
tracking information for delay-Doppler radar applications, Radar [41] Dehshibi, M.M.Vafanezhad, A., Shanbehzadeh, J., Kernel-based object
Conference (RADAR), 2012 IEEE. tracking using particle filter with incremental Bhattacharyya similarity,
[17] Isard, M., & Blake, A. (1996). Contour tracking by stochastic Hybrid Intelligent Systems (HIS), 2013 13th International Conference on
propagation of conditional density. In B. Buxton & R. Cipolla (Eds.), 4-6 Dec. 2013.
LNCS: Vol. 1064. Proceedings of the fourth European conference on [42] Dorin Comaniciu ,Visvanathan Ramesh ,Peter Meer, Real-Time
computer vision (pp. 343356). Berlin: Springer. Tracking of Non-Rigid Objects using Mean Shift ,IEEE CVPR 2000.
[18] Avidan, S. (2001). Support vector tracking. In Proceedings of IEEE [43] H. Grabner, C. Leistner, and H. Bischof, Semi-supervised on-line
conference on computer vision and pattern recognition (Vol. 1, pp. 184 boosting for robust tracking, in Proceedings of European Conference on
191). Computer Vision, pp. 234247, 2008.
[19] Williams, O., Blake, A., & Cipolla, R. (2003). A sparse probabilistic [44] B. Babenko, M.-H. Yang, and S. Belongie, Robust object tracking with
learning algorithms for real-time tracking. In Proceedings of IEEE online multiple instance learning, IEEE Transactions on Pattern Analysis
international conference on computer vision (Vol. 1, pp. 353 360). and Machine Intelligence, vol. 33, no. 8, pp. 16191632, 2011.
[20] Jepson, A. D., Fleet, D. J., & El-Maraghi, T. F. (2003). Robust online
appearance models for visual tracking. IEEE Transactions on Pattern
Analysis and Machine Intelligence, 25(10), 12961311.

Design of an Autonomous Vehicle for Precision Agriculture using

Sensor Technology
T.Sireesha1, Dr. K.Sreenivasa Ravi2,
1
(Department of ECM, K.L.University. Guntur (Dt), Andhra Pradesh-522502, India,)
2
(Department of ECM, K.L.University. Guntur (Dt), Andhra Pradesh-522502, India,)
ABSTRACT : To avoid the various problems to high demands on the processing unit of this
which affects the crop production, an autonomous design place. A number of electronic components
vehicle is to be designed and used for crop on agriculture equipment increases during the
transplanting and yielding, weed detection, crop situations in the normal field and the spray rate
protection, soil moisture properties, water status,
controllers, variable rate planter controllers, and
temperature monitoring, fertilization and pesticides
with their resource usage and also special focus on implement system controllers as well as controls
control and data monitoring with the embedded are interact with a normal vehicle operation. This
system. Some of the challenges and considerations results a creation of standard communication link
on the use of these sensors and its technologies for within all agricultural equipment. Therefore, a new
crop production are also discussed in this paper. methodology was needed in the agriculture
Fiber optic gyroscopes and multiple resolvers are autonomous vehicle which quantifies any auto-
employed to acquire the data for enhancing the
accuracy of target positioning and in order to steering system level performance and different
evaluate there is a method which describes the solutions are classified according to their accuracy
behavior of agricultural automation vehicle and reliability [1, 13].
traveling along paths of any curvature. II. EXISTING METHODOLOGY
Keywords- Precision Agriculture, Autonomous In the existing methodology of an

Vehicle, Fiber Optic Gyroscope, Sensors, Service autonomous robot was developed by the API
Unit, SLAM Algorithm. platform, which is able to survey an agricultural
field autonomously and the robot is four-wheel
I. INTRODUCTION drive and four-wheel steered as shown in Figure 1.
The autonomous navigation of the vehicle is
Precision agriculture (PA) is an obtained by the crop and weed density
innovative; the fundamental requirement of the measurements. For further processing of this
agricultural modernization is to improve the information is done by combining the data into a
efficiency of agriculture production without digital map of the field.
affecting these various factors. In the agriculture The GPS, gyros, magnetometer and
automation technology is to correct identification odometers was used in the equipment of the robot,
and positioning for the agriculture objects with in order to determine the exact location for image
trajectory of tracking problems. The data which is taking, as well as to estimate of the robots position
monitored recorded automatically, accumulated and orientation for a tracking algorithm was
easily and effectively is the basis to implement in facilitated with providing these measurements [12].
precision agriculture. But the technology is playing Actuation is obtained only by drive on four wheel
an important and increasing role from past several assemblies but not with steering. The platform is
years. Recently, the advent of different autonomous connected to a base station for enabling farmer
vehicles ranging from different fields of operations supervision and on-line data transmission.
was developed in the industry, but all these
products available are commercial and impressive
having a substantial cost (depending upon their
accuracy and functionality) to farmer.
The fully autonomous of all vehicle
operations carried out by all control routines and
this control relies on a central processing unit for
coordination which was developed by many
Fig 1: The API Platform
systems. But these systems have high unit cost due

The API platform is equipped with a high

resolution camera in order to map the growth and
density of crops and weeds and also analyzing
single plants at different growth stages implying
that the position of the inspection camera must be
accurate within few centimeters. This platform
must move between the rows to avoid the damage
on crops. So, the vehicle precision must be high, to
operate in the field and its mapping can be done in
a fixed spatial grid or by use of adaptive route
planning. For further treatment, all pictures have to
be transferred to the base station within a time.
Some of the platform requirements based on these
functionalities are considered in which robustness,
reliability, safety and accuracy are major
Fig 2: System Information.
requirements in the field because this will not
provide a satisfactory solution from the normal 3.1.2 Optimization of C/OS-II
laboratory prototype equipment [2].
The Control, Reliability and Safety Issues According to the application difference of
are considered in an open environment; the system, the operating system needs to be clipped
platform has to move autonomous where unknown and optimized, and this optimized system can not
obstacles potentially can be in the vehicle moving only save expenditure of system resource but also
direction, so there is necessity to require a high and create better overall performance. C/OS-II is an
reliable platform-self-control solution. embedded RTOS (real-time operating system) that
III. PROPOSED METHODOLOGY can be clipped and configured depending on
different needs of application, and the functions
In our proposed work, there is a need to that are not used temporarily can be closed, thus
add the additional hardware components and also not only saving memory space but also improving
consider some important aspects of software while the overall efficiency of the system [3].
designing an autonomous vehicle. 3.1.3 Flow Chart of Node Software
3.1 Software In this system there are two kinds of

In the system design, there are three kinds nodes. One is sensor node for collecting
of operating models including regular collection information of sensors; another is data central node
model, threshold alarm and real time inquiry model for receiving data sent from sensor nodes and
[3]. completing communication with computer.
3.1.1 Structure of Software System 3.1.3.1 Design of data central node software
If the system gets off the ground to
The C/OS-II is a portable, scalable and
initialize then hardware protocol begins to establish
preemptive operating system kernel that can be
network after initialization. Firstly set up network
embedded into ROM and can carry out multi-tasks.
in accordance with the sensor node list and modify
It is widely used in microprocessors,
lists through communication among sensor nodes.
microcontroller and digital signal processor. The
Then check whether there are new sensor nodes
structure of software system is shown as Figure 2.
that are waiting for accession to network and
complete the upgrade of network. Finally, operate
the protocol tasks after successful accession to
network, receive node data and sent them to PC. If
the number of times it is a failure exceeds regulate
done in the process of establishment, and the next
one should continue. The flow chart is shown as in
Figure 3.

Fig 3: The flow chart of central node.

3.1.3.2 Design of sensor node software
Sensor nodes are mainly used for
collecting sensors' data, receiving control data from Fig 4: The flow chart of sensor nodes.
data center and uploading collected data to data
central node. If there is no transceiving of data, the After completion of connection to the
nodes will enter into dormant state with the least network, the data from each node is received and
this data is sent to the embedded module via serial
power consumption.
port [4]. And then sends processed data to a remote
The hardware and protocol initialization is
control center through a network in Figure 5.
to be done after the system initialization. Protocol
tasks are carried out after successful loading. After
judgment of tasks, the sensor's data is to be
measured and are sent to data center node. Then the
next task is undertaken after the node enters into
dormant state [3]. The establishment of network
fails if the number of times exceeds the regulated
one. The flow chart is shown as Figure 4.
Fig 5: Host Device Control Flowchart Diagram

3.2 Hardware

An additional hardware components designed to be capable of insertion, as shown in

required in autonomous vehicle is discussed below: Figure 6.
3.2.1 Multimodal Sensor 3.2.2 Soil Moisture Sensor
In precision agriculture, soil Focusing on soil moisture monitoring, it

characteristics plays an important role for has been shown that the mobile agents, performing
absorption and desorption of water and nutrient data acquisition, data analysis, data aggregation
ions, nutrient solution changes in surface soil area, and decision making directly on the nodes, are able
deep soil area, and near roots area have different to respond in a timely manner to changes in the soil
behavior and time delay by the hour or day [5]. and to precisely schedule irrigation events, which
Therefore, a pinpoint measurement is required in results in a reduction of freshwater consumption
precision agriculture. So, in our proposed work and lowered irrigation costs [6]. The structure of
rather than using a normal temperature sensor in agriculture soil under irrigation is as shown in
precision agriculture, we have to replace with a Figure 7 [7].
miniaturized and insertion type multimodal sensor
used for precise control of the plants growth
conditions in medium culture measures directly.
The diffused solution in soil and the time
lag between supplying solution and soil condition
change were visualized by using the multimodal
sensor in actual cultivation environment, for the
first time [5].
Fig 7: Structure of agriculture soil under irrigation

The VG400 is a low-power and robust soil
moisture sensor, it senses volumetric water content
based on measurements of the dielectric constant of
the soil, a technique known to provide highly
accurate results. The sensor is insensitive to water
salinity and cannot corrode over time as, for
example, traditional conductivity based sensors.
And also by monitoring, crop and climate in a field
and providing the useful information which can be
used in making efficient use of water resources and
also achieving in high yield. If an additional
sensors, such as rain sensors may be integrated into
the monitoring system to further reduce the
freshwater consumption and the irrigation costs [6].
3.2.3 Rainfall Sensor
The rain sensor or rain switch is a

switching device, which is activated by rainfall and
having two main applications in rain sensors.
Fig 6: A cross-section structure image of a sensor Initially for an automatic irrigation system, a water
chip and bounded on PCB substrate conservation device is to be connected then causes
the system to shut down in the event of rainfall and
The multimodal sensor chip integrated
in the next section, by using a device protection of
with an electrical conductivity (EC) sensor and
automobile interior parts from rain was done,
temperature sensor for pinpoint measurement using
which supports the automatic mode of windscreen
Si large-scale integration (LSI) processes [15], and
wipers [8].
also the chip was bonded on the PCB package with
3.2.4 Humidity Sensor
the size of 8 mm and the length of 300 mm was

SHT11, a digital temperature and algorithm, because there is an impact with the
humidity sensor chip is widely used in fields like number of communication messages between the
heating and ventilation, air conditioning, UAV and the WSN [9].
automobile, consumer electronics and automatic 3.2.6 Attitude Sensor
control [3].
The chip integrates with a unique Due to random generating and the attitude
capacitive sensor element for measuring relative of the vehicle rapid changes, most of the
humidity and a band-gap sensor for temperature, agricultural fields are uneven, so that quick
additionally the signal processing on a tiny foot response is required for the measurement of the
print provides a fully calibrated digital output, attitude of an off-road vehicle. A low-cost
featuring in excellent reliability and long term electrolytic fluid inclinometer method is used to
stability [16]. The design of sensor module SHT11 sense tilt angles (roll and pitch), this is to be
is as shown in Figure 8. The ultimate choice for resolved with several issues such as poor accuracy
this design of SHT11 is tiny size and low power and sensor noisy response due to lateral
consumption [3, 14]. acceleration of the vehicle. But in real time, there is
a necessity of noise correction in the sensor was
required for operation [10].
Two inclinometers and three vibratory
gyroscopes are used in low-cost attitude sensor
with a quicker responses and higher signal to noise
ratios in their development while compare with the
other inclinometers which was used alone as shown
in Figure 9. These developed attitude sensor
evaluates an accurate results on field tests on a flat
field, a sloping ground and a bumpy road. So this
low-cost attitude sensor is preferable and to replace
fiber optic gyroscope which is cost effective in
Fig 8: SHT11 Sensor agriculture [10].
3.2.5 Spraying Operation
In agricultural areas, the application of

pesticides and fertilizers is having a crucial
importance for crop yields to carry out this task
mainly by using this aircrafts because of their
speed and effectiveness in the spraying operation
but some of the factors causes to reduce the crop
yield, or damage. Weather conditions, such as the
direction of the wind and its intensity during the
spraying process will add further complexity to the
problem of maintaining control. To avoid this
problem consider the architecture, which is to
address the problem of self-adjustment of the UAV
routes when spraying chemicals in a crop field. Fig 9: An overview of an attitude sensor prototype
In our proposed methodology, an 3.2.7 Service Unit
algorithm was evaluated to adjust the UAV route in
order to change the direction and wind intensity. To To improve the productivity and
adapt the path runs in the UAV, the wireless sensor efficiency in precision agriculture by doing these
network (WSN) deployed in the crop field which processes such as seeding, harvesting, weed
was obtained by input feedback. This shows the control, grove supervision, chemical applications,
sensors can use the feedback information in order etc. In an autonomous vehicle require a
to make adjustments to the routes could (unmanned) service unit which is to perform the
significantly reduce the waste of pesticides and primary or secondary tasks in the agricultural
fertilizers. Evaluation can be done in this environment.

The most important current abilities in the

autonomous vehicles are performed by the
agricultural tasks, which can be grouped into four
categories: guidance, detection, action, and
mapping, and the relation between its four abilities
are as shown in Figure 10 [11].
Fig 11: Functional structure of an autonomous

vehicle
3.2.7.1 Guidance:
The guidance requires information
regarding the surrounding environment (mapping)
and the features are currently detected (detection).
Fig 10: Relation between the four most important In the service units, the control and motion-
implementations of a service unit planning strategies are applied to drive the vehicle
within the agricultural field for specific purposes
But there exists a localization problem which are closely related to the action stage. Thus,
(i.e., not able to perform the action associated with the way the vehicle navigates within the
the agricultural task such as path-following, path- agricultural environment needs information
tracking, or trajectory-tracking activities), even regarding its location in the field (localization
though these stages are intrinsically related. If the system) and this system uses the sensors for a
localization system fails or is inaccurate, then this correct localization of the extracted features within
inaccuracy is propagated to the four abilities of the the map.
service unit. The SLAM algorithm was considered
3.2.7.2 Mapping:
as an inexpensive solution for the localization
problem. The most important stage is to plan
The simultaneous localization and feasible and safe paths or trajectories for the
mapping SLAM algorithm minimizes the navigation process by using the mapping and the
estimation and positioning errors in both the construction of a map in agricultural field will
localization and the mapping processes and this provide most relevant features. Thus in a service
algorithm concurrently estimates both the pose unit, the map of the environment is to navigate
(position and orientation) of a vehicle and the map safely, and the detected features will allow
of the environment in which the vehicle is located. appropriate planning for performing actions (e.g.,
The sensors mounted on the vehicle have an extract terrain leveling, chemical spreading, etc.).
features from the surrounding environment and During mapping, a map of the surrounding
these are located within a map, which is maintained environment is built and to maintain with the aid of
and updated by the SLAM algorithm. navigation (guidance) process. The measurements
The SLAM algorithm have an advantage acquired from the environment (detection) and the
is that they can optimally perform in places where information regarding the location of the service
other positioning systems fail and can be used to unit within such a map (for guidance and action).
further improve GPS-based localization systems. In The ability of detection is done by this stage and
addition to this, a ground station allows the tele- the localization system (the DK-GPS, the internal
operation of the vehicle. Thus, there are some sensors, and the low-level CPU).The high-level
specific strategies that are directly related to the CPU generates a map of the environment based on
environment disposition and the vehicles the exteroceptive sensors whereas the low-level
capabilities during navigation, positioning, CPU provides the localization information.
orientation, and turning maneuvers. The functional 3.2.7.3 Detection:
structure of the autonomous vehicle is as shown in In the agricultural environment, the
Figure 11. information is directly acquisted by using the

detection (i.e., the extraction of biological features Communication for an Agricultural Autonomous Vehicle,
Aalborg University, Denmark, CATA-2005, pp.494-499.
from the environment). At the mapping stage only,
[3] Zheng Ma and Xing Pan, Agricultural Environment
this information is to build and maintain an updated
Information Collection System based on Wireless Sensor
map of the surrounding environment to guide the Network, Nankai University, China, IEEE Global High Tech
navigation process (guidance) or to perform a given Congress on Electronics 2012, 978-1-4673-5085-3/12,pp:24-28.
action (e.g., weed detection, grove maturity [4] Lei Xiao, Lejiang Guo, The Realization of Precision
inspection, or agrochemical disposal). In this stage, Agriculture Monitoring System based on Wireless Sensor
Network, International Conference on Computer and
it consists of two range laser sensors, the stereo
Communication Technologies in Agriculture Engineering,
vision system and the high-level CPU processes the CCTAE2010, pp: 89-92.
sensors information. [5] M. Futagawa, Y. Ban, K. Kawashima, and K. Sawada
3.2.7.4 Action: Toyohashi, On-Site Monitoring of Soil Condition for Precision
Agriculture by using Multimodal Micro-Chip Integrated with
The action means interaction of the EC and Temperature Sensors, JST, JAPAN, M3P.025,
service unit with the agricultural field (e.g., Transducers, Barcelona, SPAIN,16-20 June 2013, IEEE,
radicchio harvesting in which the vehicle was pp:112-115.
designed for the execution of the task), but it can be [6] Kay Smarsly, Agricultural Ecosystem Monitoring based on
performed on the basis of a guidance process (e.g., Autonomous Sensor Systems, Berlin Institute of Technology,
Berlin, Germany, Agro-Geoinformatics, 2nd International
harvesting or seeding), detection (e.g., weed Conference on IEEE 2013.
removal), or mapping (e.g., agrochemical disposal [7] Wireless Sensor Network for Precision Agriculture,
based on previously acquired tree top Prakashgoud Patil, Vidya H, Shreedevi Patil, B.V.B College of
information).This stage was designed to monitor Engg. & Tech, Hubli, Karnataka, India, International
Conference on Computational Intelligence and Communication
and supervise a grove. Therefore, a robotic arm,
Systems 2011, IEEE Computer Society, Pp:763-766.
controlled by a high-level CPU, can be mounted on
[8] P.Satyanarayana, A.Gopala Krishna, J.Archana, Intelligent
the vehicle for manipulation purposes [11]. Low Cost Mobile Phone based Irrigation System using ARM,
IV. CONCLUSION International Journal of Scientific & Engineering Research,
IJSER, July-2013,Vol 4, Issue 7,1699 ISSN 2229-5518,
pp:1699-1704.
In precision agriculture the autonomous
[9] Fausto Costa, Jo Ueyama, Torsten Braun, Gustavo Pessin,
vehicle plays a most important role, to improve the Fernando Osorio, Patricia Vargas, The Use of Unmanned
efficiency of crop production without affecting the Aerial Vehicles and Wireless Sensor Networks in Agricultural
various factors in agriculture and also reducing the Applications, IEEE International and Remote Sensing
cost of production. In this paper, to develop the Symposium, Munich, Germany;01/2012.
[10] Akira Mizushimaa, Kazunobu Ishii, Noboru Noguchib,
design of an autonomous vehicle by considering
Yousuke Matsuoc, Renfu Lua, Development of a Low-Cost
the current developments and future perspectives of Attitude Sensor for Agricultural Vehicles, Japan, Computers
the Precision Agriculture (PA) for crop production. and Electronics in Agriculture 76 (2011), pp: 198-204.
It provides a better solution that optimizes product [11] Agriculture Robotics Unmanned Robotic Service Units in
quality and quantity of crop production by the cost Agriculture Tasks, IEEE Industrial Electronics Magazine,
1932-4529/13,IEEE, 19 September 2013, pp: 48-58.
minimization, human intervention and the variation
[12] Thomas Bak, Hybrid Control Design for a Wheeled
caused by environment due to unpredictable nature.
Mobile Robot, Lecture Notes in Computer Science, 2003.
V. Acknowledgment [13] Rovira-Mas, Francisco, Shufeng Han, and John F. Reid,
Evaluation of Automatically Steered Agricultural Vehicles,
The author T.Sireesha would like to express her sense of IEEE/ION Position Location and Navigation Symposium, 2008.
gratitude to Dr. M. Suman, HOD of the ECM, K.L.University, is [14] Murad, Mohsin, Khawaja Mohammad Yahya, and Ghulam
great-fully acknowledged for their cooperation and also it gives Mubashar Hassan, Web Based Poultry Farm Monitoring
me immense pleasure to thank my guide Dr. K.Sreenivasa Ravi, System using Wireless Sensor Network, Proceedings of the 6th
professor, the faculty of Embedded Systems in K.L.University International Conference on Frontiers of Information
for their valuable guidance, encouragement and help with the Technology - FIT 09 FIT 09, 2009.
data collection in connection to this work. And also I would like
[15] Futagawa, Masato, Taichi Iwasaki, Hiroaki Murata,
to express their thanks to the management of K.L.University for
Makoto Ishida, and Kazuaki Sawada, A Miniature Integrated
their support during this work.
Multimodal Sensor for Measuring PH, EC and Temperature for
References Precision Agriculture, Sensors, 2012.
[16] Xu.Bo.Yong Jun Zheng, Yan Xin Yin, and Yu Tan,
[1] Darr Matthew John, Development and Evaluation of a Temperature and Humidity Monitoring System based on
Controller Area Network based Autonomous Vehicle, (2004), CC2530, Applied Mechanics and Materials, 2013.
University of Kentucky, Paper 192.
[2] Jens Dalsgaard Nielsen, Kirsten Mlgaard Nielsen, Jan D.
Bendtzen, Design of Embedded System and Data

SMART INNOVATION
(A Smart Eco-Friendly Car system with Traffic Conjecture using
inductive loops along with Alcohol Prevention in Automobiles)
G.R.P.SIVA SUJITH G.N.V.SAI SUSHANTH
Sri Vasavi Institute of Engineering & Technology Gudlavalleru Engineering College
Ph no: 9030633480 Ph no: 7660961213
Abstract: using Google maps to forecast the traffic

through GPS to the driver at any place
The traffic jams in cities are most common regarding the traffic at the junctions. And
and remained as a uncontrollable problem. also we are using these induction loops to
This idea we developed is the result of our reduce the noise pollution at the traffic
observation in one of the major traffic intersections. We also quoted about one of
intersection at Vijayawada (Benz circle). In the smart car experience to the commuter
this paper we are going to present you the ie., a touch less mobile controlling system
best solution to eradicate heavy traffic jams that enables the driver to operate his/her
without using any constructions like bridges mobile with out even touching it. This touch
or additional lanes on the roads. We less mobile system is more useful to
observed that the commuters are getting eliminate accidents that occur due to usage
stuck in the traffic jams mainly due to no of mobile phones while driving.when a
proper estimation on the traffic on their way. driver or any passengers of car consumes
So, we thought that we can eliminate this alcohol then the car automatically stops in a
problem by intimating or instructing the safe place side of a road using traffic
driver/commuter about the level of traffic at conjecture & the front wheels of the car will
any junction all over the city.Such that the be locked out.Through GPS technology a
commuter can escape from the traffic jam complaint will be sent to the nearest police
and he/she can reach the destination in time. station.If driver talks in a mobile for more
So to intimate the density of traffic at any than 8 seconds then the system of the car
junction, we are going to use Induction activates the jammers to cut the call signals
loops to calculate the density of traffic at & It stops the car.We are also Placing a
the junctions and to display them to the piezo-electric devices on roads & zebra
commuter. In addition to this we are also crossing will generate electric energy

through the mechanical stress applied by the Flow chart:

pedestrians & vehicles while going on the
road. We are going to use this electric Installation of
energy as the power supply to the Induction inductive loops
Loops,Traffic Signals & Traffic Street
Lightening System.We are using this
technique with less amount of investment
Detecting the data given by
and more Accurate Output.
the loops
Approach:
The traffic can be controlled by using

various advanced technologies such as Estimating the traffic and
digital signal processing and various other.
calculation of time required
to reach the required
But here we are using a simple and cost
destination
effective tool to estimate the density of
traffic. And after getting the data on density
we are going to control the traffic depending
up on the density value.
Displaying the data and
Steps involved for traffic conjecture: instructions
1. Placing the induction loops at the

The main element that is needed to
junctions beneath the road.
implement this idea is the inductive loop.
2. Setting up a detector to detect the sata And also proper controlling system to attain
given by each and every loop placed on the much accuracy.
road.
Inductive loops:
3. Setting up a control system that enables
us to display the density and give the An induction loop is
instructions to the driver. an electromagnetic communication or
detection system which uses a
4. Further using the data given by the
moving magnet to induce an electrical
detectors to indicate the Google map with
current in a nearby wire. Induction loops are
traffic density.

used for transmission and reception of and information systems. Railways may use
communication signals, or for detection of an induction loop to detect the passage of
metal objects in metal detectors or vehicle trains past a given point, as an
presence indicators. A common modern use electronic treadle.
for induction loops is to provide hearing The relatively crude nature of the loop's
assistance to hearing-aid users. Vehicle structure means that only metal masses
detection loops, called inductive-loop traffic above a certain size are capable of triggering
detectors, can detect vehicles passing or the relay. This is good in that the loop does
arriving at a certain point, for instance not thus produce very many "false positive"
approaching a traffic light or in motorway triggers (say, for example, by a pedestrian
traffic. An insulated, electrically conducting crossing the loop with a pocket full of loose
loop is installed in the pavement. The metal change) but it sometimes also means
electronics unit transmits energy into the that bicycles, scooters, and motorcycles
wire loops at frequencies between 10 kHz to stopped at such intersections may never be
200 kHz, depending on the model. The detected by them (and therefore risk being
inductive-loop system behaves as a tuned ignored by the switch/signal). Most loops
electrical circuit in which the loop wire and can be manually adjusted to consistently
lead-in cable are the inductive elements. detect the presence of scooters and
When a vehicle passes over the loop or is motorcycles at the least. A different sort of
stopped within the loop, the vehicle "induction loop" is applied to metal
induces eddy currents in the wire loops, detectors, where a large coil, which forms
which decrease their inductance. The part of a resonant circuit, is effectively
decreased inductance actuates the "detuned" by the coil's proximity to a
electronics unit output relay or solid-state conductive object. The detected object may
optically isolated output, which sends a be metallic (metal and cable detection) or
pulse to the traffic signal controller conductive/capacitive (stud/cavity
signifying the passage or presence of a detection). Other configurations of this
vehicle. Parking structures for automobiles equipment use two or more receiving coils,
may use inductive loops to track traffic and the detected object modifies the
(occupancy) in and out or may be used by inductive coupling or alters the phase angle
access gates or ticketing systems to detect of the voltage induced in the receiving coils
vehicles while others use Parking guidance relative to the oscillator coil.

An increasingly common application is for indicates the free path without any traffic.
providing hearing aid-compatible "assistive The vehicle and the destination are indicated
listening" telecoil. In this application a loop above. At the first junction before the main
or series of loops is used to provide an audio junction, we are going to place a board that
frequency oscillating magnetic field in an indicates the driver about the status of the
area where a hearing aid user may be traffic and also instruct him to take diversion
present. Many hearing aids contain a telecoil or not. This makes the commuter to reach
which allows the user to receive and hear the his destination as fast as possible. To
magnetic field and remove the normal audio achieve this we need to use the inductive
signal provided from the hearing aid loop detectors and to be placed beneath the
microphone site. These loops are often road to detect the vehicle density.
referred to as a hearing loop or audio
Working:
induction loop.
An anti-submarine indicator loop was a Our Intelligent Traffic Light System is
device used to detect submarines and surface capable of changing priority level of the
vessels using specially designed submerged roads according to their traffic level. To
cables connected to a galvanometer. measure the traffic level we have several
mechanisms.
Application of this idea to control
Image Processing
traffic:
Pressure sensors, give reading when
pressure changes by the vehicles
Inductive Loops
From the above methods we choose

inductive loop.It is build on the concept of
inductance change of a coil when a metal
object come closer.You know that when you
send electrical current through a wire, it
generates a magnetic field. For a coil this
electromagnetic field is high.You can
The above is the case study that we had change the inductance of the coil and change
conducted. The red colored lane indicates the electromagnetic flux by introducing
the heavy traffic area. And the green lane additional conductive materials into the

loop's magnetic field. This is what happens installing or troubleshooting an inductive

when a car pulls up to the intersection. The loop vehicle detection system is absolutely
huge mass of metal that makes up your car critical. The preformed or saw-cut loop is
alters the magnetic field around the loop, buried in the traffic lane. The loop is a
changing its inductance. continuous run of wire that enters and exits
from the same point. The two ends of the
loop wire are connected to the loop
extension cable, which in turn connects to
the vehicle detector. The detector powers the
loop causing a magnetic field in the loop
area. The loop resonates at a constant
frequency that the detector monitors. A base
frequency is established when there is no
vehicle over the loop. When a large metal
object, such as a vehicle, moves over the
loop, the resonate frequency increases. This
increase in frequency is sensed and,
depending on the design of the detector,
forces a normally open relay to close. The
relay will remain closed until the vehicle
So we have made a coil develop the
leaves the loop and the frequency returns to
inductive loop.We have some kind of
the base level. The relay can trigger any
metering device to meter the voltage level
number of devices such as an audio
change in the coil.
intercom system, a gate, a traffic light, etc.
In general, a compact car will cause a
An inductive loop vehicle detector system
greater increase in frequency than a full size
consists of three components: a loop
car or truck. This occurs because the metal
(preformed or saw-cut), loop extension cable
surfaces on the under carriage of the vehicle
and a detector. When installing or repairing
are closer to the loop. Figures 3 and 4
an inductive loop system the smallest detail
illustrate how the under carriage of a sports
can mean the difference between reliable
car is well within the magnetic field of the
detection and an intermittent detection of
loop compared to the sports utility vehicle.
vehicles. Therefore, attention to detail when

Notice that the frequency change is greater as the loop, the greater the increase in
with the smaller vehicle. frequency. For example, a one square foot
piece of sheet metal positioned in the same
plane of the loop has the same affect as a
hunk of metal one foot square and one foot
thick. Another way to illustrate the point is
to take the same one square foot piece of
sheet metal, which is easily detected when
held in the same plane as the loop, and turn
it perpendicular to the loop and it becomes
impossible to detect. Keep this principle in
mind when dealing with inductive loop
detectors.
Some detectors provide PC diagnostics via a

communication port on the detector.
Diagnostic software gives you a visual
picture of what is happening at the loop, and
will help you troubleshoot any problems you
Also, it is interesting to note that the
may experience during installation or in the
frequency change is very consistent between
future. Detectors with this feature are
two vehicles of the same make and model,
usually in the same price range as other
so much so that a detector can almost be
detectors and can help you save time solving
designed to determine the type of vehicle
a detection problem. The PC software and
over the loop. There is a misconception that
cable is usually additional, however keep in
inductive loop vehicle detection is based on
mind that if you have multiple installations
metal mass. This is simply not true.
you need only buy the software and cable
Detection is based on metal surface area,
setup once. Diagnostics software can also
otherwise known as skin effect. The greater
help determine the depth and position of the
the surface area of metal in the same plane
loop in the pavement.

we are seeing daily in newspaper,

newschannels etc., that many people are
dieing in accidents due to Alcohol
consumption or negligence of the driver
while driving. My solution to this problem is
Alcohol Prevention in Automobiles.In this
system firstly,we will fix Alcoholic Sensors
or Breath Analyzers in stearing of
vehicle,near dashboard,back part of front
seats(rear & back),so we can easily detect
alcohol consumption of driver and
passengers in the cars,trucks,lorrys,buses
etc.,. In Bikes,we will place sensors near the
Mapping the traffic density and enabling speedometer, to detect the alcohol. After
GPS: detection,the system will give three
warnings with voice response saying you
By using the data given by the induction
are alcohol consumed please stop the
loop detector we are going to enable the
vehicle. If driver does not listen then
Google maps to give the updates of high
car(any vehicle) stops in a safe place on
traffic prone areas through GPS.
other side of the road by observing Traffic
conjecture Status on the screen in his/her
car, or if we have Wi-Fi in our mobile, then
by clicking on Traffic Conjecture App on
your Mobile you can see the Traffic Ahead
of you. After car has stopped, the Front
Wheels of vehicle will be locked, Car
Engine will be OFF and the position of the
vehicle(through Google map&through
GPS) will be given as a Compliant to the
nearest Police Station through Wide Area
Alcohol prevention in Automobiles:
Differential Global Positioning
System(WADGPS).we will also place a

camera in front of driver seat, so if driver

falls asleep or became less sensitive then it
warns first and then stops the car
Automatically after warning.
Eco-Friendly Autonomous Car:
We are planning to make our Automobile

more Eco-Friendly one. For this,we are
going to place Piezo-Electric Devices on
Roads,Zebra Crossing,Footpath etc.,thereby
we are going to generate Electric energy
from the mechanical stress applied by the
pedestrians,vehicles while going on the
Roads.we are going this developed electric
energy as a power supply to the Induction
Loops,GPS system,Traffic Signalling
System,Street Lightening System,Traffic
Conjecture Display Screens etc,.Thus we are
using renewable energy.
Conclusion:
At last, Our Aim is to Develop an Eco-

Friendly Car with Traffic Conjecture using
Induction Loops and Alcoholic Prevention
System in Automobiles. Being Engineers,
we did Our Maximum Efforts what we can
do for Automobile Field. Hence it is a
LAUNCH OF NEW ERA OF
AUTOMOBILES..

Development of a smart city using openmtc platform
N.V.K.RAMESH GADDAM SARATH CHANDU

ELECTRONICS AND COMPUTERS ENGINEERING ELECTRONICS AND COMPUTERS ENGINEERING
KLUNIVERSITY KLUNIVERSITY
VIJAYAWADA, INDIA VIJAYAWADA, INDIA
nvkr@kluniversity.in Sarath.gaddam431@gmail.com
Abstract of Things (IoT) and brings several benefits to industry and

business in general as it has a wide range of applications such
In mobile communications Machine-to-Machine
as Smart Cities, industrial automation, logistics, Smart Grid,
(M2M) communication offers a new communication approach
health etc. mostly for monitoring but also for control purposes.
that enable connecting billions of sensor objects to actuators.
Smart City concepts and implementations are
Smart City refers to behavior of a region where some
currently widely discussed. Definition of Smart City refers to
processes of activity are done automatically through
behavior of a region where some processes of activity are
communication from M2M. The openmtc platform aims to
done automatically through communication from M2M. Here,
provide a standard oriented middleware platform for M2M
M2M becomes enabler to develop smart city solutions. M2M
communication. This paper proposes a new approach on
itself is a standard concept that is already designed by several
development of a smart city using openmtc platform. For
standard communities, one of them is ETSI. This community
the development of smart city, we are mainly concentrate an
defines specification of M2M system as a standard reference
areas like new approach on managing all vehicle data using
for all developers of platform and application.
Machine-to- Machine (M2M) communication form which
Recently, the smart city concept is becoming trend
Open Machine Type Communication (OpenMTC) as
often spoken by many researchers of ICT sector, especially,
communication platform for aggregating and processing
sector of Next Generation Network Infrastructure (NGNI),
location data and managing communication between doctor
Internet of Things (loT) and Future Internet (FI). Recognizing
and patient when sudden changes in heart beat like heart
the need for reliable network infrastructures and various
attacks caught by patient. These applications are run on
Standards Developing Organizations (SDO) have recently
openMTC platform and with the help of above two
promoted several standardization activities in the M2M
applications, this paper will test the scalability and
communication domain, just to mention a few: i) the European
adaptability of the openmtc platform and in ensuing days
Telecommunications Standards Institute (ETSI) TC M2M
scalability increased as per development.
mainly focusing on the service middleware layer. ii) 3rd
Generation Partnership Project (3GPP) [8] address
Index Terms M TO M communication, openmtc, requirements and functional architecture for Machine Type
vehicle tracking, ECG Communications (MTC). iii) Telecommunications Industry
Association (TIA) [9] established the TR-50.l Smart Device
Communications. iv) oneM2M, a consortium of several
I. INTRODUCTION
standards development bodies to reduce the standardization
Machine to Machine (M2M) refers to technologies overlap by providing ongoing standards support, and increase
that allow both wireless and wired systems to communicate the capability of M2M solutions and produces to
with other devices of the same type. The concept can be interoperate.[10]
implemented in various domains such as logistics, smart One implementation of M2M ETSI standard is the
environments, energy and asset tracking. This development is Open Machine Type Communication (OpenMTC) platform
cheered by the fact of saturation and high competition in the that is developed by Fraunhofer FOKUS. This platform aims
mobile market, which raises the need to introduce new to provide a standard oriented middleware platform for M2M
potential services to fulfil the revenue gap. Additionally the oriented applications and services to facilitate research and
evolution in semiconductor industry shrinking lithography development of M2M systems. One pre-eminence of
continues to reduce chipset cost and power consumption, and OpenMTC platform is the ability to access and process
implants more sensors into devices used in different aspects. information through standardized application programming
M2M is a broad term as it does not pinpoint specific wireless interfaces (APls). The main advantage using openmtc is its
or wired networking, information and communications scalability and adaptability.
technology. M2M is considered an integral part of the Internet

Mobile tracking system is an application for (GSCL) and a network service capability layer (NSCL). Those
monitoring position of mobile vehicle. Big companies, which have been defined and specified by the ETSI Technical
have many armadas, use such systems to track and control Committee M2M. OpenMTC will also support the standards
their vehicles. Not only location information can be that are currently produced by oneM2M as an evolution of the
aggregated and managed, but also other environmental ETSI work.
measurements (e.g. temperature, humidity) can be aggregated
to help in improving shipping patterns and assets control.
Vehicle position is one piece of information that is needed in
monitoring. There is further information that is able to be
captured once this position data is available such as vehicle
temperature, opening or closing state of the door, relative
movement of the vehicle, and so on. Using M2M concepts,
this information can be gathered and sent to an administration
division for analysis prior to taking decisions aiming to
improve the monitoring and control services. Server
application has to be modified regarding variation of data
types.
ECG is a physiological signal that is used to measure
and diagnose abnormal rhythms of the heart. ECG is captured
using device that is called electrocardiograph. For some
special cases, it is required that patients ECG should be
monitored continuously. Usually in monitoring ECG,
physician uses one or two electrodes attached in patients
body using typical configuration. In previous works, we have
designed wireless LAN ECG for hearts patient tele
monitoring. Single channel ECG is connected with wireless
data transmission module to transmit ECG signal to
physicians computer or terminal devices. The data collected
by ECG is transmitted to openmtc for communication between
doctor and patient.
This paper combines these two applications to check
the scalability of openmtc and by increasing the scalability of
openmtc smart city was developed by number of applications Fig1: concentrated areas for development of smart city.
development. The remaining paper describes as follows:
section II openmtc platform for smart city development.
Section III implementation of smart city finally Section IV The GSCL is a flexible M2M gateway that supports
gives the conclusion. various M2M area network technologies and communication
protocols such as ZigBee and Wireless M-Bus. The NSCL is a
cloud-based M2M platform that aggregates and stores data
from various devices and acts as a device management and
abstraction layer providing intuitive Application Pro-
gramming Interfaces (APIs). Via the APIS and standardized
II.OPENMTC PLATFORM FOR SMART CITY data structures, fast application development can be enabled
across various M2M segments such as Smart Cities, eHealth,
DEVELOPMENT
Logistics, and Utility Metering.
OpenMTC supports a RESTful architecture with a
The OpenMTC platform is a prototype hierarchical resource tree defined by ETSI. This style governs
implementation of an M2M middleware aiming to provide a how M2M applications can exchange data with the OpenMTC
standard-compliant platform for Smart City and M2M M2M Core (NSCL) and the OpenMTC Gateway (GSCL).
services. It has been designed to act as a horizontal Each entity in the M2M system, i.e. applications, devices,
convergence layer supporting multiple vertical application data, is represented by uniquely addressable resource in the
domains, such as transport and logistics, utilities, automotive, resource tree, which can be accessed and manipulated via the
eHealth, etc., which may be deployed independently or as part CRUD operations (create, retrieve, update, delete) over
of a common platform. Fig1 shows the concentrated areas of different transport protocols (e.g. HTTP). Fig2 and fig3
smart city. shows the architecture of openmtc platform
The OpenMTC features are aligned with the ETSI TC ETSI defined three interfaces as part of the
M2M specifications. The platform mainly consists of two TC M2M specifications: mIa, dIa and mId, as depicted below,
service capability layers, a gateway service capability layer which offer generic and extendable mechanisms for

interaction with the service capability layers (xSCL). The mIa the openmtc middleware. Next we will see the implementation
interface resides between network applications (NA) and the of these two applications by using openmtc.
NSCL; the dIa interface mediates the interactions between
applications in the M2M network area, i.e. gateway
applications (GA) or device applications (DA), and the GSCL;
and the mId interface resides between the GSCL and NSCL.
FIG 3:TECHNICAL ARCHITECTURE OF OPENMTC PLATFORM
III. IMPLEMENTATION & ANALYSIS

Fig 2:a simple architecture of openmtc platform Implementation involves the operation of
two applications and implementation of these applications.
The strength of OpenMTC Platform is the ability to handle
The OPENMTC tool kit provides an SDK to make implementation for a variety of sensor and user's application.
the core assets and service capabilities available to third party Flow of messages in of the mobile tracking application is
developers. The OpenMTC SDK consists of a set of high-level showed in Fig.4. Device and sensor connected to the mobile
APIs which hide internal system details and allow the vehicle are registered into a gateway. Then, gateway informs
developer to concentrate on implementing functional M2M the OpenMTC core about registered applications and sensors.
application logic. This will stimulate the community and On the client side, user's application registers them to
facilitate innovation in the telecommunications and Internet OpenMTC, in order to be able to get data and updates from
arena helping to deliver a rich application landscape in Smart sensors thorough the system. The application can search for
Cities. devices connected to OpenMTC, and the core server will send
This paper uses this platform for the development of information about available resources that meet the search
Of both on managing all vehicle data using Machine-to- criteria. The application sends a request to the gateway to
Machine (M2M) communication form which Open Machine access required information from the available resource, or
Type Communication (OpenMTC) as communication subscribe to them, and get notifications when sensor data is
platform for aggregating and processing location data and received by gateway. If needed, the application will trigger
managing communication between doctor and patient when actuator advice to do something according to position of the
sudden changes in heart beat like heart attacks caught by tracked vehicle. Since mobile tracking is a real time system, a
patient. By this applications, we will increase the scalability of robust platform which can support real time data transfer
among sensors is needed. Real time data generated from

mobile tracking system is longitude, magnitude, and time sent

from sensors to OpenMTC platform. OpenMTC will process $GPGSV,3,3,11,29,09,301,24,16,09,020,,36,,,*76
$GPRMC,092751.000,A,5321.6802,N,00630.3371,W,0.06,31.
and store input data, and enable the access to them by
authorized client's devices via APIs. In the implemented 66,280511,,,A*45
mobile tracking system, the client application has a user
friendly graphical interface that shows moving tracked objects
on a map according to their position. The location information
is received form the OpenMTC platform via notifications. Fig.5 NMEA format data
The data generated in NMEA format has several

commands like
GGA essential fix data which provide 3D location
and accuracy data.
GSA- GPS DOP and active satellites.
GSV- satellites in view shows data about the
satellites that the unit might be able to find based on its
viewing mask and almanac data
RMC- NMEA has its own version of essential gps
pvt (position, velocity, time) data
GLL geographic latitude and longitude etc.
Fig.4 message flow in openmtc platform Similar message flow mechanism is also used in
managing communication between doctor and patient when
The data generated in the openmtc is in the form of sudden changes in heart beat like heart attacks caught by
NMEA. Fig 5 shows the data in NMEA form. The data will be patient is shown in fig.6
shown in google earth is in the kml format. In this application,
the data is converted from nmea to kml format. The output file
is shown in fig.9
$GPGGA,092750.000,5321.6802,N,00630.3372,W,1,8,1.03,6
1.7,M,55.2,M,,*76
$GPGSA,A,3,10,07,05,02,29,04,08,13,,,,,1.72,1.03,1.38*0A
$GPGSV,3,1,11,10,63,137,17,07,61,098,15,05,59,290,20,08,5
4,157,30*70
$GPGSV,3,2,11,02,39,223,19,13,28,070,17,26,23,252,,04,14,
186,14*79
$GPGSV,3,3,11,29,09,301,24,16,09,020,,36,,,*76
$GPRMC,092750.000,A,5321.6802,N,00630.3372,W,0.02,31. Fig.6 message flow graph ECG monitoring system
66,280511,,,A*43 The ECG signal produced from ECG device has two
$GPGGA,092751.000,5321.6802,N,00630.3371,W,1,8,1.03,6 signals from two leads. Example of these data is shown in
1.7,M,55.3,M,,*75 figure 7 and figure 8.
$GPGSA,A,3,10,07,05,02,29,04,08,13,,,,,1.72,1.03,1.38*0A The output of these two applications are developed
$GPGSV,3,1,11,10,63,137,17,07,61,098,15,05,59,290,20,08,5 on single middleware will the main objective
4,157,30*70
$GPGSV,3,2,11,02,39,223,16,13,28,070,17,26,23,252,,04,14,
186,15*77

Fig.7 First Lead ECG Data Signal
Fig.9 output
Fig.8 Second Lead ECG Data Signal

REFERENCES
The signal outputs in two applications are shown in [1] Shahryar, R, dkk (2009): "Intelligent Mobile Health
fig.10 Monitoring System (IMHMS). International Journal of Control
and Automation. Vol 2. No. 3.
[2] Heikkila, J. Silven, O. (2000): "A Real-Time System
for Monitoring of Cyclists and Pedestrian". Infotech Oulu and
Department of Electrical Engineering.
[3] Kumar L., Arunachalam P, Sandhya S (2009):
"Biometric Anti- theft and Tracking System for mobiles -
BATS". International Journal of Recent Trends in Engineering.
[4] Montaser N. Ramadan, Mohammad A. AI-Khedher and
Sharaf A. Al-Kheder (2012): "Intelligent Anti-Theft and
Tracking System for Automobiles". International Journal of
Machine Learning and Computing, Vol. 2, No. 1.
[5] Abdurohman, M (2013): "Mobile Tracking System
Based on Event Driven Method. International Conference on
Mechatronics and Industrial Informatics. Guangzhou 2013.
[6] ETSI TS 102 689 VI.1 .2, "Machine-to-Machine
communications (M2M); M2M service requirements," 2011.
[7] ETSI TS 102 690 vl.l.l, "Machine-to-Machine
communications (M2M); Functional architecture," 2011. [II]
ETSI TS 102 921 vI. 1.1, Machine-to-Machine
communications (M2M); mIa, dIa and mId interfaces," 2012.
[8] 3GPP, "System Improvements for Machine-Type
Communications," TR 23.888 VO.5.1. July 2010
[9]Telecommunications Industry Association
http://www.tiaonline.orgltags/m2m [IS] OneM2M [online]
[10]http://www.onem2m.org/

MOBILE ROBOTS FOR SEARCH AND RESCUE
Abstact:
Mobile robots are integrated into a groundwork and
rescue team as tools for looking victims in dangerous areas
that is harmful for human, on offer the perception information
for map building, and on follow the human entity throughout
the mission. The teleoperated management and also the autonomous
path following is enforced on the robots for the semiautonomous
navigation within the simulated firing state of affairs. This
paper focuses on the automaton entities within the PeLoTe project
(Building Presence through Localization for Hybrid
Telematic Teams). The design of the code
integration for mobile robots into the PeLoTe system and also the
experimental results ar bestowed
Introduction:
In the search and operation, mobile robots square measure
presently developed so as to help humans in dangerous
and risky tasks throughout the mission. As a team, the
corporation of human and golem are often managed by a
remote organiser, United Nations agency is found during a safe remote place
outside of the country. This hybrid telematic team is
presented in As a team member, the task of a mobile golem is to
explore the unknown space and also the dangerous space that's
not accessible or risky for the human. By victimization the
appropriate perception instrumentality, the mobile golem will
localize itself and conjointly senses the encompassing within the dark
area, during which the human has low visibility. Meanwhile,
the human entity equips the sensors, a knowledge process unit
and alittle show. He can even localize himself and sense
the surrounding within the similar method because the robots.
The mobile robots have 2 main functions throughout the
mission; one is to supply the sensing element knowledge for the mapping
module and another one is to follow the human entity.

These robots square measure supported constant management construct victimization

on the design of the microcontroller and also the PC104.
Nevertheless, the computer code interface protocols of those
robots don't seem to be identical. Thus, the mixing methodology
of the golem computer code into the PeLoTe system is meant.
This paper describes the tasks, features, functionalities,
and the integration methodology of the mobile robots in
the PeLoTe project (Building Presence through
Localization for Hybrid Telematic Teams) funded by the
European Community underneath the IST-program: Future and
Emerging Technologies. The experiments were performed
Mobile Robots:
A mobile automaton is Associate in Nursing automatic machine that's capable of locomotion.
A spying automaton is Associate in Nursing example of a mobile automaton capable of
movementin a very given surroundings.[1]
Mobile robots have the aptitude to maneuver around in their surroundings and aren't fastened to
1physical location. Mobile robots will be "autonomous" (AMR - autonomous mobile
robot) which impliesthey're capable of navigating Associate in
Nursing uncontrolled surroundings while not the necessity for physical or electro-
mechanical steerage devices. or else, mobile robots will believesteerage devices that enable them
to travel a pre-defined navigation route in comparatively controlledarea (AGV -
autonomous target-hunting vehicle). in contrast, industrial robots ar typically more-or-less
stationary, consisting of a articulated arm (multi-linked manipulator) and gripper assembly
(orfinish effector), connected to a hard and fast surface.
Mobile robots became a lot of commonplace in business and industrial settings.

Hospitals arevictimization autonomous mobile robots to maneuver materials for several years.
Warehouses haveput in mobile robotic systems to with efficiency move materials from
STOCKING shelves to order fulfillment zones. Mobile automatons also are a significant focus of
current analysis and nearly eachmajor university has one or a lot of labs that specialize in mobile
robot analysis.[2] Mobile robots also are found in industrial, military and security settings.
Domestic robots ar shopper merchandise, as well as recreation robots and people that
perform bound responsibilities like vacuuming or husbandry.
Robot Tasks:
In the modern times, most of the rescue groups solely need teleoperated robots, exception ar the
military that claim autonomous robots for map building or exploration ofenvironment. hearth
fighters ar in some cases cautious about the dependableness and flexibility of autonomy, for
these cases teleoperated robots ar wished. In any case, the robots thought to give autonomous

low-level actions and emergency handling strategies so as to satisfy the requirement of lustiness.
E.g. the mechanism finds the methodback just in case of communication interruption. These
requirements are going to be addressed . Desired autonomies for robots in an exceedingly farer
future are:
give knowledge for map building
Explore the atmosphere
explore for wounded folks in wide areas
Carry equipments
Robot Description:
There exists solely few costly rescue robots within the
market and their options don't seem to be complete consistent with the
project needs. Their functionalities square measure restricted
according to the mechanical structure and therefore the sensors on
board. Thus, the exploited mobile robots within the PeLoTe
project square measure designed supported the prevailing mobile automaton
technologies. As shown in Figs. 1 and 2, the Mobile
Experimental automaton for Locomotion and Intelligent
Navigation (MERLIN) is controlled by 80C167 atomic number 24 sixteen bitprocessor
and PC104 [2]. 2 motors square measure the servo
steering motor and dc driving motor for the automaton
orientation management and therefore the automaton speed management,
respectively. The equipments on board square measure encoders,
ultrasonic sensors, infrared sensors, 3-axis magnetic
compass, gyroscope, optical maser scanner, beacon system,
bumpers, web camera, WLAN, and white LED lamps. The
mobile robots offer device information for the upper layer
software interface. These information square measure the driven distance on
the left wheel and therefore the right wheel, the driving speed, the
angle of the automaton heading, absolutely the roll, pitch, yaw
rotational angles, the battery level detector and therefore the
obstacle distance detection.
MERLIN is remotely controlled via the programmed
client and server mistreatment java software package supported TCP/IP
communication. The wireless local area network is employed for wireless
data communication between user user interface and therefore the automaton. The
control commands and measure information from sensors square measure
transmitted between the shopper and server endlessly.

In semi-autonomous teleoperated management, the user

commands the automaton by employing a joystick or arrow cursors
as to steer the front wheels and to propel the rear wheels.
Recom Data Flow
Software Integration:
The overall software system design of PeLoTe system is
described in the detail of the mechanism software system
integration is represented during this section. From the developed
GUI and hardware interface of the mechanism as explained in
the previous section, associate interface layer is needed to
mount 2 or additional robots into an equivalent knowledge and
command protocols. From totally different protocols into one

common protocol that's recognizable for all robots,

ro the
Robot Entity management Module (RECoM) is outlined,
designed, and developed victimization java RMI for the server
communication. The RECoM may be a system
consisting of many modules. the information flow among these
Types of robots :

Different Rescue Operations:

The Problem:
There exists only few expensive rescue robots in the market and their features are not
complete according to the project requirements
requirements.
Complexity is more.
Expensive.
Future Work:
By adding encryption and decrtption algorithm ,we can make the circuit more safe and secure.
By adding SOC (System on chip ) we can reduce complexity of robot, as well as cost .
Conclusion:
This paper presents the mobile robots used in search and rescue operations to support the hybrid
telemetric teams in PeLoTe project. As a member of the team, the robots have functionalities for
searching victims, investigating dangerous area or low visibil
visibility
ity area, providing sensor data For
mapping and following human.
human.Effective for the large-scale
scale complicated search and rescue
missions.

Design Of A RF Transmitter And Receiver For A Guided Vehicle

Ch. Naveen Kumar1, Abdul Ahad2
1(
M-Tech, Department of ECM, KL University, India)
2(
Asst. Professor, Department of ECM, KL University, India)
ABSTRACT: The main motivation of this paper is to tank, returns to any angle on its axis, moving forward and
develop a robotic vehicle using RF technology for remote reverse left/right, running into opposite direction[3]. The
operation added with ultrasonic sensor for supervising Combat robot is radio controlled, auto-powered, and has all
purpose. The robot consisting with ultrasonic sensor can the commands like a general car. A pair of laser gun enclosure
transmit without wire i.e. wireless real time supervising has been installed in it, so that it is used as a fire on enemy
information. This is type of robot can be utilized for sighting remotely whenever required, this is not possible until a
purpose in battle fields. In case of ARM 7 series wireless camera is installed. Wireless camera can able to send
microcontroller is only used for the desired operation. At the real time video as well as audio signals. Which would be seen
transmitting side utilizing the MEMS, commands are sending on a remote monitor and control can be taken accordingly. It
to the receiver to control the running of the robot either to can silently enter into enemy area and send us all the
move forward/backward and left/ right etc. This new Combat information through its tiny camera eyes. It is designed for,
robot is a radio controlled, it has got two barrel enclosure fighting as well as suicide attack [4]. The Main Objectives of
through LASER can be evoked, ultrasonic sensor in a using robot are man dares not hazard Robots have generally
synchronization with the enclosure can resolve up/down, been forced to use in surroundings that are too dangerous for
left/right up to a secure eliciting limit enclosure and man [5].
ultrasonic sensor mechanism has been added which has all The motivation is to implement a model which is
the function like a tank, turning to any angle on its axis, efficiently used to reduce the terrorist actions and affects. To
moving towards and backward then turning left/ right, efficiently exhibit and achieve reliable long distance
running instantly into opposite direction. The RF transmitter communication it is necessary for the open area of research to
acts as a RF remote control for receiver and also transmitter. robotics as well as other technology domains. As the
importance in case of robotics goes forward to grow, robots
Keywords- Combat Robot, MEMS, LASER, RF transmitter are progressively being united into everyone life. The
and Receiver. responses of this integration are final users possessing less and
lower in technical knowledge of the applied technologies [7].
I. INTRODUCTION The congregation robots are unable to waffle to tread
The world-wide focus on terrorism and protection may even the dreaded terrain of battlefields. Their utilization in
have start up following the 9/11 attacks in USA. The high risk countries like Afghanistan and Iraq wars make us amazing if
of terrorist motivation can possibly never be reduced, but robots have become so efficient Battle robots of different
sensible steps can be taken to reduce the danger. The problem shapes and sizes were developed to withdraw landmines,
here is how critically the governments take the scourge of search for criminals obscuring in caves, look for bombs below
about terrorism. We cant continue to trust for the better and cars and in building, for spying and what not. These robots
dismiss the lessons [1]. Our Prime Minister also declared in a can be controlled by users such as humans or mainly by
in his words during past year that earlier Indian soldiers will military people [8].
have robots serve them to counter terror attack. We are yet to
hear more on that idea. It struck a thought in our mind, why
II. BLOCK DIAGRAM REPRESENTATION
we are unable to make a robot to tackle such type of position. Fig-1 represents the block diagram of the RF transmitter
This combat robot is an extended and effective version of and receiver in either way.
former robot [2].
The Combat robot (Battle Robot) is radio controlled, the
two barrel it has enclosures through bullet can be elicited,
radio camera on synchrony with the gun enclosure can rotate
up/down and left/right up to a safe eliciting limit. Gun
enclosure and mechanism of camera has been installed on
recent spots robot vehicle, which has all the functions like

Fig-1: (a) Transmitter
Fig-1: (a) represents the transmitter block of the

proposed system. This transmitter comprised of an ARM
Microcontroller LPC2148 as a basic element and MEMS
sensor followed by a RF transreceiver which can connect to Fig-1(d): LPC2148 Development board
receiver block and LCD is used for displaying purpose.
LPC2148 is one of the most widely used IC of ARM-
7 families which is manufactured by NXP Semi-conductors
formerly known as Philips. It uses the RISC Architecture and
Harvard Architecture. Harvard architecture has separate bus
line for both Program Memory and Data Memory which is
used for fast data Transfer. It has many features like it is
having 64-pin High-Performance ARM Microcontroller. It has
two Ports and each port consists of 32pins. In that 3pins are
reserved for future applications and 16Pins are hidden which
is used to boost up the remaining pins operation. Pins are
available up to nine edge or level sensitive external interrupt.
Fig-1: (b) Receiver There are only 45 GPIO Pins which can use for I/O
applications. .It has 512 kb of flash program memory and
Fig-1: (b) represents the Receiver block of the proposed 32kb of Static RAM Data Memory. It has two UARTs which
design. In this also LPC2148 place a cordial role. The all sub is used for Serial Communication Between two modules. It is
blocks are connected to this micro controller. Here, for and Interface width of 128bit which enables the operating
sensing ultrasonic sensor is used and the remaining frequency of 60MHZ. It is a low power Consumption with
components are laser light, buzzer, motor driver and RF Operating voltage of 3.3v.It has four stages of Pipe lining
transceiver.
which is used make an sequence flow of execution. Crystal
Operating frequency of up to 60MHz because it consists of
a. LPC2148:
Phased Lock Loop(PLL), Two I2C serial interfaces, SPI serial
interfaces, 32-bit timers, PWM unit, Watchdog Timer, Real
Time Clock with optional battery backup, Brown out detect
circuit. It even has In-System Programming (ISP) and In-
Application Programming (IAP).
Advantages of RF:
Do not exhibits or need of line of sight.
This could not block by common things: It can infiltrate
more solids and pass even through on walls.
More retentive in case of range.
It is not sensitive to the light
It is cannot closed to sensitive to the surrounding
changes and also weather conditions.
Fig-1(c): Pin diagram of LPC2148.

Precautions to taken:
Interference: communication devices utilizing customized
frequency scanners, wrist radios, wireless phones, and also
personal boundaries can interfere with transmission.
Security Deficiency: more sluttish to "eavesdrop (listen
in)" on transmissions during the signals are circulate out in
space rather than confined to a wire.
More costly than infrared type.
Federal Communications Commission (FCC) should be
given the licenses for proper utilization of some products.
Reduction in speed: The data rate during the transmission is
less than infrared and wired transmission.
Fig-3: Run process of compiled code
III. FIRMWARE IMPLEMENTATION
Firmware implementation deals in programming the
microcontroller, for which OS(Operating System) to be used
so that it can control the operation of the ICs utilized in the
implementation. In the presented work, we have taken the
Orcad type design software for PCB circuit design description
and Keil v4 software development tool is used to write and
compile the source code, the code has been written in the C-
programming language. Here, the Flash magic programmer
has been utilized to compose this compiled code into the
microcontroller. Software tools involved are Flash Magic
Orcad, Orcad, Keil Vision4 is used for drafting the
schematic diagram, it has been mentioned in above. Keilv4,
Flash magic are the two software tools are utilized to program
microcontroller.
Fig-4: Dump process
Fig-2: Source code compilation
Fig-5: Transmitter implementation

Fig-6: Receiver implementation
Working:
The main aim of this work is to navigate the robotic Fig-7: Hardware
vehicle and fixing the target by using RF technology. MEMS
sensor is connected to the Microcontroller. The key data is
transmitted through UR24A transceiver module to the
receiving microcontroller. MEMS sensor is having the four
ways of directing forward, reverse, left and then right. This
commands data is transferred to the receiver kit with the help
of RF transceiver. At the receiver kit ultrasonic sensor find out
the targets and firing laser fix the target through LASER beam
and then fire by alerting buzzer. The code was written in
EmbeddedC language using KEIL compiler and for dumping
the code into the microcontroller we use the relevant hex file
using flash magic software.
IV. HARDWARE IMLEMENTATION AND

RESULTS
In hardware implementation we assemble the circuit on the Fig-8: sensor working up/down
PCB as shown in above circuit diagram. After assembling the
circuit on the PCB, check for the proper connections made
before switching ON the power supply. With the help of this
circuit diagram and the necessary precautions taken we had
implemented RF based target firing Robot. We controlled the
speed by using RF technology.
Fig-9: with ARM-7 MEMS sensor

face terror attacks and to get more secured at the border as

well as high density areas its wise to maintain a world class
military technology in assisted with belligerent needs. In this
paper the proposed combat robot can uses the technological
power for the efficient facing of the terror attacks which can
give more precise response than man power
REFERENCES
[1]. Khamis, M. Prez Vernet, K. Schilling, A Remote Experiment On
Motor Control Of Mobile Robots, 10thMediterranean Conference on
Control and Automation MED2002.
[2]. T. DOrazio, M. Ianigro, E. Stella, F. P. Lovergine, and A. Distante,
Mobile robot navigation by multi-sensory integration, in Proc. IEEE
Int. Conf. Robot. Autom., May 1993, vol. 2, pp. 373379.
[3]. Pete Miles & Tom Carroll, Build Your Own Combat Robot, (2002).
Fig-10: operating with MEMS sensor [4]. Md. Suruz Miah & Wail Gueaieb, An Intelligent Mobile Robot
Navigation Technique Using RFID Technology[3], IEEE
The communication between different modules in the TRANSACTIONS ON INSTRUMENTATION AND
MEASUREMENT, VOL. 57, NO. 9, SEPTEMBER 2008
design is properly done without any interference of any other [5]. Asaro,P. How just could a robot war be?, Frontiers in Artificial
modules and is done perfectly to meet the all system Intelligence and Applications, 75, 50-64.
[6]. S. Y. Harmon & D. W. Gage, Current Technical Research Issues of
specifications and requirements as the need to full fill the Autonomous Robots Employed In Combat, 17th Annual Electronics
project. Software tools like Keil vision Simulator for writing and Aerospace Conference.
the code in the language of C or C++ or Embedded C , Flash [7]. T.Iwamoto, H.Yamamoto, Mechanical Design of Variable
Configuration Tracked Vehicle, J. of Mechanical Design , 112 , 289-
magic to dump Hex file into the microcontroller, Orcad Lite 294, 1990.
software for the schematic diagram are been used to develop [8]. Iovine John, McGraw -Hill, Second Edition, Robots,Androids, and
Animatrons 12 Incredible Projects You Can Build. Chapter-5 (DTMF
the software implementation before realizing the hardware. IR Communication/ Remote Control system) pp. 70-75.
Circuit is implemented in Orcad and also in the hardware i.e., [9]. Robert L.Boylestad and Louis Nashel sky, Electronic Devices and
Circuit Theory, 8th Edition, 2006 .
on the microcontroller development board. The performance
has been verified both in software simulator and hardware Authors Profile:
design. The total circuit of the project is completely verified
functionally and is following all the software applications. Naveen was born in Prakasham District,
Andhra Pradesh, India in 1992. He
received B.Tech Degree in Electronics
and Instrumentation Engineering from
JNT University, Hyderabad. Presently he
is pursuing M.Tech (Wireless
Communications) degree from K.L.
University, Guntur A.P., and India. His research
interests include real time applications in Embedded
Wireless Technology.
Abdul Ahad was born in Guntur City, A.P., INDIA in

1975. He received the M.Tech. Degree in computer
science engineering from Jawaharlal Nehru Technical
Fig-11: Working of hardware University, Kakinada, A.P., INDIA in 2009. He is
The figure (11) depicts the hardware implementation of the currently pursuing the Ph.D. degree in computer science
proposed design in the desired manner. It shows the MEMS engineering at Acharya Nagarjuna University, Guntur,
sensor and ARM7 controller which can perform the desired A.P.,INDIA. Since 2011, he has been an Assistant
operation in better way. Professor with the Electronics and Computers
It can be concluded that the design implemented in Engineering Department, KL University, Vaddeswaram,
A.P., INDIA. His research interest includes the data
the present work provide portability, flexibility and the data
mining and Data warehousing, Operating Systems,
transmission is also done with low power consumption. Software Engineering, and wireless Sensor Networks.
V. CONCLUSION
In present days India is suffering from massive terror
attacks, bomb explosions at many of the lavish resorts. So, to

Human Encroachment Perception In Light Of Remote

Correspondences
K. Charmila (1) P. Vikram (2)

1(M.Tech-WCSN, K L University, India)
2(Assistant Professor-ECM, K L University, India)
Abstract: Wireless operations permit services, such Sensors that are interfaced with a microcontroller that
as long-range communications, that are impossible dependably screens the House and when interruption
or impractical to implement with the use of wires. has been happened it makes an impression on the
Wireless Communications marks a prominent role predefined client utilizing GSM and by utilizing
in the era of modern technology, which facilitated Voice Recorder and Play back module and speaker
the capability of data transfer between devices lingerie to the Neighbours to alarm them about the
without any physical connections. This lifted the interruption.
existing technology one step higher.
By utilizing GPS joining to the spot where
The break-ins are increasing day-by day
interruption is conceivable we can likewise discover
which leads in financial deprivement. This work
the area. We can likewise secure the Intruder in the
presents an idea to detect human intrusion based on
house by utilizing engines which naturally shuts the
Wireless Communication. There is a substantial
entryway and we can open the entryway by sending
need to control these intrusions. This work proposes
message through GSM to open the entryway.
a novel approach to detect the intrusion and inform
Gatecrasher pictures are caught utilizing cam and
to the corresponding authorities.
pictures are put away in web utilizing a little PC
In this several sensors are used to detect
Raspberry Pi.
the intrusion and message is sent to the predefined
user using GSM and location of that intruder is The Technology that is used in this project is
found using GPS. When the intrusion is detected an Wireless Communication. Wireless Communication
alert sound will be given to make an indication to the transfers information or data between two or more
neighbours. Images of the intruder are captured by points that are not connected by an electrical
camera and stored in Web using Raspberry Pi. conductor. Wireless operations permit services, such
Keywords: GSM, GPS, Sensor, Raspberry Pi. as long-range communications that are impractical to
implement with the use of wires. Information can be
I. Introduction transferred in both short and long distances. Mostly
wireless technologies use Radio Waves. Applications
Presently a-day's the same number of
of Radio Wireless Technology are GPS units, garage
individuals are occupation holders nobody will arrive
door openers, radio receivers, satellite television.
at home in light of that Human Intrusions are
expanding and it is likewise hard to discover the II. Methodology
interlopers. Due to these Intrusions, the burglaries are
When supply is switched on and Ultrasonic
expanding where that can't be diminished or
Sensor detects the intruder the door will be
recovered. So as to defeat these issues this paper is
automatically closed. Automatically door closing is
proposed. As now a day's most extreme houses are
done the DC motor which is connected to L293D. The
having Home Automation Technique, we utilize

location of the Intrusion will be found by GPS and the connected to the internet attached with a motion
location is sent to the predefined user through GSM. detector and Web camera. Motion detector is
continuously monitored and when HC detects the
By using SMS Google Maps the location
motion it immediately starts a video call with the
can be seen in Android Mobile Phone. By using Voice
wireless computer the user is currently at which must
Record Playback Module we will record an alert tone
be connected to the Net and user should be logged on
for intimating the neighbours that intrusion is
to the IPTEL Session Initiation Protocol Registrar
occurred. Speaker is attached to voice module and it
Server[10].
sounds the previous recorded alert tone. Raspberry Pi
This paper proposes a automated intrusion
a small computer is used to capture the images of the
detection system which uses alarm technique that
intruder using the USB Camera. The captured images
alerts the human security. Closed Circuit Television
can be seen in the monitor using web.
(CCTV) is used for video surveillance. This provides
a means of identifying the culprit in case the intruder
is able to escape before a response team arrives [1].
Power Supply GPS
IV. Proposed Technique
This proposed strategy does not require any
Ultrasonic Voice Play Back systems administration method to impart. By
Sensor Module connecting GPS Module to the spot where
interruption happens the area can be discovered even
L293D Microcontroller Speaker
the interloper moves far from home.
Raspberry Pi As present days each one has web office in

M
their mobiles the area can be seen regardless of the
fact that the client is not accessible close to
GSM Camera
interruption. The area data is sent to the client through
GSM and Door is additionally shut utilizing L293D
driver. Pictures are caught utilizing cam alongside
Fig 1: Block Diagram Raspberry Pi. The images of the intruder can be in PC
using the IP address of the PC. The Web page that
III. Previous Techniques
you want should be predefined so that it appears in the
This paper proposed a home security system way that we had selected. We can set the date time
using Microcontroller. In this door locking system is and year for the pictures of the intruder so that the
used which is password Protective and has LED based intrusion can be rectified easily as we will not be there
resistive screen input panel. For displaying 16x2 LCD in the Intruded area.
is used. IR Sensors are used for detection of intruder
at the windows and doors in the night time and when
the people are away. Fire alarm system uses LM35
Temperature sensor for detecting changes in
[3].
temperature and raises an alarm
This paper proposes a home surveillance

system that consists of a home computer (HC) that is

V. Results References
[1] Awodele Oludele, Ogunnusi Ayodele, Omole Oladele, Seton

Olurotimi Design of an Automated Intrusion Detection System
incorporating an Alarm JOURNAL OF COMPUTING,
VOLUME 1, ISSUE 1, DECEMBER 2009, ISSN: 2151-9617.
[2] Xiangjun Zhu, Shaodong Ying and Le Ling, Multimedia
sensor networks design for smart home surveillance, Control and
Decision Conference, 2008, Chinese, 2-4 July 2008, pp. 431-435.
[3] Nikhil Agarwal and G. Subramanya Nayak, Microcontroller
based Home Security System with Remote Monitoring Special
Fig 2: Practical Kit
Issue of International Journal of Computer Applications (0975
This Kit shows how the connections are done 8887) International Conference on Electronic Design and Signal
Processing (ICEDSP) 2012
between the modules, controller and other
[4] Mrs Smita Parte1 ,Ms Noumita Dehariya, A Review on
Components. Pictures of the intruder are taken using Intrusion Detection techniques for Wireless Local Area Network
Camera attached to Raspberry Pi and stored in the SSRG International Journal of Computer Science and Engineering
web. The images of the intruder are seen in using (SSRG-IJCSE) volume1 issue 8 October 2014
[5] Mikko Nieminen, Tomi Raty, and Mikko Lindholm, Multi-
web. As the connections are connected from
Sensor Logical Decision Making in the Single Location
controller to the remaining components single Surveillance Point System, Fourth International Conference on
external power supply is enough to the Systems, France, 1-6 March 2009, pp. 86-90.
Microcontroller. Supply for the camera is taken from [6] Ying-Wen Bai, Li-Sih Shen and Zong-Han Li, Design and
Implementation of an Embedded Surveillance System by Use of
the Raspberry Pi and the Supply to Raspberry Pi is
Multiple Ultrasonic Sensors, The 28th IEEE International
taken from the lap or system that we are using. A Conference on Consumer Electronics, Las Vegas, Nevada, USA,
cable is used to open the web page that had been used 11-13 Jan. 2010, 11.1-3, pp. 501-502.
for storing the images and using HTML default web [7] Ghanshyam Prasad Dubey, Prof. Neetesh Gupta, Rakesh K
Bhujade, A Novel Approach to Intrusion Detection System using
page is created.
Rough Set Theory and Incremental SVM International Journal of
Soft Computing and Engineering (IJSCE) ISSN: 2231-2307,
VI. Conclusion
Volume-1, Issue-1, March 2011
[8] Hai-Wen Zhao, Hong Yue, and He-Gao Cai, Design of a
This paper finally concludes that there is no
Distributed Ultrasonic Detecting System Based on Multiprocessor
need of learning various algorithms and transfer data for Autonomous Mobile Robot, Proceedings of tbe 2007 WSEAS
between nodes. It is enough if we are having the basic Int. Conference on Circuits, Systems, Signal and
idea of what we are using and how to use them. So Telecommunications, Gold Coast, Australia, January 17-19, 2007,
pp. 59-64.
that it is very easy to implement. It can be applicable
[9] Zi-LI Xie and Zong-Han Li, Design and implementation of a
in places like closed homes, Banks, Industries. home embedded surveillance system with ultra-low alert power,
IEEE Transactions on Consumer Electronics, Feb. 2011, pp. 153-
VII. Future Scope 159.
[10] Rathika Rajaravivarma Prototype model for real time home
As we are not using any nodes here there will surveillance system over the internet Proceedings of The 2008
be no communication between several members to IAJC-IJME International Conference ISBN 978-1-60643-379-9
alert about the intrusion. This model is also used for

closed systems only. Intrusions will be in open areas
also so in order to reduce those intrusions some
simple technique should be implemented.

A survey on elliptic curve cryptography
K. Bhanu Prakash Reddy

T. Narendra babu
Department of Electronics and Computer Science Engineering
K L University, Guntur, A.P., India
ABSTRACT: used for encryption and decryption. It also

provides digital signature methods for
In this, a cryptographic method of encryption. ECC can provide higher security and
hiding and securing data or images for the better performance than the well-known RSA
purpose of secure communication is discussed. technique at a smaller key size. In this way it can
Cryptography is a method of encrypting and reduce processing time and computational costs.
transmitting data in a hidden and secure form, so ECC makes use of elliptic curve defined over a
that no other third party person can read it and finite field which restricts the variable and
process it. It analyzes protocols and provides coefficients to its elements.
various aspects for the sake of information
security. Elliptic Curve Cryptography is the 2.LITERATURE REVIEW:
most powerful cryptographic technique being
used widely. It is a public key encryption I. Many researchers have been working on
technique which is capable of creating much various Elliptic Curve Cryptographic methods to
efficient cryptographic keys. The method is ensure the best and reliable protocols for
based on elliptic curve theory. The use of this wireless communication networks. Ray C.C.
technique will provide greater security using Cheung (2005), has proposed Customizable
fewer bits and requires less bandwidth than other Elliptic Curve Cryptosystems[2]. In this, a
protocols. method for producing hardware designs for
elliptic curve cryptography (ECC) systems over
Keywords: Elliptic Curve Cryptography (ECC), the finite field GF(2 ), using the optimal normal
discrete logarithm Elliptic Curve (EC), public basis for the representation of numbers is
key cryptography. proposed. This field multiplier design is based
on a parallel architecture containing multiple bit
1.INTRODUCTION: serial multipliers; by changing the number of
such serial multipliers, designers can obtain
The progress in wireless technologies implementations with different tradeoffs in
and communication systems has provoked new speed, size and level of security. To facilitate
security questions because the wireless data performance characterization, they have
transfer is not that much secure and can be easily developed a parametric model for estimating the
get exposed to the hackers or unauthorized number of cycles for the generic ECC
users. To meet the increasing needs for secured architecture. ECC has been adopted to a wide
wireless communication, advanced and high spectrum of applications from digital certificates
secure protocols are to be developed that in webserver authentication [12] to embedded
requires low power and have less overhead. processors [13] in wearable devices.
Some measures like authenticity, confidentiality
and integrity are to be taken to obtain reliable
and secured wireless communication[1]. The
transmitter and receiver should authenticate
one another using a session key, to encrypt
the data that is being sent over the channel. The
Elliptical Curve Cryptography is a public-key
encryption method in which separate keys are

Step2. A encodes the sw image as sw P = (x, y)

= (g5,g3).
Similarly others points are calculated using

equation (4) with generator g.
Step3. A choose a random number K and
produce the
Ciphertext Csw =[k G,Psw +k PB ] and
sends this ciphertext sw C to B.
Step4. To decrypt the sw image, B computes nB
k G.
Step5. B again computes
Psw+KPB-nBKG=Psw-K(nBG)+KPB
Fig: Interactions between different =Psw-KnB+KnB=Psw.
operations In other words, we can say B picks the first co-
ordinate KG of Csw multiply that with his
In the above figure the key size for our ECC private key and then subtract this form the
architecture, is a characteristic of the field. In second point .Psw=KPB.
this subsections under cover field multiplication, To achieve higher security of digital image RSA
field inversion, field squaring, point scheme with MRF and ECC proposed for image
multiplication, point addition and data encryption [9]
embedding respectively.
The scheme consists of the important
II. In Public Key Cryptosystem Technique algorithms namely encryption algorithm which
Elliptic Curve Cryptography with Generator g is used to create every 2-D image pixels of the
for Image Encryption (2012), an efficient original image into the ECC points in a finite
technique of transmitting the image securely abelian group over GF(2m). These ECC points
using ECC is proposed. It has been shown convert into cipher image pixels at sender side
though the image encryption by ECC the image and decryption algorithm is used to get original
can be transmitted secretly and efficiently and image within a very short time with a high level
can be recovered the same at the receiver end. of security at the receiver side. The Mixed image
An image encryption for secure internet in element encryption using elliptic curve
Multimedia application was proposed in [6]. cryptography has been proposed in [10].
Image encryption and decryption:
III. In Encryption using Elliptic Curve

Cryptography using Java as Implementation
tool (2014), the implementation of ECC for
encryption/decryption and authentication
process is discussed, using JAVA as the
implementation tool[4]. It is worth noting that
brute force attack on ECC is infeasible due to
the discrete logarithm problem it possesses.
Fig: Encryption Model Kristin Lauter as provided an overview of ECC
The above figure shows that ECC can be used for wireless security [11].
for both encryption and decryption. Imagine that
user A want to encrypt a sw image for the user
B, and then the following steps are involved. Image encryption using elliptic curve
cryptography, (2013) deals with the encryption
Step1.Take any RGB color image as sw. of an image using ECC with generator G. ECC
points converts into cipher image pixels at

sender side and decryption algorithm is used to on the performance benefits within the
get original image within a very short time wireless environment by victimisation ECC
with a high level of security at the receiver rather than the normal RSA
side. cryptosystem.The customary standard
specifications for public key cryptography is
outlined in . In the paper as projected by
Elliptic Curve Cryptography with Hill Jaewon Lee presents three algorithms to
Cipher Generation for Secure Text perform scalar multiplication on European
Cryptosystem, 2014, proposed an idea in which Union defined over higher characteristic
hill cipher is generated with Elliptic Curve finite fields like OEA (Optimal Extension
Cryptography to provide better security and
Field). Liu Yongliang showed that Aydos
proper security coverage. Hill Cipher is harder
to break due to its linearity and ECC is a smaller et al.' s protocol is at risk of man-in the-
key size algorithm, which provides fast middle attack from any aggressor however
computations. not restricted on the within attacker.
Achieving Authentication and Integrity
using Elliptic Curve Cryptography Architecture
V. SECURITY MEASURES:
(2013), presents the security protocol for
securing data between two communicating Some of the security measures that are
nodes in mobile adhoc networks. The proposed to be taken to obtain secured communication are
scheme would solve the issues of space in terms as follows:
of key management in adhoc networks as well as
it will provide integrity and authentication for Authentication:
the communicating entities.
In Use of elliptic curve cryptography In order to eliminate the fraudulent
for multimedia encryption (2013), the ECC is activities in wireless communication, mutual
used to perform encryption along with authentication of users and servers is a must.
multimedia compression, and two ECC-based The user and the server should authenticate one
encryption algorithms are introduced and another before the communication takes place.
applied before and during compression. The first
algorithm performs selective encryption on the Confidentiality:
transform coefficients during compression, With the advancement of technology
whereas the second algorithm achieves there are commercially available scanners that
perceptual encryption based on selective bit- can easily intercept radio signals traveling over
plane encryption before compression. the air in wireless communication systems[5].
Speeding up Video Encryption using Although digital systems are advantageous over
Elliptic Curve Cryptography (2013), proposed analog systems in terms of security, they are also
a computationally efficient and secure video prone to eavesdropping. To avoid this problem,
encryption algorithm. This makes secure video subscribers and servers must agree on a key that
encryption feasible for real time applications can be used to encrypt information. Key
without any extra dedicated hardware. It uses agreement is usually the last part of the
RC5 for encryption of the DCT coefficients and authentication process between subscribers and
ECC for small key sized generation. servers. Session keys must differ for each
transfer to enhance security. There may be
IV. In the literature, several authors have tried to situations that some users may not want to
take advantage of the options of global expose their identity or other information to
organization field to deploy for security third parties. So every user can be assigned with
applications. We have printed a number of the a temporary identity without revealing the
highlights of the relevant work in this original identity.
section.Kristin Lauter has provided an outline
of ECC for wireless security [3]. It focuses

[6] Philip P. Dang and Paul M. Chau, Image

Physical requirements: Encryption for Secure Internet Multimedia
Beyond security, there are also some Application, IEEE Transaction on Consumer
necessary elements for any protocol. They are Electronics, Vol. 46, No.3 pp. 395-403,
efficiency, power consumption, computational Aug.2000.
behavior, bandwidths used, battery power and [9] Chaur - Chin Chen RSA scheme with MRF
storage. These all should be well defined in and ECC for Data Encryption, 0-7803-8603-
order to design an efficient protocol. 5/04 IEEE, 2004.
[10] Guiliang Zhu and Xiuaoqiang Zhang,
Mixed Image Element Encryption System ,
3.CONCLUSION: 9th IEEE International Conference for young
computer Scientists ISSN 978-0-7- 3398, pp.
In this paper text based elliptic curve 1995-1600,Aug.2008.
cryptosystems is implemented. Transformation [11] Kristin Lauter, "The Advantages of Elliptic
of the plaintext ASCII value by using an affine Cryptography for Wireless Security", IEEE
point is one of the contributions of this work. Wireless Communications, pp. 6 2 - 67, Feb.
Cryptography is a method of encrypting and 2006.
transmitting data in a hidden and secure form, so [12] N. Nguyen, K. Gaj, D. Caliga, and T. El-
that no other third party person can read it and Ghazawi, Implementation of
process . elliptic curve cryptosystems on a reconfigurable
This design supports various parameters, such as computer, Proc. IEEE Field Programmable
the key size and the degree of parallelism, to Technology, pp. 6067, 2003.
enable tradeoff between level of security, design [13] T.Wollinger, J. Guajardo, and C. Paar,
size and speed. It is a public key encryption Cryptography in embedded systems: An
technique which is capable of creating much overview, in Proc. Embedded World Exhibition
efficient cryptographic keys. and Conf., 2003, pp. 735744.
REFERENCES
.
[1]M.AyDos, Elliptic curve
cryptography,Dallas,Texas,0ct 30 1998,pp.1
[2] Ray c.c cheung, Customizable

Elliptic Curve Cryptosystems,student
member,IEEE, Nicolas Jean-baptiste Telle,
Wayne Luk, Member, IEEE, and Peter Y. K.
Cheung, Senior Member, IEEE,VLSI sytems
vol.13,2005,pp.1.
[3] s. Maria Celestin Vigila1 , K. Muneeswaran,
Implementation of Text based Cryptosystem
using Elliptic Curve Cryptography, 978-1-4244-
4787,2009 ieee.
[4] Ikshwansu Nautiyal, Encryption using
Elliptic Curve Cryptography using Java as
Implementation tool, Volume 4, Issue 1, January
2014.
[5] Vinod Kumar Yadav, Public Key
Cryptosystem Technique Elliptic Curve
Cryptography with Generator g for Image
Encryption, ,Int.J.Computer Technology &
Applications,Vol 3 (1), 298-302. JAN-FEB 2012.

AN SHREWD APPROACH TO IMPROVE THE ACOMPLISHMENT OF

LINEAR ALGEBRA CORES
Sumadeep Juvvalapalem M.Venkateswara Rao
Embedded systems Asst.Prof
KL University KL University
Guntur,Andhra Pradesh Guntur,Andhra Pradesh
AbstractThis paper analyze the mapping of calculations distributed as a Fortran library are still utilized as a building
experienced when illuminating thick direct frameworks and straight piece as a part of larger amount math programming dialects
minimum squares issues to a custom direct variable based math and libraries, including MATLAB, LINPACK, LAPACK[5].
processor. particularly the emphasis is on level-3BLAS calculation,
which used to enhance the force and execution proficiency and to Past work concentrated on upgrades for the internal parts
coordinate the heterogeneous framework on GPU and CPU. A level-3 of LU factorization, Cholesky factorization, and QR
BLAS used to accomplish the high productivity up to 64GFLOPS, factorization to enhance the calculation/building design co-
our augmentation to the MAC unit can accomplish the rate up to 30 outline conveys to the low-level inward bit of QR factorization
percent for the blocked calculations using single flux quantum general blocked calculations of each of the three lattice
circuits. factorization.
Keywords-Level 3-BLAS,heterogeneous system, floating point In gadgets, quick single flux quantum (RSFQ) is an
arithmetic, single flux quantum circuits advanced hardware innovation that depends on quantum
impacts in superconducting gadgets, in particular Josephson
I. INTRODUCTION intersections, to process computerized signs. Josephson
Lessening force utilization is an undeniably essential issue intersections are the dynamic components for RSFQ gadgets,
both in the installed areas, where frameworks need to work in pretty much as transistors are the dynamic components for
exceedingly confined situations, and as a rule reason semiconductor hardware. Be that as it may, RSFQ is not a
registering, where physical breaking points of innovation quantum registering innovation in the customary sense. Indeed,
scaling have made force dividers the fundamental barrier to even along these lines, RSFQ is altogether different from the
supported execution. Moreover, numerous rising frameworks, conventional CMOS transistor innovation utilized as a part of
progressively request both elite and force effectiveness on non- each day PCs:
customary platforms[2]. it is taking into account superconductors, so a cryogenic
The factorization calculations are ordinarily the first and situation is needed;
most figure serious steps experienced towards when fathoming the computerized data is conveyed by attractive flux quanta
thick direct frameworks and straight minimum squares issues that are created by Josephson intersections rather than
to a custom Linear Algebra processor[1]. transistors in semiconductor hardware;
The objective of our task is to outline elite, low-control the attractive flux quanta are conveyed by picosecond-term
Linear Algebra Cores(LACs) that understands the level-3 voltage beats that go on superconducting transmission lines,
BLAS portions straightforwardly in particular hardware.[2]. In rather than static voltage levels in semiconductor gadgets.
this paper, we sum up our outline to other level 3-BLAS,
showing that with little small scale building alterations, the Therefore the region of the quantized voltage beats that
Linear Algebra Cores can be reached out to bolster the full convey single attractive flux quanta is steady. Contingent upon
arrangement of BLAS operations without misfortune in the parameters of the Josephson intersections, the beats can be
proficiency. as thin as 1 picosecond with an adequacy of around 2 mV, or
more extensive (ordinarily 510 picoseconds) with a lower
Moreover, we outline and coordinate a particular MAC unit sufficiency;
and incorporate it and other key segments of LAC miniaturized
scale structural engineering [3][4], to achieve the execution of since heartbeats ordinarily engender on superconducting
60 GFLOPS further expansion to the MAC unit can lines, their scattering is constrained and generally insignificant
accomplish the pace up to 30 percent for the blocked if no unearthly segment of the beat is over the recurrence of the
calculations. This speaks to two requests of size change over vitality crevice of the superconductor;
current CPU architectures and a request of greatness change in 2010, the commonplace estimations of the most extreme
over current GPUs. heartbeat abundancy, generally called the IcRn item, is of the
BLAS are indicated situated of low-level subroutines that request of 0.5 to 1 mv. Rn is the typical resistance of the
perform normal direct variable based math operations, for Josephson intersection that produces the voltage heartbeats,
example, vector scaling, copying,vector spot items, straight while IC is its basic current.
mixes, and lattice duplication's. Furthermore, were initially

On account of beats of 5 picoseconds, it is ordinarily quantum registering innovation in the customary sense. Indeed,
conceivable to time the circuits at frequencies of the request of even along these lines, RSFQ is altogether different from the
100 GHz (one heartbeat each 10 picoseconds) conventional CMOS transistor innovation utilized as a part of
each day PCs.
II. RELATED WORK

B. Multiply and Accumulator
LINEAR ALGEBRA CORE WITH LEVEL 3-BLAS
In computing, especially digital signal processing, the
Lessening force utilization and expanding effectiveness is a multiplyaccumulate operation is a common step that
key sympathy toward numerous applications. It is very much computes the product of two numbers and adds that product to
acknowledged that heterogeneous and specialization are an accumulator. The hardware unit that performs the operation
essential methods to enhance both power and performance.[6] is known as a multiplieraccumulator (MAC, or MAC unit);
Yet, to outline exceedingly effective preparing components the operation itself is also often called a MAC or a MAC
while keeping up enough adaptability inside of an area of operation. The MAC operation modifies an accumulator a:
utilization is a major inquiry. Here , we show the configuration
of a specific LAC for an imperative class of computational bits,
the level-3 BLAS. We show a nutty gritty calculation/structural
engineering co-outline for mapping various level-3 BLAS
When done with floating-point numbers, it might be
operations onto the LAC and the outcomes demonstrate that
performed with two roundings (typical in many DSPs, or with
our model LAC accomplishes an execution of around 64
a single rounding. When performed with a single rounding, it is
GFLOPS (twofold accuracy) for these operations,and while
called a fused multiplyadd (FMA) or fused multiply
expending under 1.3 Watts in standard 45nm CMOS
accumulate (FMAC).
innovation with a full-hand craft and up to 50 and 10 better
regarding force effectiveness than CPUs and GPU Modern computers may contain a dedicated MAC,
consisting of a multiplier implemented in combinational
followed by an adder and an accumulator register that stores
the result. The output of the register is fed back to one input of
the adder, so that on each clock cycle, the output of the
multiplier is added to the register. Combinational multipliers
require a large amount of logic, but can compute a product
much more quickly than the method of typical of earlier
computers. The first processors to be equipped with MAC units
were digital signal processing, but the technique is now also
common in general-purpose processors.
C. Basic linear Algebra Core
BLAS (Basic Linear Algebra Subprograms) is a
specification that prescribes a set of low-level routines for
performing common linear algebra operations such as vector
Fig.1.LAC architecture is optimized for rank-1 updates to perform matrix addition, scalar multiplication, dot product, linear
multiplication[7]. combinations, and matrix multiplication They are the de facto
API for linear algebra libraries, with bindings for both C and
FORTRAN.
III. BASE ARCHITECTURE It originated as a Fortran library in 1979 and its interface
We concentrate on the open doors for coasting point was standardized by the BLAS Technical (BLAST) Forum,
expansions to an exceedingly advanced quickening agent as of whose latest BLAS report can be found on the Netlib site. This
now. The smaller scale structural planning of LAC(Linear Fortran library is known as the reference implementation
Algebra Core) is outlined in Fig.1. A Linear Algebra Core (sometimes confusingly referred to as the BLAS library) and is
accomplishes requests of greatness which has better range not optimized for speed.
utilization and proficiency in force contrasted with ordinary
broadly useful architectures[7]. Most libraries that offer linear algebra routines conform to
the BLAS interface, allowing library users to develop programs
A. Single Flux Quantum Circuits that are agnostic of the BLAS library being used. Examples of
In gadgets, quick single flux quantum (RSFQ) is an such libraries include: AMD core math library (ACML),
advanced hardware innovation that depends on quantum ATLAS, Intel Math Kernel Library (MKL), and Open BLAS.
impacts in superconducting gadgets, in particular Josephson ACML and MKL are proprietary vendor libraries, optimized
intersections, to process computerized signs. Josephson for their respective brands of CPUs. Open BLAS is an open-
intersections are the dynamic components for RSFQ gadgets, source library that is hand-optimized for many of the popular
pretty much as transistors are the dynamic components for architectures. ATLAS is a portable library that automatically
semiconductor hardware. Be that as it may, RSFQ is not a optimizes itself for an arbitrary architecture. The LINPACK

benchmarks rely heavily on the BLAS routine gemm for its

performance measurements.
V. REFERENCE
[1] G. H. Golub and C. Van Loan, An analysis of the total least squares
problem, SIAM Journal on Numerical Analysis, vol. 17,no. 1, pp. 883
893, Dec. 198
IV. CONCLUSION [2] A. Pedram, R. van de Geijn, and A. Gerstlauer, Codesign trade-offs for
high-performance, low-power linear algebra architectures, IEEE Trans.
Comput., Special Issue Power Efficient
In this paper, we presented the mapping of both inner ker- [3] Comput., vol. 61, no. 12, pp. 17241736, Dec. 2012.[3] E. Agullo, C.
Augonnet, J. Dongarra, M. Faverge, H. Ltaief, S.Thibault, and S.
nels and blocked matrix factorization problems onto a highly Tomov, QR factorization on a multicore node
efficient linear algebra accelerator. We propose two [4] enhanced with multiple GPU accelerators, in Proc. IEEE Int. Prallel
modifications to the MAC unit designs to decrease the com- Distrib. Process. Symp., 2011, pp. 932943.[4] V. Volkov and J. W.
plexity of the algorithms. We also showed how existing Demmel, Benchmarking GPUs to tune dense linear algebra, in Proc.
processing elements can be enhanced to perform special ACM/IEEE Conf. Supercomput., 2008,
functions such as divide and square-root operations. To [5] pp. 111. A. Pedram, A. Gerstlauer, and R. A. van de Geijn, Floating
point architecture extensions for optimized matrix factorization, in
demonstrate the effectiveness of our proposed extensions, we Proc. IEEE Symp. Comput. Arithmetic, 2013, pp. 4958.
applied them to the mapping of Cholesky, LU and QR [6] A. Pedram, A. Gerstlauer, and R. A. Geijn, A high-performance, low-
factorizations on such an improved architecture. We studied power linear algebra core, in Proc. IEEE Int. Conf. Appl.-Specific
both inner kernels and blocked-level algorithms and presented Syst., Archit. Processors, 2011, pp. 3542.
the resulting performance and efficiency benefits. Results [7] A. Pedram, S. Z. Gilani, N. S. Kim, R. v. d. Geijn, M. Schulte, and A.
Gerstlauer, A linear algebra core design for efficient Level-3 BLAS,
show that our extensions significantly increase efficiency and in Proc. IEEE 23rd Int. Conf. Appl.-Specific Syst., Archit. Processors,
performance of inner kernels and are effective for bigger 2012, pp. 149152.
problem sizes that fit on the LAC. Future work includes the [8] J. Kurzak, A. Buttari, and J. Dongarra, Solving systems of linear
integration of the LAC into a heterogeneous system equations on the cell processor using Cholesky factorization,IEEE
architecture next to general purpose CPUs and a Trans. Parallel Distrib. Syst., vol. 19, no. 9, pp. 11751186, Sep.2008.
heterogeneous shared memory systems, which will allow com.

IMPLEMENTATION OF HEART BEAT PER MINUTE RATE

P.Lakshmipallavi1, K.Ravikumar2
1
M.Tech Student, Department of ECM, K L University, Guntur, Andhra Pradesh
2
Assistant professor, Department of ECM, K L University, Guntur, Andhra Pradesh
ABSTRACT
With the upcoming ageing society and emerging of some newly discovered chronic diseases, the demand of
hospital nursing for people has significantly increased. For every disease knowing the heart beat per minute rate is
mandatory. Because the heart beat rate reflects the functioning of the organisms. Knowing the heart beat without a
doctor or stethoscope is a tough task for normal people. So in this paper we are implementing a wireless device to
know the heart beat rate per minute and temperature. If any of these parameters are abnormal then the data will be
automatically sent to the doctors mobile through GSM. So we can know the details whenever we want.
Keywords Microcontroller, Optical sensor, Bluetooth, Temperature Sensor, GSM MODEM.
INTRODUCTION GSM MODEM
Heartbeat is an exceptionally indispensable A modem is a GSM remote modem that

wellbeing parameter that is specifically identified works with a remote GSM framework. A remote
with soundness of human cardio-vascular system. modem proceeds as a dial-up modem. The central
The heartbeat of a person can be measured through a difference between them is that a dial-up modem
fingertip. While the heart is pulsating it is really sends and gets data through a phone line modified
pumping the blood all through the body, and that while a remote modem sends and gets data through
makes the blood volume inside the finger supply radio waves.
route to change time to time. This fluctuation is
BLOCK DIAGRAM SHOWN BELOW WITH
detected by an optical sensing mechanism. This
HARDWARE BLOCKS
design is accurate and low cost implementation. The
body temperature of the patient can be noticed by
the general purpose temperature sensor. When the
heartbeat and body temperature of a person reaches
to abnormal levels it sends an indication to the
doctor (or any other concerned person) through
SMS or call. The normal body temperature is 37oc
and normal heart Beats per Minute (BPM) is 72.
SYSTEM HARDWARE BLOCKS
Optical Sensor
It consists of IR LED (Infrared ray emitting

LED) and photo diode placed side by side. The
fingertip is place over this assembly. Fig.1.Block diagram of implementation of bpm
ARM7-based embedded system rate with GSM Module
The ARM7-based embedded system is heart HARDWARE DESIGN

of entire System. It is designed based on low power
32-bit ARM7 (LPC2148). It is RISC architecture The hardware implementation shown below
and can use oscillators, thus it is ideal to be used as and the following components are used each of them
an embedded system. The LPC2148 is an32k explained below.
instructions program buffers, 512kb of RAM, three
timers and 32 bit A/D converter microcontroller. It
is highly performance and low cost solution for
embedded applications.

temperature scope of -55 C to + 150 . This

straightforward sign is given to ADC, which
changes with the basic sign and holds progressed
microcontroller.
ARM Microcontroller Unit
As said earlier the output of the signal

conditioning circuit changes from Vcc to 0V
continuously due to heart beat. A timer is made on
and after 6 seconds of time is made off. In this time
span the number of fluctuations in conditioning
circuit is counted and this is multiplied by 10 to get
the beats per second value. The ADC is used to get
the temperature value from the temperature sensor.
These values are displayed on the LCD display.
When these values exceed the particular range data
is sent to GSM module through serial port.
Fig.2.Indication of heart beat and temperature Transceiver and Serial Port Interface
The output of the TTL logic is always 5V

and 0V. So these values are needed to be amplified
Optical Sensor Unit before being transmitted to the GSM Module for
this purpose MAX232 is used. It has incorporated
The IR LED emits the IR rays and when switched capacitor charge pump circuits to generate
there is no obstacle or less volume of blood the required voltage levels. It has in-built voltage
photo diode receives less IR rays and the current doublers and inverter circuits. MAX 232 has two
through the reverse biased p-n junction of photo pairs of such transceivers.
diode decreases. When the volume of blood is more
due to pumping of blood by heart the IR rays are GSM MODEM Module
reflected back to the photo diode and the current
through it increases. Thus there are continuous An external GSM module is connected
fluctuations in photo diode current. through serial port. Through this the extended AT
commands for writing the SMS message can be
controlled. Whenever the sensed values exceed the
predefined values a signal is sent to GSM Module
Signal Conditioning Circuit through transceiver and a serial port.
The reflected IR signal detected by the SOFTWARE DESIGN
photo diode is fed to a signal conditioning circuit
that filters the unwanted signals and boosts the Keil c
desired signals. For this purpose a general purpose
Op-Amp is maintained high for necessary
amplifications. At the output a LED is connected Keil programming is the main merchant for 8/16-
that will blink with heartbeat. Here a transistor is bit advancement devices. Keil programming speaks
used at the cathode of LED which is turned on for to everyone in more than 40 nations, from the
every 6 seconds. Through this the actual heartbeat presentation of business in 1988; the keil C51
can be obtained by multiplying the obtained value compiler is the accepted business standard and
backings more than 500 present gadget variations.
with 10.
Presently, keil programming offers improvement
Temperature Sensor devices for ARM. Keil programming makes C
compilers, full scale constructing agents, constant
bits, debuggers, test systems, coordinated situations,
Body temperature sensor LM 35 is accomplished.
and assessment sheets for 8051, 251, ARM and
This is an assembled exactness temperature sensor XC16x/C16x/ST10 microcontroller families.
circuit is little and can be set anyplace in the body.
LM execution is the voltage 35 is on body FLASH MAGIC
temperature. The external LM35 obliges no slicing
or arrangement to normal correct nesses of 1/4 C Flash magic can control the territory in the
to room temperature and 4.3 spread a wide ISP procedure for some microcontrollers gadgets

utilizing handshaking signs COM port to control the

gadget. Normally, signs handshake sticks are
utilized to control and reset, PSEN and VCC. The
careful pin utilized relies on upon the particular
gadget. Generally as this highlight, Flash Magic
remains consequently put the gadget into ISP mode
to the begin of an ISP operation. Enchantment
Streak then subsequently makes the gadget run code
toward the end of ISP operation.
RESULTS
Fig.5. Displaying temperature
After initializing of hardware, at first it

displays the room temperature. Whenever we put
the finger into the optical sensor it displays the heart
beat per minute rate along with the temperature as
shown below.
Fig.3.Hard ware
The above Figure 3 Shown is an Hardware Kit

Module of this Project. GSM Module, Bluetooth
Module, and optical Sensor is interfaced with
LPC2148.
Fig.6.Displaying heartbeat along with

temperature
For continuous monitoring of heart rate we used

Bluetooth, and it shown below.
Fig.4.intialization of hardware

Fig.7.Displaying heart rate on hard ware and CONCLUSION

mobile through Bluetooth
In this paper mechanism to continuously
This feature is available for all android mobiles monitor the heart BPM and body temperature of a
through an Bluetooth controller app. This will person is illustrated using a microcontroller
shows continuous monitoring of heart rate and interfaced with an optical sensor unit. This design is
temperature till we disconnect the Bluetooth or interfaced with a GSM module so that when the
optical sensor. And Gsm is used for abnormal heartbeat and body temperature reaches to abnormal
conditions,i.e. if the heart rate is not in between level a message is sent to the doctor. It is an
given range then only the message will send to the intrusive process and the patient can be monitored
doctors mobile through gsm modem, and it shown 24x7 without much medical assistance from
below. outsiders.
FUTURE SCOPE
The implementation will not be accurate

when the blood levels of the person to be monitored
are low. So in place of normal optical sensor
powerful IR sensors can be replaced so that the
heartbeat of those persons can also be monitored
easily. Even continuous monitoring of blood
pressure can also implemented as it is also important
parameter while diagnosing the diseases.
REFERENCES
Above figure shows the heart rate exceeds the given [1] How the Heart Works, December 26, 2011.
range then the gsm is activate and sends an message Available:http://www.webmd.com/heart-
whatever we given in the program. Example shown disease/guide/how-heartworks
in below figure,i.e. patient condition is abnormal,
[2] John R. Hampton, The ECG in Parctice
then the corresponding doctor or person will
respond. [3] M. Malik and A. J. Camm., Heart Rate
Variability.Futura Publishing
Co.Inc., sept. 1995.
[4] Heart rate measurement from finger trip.
Available: http://embedded-lab.com/blog/?p=1671
[5] K.Ramesh, S.V.Aswin Kumar, Efficient health
monitoring system
Using sensor networks, International Journal of
Scientific &
Engineering Research Volume 3, Issue 6, June-
2012.
[6] Deign of Low Cost Blood Pressure and Body
Temperature interface
Johevajile K.N Mazima, Michael Kisangiri, Dina
Machuve International Journal of Emerging Science
Fig.8. Message sent through gsm to mobile and Engineering (IJESE) ISSN: 23196378,
Volume-1, Issue-10, August 2013.
FEATURES
This system consumes low cost to develop and

having good sensitivity of measuring pulse more
over it is simple to construct and set up. Finally it
gives more accurate results.

Collision Avoidance of Trains by Using ARM7
B.Indira Priya darshini1 , K.Ravi Kumar2

1 2
M.tech Student, Department ofECM, Asst.Professor, Department ofECM,
KL university, vaddeswaram,guntur, KL university,vaddeswaram, Guntur,
Andhra Pradesh, INDIA Andhra Pradesh, INDIA
ABSTRACT : Railways are the popular mode of Transport in almost all major cities of the World. Railways are the
most widely used and comfortable modes of transportation system. The major cause for railway accidents is collision
of trains on the same track. The main aim of this anti collision system is to identify collision points and to report
these error cases to main control room nearer to the station as well as grid control station. Majority of accidents
occurred due to improper communication among the network between drivers and control room, due to wrong
signaling, worst atmospheric condition, immediate change of route . The train driver doesnt get proper information
in time leading to hazardous situations. So this system by using zigbee protocol provides communication in between
trains, which provide information or track id of one train to another train to avoid collision.
Keywords ARM7LPC2148, sensor, track switches, fire alert, zigbee technology.

Van), Station ACD with console (fitted in Station
I.INTRODUCTION Masters' Cabin), Manned(manually operated level
Today Railways is one of the efficient and largest crossing gate) and Unmanned Gates ACD with
transports in world. But railway system is also facing hooters and flashers (in each location) and
some problems. In this paper we are providing some Repeater ACDs (ensuring communication in the
electronic element to provide some safety measure for defined zone of 3km) which work in concert to
collision of trains by using zigbee protocol. It may prevent the following kinds of collisions and
helps to save the human life from accidents by accidents like
implement this paper in railway transport. By a. Head on collisions.
implementing this automatic system one could avoid b. Rear end collisions.
human error. In this implementation, every train sends c. Collisions due to derailment.
its track id to nearby trains, if the one train goes in a d. Accidents at the level crossing gates.
first track, the signal is given to the other train, if any Head on collision occurs when one front
another train comes in same track and it also sends end of the train/vehicle hits the another front end of
first track to other, then two trains receives same track train or any vehicle. Rear end collision occurs when
id alerting two train drivers so that they can stop train the front of train or any vehicle hits the back end of
at a distance to avoid train collisions. This system train or any vehicle.
uses buzzers, switches, ARM7LPC2148, LCD ACDs have knowledge embedded
display, MAX232, DC motor, Zigbee receiver and intelligence. Receive inputs from GPS satellite system
transmitter. for position updates and network among themselves
Collisions and communication can be done through radio
Collisions are most dreaded accidents. In case of fault modems to take decisions for timely auto-application
signaling it is very difficult to stop both the trains on of brakes to prevent dangerous collisions. But
the same track because generally there would be drawback of ACD is as it is made from IR system
travelling very fast. Collisions are made by human (works only on LOS) fails on curved tracks and here
errors or by equipment fault but by the collision there we are using radio modem for communication which
would be a lot of loss to property and the death toll in may get obstacles in places like hill stations
such accidents would be very high.
Related Methods
II. EXISTING METHODS AT PRESENT Railway Collision Avoidance System (RCAS) is the
SYSTEM another system which identifies the train position and
Detection of Train Paths also the movement of direction, but they are not trust
This system continuously tracks the path of worthy. Train Collision Avoidance System (TCAS)
train. The proposed Solution is based on IR Rays & which has signal aspect in cab and also display
Sensors. prevention of SPAD(signal passing at danger)with
Anti-collision Device (ACD) emergency break application
The Anti-Collision Device (ACD) is a
automatic train invented by RAJARAM BOJJI III.PROPOSED EMBEDDED SYSTEM DESIGN
director of Konkan railways. The system consists of VIEW
Loco ACD (locomotive of train) with a console The proposed system is used to protect the accidents
between the trains automatically which helps for
(message display) for the driver (in each Loco safety purpose by using buzzers, switches,
Engine), Guard ACD with remote (fitted in Guard ARM7LPCs2148, LCD,MAX 232 serial

communication, Temperature Sensor, DC Motor, life. The other wireless techniques such as Bluetooth,
Motor drive, Zigbee transmitter and receiver. infrared etc are having the limitations of range and
also of the efficiency. The wireless system based on
GSM/GPRS is well known. But the fee is needed of
using GSM/GPRS network, and also the cost of
hardware system is very high. The following are the
some of the features of the zigbee technology.
Features:
1. Good sensitivity
2. Excellent stability
3. Good pinpointing ability
4. Loudspeaker output
5. Low cost.
TEMPERATURE SENSOR
As there are many types of temperature sensors but
we are using LM35temparature sensor.LM35 is an IC
which measures temperature with an electrical output
proportional to Celsius. Compared to the thermistor
LM35 measures temperature more accurately. It has
three pins which are positive power supply, analog
Fig-1train module I output, ground or negative power supply. This LM35
is less cost effective and it is suitable for remote
applications such as power supplies, battery
management. It is a Waterproof temperature sensor.
16X2 LCD
DISPLAY
MAX232 SERIAL COMMUNICATION
TRACK
The output of the TTL logic is always 5V
SWITCH
and 0V. So these values are needed to be amplified
before being transmitted to the GSM Module for this
ZIG purpose MAX232 is used. It has incorporated
BE
ENGI
DRI
M switched capacitor charge pump circuits to generate
NE
CTR
VER A required voltage levels. It has in-built voltage doubler
AR
X2
M7 and inverter circuits. MAX 232 has two pairs of such
LPC
2148
DRI BUZ
transceivers.
VER ZER
R
S
OSC
T
Fig-2 train module II RESULTS
ARM7-BASED EMBEDDED SYSTEM

The ARM7-based embedded system is heart
of entire System. It is designed based on low power
32-bit ARM7 (LPC2148). It is RISC architecture and
can use oscillators, thus it is ideal to be used as an
embedded system. The LPC2148 is an32k
instructions program buffers, 512kb of RAM, three
timers and 32 bit A/D converter microcontroller. It
is highly performance and low cost solution for
embedded applications.
ZIGBEE TECHNOLOGY
Zigbee is the latest technology and also
provides an efficient way to convey the information to
the authorized official at low cost as compare to that FIG-4: HARD WARE OF TRAIN COLLISION
of the GSM Modem and also utilizes a cell-phone to AVOIDANCE SYSTEM
send the message to the officials having a long battery

FIG-5: INDICATION OF TRACKS OF THE TWO

TRAINS
FIG-8: DISPLAY OF MESSAGE ALERT WHEN

TWO TRAINS COME TO SAME TRACK TO
STATION MASTER.
FIG-6: DISPLAY OF ALERT MESSAGE WHEN
TWO TRAINS ARE IN SAME TRACK
FIG-9: DISPLAY OF LONGITUDE AND

LATITUDE WHEN FIRE ACCIDENT IS
OCCURED
FIG-7: DISPLAY OF LONGITUDE AND

LATITUDE WHEN TWO TRAIN S COME IN TO
THE SAME TRACK

[3] Indian railway vision 02r0.Government of

India, Ministry of Railway, (Railway Board)
December 2009
[4] Comparison of Collision Avoidance
Systems and Applicability to Rail
Transport, in Institute of Navigation.
[5] Lim, Y.-s., S. Lim, et al. (2007). A Fire
Detection and Rescue Support Framework
with Wireless Sensor Networks.
Convergence Information Technology.
[6] Bhatt, Ajaykumar A, An Anti-Collision
Device Network A train Collision
Prevention System (TCPS).
[7] Z. Tang and J.J.Gercia-Lune- Aceves, A
protocol for topology dependent
transmission scheduling in wireless
FIG-10: DISPLAY OF MESSAGE ALERT FIRE
networks, proceedings of IEEE WCNC
ACIDENT IS OCCURED. '99,pp.1333-13337, September 1999.
[8] http://www.konkanrailway.com/node/392
CONCLUSION
In this paper, an anti collision device has been
designed ,in this system if two trains are travelling g
in same track with the speed of 120kmph then
immediately the information signal is sent to the train
driver and the train can stop at the minimum safe
distance of 920m also have safety valves at each
window. So that when fire accident is occurred every
one can escape from the train without any rush. By
implementing this system in railways we can save
many human lives
FUTURE SCOPE
Zigbee covers up to a distance of 1km whereas by
using Wi-Fi we can cover over long distances so that
we can easily avoid the accidents and can have the
safest mode of transportation.
REFERENCES
[1] Bhatt, Ajay Kumar A, An Anti-Collision
Device (ACD) Network A train Collision
Prevention stem (TCPS).
[2] A concept for reducing railway accidents. H
Ben Brown, Jr. Gregg Ponder, Mel Siegel,
February, 2005.

A ROBUST WIRELESS SENSOR NETWORK FOR

DISASTER COMMUNICATION
A.NAGA SIRISHA[1] N.VENKATRAM[2]
1(M.Tech-WCSN, K L University , India)
2(Professor-ECM, K L University, India)
ABSTRACT: Wireless Sensor Networks has seen Introduction

an explosive growth in the recent years due to their
A calamity can be of many types like floods,
independent sensing capabilities and the ability of
tornadoes, earthquakes, tsunamis etc. Any natural or
small autonomous nodes to perform routing and
man-made disaster would ultimately result in
networking capabilities. Wireless Sensor Networks
uprooting of trees, fallen electric poles, blocked or
finds its applications in a wide variety of fields such
cracked roads, communication cut from external
as target detection and identification by military,
world, ruins, debris, persons caught in rubble. Fires
environmental applications such as disaster
erupted in some locations. Communication and
mitigation and agriculture monitoring, green house
electricity towers in several areas got broken. This
monitoring, consumer applications like home
resulted in people getting trapped in several
automation etc. There is a dire need of information
locations; with some having injuries of varying
collection and transfer after a natural calamity. If
degrees of seriousness. This might be the scenario
the then existing communication system has totally
post disaster. There might be times when people of
collapsed, then there will be a requirement for
rescue team may not have clear information about
temporary communication system to be established.
victims in the area of disaster. This might limit their
This want can be sufficed by the use of Wireless
chances to help people caught in wreckage. This
Sensor Networks. In this project a prototype of a
paper mainly focuses on retrieving the data of needy.
data collection module, to collect information from
Since there is no established communication system,
disaster survived victims, is proposed which relies
a temporary communication system must be
on Zigbee for communication. The Human Module
established. There is a need for the information of
is the end module at the receiver which is used to
victims to be communicated to some main location or
sense and communicate human location through an
a fixed base station, from where help might be sent. It
Intermediate node to the Base Station. The
might take all effort, money, time to establish a wired
communication here is done via a Zigbee module
communication, even though it is not a timely idea.
which is used for transmitting and receiving the
In order to meet the requirement of establishing a
data to Human module and from Human module
temporary communication instantly, the technology
respectively. This project finds application in many
that would suffice this need is Wireless
areas where data collection is required but human
Communications and Sensor Networks.
intrusion is not possible.
Key Words: WSN, Zigbee, Human Module, Master

Node, GPS.

Introduction to wireless sensor networks architecture applied WSN and Global Positioning
System (GPS) dual-network integration technology.
Over the late years the business for remote
Where By this work mainly focuses on
sensor systems has seen a phenomenal development.
conveying the information, thereby assisting rescue
Despite the fact that it has history of more than a
team, of life in debris, which cannot be readily
century, remote transmission has discovered far
recognized or where a human intrusion is impossible
reaching use in correspondence frameworks just in
to locate the needy. In this prototype, the work tried
the most recent 1520 years. Remote systems are
to implement is, how wireless sensor networks can be
additionally helpful in diminishing systems
used to transmit data collected from a sensor through
administration costs in a few cases. This stems from
a zigbee to a main base station from which help is
the way that a general establishment of a remote
inevitable or help can be reached.
system requires essentially less cabling than a wired A Human Module which can also act as an
one, or no cabling by any means. This can be to a
intermediate node, to transmit data to the Master
great degree valuable:
Node or Base Station when the Human Module is not
Network deployment is difficult. Such is the case
in range of Master, is used to collect the information
for cable placement in rivers, oceans, etc. Another
of survivors under the debris with the help of sensor.
example of this situation is the asbestos found in old
These Human Modules are movable devices which
buildings. Inhalation of asbestos particles is very
can act as a mobile sensor node, which are randomly
dangerous and thus either special precaution must be
deployed in the disaster area.
taken when deploying cables or the asbestos must be
These modules scan for victims from place
removed. Unfortunately, both solutions increase the
to place with a specialized methodology by scanning
total cost of cable deployment.
for sufferers and when there is detection, the node
Prohibition of cable deployment. This is the
will transmit the GPS value to the master node via a
circumstance in network deployment in several cases,
zigbee module. Based on this information the
example chronicled structures.
individuals monitoring base station can readily
Deployment of a temporary network. For this
provide help at the specific location.
situation, link organization does not bode well,
Block Diagram
since the system will be utilized for a brief while
period. This is the circumstance for system
organization in instances of calamities [2]
Methodology:
According to time-consuming, difficulty and

network building complication problem for
traditional geological disaster rescue methods, on the
basis of the study of wireless sensor network (WSN)
and Zigbee standard protocol, this paper presents a
real-time and geological disaster rescue system
Fig 1:Block Diagram

The Block Diagram shown in Fig 1 shows cellular network. In A Framework for Data
the organization of system. Base Station is fixed node Collection and Wireless Sensor Network Protocol for
which receives data collected from different modules. Disaster Management[4] ARS are used, this paper
HM/IN stands for Human Modules/ Intermediate even describes disaster mitigation strategy. ARS
Nodes and these are mobile nodes. The name is stands for Ad-hoc Relay Stations which also act as a
because these human modules should be capable of medium to convey data to base station. Same
detecting humans stuck in ruins and wreckage. question arises again. However a last minute strategy
is developed which suffers with the problem of local
As a disaster would occur in a larger area
minima [4]. In paper [5] a hybrid network of MANETS
and here it is relied on Zigbee for communication
and WSN is used. MANET is composed of simple
which has practical limitations (Zigbee is capable of
portable devices like PDAs carried by rescue
transmitting within a diameter of 20 meters in closed
personnel. System works on System initialization,
room and 60 meters in outer space). This will
service request, and service discovery.
certainly result in nodes being out of range or not in
direct line of sight of Master Node. There comes the Comparison Statements
concept of traversing data through an intermediate
human module there comes the name intermediate The proposed system doesnt require any
node. manual assistance to locate the needy since the nodes
are smart enough to locate. Since a completely new
In this particular work a hardware
and makeshift correspondence framework is built
simulation module is built to illustrate the situation.
there is no requirement for considering existing cell
A paradigm is built using Atmega8 and LPC 2148
base station for correspondence. So regardless of the
micro controllers to work out entire processing; an possibility that a base station is broke down this
ultrasonic sensor is used for detection, GPS to locate
viewpoint doesn't trouble. Zigbee is used for
the spot and this GPS location value is what that is
communication which is simple, apt and isnt a bulky
transmitted over Zigbee to Master. The entire code is
system to deploy and easily adapts in the existing
written in embedded C language
network so any no. of nodes can be added at any
point of view. Since the location information of
Literature Survey
person is precisely communicated, rescue personnel
[3]
This technique makes an attempt to can directly approach the needy.
evacuate from sensor nodes by an hybrid network
which is a combination of cellular and ad-hoc
Results
networks. In this method the disaster area is divided

into cells and data from different nodes in a cell is
communicated to the corresponding cellular base
station in that cell. Here comes the question what
happens if a base station is damaged in a natural
disaster? This results in incomplete and impractical
solutions. A catastrophe would always result in
Fig 2. Simulated output Result
communication cut due to destructed established

Figure 2 shows the simulation result when a This Master node does the entire processing
person is detected in debris. The latitude and after receiving data from different nodes. This node
longitude location recognized at which particular will have information of life detection at a particular
node is displayed at the master nodes terminal node.
equipment
Applications
This sort of framework association can be

utilized as a part of applications where information
gathering is prerequisite however human interruption
is impractical. This framework can likewise discover
application in reconnaissance, to continually screen
information at hot springs, dynamic volcanoes etc
Fig 3. Rover released into disaster site

Conclusion
Figure 3 shows a land rover that will be released into
This model is made simpler in comparison
disaster site for locating victims.
with previous techniques however in this model
precise human identification sensor is not exhibited,
and routing can be implemented when more number
of nodes is to be induced. A routing technique is
proposed in the future Scope which can be done by
interfacing Matlab to Hardware. Some simulation
results are shown. Human detection can be made
possible with micro wave detection techniques,
[6]
sensing is made using heart beat and breath . Many
new techniques are proposed and some practical
[7]
Fig 4. Top view of Rover research works are used in real time. However the
practical workings of the prototype have provided
Figure 4 shows the top view of Rover consisting
ultrasonic sensor, Zigbee module, 12V Battery, sufficient results as shown above.
Atmega8 microcontroller board, GPS module. Future Scope
The mobile nodes are randomly released

[8][9]
into disaster but if a quad copter could be
employed to make an aerial survey of survivors and
then release mobile nodes with robotic arm, that
would surely serve the purpose. These mobile nodes
should be light weight, embedded into a more precise
and compact shape and should be capable of sensing
[6],[7]
and transmitting some more useful parameters
about the victim condition. In order for this data to
reach the base station at an earliest possible shortest
Fig 5 Master Node

[11]
path routing algorithms can also be made use of.
One proposed method with some simulation results is
Dijkstras Algorithm [10]
to find the shortest path
among given number of nodes.
Fig:9. Shortest Distance Between Nodes
References
Fig:6. Node Description
[1]. ADHOC and Sensor Networks by Carlos De
Morais Cordeiro, Dharma Prakash Agarwal.
[2]. Wireless Networks by P. Nicopolitidis Aristotle
University, Greece, M. S. Obaidat Monmouth
University, USA, G. I. Papadimitriou Aristotle
University, Greece, A. S. Pomportsis Aristotle
University, Greece
[3]. Efficacy of Power of Two Choices Technique for
Data Evacuation Process in Sensor Networks for
Post-Disaster Relief Operations By Chittaranjan
Fig:7. Distance Between Nodes
Debata, Rajarshi Roy
[4].A Framework for Disaster Management System
and WSN Protocol for Rescue Operation by Suman
Saha, Mitsuji Matsumoto GITS, Waseda University
Tokyo, Japan.
[5]. A WSN-Driven Service Discovery Technique for
Disaster Recovery Using Mobile Ad Hoc Networks
By Yasser Gadallah, M. Adel Serhani
[6]. Microwave Life-Detection Systems for
Searching Human Subjects Under Earthquake Rubble
or Behind Barrier Kun-Mu Chen*, Life Fellow,
IEEE, Yong Huang, Member, IEEE, Jianping Zhang,
Fig: 8. MAT LAB Output for Dijkstras Student Member, IEEE, and Adam Norman,
Member, IEEE, IEEE TRANSACTIONS ON

BIOMEDICAL ENGINEERING, VOL. 27, NO. 1,

JANUARY 2000
[7].http://www.independent.co.uk/life-style/gadgets-
and-tech/news/nasa-uses-microwave-technology-to-
detect-heartbeats-of-people-stuck-under-nepal-
earthquake-rubble-10235188.html
[8].http://andrew.gibiansky.com/downloads/pdf/Quad
copter%20Dynamics,%20Simulation,%20and%20Co
ntrol.pdf
[9].http://eprints.uthm.edu.my/2926/1/mongkhun_qet
keaw_1.pdf
[10].http://pages.pacificcoast.net/~cazelais/222/dijkst
ra.pdf
[11]. Rachika Gupta et al. Mobile Adhoc
Network(MANETS) : Proposed solution to Security
Related Issues, Indian Journal of Computer Science
and Engineering (IJCSE), ISSN : 0976-5166 Vol. 2
No. 5 Oct-Nov 2011.

An Energy Efficient Routing Protocol for Wireless Sensor Networks

G.Bindu Bhargavi(1),Dr.K.Raghava Rao(2)
1. M.Tech Student (WCSN), K L University, Guntur.
2. Professor (ECM), K L University, Guntur.
ABSTRACT: The wide-ranging applications of aim communication protocols which help to usethe
wireless sensor networks in various fields such as energy source effectively and reduce the network
Military, Surveillance, Monitoring, Agriculture, Road delay. Sensor nodes senses the environment, gather
Safety etc. has made it a supportive era for various the data from the environment and communicates it
technologies. In such networks, the data delivery to the central. Communication requiresmore energy
from source to sink is significant with proper when compared to remaining tasks and hence the
route/path between them. Hence an efficient routing major concern is the communication task. Sensor
is to be provided to provide energy efficiency for the nodes in WSN are deployed randomly in
nodes and to enhance the network lifetime. In the environment or space. In sensor network there is a
proposed paper, an efficient routing protocol is BS (base station) which is located distantly from
proposed for WSNs.In the protocol concept of head- sensor field. Sensor nodes send the sensed data to the
setis presented that is liable for controlling and BS which requires much energy. If energy
managing the network. The controlling nodes can be consumption is high then the nodes ability to work
adjusted consistently for the given data collecting reduces and hence the network lifetime decrease.
nodes in order to minimize the energy intake and to Energy management and network lifetime
enhance the network existence. The proposed method enhancement are the significant tasks in the scheme
is analyzed in MATLAB Simulation Tool that of WSNs.Communication protocols are developed
provides the results regarding the network considering the MAC as the essential concept as the
throughput and lifetime. need is to develop a protocol that provides contention
Keywords: Wireless Sensor Networks, cluster, less, effective and reliable data transmission and
delivery.The schedule based protocols do not report
cluster-head, head-set.
idle listening prevention, for example, by using
I. INTRODUCTION
TDMA schemes, which explicitly allocate
WSNs has emerged as a significant expertise
communication prospects to the nodes and let them
for wide area of applications such as Military
sleep at remaining periods. They also provide
surveillance, environment monitoring, border
collision free reception.
protection, health care monitoring, and weather
The Low Energy Adaptive Clustering
monitoring applications. These require data without
Hierarchy protocol adopts a compact network of
adjournment and energy effective utilization and
sensor nodes, which sense the parameters specified
functioning. WSNs are also deployed in harsh
based on the requirement and report the information
environment where it is not possible to replace or
to a sink node. In LEACH, a time division based
charge battery of sensor nodes. So it is desirable to
medium access control protocol is united with

clustering and a routing protocol.It screens the nodes converts to a cluster head once at the time of epoch.
into clusters and a cluster-head is selected in each Every round involves several iterations. In each
cluster based on the signal strength of the nodes. round, every sensor node turns into a head-set
Time slots are dedicated to the cluster members that member for one time.
can be used to transmit information to the cluster-
head. When there is no information for the node to
transmit, then it remains in the in-active mode thus
saving the energy.The cluster-head collects the data
from all its members and transfers to the central.
II. PROPOSED METHOD

The proposed protocol can be defined as
Energy Efficient Cluster Based Hierarchical Routing
Protocol (EECBHRP), suitable for applications such
as habitat and environmental monitoring. It is
dependable on the statistic that the energy Figure 1: Stages in the protocol
expendedfor distant transmission is extremely more The sensor states that are out of order are
than that required for short range transmission. The discarded. Each node joins the network as a
proposed protocol is the extension of LEACH candidate. At the flinch of each iteration, stable nodes
protocol with the use of head-set instead of cluster are elected as cluster heads that attain active state. By
head. At the time of each selection, a head-set of the end of selection phase, some nodes are elected as
several nodes is considered whose members are liable head-set members acquiring associate state and one
for transferring messages to the central. Single head- head-set member is in active state. In a period of a
set member is active every time while remaining data transmission phase, the active sensor node
would be in inactive mode. The responsibility of communicates the frame to the central and enters into
communication to the central is consistently passive associate state. The succeeding associate
disseminated among all members of the head-set. node that needs to transfer the data to central attains
Cluster head transmits collected data to the central. active state. During an epoch, head-set members are
Non-cluster heads convey the composed data to disseminated as follows: one in active state, few in
theircluster heads. Every cluster consists of head-set associate state, and some other in passive associate
comprising of numerous fundamental cluster heads. state. There is no node present in associate state at the
Each iteration involves two stages: (1) Election phase end of epoch. During the next epoch, all head-set
(2) Data transfer phase. In the first stage, head-sets members convert into associates and one is selected
are elected for purposeful number of clusters. In to be active. At the end of each iteration, all head-set
second stage, head-set members transfer collected members will be preset in non-candidate state;
data to the central. Each data transfer phase involves ineligible as a head-set member. During the period of
more than one epoch. Each head-set member a next round, all non-candidate nodes obtain

candidate state; a new round initiates when all nodes the inactive mode and do not obtain any data. Aftera
obtain non-candidate state. period of time, the following associate turns out to be
The analysis is based on the following assumptions: a cluster head and the current one turns out to be a
Central is located distant from sensing field and the passive. As a final point, at the end of each round, all
sensor nodes are immobile once they are deployed. nodes become non-candidate members. At this stage,
Sensor nodes are homogeneous i.e. all the sensor next round starts and all nodes turn into candidate
nodes have same initial energy, a battery. members.
Radio channel is symmetric means the energy
consumed for transmitting data is the same for
IV. SIMULATION MODEL
receiving it.
All the sensor nodes have knowledge of its location Radio Communication Model
and energy. A radio model where in the instance of short
distance communication within the cluster, the
energy expended by a transmit amplifier is
III. PROTOCOL DESCRIPTION
proportional to d2. For distant communication, i.e.,
The protocol functioning has two phases
cluster head to the central, the energy expended is
First Phase-Election proportional to d4. Hence the energy expended to
Predefined numbers of clusters-k are considered. At transfer an l-bit message for distant transmission is
the beginning, set of cluster heads are preferred specified as:
randomly that direct a short range announcement
broadcast message. The other nodes obtain broadcast
messages and select the corresponding cluster heads The energy expended to transfer l-bit message for
based on signal strength of the messages. Every node short distance is specified as:
transmits an acknowledgment to the corresponding
cluster head. During each iteration, cluster heads
Energy expended to receive l-bit message is
electset of associates centered on signal strength of
specified as:
the acknowledgments. A head-set comprising of a
cluster head and the associates, responsible to
transfer data to the central is selected for an iteration It includes the data accumulation approach.
of a round.
Second Phase-Data Transfer Election Phase for EECBHRP
As soon as the clusters, head-set members and head-
Considering that all nodes have equal
sets are elected, schedules are shaped, data
energy, the energy expended is similar for entire
communication initiates. The nodes other than
cluster. For evenly disseminated clusters, each
cluster-heads aggregate the information and send to
consists of n/k nodes. The energy expended by a
the cluster head, in the allotted time slots. The
cluster head is predicted as
cluster-head have to keep its radio active for
receiving the data. The associate members persist in

Energy consumed by the nodes other than cluster- For a single non-cluster head;
heads are expected as
The preliminary energy, Einit should be adequate for

at least a single round. A node turns into a head-set
Data transfer phase for EECBHRP member once andinto a non-cluster head for (n/mk-1)
Nodes transfer data to corresponding cluster times in each round.
heads which transfer the gathered message to the An assessment of Einitis
central. The energy spent by a cluster head is
specified as:
The energy, Enon-CH/frame, expended by a node to

transfer data to cluster head is specified as:
Essential Time required for each round
The frame time, tframe is the combination of frame
In each iteration, Nf frames are transferred. Each transferred time by all cluster nodes. For Rb
cluster sends Nf/k frames, consistently distributed bits/second data transfer rate and message of l bit
among n/k nodes. Hence, a cluster head sends the length, the time to transmit a message, tmsg, is
frame of (n/k-m) non-cluster head data.
The energy intake in a data transfer phase in the case
of each cluster is specified as: Since each time only one head-set member is active,
the remaining (m-1) do not transfer any data.
The time for single frame is
where
As Nf frames are transferred in everyperiod, the time

for each iteration, t iteration is:
Primary Energy for each round

In every iteration, m nodes are selected as As there are n/km iterations in each round, the time
associate head for every cluster. Hence, km nodes are for each round, tround is:
selected as head-sets members.

For a single cluster head;
Likewise, there are (n m) nodes other than cluster-

heads in a cluster.

OPTIMUM NUMBER OF CLUSTERS cluster head which further transmits to the Base
In a cluster, the energy expended to transfer an Station.
accumulated data to the central is: The proposed method is energy efficient in
The total energy spent by k cluster heads is the sense that the nodes in the head-set member
equally distribute the responsibility of data
aggregation and data transmission to the central such
The total energy spent by k cluster heads is
that the energy of the node is optimized such that the
nodes life time can be optimized and hence the
network lifetime.
The optimum value of k for the slightest degeneracy
of frame energy is
VI. RESULTS
The Results shown are the simulations results of the
proposed routing protocol that is analyzed using the
MATLAB tool.
V. RELATED WORK Maximum optimum number of clusters
15
no of Cluster Heads
LEACH (Low-Energy Adaptive Clustering

Hierarchy), [1] a protocol is based on the clustering # of Clusters 10
mechanism that employsvariation of cluster-heads in

a random manner to uniformlyallot the work of data 5
aggregation among the sensors in the network. It

0
1 2 3 4 5 6 7 8 9 10
provides scalability for the networks that are Head-Set Size
dynamic, and includes data fusion into the routing Figure 2: Optimum number of Clusters versus Head-
protocol to reduce the transmission information to the set size with d=150m and n=1000.
central. A centralized LEACH protocol is proposed
in this paper [2] that uses a centralized clustering 10
algorithm and forms clusters and elects cluster-heads 8

# of Clusters
based on the decision of the Base Station. A new 5
improved algorithm based on LEACH [3] is proposed 3
1
250
that defines the idea of electing a secondary cluster 4

4.5
5
200 3.5
3
2.5
head that gathers the data from the cluster nodes and 150 1
1.5
2
Distance Head-Set Size
transmits this data to the primary cluster head which

Figure 3: Maximum cluster size versus distance from
further transmit to the central An improved protocol
the central and head-set size
of LEACH [4] is proposed that is based on the
From the graph it can be seen that larger
concept that the data received from the nosed is
cluster sizes can be achieved for greater values of
aggregated by the corresponding cluster-head and it
head-set sizes. For minimum head-set size, only
transmits this data directly to the sink if it is
minor numbers of clusters are conceivable. When the
reachable else the data is transmitted to the next hop
distance with respect to central is improved, further

energy is utilized for distant transmission. Thus, for than optimum, further transmissions to the base
same head-set size, extreme number of clusters station takes place.
decrease while the lengthwith respect to the central
increases.
0.12
Maximum optimum number of clusters
6 0.1
ECluster-ele
ENon Cluster-ele
0.08
Estart (J)
5
0.06
0.04
4
0.02
Energy (J)
0
3 20
18
16
200
14 180
12 160
10 140
8 120
100
2 6 80
4 60
2 40
20
0 0
Head-Sest Size Network Diameter (m)

1
Figure 6: Variation of Estart based on head set size and

0
0 2 4 6 8 10 12 14 16 18 20
Number Of Clusters
network diameter
Figure 4: Energy consumption versus number of Using a head set is advantageous than a single one.
clusters As the head set size increases the energy intake gets
The energy spent is scaled-down with reduced.
increase in clusters number. The rate of decline in

energy consumption is condensed for large cluster 4
x 10
sizes. Furthermore, there is less energy consumption 1.8

Time for one iteration (sec)
1.6
1.4
when head-set size is 3 than that of 1. 1.2
0.8
0.6
0.4
0.2
1 100
90
0.9 80
200
70 180
0.8 60 160
50 140
0.7 40 120
100
30 80
0.6 20 60
Estart (J)
10 40
20
0.5 0 0
0.4 Head-Set Size (%) Network Diameter
0.3
0.2
0.1
0
Figure 7: Variation of time for one iteration based on
50
45
40
35
30 140
160
180
200
head set size
25 120
20 100
80
15
10
5 0
20
40
60
The head set size is preferred as percentage of cluster
# Of Clusters Network Diameter (m)
size. If it is 50% of the cluster size then each iteration
Figure 5: Energy consumed per node versus cluster duration increases. If it is less than 50% there will be
number and network diameter few transmissions and more iterations. If it is greater
than 50% there will be more transmissions and less
The energy exhausted reduceswith increase iterations.
in the number of clusters. The energy intake increases
below the optimum range with increase in the
number of clusters. When clusters number is superior

the equal number of data collection sensor nodes, the

4
x 10
controlling and supervising nodes can be accustomed
7
6
based on network environment. The iteration time
ime for one iteration (sec)
4
and data transfer rate intensifies with head-set size.
3
Thus, overall network lifetime is intensified. Hence
2
1
the proposed protocol presents an energy-efficient
0
-1
20
hierarchical routing design suitable for
18
16
50
14
12
10
30
35
40
45
enormousapplications.
8 25
6 20
4 15
2 10
5
0 0
Head-Set Size # of clusters, k
VIII. FUTURE WORK

Figure 8: Variation of time for one iteration based on
The protocol is analysed for one iteration. Hence
head set size and cluster number.
the analysis is to be made for more iterations to
When head set size increases, each iteration lasts for
analyse the protocol performance and network
a long time and this leads to the improvement in the
lifetime. As the adhoc networks and wireless sensor
network life time.
networks have some significant similarities, the
application of the proposed protocol to the adhoc
x 10
4
networks such as MANETs, VVANETs need to be
18
16
studied.
14
12
# of Frames
10
6
IX. REFERENCES
4
2
[1] Wendi Rabiner Heinzelman, Anantha
Chandrakasan, and Hari Balakrishnan Energy-
0
20
18
16
200
14 180
12
10
8 120
140
160
Efficient Communication Protocol for Wireless
100
6 80
60
Microsensor Networks, Proceedings of the 33rd
4
2 40
20
0 0
Head-Set Size Network Diameter (m)

Hawaii International Conference on System Sciences.
Figure 9: Variation of number of frames transmitted [2] Wendi Rabiner Heinzelman, Anantha
based on the head set size. Chandrakasan, and Hari Balakrishnan An
As the head set size increases, frame number also Application-Specific Protocol Architecture for
increases. The iteration time is increased which leads Wireless Microsensor Networks, IEEE
to the improvement in thelife time of the network. TRANSACTIONS ON WIRELESS
VII. CONCLUSION COMMUNICATIONS.
(EECBHRP) is intended to provide efficient [3] Chunyao FU, Zhifang JIANG, Wei WEI and
energy utilization and to enhance network lifetime. Ang WEI An Energy Balanced Algorithm of
The analytical results specify that the energy LEACH Protocol in WSN, IJCSI International
utilization can be significantly abridged by Journal of Computer Science Issues.
comprising more sensors as cluster heads in a head- [4] LalitaYadav, Ch. Sunitha Low Energy Adaptive
set instead of single cluster head within a cluster. For Clustering Hierarchy in Wireless Sensor Network

(LEACH), (IJCSIT) International Journal of [14] Study and Simulation of Vehicular AdHoc
Computer Science and Information Technologies. Networks using MATLAB.
[5] B.Zarei, M.Zeynali and V.Majid Nezhad, Novel [15] ADHOC On Demand Distance Vector Routing
Cluster Based Routing Protocol in Wireless Sensor Protocol for WSNs.
Networks, IJCSI International Journal of Computer [16] Holger Karl and Andreas Willig PROTOCOLS
Science Issues. AND ARCHITECTURES FOR WIRELESS
[6] F.Bajaber and I.Awan, Energy efficient SENSOR NETWORKS.
clustering protocol to enhance lifetime of wireless
sensor network, Journal of Ambient Intelligence and
Humanized Computing.
[7] W. Heinzelman, A. Chandrakasan and H.
Balakrishnan, An application-specific protocol
architecture designed for wireless micro sensor
networks.
[8] D. Mahmood, N. Javaid, S. Mahmood, S.
Qureshi, A. M. Memon, T. Zaman, MODLEACH:
A Variant of LEACH for WSNs.
[9] Nguyen Duy Tan, Longzhe Han, Nguyen Dinh
Viet, and Minho Jo An Improved LEACH
Routing Protocol for Energy-Efficiency of Wireless
Sensor Networks, Smart Computing Review.
[10] F. Borgonovo, A. Capone, M. Cesana, and L.
Fratta, ADHOC MAC: New MAC architecture for
ad hoc networks providing efficient and reliable
point-to-point and broadcast services, Wireless
Netw., vol. 10, pp. 359366, 2004.
[11] H. Omar, W. Zhuang, and L. Li, VeMAC: A
TDMA-based MAC protocol for reliable broadcast in
VANETs, to be published.
[12] P. Liu, Z. Tao, S. Narayanan, T. Korakis, and S.
S. Panwar, CoopMAC: A cooperative MAC for
wireless LANs, IEEE J. Sel. Areas Commun., vol.
25, no. 2, pp. 340354, Feb. 2007.
[13] Sailesh Bharati, Student Member, IEEE, and
Weihua Zhuang, Fellow, IEEE CAH-MAC:
Cooperative ADHOC MAC for Vehicular Networks.

Ncaessn Part 4

Hochgeladen von

Dokumentinformationen

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Ncaessn Part 4

Hochgeladen von

Copyright:

Verfügbare Formate

National Conference on Advancements in Embedded Systems and Sensor Networks (AESSN - 2015)

HARDWARE AND SOFTWARE CO-DESIGN ISSUES IN H.264

INTRODUCTION H.264 DECODING PROCESS

H.264/ AVC codec is a multimedia application produces high-

ISSN: 2348 - 8387 www.internationaljournalssrg.org Page 283

Memory model is used for input and output to minimize

Software implementation is necessary before going to the

Module complexity. De-blocking filter consumes 30% of

ISSN: 2348 - 8387 www.internationaljournalssrg.org Page 284

ISSN: 2348 - 8387 www.internationaljournalssrg.org Page 285

[13] Damak T., Werda I., Ben Ayad M-A, Masmoudi N, An

ISSN: 2348 - 8387 www.internationaljournalssrg.org Page 286

A NOVEL SYSTEM TO PROVIDE CAB SAFETY FOR WOMEN

AND IMPLEMENTING WITH INTERNET OF THINGS

KONA GOWTHAMI K.SRIPATH ROY

ABSTRACT: Constructing an intelligent monitoring device is embedded with our

ISSN: 2348 - 8387 www.internationaljournalssrg.org Page 287

common place communication protocols, network formation of knowledge transport layer

Fig 1. Three-tier architecture of internet of 2.3 GPRS:

ISSN: 2348 - 8387 www.internationaljournalssrg.org Page 288

services for GSM users. GPRS provides high-

The objective of intelligent vehicle tracking

Fig 4. Network architecture

ISSN: 2348 - 8387 www.internationaljournalssrg.org Page 289

Transportation Research Part B 40 (2006) 857

ISSN: 2348 - 8387 www.internationaljournalssrg.org Page 290

DEVELOPMENT OF SMALL WIRELESS SENSOR NETWORK

FOR WATER MANAGEMENT USING XBee BASEDTECHNOLOGY

ISSN: 2348 - 8387 www.internationaljournalssrg.org Page 291

Recent advances, through in wireless sensor

ISSN: 2348 - 8387 www.internationaljournalssrg.org Page 292

microcontroller establishes communication with the

A sensor unit is basically consists of several

ISSN: 2348 - 8387 www.internationaljournalssrg.org Page 293

ISSN: 2348 - 8387 www.internationaljournalssrg.org Page 294

[6]. Sensor networks for monitoring and control of water

[7]. Implementation of Wireless Data Transmission in Monitoring

ISSN: 2348 - 8387 www.internationaljournalssrg.org Page 295

HANDLING LOCK HOLDER PREEMPTION FOR

ISSN: 2348 8387 www.internationaljournalssrg.org Page 296

ISSN: 2348 8387 www.internationaljournalssrg.org Page 297

ISSN: 2348 8387 www.internationaljournalssrg.org Page 298

the example described above, if the RTOS is assigned SPECIFIC

ISSN: 2348 8387 www.internationaljournalssrg.org Page 299

ISSN: 2348 8387 www.internationaljournalssrg.org Page 300

on Consumer Communications and Networking Conference, 2009.

ISSN: 2348 8387 www.internationaljournalssrg.org Page 301

ISSN: 2348 8387 www.internationaljournalssrg.org Page 302

1.1 CLOUD COMPUTING for what you use.

years become a serious platform for

dynamically provision and dimensional completely different in designs, capabilities,

Cloud computing will be able to use the embedded systems. Now-a-days an

remote location. From mainframe is created for specific controlling functions

computing to client-server model, it has which is installed in a larger system with

cleverly picked up some of the similarities real-time computing constraints. It is

have been the major contributors of this a general-purpose computer, such as a

control various number of devices in protocols, signal processing, user interface

1.4 CURRENT CLOUD SERVICES or pay-for-use model. It helps an enterprise

the operating system, hardware or network

ISSN: 2348 8387 www.internationaljournalssrg.org Page 306

software partitioning, fabrication of

Figure 2.2: An example topology of a professional rapid prototyping.

Embedded System Design is unique because

ISSN: 2348 8387 www.internationaljournalssrg.org Page 307