Conference Title

The Fifth International Conference on Digital Information Processing, E-Business

and Cloud Computing (DIPECC2018)

&

The Fifth International Conference on Technological Advances in Electrical,

Electronics and Computer Engineering (TAEECE2018)

Conference Dates
July 05-07, 2018

Conference Venue
Avrasya University, Trabzon, Turkey

ISBN
978-1-941968-47-5 ©2018 SDIWC

Published by
The Society of Digital Information and Wireless Communications (SDIWC)
Wilmington, New Castle, DE 19801, USA
www.sdiwc.net
 Table of Contents

DIPECC2018

How Does Cloud Computing Change the Strategic Alignment Between Business and IT?……………… 1

TAEECE2018

Design of Modular Multi-level Converter and Permanent Magnet Synchronous Generator-based

Wind Energy Conversion System …………………………………………………………….……………………………………. 1

Performance Evaluation of Underwater Acoustic Communication using Trigonometric Chirp

Modulation …………………………………………………………………………………………………………..……………………... 7

Multi-keyword Search Employing Identity-Based Encryption Technique (MKS-IDE) ……………………... 13

Image Processing Based Anti-Sleep Alarm System for Drowsy Drivers ………………………………………….. 19

Design and Implementation of Energy Management System for Buildings ………………………….……… 23

Proceedings of the Fifth International Conference on Digital Information Processing, E-Business and Cloud Computing (DIPECC2018), Trabzon, Turkey, 2018
How Does Cloud Computing Change the Strategic Alignment Between
Business and IT?
Peter Fuzes
Corvinus University Budapest
Budapest, Hungary
peter.fuzes@uni-corvinus.hu
ABSTRACT
Strategic alignment between business functions of
the company and the Information Technology (IT)
department has been in the focus of strategic
management research for decades. Several studies
showed that IT systems could be a source of
competitive advantage. However, due to lack of
alignment with the company strategy, IT often does
not fill this role.
Cloud computing changes the way companies use
IT. Instead of building own IT systems, companies
can use IT as a service, from external providers.
In this paper, we analyze the impact of cloud
computing on the strategic alignment between
business and IT. We conclude that cloud computing
changes the role of IT department and use of cloud
services can improve the alignment.
KEYWORDS
Business – IT Alignment, Cloud Computing, Cloud
Services, IaaS, PaaS, SaaS, Business Needs
1 INTRODUCTION
Strategic alignment between business and the
IT department has been widely researched for
nearly four decades. Since the ‘90s several
studies tried to understand the linkage between
use of IT and company performance and prove
that increased investment in IT systems results
in superior company performance. However,
research showed inconsistent result. While
several studies showed positive correlation
between IT investment and company
performance [1] in many cases increased IT
investment did not improve the company
performance [2]. That phenomenon has been
named as the ‘IT Paradox’ [3].
ISBN: 978-1-941968-48-2 ©2018 SDIWC
Scholars have offered two explanations for the
IT paradox:
a. IT investment is not always aligned
with organizational goals, strategy,
resources or capabilities
b. The value of IT investment has not
been captured properly by the studies.
The real value of IT investments can be
difficult to measure, for example in
case of increased customer value.
The first is the more commonly mentioned
reason and highlights the importance of
alignment between the business functions and
the IT department. Despite the understanding of
its importance, business-IT alignment remains a
top concern for IT executives and an area of
interest of researchers.
2 REASONS FOR SUB-OPTIMAL
ALIGNMENT
Based on the existing literature, the reasons for
sub-optimal business-IT alignment can be
organized into three categories [4].
a. The IT department and the business
leaders do not understand each other’
domains. Often, they are unable to
express themselves in common
language, explain their goals and
complexities.
b. Today’s rapidly changing environment
business requirements are subject to
frequent change. Sometimes it is a
result of conscious change, but
sometimes it is due to unforeseen
happenings, failed promises and human
errors. Business users often learn
during the project what are the
possibilities and they change the
requirements for the IT system. As a
1
Proceedings of the Fifth International Conference on Digital Information Processing, E-Business and Cloud Computing (DIPECC2018), Trabzon, Turkey, 2018
result, a gap opens between the stated
objectives at the conception of the
project and the requirement during and
after execution.
c. Lack of flexibility of IT systems. IT
systems create a complex architecture,
where different layers built on each
other and must work together
seamlessly. To change a specific
application in the complex architecture
may require significant development.
3 CLOUD COMPUTING
Traditionally, companies owned the IT systems
they used. They built their own data center,
purchased the hardware (servers, storage) and
software components for different layers of the
software architecture, and developed
customized solutions according to their
business needs.
During the past decade, a new way of using IT
has emerged as disruptive innovation [5, 6].
Due to newly developed IT technologies (such
as virtualization) and availability of high speed,
reliable internet connection, consumers do not
need to have their own IT system; they can use
IT as a service. This model is called ‘cloud
computing’.
There are three main service models of cloud
services: [7-9]
a. Infrastructure as a Service (IaaS)
b. Platform as a Service (PaaS)
c. Software as a Service (SaaS)
4 ANALYSIS OF BUSINESS - IT
ALIGNMENT IN CLOUD
ENVIRONMENT
In cloud computing, part of the IT related tasks
is transferred from the consumer’s IT
department to the cloud provider [10]. This has
an impact on the role and responsibilities of the
consumer’s IT department. This modification
of the role of IT department may have an
impact on the business – IT alignment. In our
analysis, we seek answer to the research
question: how does cloud computing changes
the strategic alignment between business and
IT?
ISBN: 978-1-941968-48-2 ©2018 SDIWC
4.1 Methodology
As described earlier, researchers identified
three key reasons for problematic IT-Business
alignment:
a. Lack of understanding of each other’s
domains
b. The expressed need of business keeps
changing
c. IT systems are not flexible enough to
support the changing need of business
To answer the research question, we analyze
what impact of cloud services on the three
listed reasons and examine what the impact of
the cloud on those problematic areas is. We
analyze the different cloud service models
(IaaS, PaaS, SaaS) for each of the three
reasons. Our analysis is based on thorough
review of the existing literature and deep
understanding of cloud computing through
industry experience.
4.2 Lack of Understanding Each Other’s
Domains
Impact of IaaS and PaaS
Traditionally, purchasing and installing
hardware elements, operating system, database,
and middleware software was the task of IT
department. Those are the IT functions which
can be covered by IaaS and PaaS cloud service.
When those tasks are transferred to the cloud
service provider, IT department remains in
charge of those services from the company side
as the key client [11].
The shift from on-prem to cloud does not have
a direct impact on how the IT staff understands
the business requirements of the functional
users (HR, Finance, Customer Service, etc).
The use of IaaS and PaaS cloud service instead
of on-premise itself does not improve the (lack
of) understanding of the business needs by the
IT staff and vice versa.
The use of IaaS and PaaS can, however, have a
long-term impact on how the IT staff
understands the business. When part of the
technical tasks is transferred to the cloud
provider, the roles within the IT department can
be reassigned, and the IT staff can focus more
on strategic and business-related issues [12-14].
With reassigned roles, a ‘collaborative partner’
IT should have a better understanding of the
2
Proceedings of the Fifth International Conference on Digital Information Processing, E-Business and Cloud Computing (DIPECC2018), Trabzon, Turkey, 2018
strategy and goals of functional units, therefore
improve the understanding of the business
domain.
Impact of SaaS
The key clients of SaaS services are the
business users. When purchasing an ERP,
HCM (Human Capital Management),
recruitment or marketing solution, business
users can articulate their needs and engage in
direct discussion with the cloud supplier, who
has experts with business knowledge on the
specific solution field. Business users can
bypass company IT, and order business
application as a service (SaaS) from a cloud
provider directly. In this case, understanding
each other’s domain between the IT department
and business becomes irrelevant; the business;
they can get access to business applications
faster directly from the SaaS provider.
4.3 The Expressed Need of Business
Keeps Changing
Business needs may change during an IT
project due to the change in the competitive
environment [4]. Also, business users may
realize during the implementation project that
the IT solution can deliver additional value for
them with extensions or customization, and as a
result they may change the requirements. In
such situation, IT is ‘shooting to a moving
target’.
Even if the necessary budget is available to
address the changed business need, the on-
premise IT system may not be flexible enough
to accommodate the required changes short
term, which can negatively impact the
business-IT alignment.
Impact of IaaS and PaaS
Using IaaS and PaaS cloud service makes the
IT system more elastic and allows to rapidly
scale up and down. With IaaS and PaaS, the IT
department can respond to the changing need of
the business better than with on-prem
environment, therefore the alignment can
improve.
Impact of SaaS
Increased elasticity also applies to SaaS
solutions. However, SaaS has another aspect
which can help to limit the customization
required by the business users. Cloud providers
ISBN: 978-1-941968-48-2 ©2018 SDIWC
offer limited customization options for their
SaaS business applications [11, 15]. The public
cloud SaaS model is based on provisioning
standardized solution to large number of
consumers with minimal customization.
Limited room for customization does not allow
the business users to keep changing the
requirements; they have to accept the standard
solution.
There is a trade-off between implementation
time (and cost) and functionality. Accepting
standard solutions may lead to shorter
implementation cycle, however, the result may
not be fully in-line with what the business users
want or need. The market success of SaaS
solutions shows that large portion of customers
is willing to accept standardized solutions in
exchange for rapid and less expensive
implementation.
Cloud providers can also influence the
customer needs by sharing best practices and
experiences with other customers. This can
help the business users to realize their latent
needs earlier during the implementation process
or give up requirements for unnecessary
features.
Overall, using cloud-based business
applications (SaaS) can have a positive impact
on preempting and limiting the change of
expressed business needs.
4.4 IT Systems Are Not Flexible Enough
to Support the Changing Need of
Business
The limited capability of the on-prem IT
infrastructure a key reason for the inflexibility
of company IT systems. The cycle of allocating
financial resources, run the procurement
process and implement the new hardware or
software parts may take a long time. That long
cycle can be seen by business users as
inflexibility and lack of ability to address the
business needs.
Impact of IaaS and PaaS
Cloud providers build large capacity data
centers, and from those data centers they can
rapidly scale up and down the service allocated
to the customers. Therefore, when the company
IT department uses IaaS and PaaS services, the
existing IT infrastructure capacity will not be a
3
Proceedings of the Fifth International Conference on Digital Information Processing, E-Business and Cloud Computing (DIPECC2018), Trabzon, Turkey, 2018
limiting factor to serve the business user’s
needs. Cloud provides elasticity, which is not
the case with on-prem solutions [16]. As a
result, IT systems become more flexible from
the business point of view and able to support
the changing business needs, thus the alignment
may improve.
Impact of SaaS
Another reason for IT inflexibility is the
complexity and inhomogeneity of existing
systems. When integration is required with the
existing systems, cloud services do not resolve
this problem, however, those can be used for a
new application outside of the current
architecture.
For example, a SaaS solution for recruiting new
employees (advertising available positions on-
line, accepting and sorting uploaded CVs,
providing a workflow for the selection process)
does not necessarily have to be closely
integrated with the company’s ERP system.
Instead of waiting for the upgrade of the on-
premise ERP system to add a new recruitment
module to it, the HR department may use a
recruitment SaaS solution from an external
provider, without dealing with the problem of
inhomogeneity of the internal IT systems.
With SaaS, new business solutions can be
provided in a shorter timeframe, which
improves the alignment between the business
needs and the IT solutions supporting those
needs.
Based on the analysis, Table 1 summarizes the
possible impact of cloud services on strategic
alignment problems.
5 DISCUSSION
Based on our analysis, the use of cloud services
may improve the alignment between business
and IT. However, changing role of the IT
department poses a threat as well: coordination
of different IT systems may fall out of their
hand. There is a risk that several systems will
be used by the company – as result of the direct
purchase by business units without the
involvement of IT department – which will not
be integrated or linked. The appearance of
shadow IT can increase the risk of loss of
valuable and confidential data [17].
ISBN: 978-1-941968-48-2 ©2018 SDIWC
IT department’s loss of control over the
company’s IT system can increase the ‘chaos’
of existing systems, instead of decreasing it.
Without coordination by the IT department, the
Enterprise Architecture of the company may
disintegrate [18, 19], which may have a
negative impact on the company’s operation.
Therefore, it is imperative for the IT
department to become a service integrator in a
new governance model, linking and integrating
different on-prem and cloud services.
The appearance of ‘shadow IT’ - when business
users directly ordering cloud services from
cloud providers without involving the IT
department – is not only a possibility, but
already reported on the market. Based on the
response from 245 companies, a survey run by
Select Hub concludes that “There is a
movement away from allowing the experts in
IT departments to exercise complete control
(due to their expertise), and turn towards the
users themselves, as the software environment
encourages their participation” [20].
Tim Killenberg, senior vice president of N3 (an
outsourced, integrated sales and marketing
execution firm), says that “We are seeing a
sharp increase in the number of line-of-business
buyers who are empowered to make technology
purchase decisions themselves. Reaching these
buyers requires a new sales mindset and new
skills” [21].
6 CONCLUSION
The role of the company’s IT department is
changing due to the growing use of cloud
computing services, and this change could
positively impact some of the problematic areas
of business-IT alignment.
IT department has an important role to
coordinate and integrate between different on-
premise and cloud-based solutions. By
outsourcing basic tasks to cloud providers and
reassigning roles, IT department has better
possibility to focus on higher level business
solutions and to become the strategic partner of
the business.
4
Proceedings of the Fifth International Conference on Digital Information Processing, E-Business and Cloud Computing (DIPECC2018), Trabzon, Turkey, 2018
7 TABLES
Table 1 - The possible impact of cloud services
on strategic alignment problems
REFERENCES
[1] Y. E. Chan, S. L. Huff, D. W. Barclay, and D. G.
Copeland, "Business strategic orientation,
information systems strategic orientation, and
strategic alignment," Information systems research,
vol. 8, no. 2, pp. 125-150, 1997.
[2] A. S. Bharadwaj, "A resource-based perspective on
information technology capability and firm
performance: an empirical investigation," MIS
quarterly, pp. 169-196, 2000.
[3] K. Celuch, G. B. Murphy, and S. K. Callaway,
"More bang for your buck: Small firms and the
importance of aligned information technology
capabilities and strategic flexibility," The Journal of
High Technology Management Research, vol. 17,
no. 2, pp. 187-197, 2007.
[4] H. Benbya and B. McKelvey, "Using
coevolutionary and complexity theories to improve
IS alignment: a multi-level approach," Journal of
Information technology, vol. 21, no. 4, pp. 284-298,
2006.
ISBN: 978-1-941968-48-2 ©2018 SDIWC
[5] N. Sultan and S. van de Bunt-Kokhuis,
"Organisational culture and cloud computing:
coping with a disruptive innovation," Technology
Analysis & Strategic Management, vol. 24, no. 2,
pp. 167-179, 2012.
[6] N. Sultan, "Knowledge management in the age of
cloud computing and Web 2.0: Experiencing the
power of disruptive innovations," International
journal of information management, vol. 33, no. 1,
pp. 160-165, 2013.
[7] D. Zissis and D. Lekkas, "Securing e-Government
and e-Voting with an open cloud computing
architecture," Government Information Quarterly,
vol. 28, no. 2, pp. 239-251, 2011.
[8] P. Mell and T. Grance, "The NIST definition of
cloud computing," 2011.
[9] N. Sultan, "Servitization of the IT industry: the
cloud phenomenon," Strategic Change, vol. 23, no.
5-6, pp. 375-388, 2014.
[10] S. Floerecke and F. Lehner, "A revised model of the
cloud computing ecosystem," in International
Conference on Grid Economics and Business
Models, 2015, pp. 308-321: Springer.
[11] S. Schneider and A. Sunyaev, "Determinant factors
of cloud-sourcing decisions: reflecting on the IT
outsourcing literature in the era of cloud
computing," Journal of Information Technology,
vol. 31, no. 1, pp. 1-31, 2016.
[12] S. Marston, Z. Li, S. Bandyopadhyay, J. Zhang, and
A. Ghalsasi, "Cloud computing—The business
perspective," Decision support systems, vol. 51, no.
1, pp. 176-189, 2011.
[13] M. G. Avram, "Advantages and Challenges of
Adopting Cloud Computing from an Enterprise
Perspective," Procedia Technology, vol. 12, no.
Supplement C, pp. 529-534, 2014/01/01/ 2014.
[14] S. Jones, Z. Irani, U. Sivarajah, and P. E. Love,
"Risks and rewards of cloud computing in the UK
public sector: A reflection on three Organisational
case studies," Information Systems Frontiers, pp. 1-
24, 2017.
[15] Q. Li, C. Wang, J. Wu, J. Li, and Z.-Y. Wang,
"Towards the business–information technology
alignment in cloud computing environment:
anapproach based on collaboration points and
agents," International Journal of Computer
Integrated Manufacturing, vol. 24, no. 11, pp. 1038-
1057, 2011/11/01 2011.
[16] P. Manuel, H. Al-Hamadi, and K. Qureshi,
"Challenges, strategies and metrics for supply-
driven enterprises," Annals of Operations Research,
vol. 233, no. 1, pp. 293-303, 2015.
[17] M. Silic, J. B. Barlow, and A. Back, "A new
perspective on neutralization and deterrence:
Predicting shadow IT usage," Information &
management, no. in press, pp. 1-15, 2017.
[18] R. Butterfield, S. Maksuti, M. Tauber, C. Wagner,
and A. Bicaku, "Towards Modelling a Cloud
5
Proceedings of the Fifth International Conference on Digital Information Processing, E-Business and Cloud Computing (DIPECC2018), Trabzon, Turkey, 2018

Application's Life Cycle," in 6th International

Conference on Cloud Computing and Services,
2016.
[19] M. Alaeddini, H. Asgari, A. Gharibi, and M. R.
Rad, "Leveraging business-IT alignment through
enterprise architecture—an empirical study to
estimate the extents," Information Technology and
Management, vol. 18, no. 1, pp. 55-82, 2017.
[20] G. Gheorghiu, "The CRM Buyer’s Profile for SMBs
", 08.30.2017. Accessed on: 11.23.2017Available:
https://selecthub.com/customer-relationship-
management/crm-software-buying-trends/
[21] L. Calhoun, "10 Trends Shaping Software Sales in
2016," Accessed on: 11.22.2017Available:
https://www.inc.com/lisa-calhoun/10-trends-
shaping-software-sales-in-2016.html

ISBN: 978-1-941968-48-2 ©2018 SDIWC 6

Proceedings of the Fifth International Conference on Digital Information Processing, E-Business and Cloud Computing (DIPECC2018), Trabzon, Turkey, 2018
A Private Cloud Implementation for Jordanian Universities Cloud
Computing: E-Learning Center of Excellence as a Service
Mokhled S. AlTarawneh
Mutah University
Faculty of Engineering, Computer Engineering Department P.O.Box (7),
Mutah 61710, Jordan
mokhled@mutah.edu.jo
ABSTRACT
There is a need to design a collaboration
infrastructure for Jordanian Universities‟ to meet
their technological needs, use benefit of
sophisticated software to solve many complexes
problems fast and at a lower cost. This paper
introduces a suggested solution on based of cloud
computing concepts and describes the architecture
of suggested cloud computing platform for
establishing an E-Learning center of excellence as
service. The author has tried to introduce cloud
computing to e-learning service, build an e-
learning center of excellence cloud, and prepare
for an active research and exploration for it from
the following aspects: architecture, construction
method and external interface with the model,
investigate the factors that influence universities‟
willingness to use cloud computing in many
educational services by using a quantitative
research approach as well empirical studies.
KEYWORDS
Cloud computing, E-learning, Cloud deployment
models, Center of excellence.
1 INTRODUCTION
Cloud computing provides users with the
ability to use resources that exceed the
capacity of their own devices, such as
processing capacity, storage space and
internet applications; at lower prices than if
they had purchased these resources. It gives
anyone with internet access the opportunity to
use resources whenever they need them
without having to install new software or buy
new equipment. Cloud computing has
emerged as a new and easy-to-use economic
environment for all types of users, whether in
ISBN: 978-1-941968-48-2 ©2018 SDIWC
business, education or even personal use.
Today, Jordanian educational institutions face
many problems in coping with changes in
information technology and rapid
communication, where the development of IT
used in the educational and training process
requires significant costs as well as the cost of
new hardware and software. Cloud computing
is becoming an attractive technology due to
its dynamic scalability and effective usage of
the resources; it can be utilized under
circumstances where the availability of
resources is limited [1]. Cloud computing has
become a research hotspot among modern
technologies; researchers pay more attentions
to its applications. A lot of problems had been
studied, such as the technology for future
distance education cloud [2], teaching
information system [3, 4], teaching systems
development[5], teaching resources
integration[6]. It has been well-recognized
that educational institutions can significantly
benefit from Cloud Computing technology
and get benefit of cost reduction, full
operation, good maintenance, high
management, and out of administrative
challenges, all with high satisfactory Quality-
of-Service (QoS) [7]. This paper will
investigate and demonstrate how Jordanian
Universities can take advantage of the
evolving Cloud Computing paradigms in
providing high quality cost-effective e-
learning services. A proposed solution will
serve all public Jordanian Universities that
spread on three regions: Northern region (
Yarmouk University (YU), Jordan University
for Science and Technology (JUST), Al Al-
Bayt University (AABU)); middle region
(University of Jordan (JU), The Hashemite
7
Proceedings of the Fifth International Conference on Digital Information Processing, E-Business and Cloud Computing (DIPECC2018), Trabzon, Turkey, 2018
University (HU), AlBalqa Applied University
(BAU)); Southern region (Mutah University
(MU), Tafelah technical university (TTU),
Al-Hussein Bin Talal University (AHU)).
This paper is divided into five main sections.
This section introduces the main theme of the
paper. A previous work is given in Section 2.
Section 3 presents a detailed description of
proposed Architecture E-learning center of
excellence based on private Jordanian
Universities. Finally, in Section 5,
conclusions recommendations for future work
are pointed-out.
2 PREVIOUS WORKS
Cloud computing is being deployed in many
organizations as well as in educational
institutions, research in this field could be
categorized into two groups: education as
service and learning as service. In the first
group, [8] survey the state of the art on the
use and research of cloud computing in
education taking in account the main
educational stakeholders, exemplified and
categorized the advantages and risk of CC in
education, identify and discuss promising
technical issues following with systematic
methodology, they reviewed 112 scientific
works on cloud computing in education up to
year 2012. [9] investigated whether there is a
need between students for applications and
services in the “cloud” software as services
(SaaS), the extent to which they use them and
what types of applications and services are
leading. They analyzed and interpreted the
results which provide indications of students'
willingness to “move to the cloud”. [10]
discussed the importance of online training
and emphasizes on its qualitative and
quantitative development for some
organizations. They concentrated on utilizing
online education based on cloud computing
environments. They discussed the necessity of
cloud-based educational systems for
organizations and countries. [11] reviewed the
adoption of CC in education in developing
countries and Arabic countries in particular.
The second group of literature survey, e-
learning as service tackled in many articles,
ISBN: 978-1-941968-48-2 ©2018 SDIWC
these will include, but not be restricted to:
[12] claimed that E-learning system based on
CC infrastructure is feasible and it can greatly
improve the efficiency of investment and the
power of management, which can make E-
learning system development into a virtuous
circle and achieve a win-win situation for
suppliers and customers. [13] show that, it is
essential for an educational and learning
organization, with its budget restrictions and
sustainability challenges, to use the cloud
formations best suited to its IT activities. A
cloud computing formation model was
proposed. [14] focused on the ways of
applying cloud computing to transfer
traditional e-Learning to a cloud based e-
Learning with overall half less costs (i.e.
implementation, maintenance, update and re-
engineering). [15] described the concept of e-
learning and cloud computing, how to
implement it, the efficient utilization of
hardware resources and software in a learning
environment.
3 PROPOSED ARCHITECTURE
Private cloud is proposed to use benefit of
existed, developed and maintained services
and infrastructure on a Jordanian private
network, it could be built on Jordanian
Universities Network (JUNET) infrastructure;
therefore, the adoption of E-Learning center
of excellence (ELCE) could be established on
geographical determined regional covered
area, i.e. to chose one data center in north part
of Jordan as well one in middle another in
southern region, and according to technical
readiness and geographical location, JUST,
JU and MU could be suggested for north,
central and southern region respectively,
where the north part of Jordan covered Irbid,
Ajloun, Jerash and Mafraq governorates,
central part covered Balqa, Amman, Zarqa
and Madaba, while southern part covered
Karak, Tafilah , Maan and Aqaba, Figure 1.
Using this scenario will raise the
availabilities, reliabilities of planning
services. This scenario will be used as a
redundant infrastructure for all universities
members; as well they will use benefit of
8
Proceedings of the Fifth International Conference on Digital Information Processing, E-Business and Cloud Computing (DIPECC2018), Trabzon, Turkey, 2018

replication approaches to reduce the service increasing the connection bandwidth and
risk. developing their data centers to achieve
scalability, persistent storage, distributed
access, efficient resource usage and
interoperability of the E-learning center of
excellence objects. The proposed private JU
cloud offers greatest level of privacy,
security, reliability and control, but it is
required the core data centers to purchase and
maintain all needed software, licenses‟ and
needed infrastructure hardware which slightly
increases the initial cost of proposed cloud.

Figure 1. Map of the governorates of Jordan

According to the figure 1, Table 1. shows the

universities distribution on the regions; where
JUST chosen to be part of private cloud in
north, JU in central and MU in the southern
part. These three universities will be the core
infrastructure of the suggested private cloud,
Figure 2. Proposed private JU cloud
figure 2. The core of proposed infrastructure
responsible for various aspects including 4 ELCE PRIVATE CLOUD
running the major database application
services, housing, maintaining and operating ELCE as service will be an Internet based
the various equipment on behalf of the learning process, using built JU private cloud
member universities. to design, implement, manage, support and
extend learning, which will be greatly
Table 1 universities distribution on the regions
improve the efficiency of proposing service.
Number Region Governorate University
Universities will use benefit of designed
1 North Irbid YU, JUST content, share architecture infrastructure as
2 North Ajloun well use virtualization technology. JU private
3 North Jerash cloud in this case supposed to be marketing
4 North Mafraq AABU term for a proprietary computing architecture
5 Central Balqa BAU that provides hosted services such as ELCE
6 Central Amman JU Figure 3. where advances in virtualization and
distributed services environments will allow
7 Central Zarqa HU
corporate infrastructure network and proposed
8 Central Madaba
data centers as core (JUST, JU, MU) to
9 South karak MU become service providers that meet the needs
10 South Tafilah TTU of universities‟ clients within the corporation.
11 South Ma'an AHU Proposed scenario ensures that learning
12 South Aqaba universities do concentrate more on research
and learning, rather than on implementing
Suggested universities (JUST, JU, MU) must complex IT infrastructure. ELCE will form
be fully connected and they could use the the basis of future IT infrastructure in Jordan
Jordanian Universities network (JUNet) after to ensure the development of hardware and

ISBN: 978-1-941968-48-2 ©2018 SDIWC 9

Proceedings of the Fifth International Conference on Digital Information Processing, E-Business and Cloud Computing (DIPECC2018), Trabzon, Turkey, 2018
software environment. It will do so by
enhancing the kind of resources utilized for
education and e-learning purposes, reducing
the cost, enhancing the security of the
information.
Figure 3. part of private JU cloud as service provider
5 EXPECTED BENEFIT FROM THE
ARCHITECTURE
The following advantages derived from the
proposed architecture are as follows but the
will be checked in future research:
a. High availability. Through the integration
of data storage and high-performance
computing power in proposed data centers;
this scenario can provide a higher quality
of service. JU private cloud computing
architecture can automatically detect the
node failure and exclude it, use benefit of
redundancy approach in the scenario it will
keep the normal operation of the solution.
b. Private JU cloud based on powerful
computing and storage capacity: ELCE
architecture locates the computing and data
in a large number of distributed computers,
the proposed private cloud based on three
data centers provide powerful computing
resources and huge data storage space, puts
it as a service available to universities
students and staff.
c. High level of security. In the private JU
cloud computing model, data is storied
intensively. Relying suggested data
centers, the managers manage the unified
data, allocate the resources, balance load,
deploy the software, control security, and
ISBN: 978-1-941968-48-2 ©2018 SDIWC
do the reliable real time monitoring, thus
guarantee the users‟ data security to the
greatest possible degree. In addition this
solution will have it is security circle with
all its keys and algorithm suggested by
universities which will be hot topic of
future research, term of security and
privacy in data using some encryption
techniques, authentication and
authorization.
6 CONCLUSIONS AND FUTURE
WORKS:
A proposed private JU cloud scenario is a
new approach to produce a solution for many
problems. This solution offers many benefits
to Jordanian Universities such as: managing
effectively their technological needs on base
of delivery of services providing as well
development platforms. They will use benefit
of solution opportunities, advantages‟ such as
accessing the file storages, data bases, and
educational resources as well it will be a core
for many „as services‟ and core of center of
excellences in many research fields. Future
research will include a study regarding the
attitude and strategy for migration to the
proposed architecture, from point of views of
cost reduction, efficiency increasing, and
flexibility of implementation, scalability,
accessibility, redeployment and sustainability.
REFERENCES
[1] Masud, A. and X. Huang, An E-learning System
Architecture based on Cloud Computing.
International Journal of Computer, Electrical,
Automation, Control and Information Engineering,
2012. 6: p. 255-259.
[2] Pranav, P., A Study of E-Learning in Distance
Education using Cloud Computing. International
Journal of Computer Science and Mobile
Computing, 2016. 5(8): p. 110-113.
[3] Juan, Y. and S. Yi-xiang, The Initial Idea of New
Learning Society which Based on Cloud
Computing. Modern Educational Technology
2010. 20(1): p. 14-17.
[4] Yongbin, Z., L. Ronghua, and Y. Shulin, Cloud
Based Working and Teaching System in Higher
Education A Case Study of Teaching Computer
Networking in International Conference on
Education Technology and Information System.
2013. p. 568-572.
[5] Mahmoud, O., W. Kevin, and G.-P. Alexeis, The
Impacts of Cloud Computing Adoption at Higher
Education Institutions: A SWOT Analysis.
10
Proceedings of the Fifth International Conference on Digital Information Processing, E-Business and Cloud Computing (DIPECC2018), Trabzon, Turkey, 2018

International Journal of Computer Applications

2015. 127(4): p. 15-21.
[6] Mohammed, B., L. Ouidad, and E. Najib, Design
of a Cloud Learning System Based on Multi-
Agents Approach. International Journal of
Advanced Computer Science and Applications,
2015. 6(3): p. 20-26.
[7] Alzoubaidi, A., Multi-Campus Universities
Private-Cloud Migration Infrastructure.
International Journal on Cloud Computing:
Services and Architecture, 2016. 6(3): p. 1-13.
[8] Gonzalez-Martinez, J.A., M Bote-Lorenzo E
Gómez-Sánchez and R Cano-Parra, Cloud
computing and education: A state-of-the-art
survey. Computers & Education An International
Journal, 2015. 80: p. 132-151.
[9] Kurelović, E.K., S. Rako, and J. Tomljanović.
Cloud computing in education and student's needs
in 36th International Convention on Information
and Communication Technology, Electronics and
Microelectronics. 2013. Croatia.
[10] Bouyer, A. and B. Arasteh, The Necessity Of
Using Cloud Computing In Educational System.
Procedia Social and Behavioral Sciences 2014.
142: p. 581 – 585.
[11] Karim, F. and G. Rampersad, Cloud Computing in
Education in Developing Countries. Computer and
Information Science, 2017. 10(2): p. 87-96.
[12] Riahi, G., E-learning Systems Based on Cloud
Computing: A Review. Procedia Computer
Science, 2015. 62: p. 352-359.
[13] Alabbadi, M. Cloud computing for education and
learning: Education and learning as a service
(ELaaS). in 14th International Conference on
Interactive Collaborative Learning. 2011.
Slovakia.
[14] Mohammadi, S. and Y. Emdadi, E-Learning Based
on Cloud Computing. International Journal of
Basic Sciences & Applied Research, 2014. 3(11):
p. 793-802.
[15] Gangi, B. and Z. Jafari, Sol Gangi, B. and Z. Jafari,
Solution to implement e- learning system based on
cloud computing. International Journal of
Scientific & Engineering Research, 2015. 6(1): p.
89-92.

ISBN: 978-1-941968-48-2 ©2018 SDIWC 11

Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018

Design of modular multi-level converter and permanent magnet

synchronous generator-based wind energy conversion system

Hakam H. Abdul Hakeem Ahmet Mete Vural

Department of Electrical and Electronics Engineering Department of Electrical and Electronics Engineering
Gaziantep, Turkey Gaziantep, Turkey
hakam.hikmat@yahoo.com mete.vural@gaziantep.edu.tr

Abstract—In this paper, a back-to-back (BTB) PSC-PWM technique [11]. Combining the averaging
modular multi-level converter (MMC) is designed for control with capacitor voltage balancing control
a permanent magnet synchronous generator (PMSG) enables capacitor voltage balancing issue without any
based wind energy conversion system (WECS). The external balancing circuit.
switching of power semiconductors of BTB-MMC is
based on phase-shifted carrier pulse width modulation II. PMSG BASED WECS DESIGN
(PSC-PWM). Capacitor voltage balancing issue of the
Two MMCs can be connected from their DC link
BTB-MMC is realized with a control system working
to form a BTB arrangement. Thus, BTB-MMC can be
in accordance with PSC-PWM technique. The
used in a WECS. The MMC near wind generator
designed PMSG based WECS is verified through case
operates as an AC-DC converter, while the other one
studies using PSCAD/EMTDC.
is connected to AC grid and operates as a DC-AC
converter. By this way, the real power generated by
Keywords: Permanent magnet synchronous
the wind turbine is transferred to the AC grid in a
generator; Wind energy conversion system; Modular
controllable way. The sub-module (SM) or cell is the
multi-level converter; Back-to-back converter.
fundamental building block of an MMC. Each SM can
be constructed by a half-bridge (HB) having two
I. INTRODUCTION power semiconductors (IGBT) and a capacitor [12],
[13]. Fig. 1 shows the switching states of a SM. Table
With the quick exhaustion of conventional fossil
1 lists the SM capacitor cases. The PMSG based
fuels, many renewable energy sources have emerged
WECS design using a BTB-MMC arrangement is
as substitutions for conventional sources [1]. In recent
shown in Fig. 2. Each MMC has three legs and each
years, the wind energy has become the fastest
leg contains an upper and a lower arm [12], [14]. Four
expanding renewable source [2], [3]. The permanent
sub-modules (SM) or cells are connected in series in
magnet synchronous generator (PMSG) is one of the
each arm. With this design, line-to-line AC voltage of
growing trend in wind energy conversion system
each MMC has 2n+1=17 levels, where n is the
(WECS) design because of many advantages such as
number of cells per arm).
no need for gearbox and rotor excitation, brushless
structure, low maintenance cost/weight, and wide
range of speed control [4],[5],[6]. It is necessary to use S1 D1 S1 D1

a power converter in WECS to control real/reactive +

C
+

+ C Vcmp + Vcmp

power efficiently [7], [8]. Consequently, PMSG needs - -

full-scale back-to-back (BTB) converter for grid- Vmp S2 D2

Vmp S2 D2

connection. In recent years, multilevel converters such - -

(a) Case 1 (i>0) (b) Case 2 (i>0)

as diode-clamped, flying-capacitor, cascaded H-
bridge, and modular multilevel converter (MMC) have S1 D1
S1 D1

gained popularity in almost all voltage and power + C

+
Vcjp + C
+

levels due to advantages such as low harmonic content, Vjp S2 D2

-

Vjp S2 D2

small filtering requirement, and no need for a coupling

-

transformer [9]. The MMC was first proposed by (c) Case 3 (i<0) (d) Case 4 (i<0)
Marquardt, R. for high voltage direct current Fig. 1. Switching states of SM.
transmission [7]. MMC is very suitable for medium
voltage/high power WECS applications due to modular Table 1. SM capacitor cases.
circuit topology and independent control capability of Current direction SM state Capacitor case
each MMC [10], [11]. In this paper, a back-to-back Positive i>0 ON Charge
(BTB)-MMC topology is applied to a PMSG based Negative i<0 OFF Bypass
WECS. On the basis of the works realized in [4], the Positive i>0 OFF Bypass
control strategy of the BTB-MMC is realized based on Negative i<0 ON Discharge

ISBN: 978-1-941968-47-5 ©2018 SDIWC 1

Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018

Fig.2. BTB-MMC based PMSG WECS.

III. PHASE-SHIFTED CARRIER PULSE WIDTH 1

𝑣𝑐𝑣𝑔 = ∑8𝑚=1 𝑣𝑐𝑣𝑚 (1)
MODULATION (PSC-PWM) 8

Phase-shifted carrier pulse width modulation As shown in Fig. 2, 𝑖𝑝𝑣 is a positive current of phase-
(PSC-PWM) has been considered in this work due to v, and 𝑖𝑛𝑣 is a negative current of phase-v. From the
easy capacitor voltage balancing, equally distributed block of the averaging control in Fig. 3, the circulating
switching frequency in each SM [16], low harmonic current 𝑖𝑧𝑣 , can be defined as:
distortion, and lower semiconductor losses [15], [16].
𝑖𝑣 𝑖𝑣
PSC-PWM has two waveforms. If the triangular 𝑖𝑧𝑣 = 𝑖𝑝𝑣 - = 𝑖𝑛𝑣 +
2 2
carrier is higher than the reference waveform, the SM 1
is bypassed, otherwise, the SM is inserted. The phase = (𝑖𝑝𝑣 + 𝑖𝑛𝑣 ) (2)
2
shift between these two waveforms is 2π/n. PSC-
PWM technique needs to accomplish two tasks 𝑖𝑧𝑣 is a loop current that is impossible to measure
(average control and balancing control) in order to directly whereas 𝑖𝑣 is a branch current [11]. As shown
complete its function completely. in Fig. 3, the 𝑖𝑧𝑣𝑔 is the current command of the 𝑖𝑧𝑣 ,
and can be defined by
A. Average Control

The circulating current occurs due to the
imbalance of the average capacitor voltages between
the phases of the MMC [9], [11]. The circulating
current has no influence on the AC or DC side; it only
flows through the phases of the MMC. Consequently,
the circulating current causes power losses and ripples
observed in capacitor voltages of the SMs [15], [17].
The arm inductors as shown in Fig. 2 are designed to
lower the circulating current. Fig. 3 shows the overall
control scheme of each MMC, including DC capacitor
voltage control and circulating current control for v-
phase in AC-DC MMC. The control scheme for other
phases is the same. Fig. 3 shows the block of the
averaging control. It forces the v-phase average
voltage 𝑣𝑐𝑣𝑔 to follow its command 𝑣∗ 𝑐 , where the
𝑣𝑐𝑣𝑔 is given by
𝑖𝑧𝑣𝑔 = 𝑘1 ( 𝑣∗ 𝑐 - 𝑣𝑐𝑣𝑔 ) + 𝑘2 ∫( 𝑣 ∗ 𝑐 − 𝑣𝑐𝑣𝑔 )𝑑𝑡 (3)
The driving voltage 𝑣𝐴𝑣𝑔 acquired from the averaging
control is shown in Fig. 3, given by
𝑣𝐴𝑣𝑔 = 𝑘3 (𝑖𝑧𝑣 - 𝑖𝑧𝑣𝑔 ) + 𝑘4 ∫(𝑖𝑧𝑣 − 𝑖𝑧𝑣𝑔 )𝑑𝑡. (4)
When 𝑣𝑐𝑣𝑔 ≤ 𝑣∗ 𝑐 , 𝑖𝑧𝑣𝑔 increases. The current miner
loop is used to force the actual DC-loop current 𝑖𝑧𝑣 to
follow the 𝑖𝑧𝑣𝑔 .
B. Balancing Control
Fig. 3 shows the balancing control block for
phase-v. The balancing control is used to force the
individual DC voltage to follow its command 𝑣∗ 𝑐 [11].
As shown in Fig. 3, the gained output voltage from the

ISBN: 978-1-941968-47-5 ©2018 SDIWC 2

Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018

Fig. 3. Control scheme of MMC1.

balancing control is 𝑣𝐵𝑚𝑣𝑔 . The balancing control is As shown in Fig. 3, 𝑣∗ 𝑚𝑣 is the AC-voltage that is
based on the phase current that is either positive or gained from the balancing after passing through
negative. The polarity of 𝑣𝐵𝑚𝑣𝑔 must be due to 𝑖𝑝𝑣 or positive and negative command arm. Every SM
𝑖𝑛𝑣 [17], [18]. When 𝑣∗ 𝑐 is greater than or equal the capacitor voltage 𝑣𝑐𝑣𝑚 normalizes voltage
𝑣𝑐𝑣𝑚 (m: 1−4) in the upper arm. The positive active command 𝑣 ∗𝐴𝑣 [11], [17].
power has to be taken from the DC-link into the four
IV. SIMULATION STUDIES
SM. When 𝑖𝑝𝑣 is positive, the product of 𝑣𝐵𝑚𝑣 and 𝑖𝑝𝑣
generates the positive active power. When 𝑖𝑝𝑣 is BTB-MMC based PMSG WECS, presented in
negative, the polarity of 𝑣𝐵𝑚𝑣𝑔 must get reverse to Fig. 1, is simulated in PSCAD/EMTDC environment.
occupy the positive active power. At last, for m =1~4 The model parameters used for simulation are listed in
is represented as: Table 2. The data of the wind turbine to produce the
mechanical torque to the PMSG are given in Table 3.
𝐾5 ( 𝑣∗ 𝑐 − 𝑣𝑐𝑣𝑚 ) (𝑖𝑝𝑣 > 0) The output of the WECS is connected to the AC grid
𝑣𝐵𝑚𝑣𝑔 = { ∗ (5) whose parameters are given in Table 4.
−𝐾5( 𝑣 𝑐 − 𝑣𝑐𝑣𝑚 ) (𝑖𝑝𝑣 < 0)
Table 2. Model parameters used for simulation.
where m, is the number of cell. However the equation Rated active power 5 MW
of 𝑣𝐵𝑚𝑣𝑔 for m = 5 ~ 8 is: Rated frequency 50 Hz
DC-link voltage 9 kV
𝐾5 ( 𝑣∗ 𝑐 − 𝑣𝑐𝑣𝑚 ) (𝑖𝑛𝑣 > 0) arm inductance 3 mH (6.4%)
𝑣𝐵𝑚𝑣𝑔 = { (6)
SM capacitor voltage 2.25 kV
−𝐾5( 𝑣∗ 𝑐 − 𝑣𝑐𝑣𝑚 ) (𝑖𝑛𝑣 < 0)
Carrier frequency 2 kHz
The voltage command of every sub-module 𝑣∗ 𝑚𝑣 is Coupling inductance 6 mH
shown below for positive-arm and negative-arm
Table 3. Wind turbine parameters.
commands:
Wind speed 12 m/m
𝑣∗ 𝐸 Frequency 50 Hz
𝑣∗ 𝑚𝑣 = 𝑣∗ 𝐴𝑣 + 𝑣𝐵𝑣𝑚𝑔 - 𝑣
+ (m: 1 ~ 4) (7) Rotor radius 35 m
4 8
𝑣∗ 𝑣 𝐸
𝑣∗ 𝑚𝑣 = 𝑣∗ 𝐴𝑣 + 𝑣𝐵𝑣𝑚𝑔 + + (m: 5 ~ 8) (8) Active power 5 Mw
4 8

ISBN: 978-1-941968-47-5 ©2018 SDIWC 3

Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018

Table 4. AC grid parameters. Fig. 5 (a) Active and reactive power output of PMSG, (b) DC-link
Frequency 50 Hz voltage of BTB-MMC, (c) Circulating current.
Rated voltage L-L rms 5.5 kV Fig. 5 shows the active and reactive power output of
the PMSG, DC-link voltage of BTB-MMC, and
In this work, the controllers of the WECS are circulating current, respectively. As shown, the
examined under two cases: under steady-state rated circulating currents are minimized. Fig. 6 shows the
conditions and under dynamic state when DC-link active and reactive power of DC-AC MMC and line-
voltage is dropped. to-line voltage of DC-AC MMC, respectively. From
these waveforms, we can see that the DC-AC MMC
A. Case 1: Under Rated Conditions side is able to output 17 levels at the line-to-line
voltage converted from the DC-link voltage that has
Fig. 4 shows the simulated waveforms of line-to- been delivered by AC-DC MMC.
line voltages, phase currents, and line-to-neutral
voltages of AC-DC MMC under rated conditions,
respectively.

(a)

(a)

(b)
Fig. 6 (a) The active and reactive power of DC-AC MMC, (b) Line-
to-line voltage of DC-AC MMC.
(b)
In order to generate high quality AC voltage, the DC
capacitor voltage of SM per each arm must be equal.
Fig. 7(a) shows the DC capacitor voltages of eight
SMs in MMC1. As observed, DC capacitor voltages in
phase-u of MCC1 has no ripples. Fig. 7(b) shows the
DC capacitor voltages of eight SMs in phase-w of
(c)
Fig. 4 (a) Line-to-line voltage of AC-DC MMC, (b) Phase currents MMC2. Theoretically, DC capacitor voltage of each
of AC-DC MMC, (c) Line-to-neutral voltage of AC-DC MMC. SM is 2.25 kV. This value can be calculated as:

DC supply voltage 9
𝑣∗ 𝑐 = = (9)
number of SM in upper arem or lower 4

(a)

(a)

(b)

(b)
Fig. 7 (a) DC capacitor voltages of AC-DC MMC, (b) DC capacitor
(c) voltages DC-AC MMC.

ISBN: 978-1-941968-47-5 ©2018 SDIWC 4

Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018
B. Case 2: Under Dynamic Case
In this case study, the dynamic performance of the
control scheme is examined when DC link voltage set
point is dynamically changed for a duration of 0.05s
as shown in Fig. 8(a). As observed from the simulated
waveforms presented in Fig. 8, the control system of
the WECS was able to ride through these dynamic
states smoothly.
(a)
(b)
(c)
(d)
(e)
Fig. 8 (a) DC link voltage set-point change, (b) Line to line voltage
of MMC1, (c) Active and reactive power of DC/AC side, (d) Phase
currents of MMC2, (e) Active and reactive power of PMSG.
After the drop of DC link voltage, three-phase AC
voltage of both MMCs are also diminished. Since, the
DC link voltage is used as reference voltage in the
controller for both MMCs. Thus, the supplied and
absorbed active power of MMC1 and MMC2 are
respectively decreased. Fig. 9(a) shows the voltages of
eight SMs in phase-w of MMC1 when DC link voltage
set-point change is applied. The voltages of each SM
ISBN: 978-1-941968-47-5 ©2018 SDIWC
has returned to their nominal values after some
disturbances.
(a)
(b)
(c)
Fig. 9 (a) DC-capacitor voltages of MMC1, (b) DC-capacitor
voltage of MMC2, and (c) voltages of AC three phase voltage source
V. CONCLUSION
The design and control of the power converters in
a WECS is still challenging. The design and
simulation of BTB-MMC based PMSG WECS is
proposed in this work. BTB-MMC arrangement brings
many advantages to WECS design such as being
flexibility and modularity. Moreover, MMC is able to
generate higher voltages with diminished voltage stress
on each power electronic semiconductor. Harmonic
levels can also be significantly decreased thanks to the
increased number of voltage levels. As a result of this,
filtering requirements are becoming less. MMCs have
also better fault ride through capabilities. PSC-PWM
method is used to trigger power semiconductors in
each SM. Balancing and average control methods are
designed to achieve the balancing of capacitor voltage
in each SM in both MMCs. In order to verify the
efficiency of the proposed control scheme, two cases
studies were performed in this work. As observed from
the simulation results, the dynamic performance of the
proposed control scheme is satisfactory.
REFERENCES
[1] Yaramasu, Wu, Sen, Kouro, and Narimani, "High power Wind
Energy Conversion Systems: State-of-the-Art and Emerging
Technologies, " IEEE. Vol. 103, No. 5, May 2015, pp. 736-
739
5
Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018

[2] Henk, P., Jan Abraham, "Trends in Wind Turbine Generator [17] Suman Debnath, Jiangchao Qin, Behrooz Bahrani, Maryam
Systems,” IEEE Jornal of emerging and selected topics in Saeedifard, and Peter Barbosa, “Operation, Control, and
power electronmics Vol. 1, No. 3, Sep 2013, pp. 174-185 Applications of the Modular Multilevel Converter: A
[3] Yaramasu, Wu, Sen, Kouro, and Narimani, "High-Power Review,” IEEE Transactions on Power Electronics., vol. 30,
Wind Energy Conversion Systems: State-of-the-Art and Jan. 2015, pp. 37–53
Emerging Technologies,” Proceedings of the IEEE., Vol. 103, [18] Hirofumi Akagi, Shigenori Inoue, and Tsurugi Yoshii,
No. 5, May 2015, pp. 736-739 “Control and Performance of a Transformerless Cascade
[4] Mian, W., Yaowei, H., Wenjian, Z., Yue W, Guozhu C, PWM STATCOM With Star Configuration,” IEEE
"Application of modular multi level converter in meduim Transactions on industry Application, VOL. 43, NO. 4,
voltage high power permanent magnet synchronous generator July/August 2007, pp. 1041-1049
wind energy conversion system,” IET Power. GEN.
Distrib.,Mar 2016, pp. 1–10
[5] M.S. Camara, M.B. Camara, B. Dakyo and H. Gualous,
"Permanent Magnet Synchronous Generators For Offshore
Wind Energy System Linked To Grid Modeling And Control
Strategies", IEEE, 16th International Power Electronics and
Motion Control Conference, PEMC2014, Antalya/Turkey,
September, 2014
[6] Tan Luong, Thanh Dung, Thanh Trang, and Huy Dung
"Advanced Control Strategy of Back-to-Back PWM
Converters in PMSG Wind Power System,” Power
Engineering and Electrical Engineering. Vol. 13, June, 2015,
pp. 81-95
[7] A. Lesnicar and R. Marquardt, "An innovative modular
multilevel converter topology suitable for a wide power
range," in Proc. 2003 Power Tech Conference Proceedings,
2003 IEEE Bologna
[8] W. Abd Halim, S. Ganeson, M. Azri, T.N.A. Tengku Azam,
“Review of Multilevel Inverter Topologies and Its
Applications,” Journal of Telecommunication, Electronic and
Computer Engineering, Vol. 8 No. 7, pp. 51-56
[9] Makoto Hagiwara, Kazutoshi Nishimura, and Hirofumi
Akagi, “A Medium-Voltage Motor Drive With a Modular
Multilevel PWM Inverter,” IEEE Trans. Power Electronic.,
Vol. 25, No. 7, Jul 2010, pp. 1786-1799
[10] Ahmed, N.; Haider, A.; van Hertem, D.; Zhang, L.; Nee, H.-
P. Prospects and challenges of future HVDC Super Grids with
modular multilevel converters. In Proceeding s of the 14th
Conferenceon Power Electronics and Applications (EPE
2011), Birmingham, UK, 30 August–1 September 2011; pp.
1–10.
[11] M. Hagiwara and H. Akagi, “Control and experiment of pulse
width modulated modula rmultilevel converters,” IEEE Trans.
Power Electron., vol. 24, no. 7, Jul. 2009, pp. 1737–1746
[12] H. Vahedi and K. Al-Haddad, "Half-bridge based multilevel
inverter generating higher voltage and power," in Electrical
Power & Energy Conference (EPEC), 2013 IEEE, 2013, pp.
1-6. Makoto Hagiwara, and Hirofumi Akagi, “Control and
Experiment of Pulsewidth-Modulated Modular Multilevel
Converters,” IEEE Trans. Power Electron., Vol. 24, No. 7, Jul
2009, pp. 1737–1746
[13] R. H. Baker, "Switching circuit," ed: US Patent 4,210,826,
1980
[14] Pracha Khamphakdi, Kei Sekiguchi, Makoto Hagiwara, , and
Hirofumi Akagi, “Design and Experiment of a Back-To-Back
(BTB) System Using Modular Multilevel Cascade Converters
for Power Distribution Systems” IEEE Trans. Power Electron,
2013, pp. 311-317
[15] Debnath, S., Qin, J., Bahrani, B., et al.:‘Operation, control, and
applications of the modular multilevel converter: a review’,
IEEE Trans. Power Electron., 2015, 30, (1SI), pp. 37–53
[16] Manish V. Kurwale ,Palak G. Sharma , Gautam Bacher,
“Performance Analysis of Modular Multilevel Converter
(MMC) with Continuous and Discontinuous Pulse Width
Modulation (PWM) ,” International Journal of Advanced
Research in Electrical, Electronics and Instrumentation
Engineering., Vol. 3, Issue 2, February 2014, pp. 7463-7474

ISBN: 978-1-941968-47-5 ©2018 SDIWC 6

Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018
Performance Evaluation of Underwater Acoustic
Communication using Trigonometric Chirp
Modulation
Ali Emadi
School of Electrical and Computer Engineering
Shiraz University
Shiraz, Iran
a.emadi@shirazu.ac.ir
Abstract—In this paper, we propose an underwater acoustic
communication based on trigonometric chirp modulated
waveforms. We show that a usual system with linear chirp
modulation requires wide bandwidth. Therefore, a Doppler
resilient and robust underwater digital communication system
based on a non-linear frequency modulation as trigonometric
chirp technique has been presented.
To evaluate the performance of trigonometric chirp modulation
technique over AWGN channel, Rayleigh fading and Rician
channels with and without Doppler effect, we have conducted
some simulations. It is demonstrated that the performance of
proposed method, even in lower spreading factors gives better
performance than an LFM chirp modulated system in terms of
bit error rate.
Keywords—Underwater Acoustic Communication, Linear
frequency Modulation (LFM), Trigonometric Chirp, Doppler
Shift, Rayleigh Fading, Rician Fading.
I. I NTRODUCTION
Over past decades, many underwater wireless
communication technologies have been proposed and
applied as ocean exploration, oceanography data collection,
control over autonomous underwater vehicles, undersea
navigation and etc.
When studying problems of sending data from transmitter
to receiver, the distortion induced by the channel lead us
to analyze for then being able to choose a right signal
waveform. A reliable underwater acoustic communication is
challenging because of some channel characteristics such as;
small bandwidth, high power attenuation, fast time variation
of the channel response, fading along of multipath propagation
and Doppler effect due to relative speed between the receiver
and the transmitter.
Actually in UWA communication, due to the low velocity
of acoustic waves (roughly 1500 m/s) Doppler shift is larger
than terrestrial radio frequency communication. But this is
not the only impact of Doppler in acoustic waves. Unlike
the case of radio transmission which Doppler shift is usually
modeled as a frequency offset, the effect of Doppler shift on
the symbol duration cannot be neglected. Significant Doppler
ISBN: 978-1-941968-47-5 ©2018 SDIWC
Ali Jamshidi
School of Electrical and Computer Engineering
Shiraz University
Shiraz, Iran
Jamshidi@shirazu.ac.ir
shift results loss of information and frame desynchronization,
as well. In addition to relative speed between the source and
the receiver, different sampling frequencies also yield to signal
compression-expansion. This means that for example, if our
DAC is not accurate, we have to consider this problem in the
system.
Recently, there have been considerable attempts to design
robust receivers which prosperously equalize the channel and
also increase transmission rate.
Chirp spread spectrum (CSS) techniques use chirp signals
for data transmission. Chirp is a sinusoidal signal whose
frequency decreases or increases over a specified time
duration. Chirps as another spread spectrum signals are
commonly used in radar applications and many years later in
sonar due to its good temporal resolution of auto correlation
function and also, considerable processing gain obtained
from correlation measurement in the receiver [1]–[9]. In
march 2007, IEEE introduced CSS physical layer in its new
standard as 802.15.4 a, allowing CSS to be used in various
application such as industrial control, sensor networking, real
time location systems and medical devices [10].
Some characteristics of chirp modulations make it robust to
multipath with low Doppler sensitivity. We know that linear
frequency modulation (LFM) is the simplest chirp waveform.
It has some advantages in comparison to non-linear chirps such
as easy generation by different technologies, mostly simple to
process by a matched filter or similar techniques. Accordingly,
more applications for the CSS system based on linear chirps
have been developed than that for non-linear types. But the
main drawback is obtaining orthogonality between two LFM
signals. A large time-bandwidth product, known as spreading
factor, is needed. Either considerable symbol duration or wide
bandwidth causes large spreading factor. The time-bandwidth
product of a linear CSS should be more than 70 to achieve
relatively acceptable orthogonality between its symbols for
binary signaling. As we know, significant bit rate is more
preferable [11]. Thus, by necessity a classical LFM signal will
occupy undesirable large bandwidth especially at high speed
data rate. So it is important to choose a pair of chirp signals
7
Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018

which will be sufficiently orthogonal in small spreading factors fc is the carrier frequency. Both signals have a common
to sending information. Nevertheless, non-linear chirps have bandwidth B, and Ts is the chirp duration in second. We
not attained high prevalent like linear type, so far. Probably it will send s0 as a bit 0 and s1 as a bit 1. At the receiver
is because of limited availability of non-linear-FM generation as shown on Fig.1, where an AWGN channel considered with
and processing devices, more complexity of system, derivation n ∼ N (0, N20 ), two decision variables are produced:
of performance evaluation can be troublous, as well. By
Γi = khr, si ik, i = 0, 1 (2)
accepting these issues, desired performance can be achieved
with considerable reduction in bandwidth. Better performances Where kh∗, ∗ik is the absolute value of the inner product
due to some intrinsic properties of non-linear chirps were between observation r and transmitted signal si . So, we can
motivation of our research. consider the hypothesis test as follows
In this paper, we use trigonometric chirp signals to modulate 
binary data. To obtain a better performance, it is preferable H0 : r = s0 + n
(3)
to consider two orthogonal chirp signals. At the receiver, H1 : r = s1 + n
we consider a matched filter to correlate the received signal
According to the non-coherent maximum a posteriori
with a known signal. We illustrate that a trigonometric chirp
(MAP) detector, the decision rule will be
modulated system has better performance over AWGN and
H0
different fading channels in contrast with a system based on khr, s0 ik − khr, s1 ik>
<H1 0 (4)
LFM chirp.
The remainder of this paper is organized as follows. In section Choosing H1 when H0 occurs and also vice versa, means
II, the orthogonality of LFM and trigonometric chirp as a type an error happens. By totally analyzing expression (4), we
of non-linear frequency modulation have been investigated. In know that inner product of LFM signals is not equal to zero
section III, performances over additive white gaussian noise and performance degradation can be occurred due to this non
(AWGN) and two fundamental fading channels for underwater orthogonality. So, we look for orthogonal signals [11].
acoustic systems as Rayleigh and Rician have been evaluated. Subsequently, the orthogonality between symbols can be
In section IV, a real analytical measurement is given which considered through evaluation of cross correlation coefficient.
predicts our simulated results. Finally, section V concludes Let Eb be the bit energy and defined as
this paper. s * (t ) Z Ts
0
Ts ☎0 Eb = s2i (t)dt, i = 0, 1 (5)
dt ✁
text 0
0
Decision

To evaluate the performance of chirp signaling techniques,

r (t )
the cross correlation coefficient is one of the important factors.
Ts
If we define ρ as the cross correlation coefficient between
dt
text ✂
text
✄1 down-chirp and up-chirp signals, we have:
0
Z Ts
s1* (t ) 1
ρ= s0 (t)s1 (t)dt (6)
Fig. 1: Block Diagram of Optimum Non-Coherent Detector. Eb 0
By considering the signals in base band, ρ will be
II. S YSTEM M ODEL Z T2s
2 B 2 B 2

In this section, signal waveforms have been expressed ρ= (e∓jπ 2T t )∗ e±jπ 2T t dt (7)
Ts 0
mathematically. The orthogonality and Doppler effect on the
signals will be discussed, too. Using some straightforward manipulations, it can simplified
as
A. LFM s √
BTs
√
BTs
πx2 πx2
Z Z
1
|ρ| = √ ( sin dx)2 + ( cos dx)2 (8)
As we know, the chirp signals can be classified into two BTs 0 2 0 2
types; ”linear chirp”, in which the instantaneous frequency
decreases or increases with time in a linear form and ”non-
The above equation is composed of two famous integration
linear chirp”, in which the instantaneous frequency is a non
as Fresnel integrals. From (8), we can see ρ depends on
linear function of time. LFM probably is the most popular
spreading factor BT. As shown in Fig.2, the cross correlation
frequency modulation due to its low complexity. It is widely
coefficient decreases rapidly from 0 to 20 (secHz). For ρ ≤
used before in radar and sonar applications. Here, it is applied
0.1, we should choose BT more than 70 (secHz). This is why
to perform an underwater digital communication. Linear Up
LFM system requires a wide bandwidth for high data rate
and Down chirp can be expressed as
communication.
In case of Doppler effect, since the source and the receiver
B 2
  
 Down :
 s0 (t) = cos 2π(fc t − 2Ts
t ) 0 ≤ t ≤ Ts move with relative speed ν, the signal waveform suffers a
(1)
  B 2

Up : s1 (t) = cos 2π(fc t + t ) 0 ≤ t ≤ Ts

2Ts

ISBN: 978-1-941968-47-5 ©2018 SDIWC 8

Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018
1

Value of Cross Correlation Coefficient (ρ)

Trigonometric Chirp III. S IMULATION R ESULTS
LFM
0.75 Although trigonometric chirp signals have better
orthogonality in comparison to LFM, but the performance
0.5 of trigonometric chirps in underwater environment have
to be investigated over different channels. we will not go
0.25
into detail of theoretical BER. Here, for showing that the
non-coherent detector of trigonometric chirp systems have
better performance, it is adequate to illustrate the experimental
0
results and also, the quiet performances have been carried
out by simulation for LFM and trigonometric chirp signal
−0.25
comparatively.
In the simulations, we have considered the underwater
−0.5
0 20 40 60 80 100 communication system impaired by additive white Gaussian
Time−Bandwidth Product noise, Doppler effect and multipath fading. Some fading
environments such as Rayleigh and Rician, that represent
Fig. 2: Correlation Coefficient for LFM and Trigonometric Chirp.
most practical aspects of underwater acoustic channels are
considered.
compression-expansion. This phenomena can be modeled as
 
ν A. AWGN Channel
sd (t) = si t(1 − ) i = 0, 1 (9)
c
The performance as BER vs. Eb /N0 in noisy channel for
Where c is the velocity of acoustic waves in water the linear and full period trigonometric chirp with different
environment, si is transmitted signal, sd is transmitted signal bandwidth is shown in Fig.3
which is impressed only by Doppler effect. Also, we know
that Doppler frequency shift is ∆fd = νc fc .
0
10

B. Trigonometric Chirp
−1
Generally, a pair of chirps can be written as following 10
Bit Error Probability

 
f0 (t) = cos 2π(fc t − Bθ(t)) 0 ≤ t ≤ Ts Trigonometric Chirp, BW= 400 Hz
−2
(10) 10 Trigonometric Chirp, BW= 500 Hz
  Trigonometric Chirp, BW= 600 Hz
f1 (t) = cos 2π(fc t + Bθ(t)) 0 ≤ t ≤ Ts Linear Chirp, BW= 400 Hz
Linear Chirp, BW= 500 Hz
Linear Chirp, BW= 600 Hz
−3
10
The instantaneous frequency of a chirp as ”chirp rate
µ” which is defined by µ(t) = dθ(t) dt , denotes the type
of chirp signal. Now, it is necessary to express that the −4
10
trigonometric chirp signal has different types based on its
frequency spectrums. Here, we use full period trigonometric
−5
chirps whose frequency spectrum is a full period cosine or 10
−2 0 2 4 6 8
sine curve, just because of better performance in comparison Eb/N0 (dB)

to other types. Then, full period trigonometric chirp can be

presented as Fig. 3: Linear and Trigonometric chirp BER over AWGN Channel, T = 20 ms.
"  #
BTs
s0 (t) = cos 2π fc t − 4π
sin(2π Tt ) 0 ≤ t ≤ Ts We observe lower BER for trigonometric chirp than LFM
s
with different bandwidth. Also, as shown in Fig.3, we should
(11)
"  # pay attention that increasing bandwidth does not mean
BTs
s1 (t) = cos 2π fc t + 4π
sin(2π Tt ) 0 ≤ t ≤ Ts better performance, necessarily. In the following Fig.4, the
s
simulated results of the detector over a noisy channel with
The cross correlation between two full period trigonometric and doppler effect is shown. As we know, Doppler shift is
chirps can be evaluated numerically [12]. The result is shown impacted by several parameters such as carrier frequency fc ,
in Fig.2. Under the same condition of time-bandwidth, the sound velocity in water c and relative speed between the
system based on the full period trigonometric chirp requires source and the receiver ν.
much narrow bandwidth in comparison with an LFM chirp In the simulations, a carrier frequency of fc = 8 kHz and
modulated system for a predefined data rate. different relative speeds were used.

ISBN: 978-1-941968-47-5 ©2018 SDIWC 9

Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018
0 0
10 10

−1 −1
10 10
Bit Error Probability

Bit Error Probability

−2 −2
10 10

−3 −3
10 10

−4 −4
10 10
Trigonometric Chirp, Fd=50 Hz
Trigonometric Chirp, Fd= 150 Hz
−5 Trigonometric Chirp, Fd=150 Hz
10 −5 Trigonometric Chirp, Fd= 50 Hz
Linear Chirp, Fd=50 Hz 10
Linear Chirp, Fd= 50 Hz
Linear Chirp, Fd=150 Hz Linear Chirp, Fd= 150 Hz
−6 −6
10 10
−2 0 2 4 6 8 −2 0 2 4 6 8
Eb/N0 (dB) Eb/N0 (dB)

Fig. 4: Linear and Trigonometric chirp Bit Error Rate over Noisy and Doppler Fig. 6: Linear and Trigonometric chirp Bit Error Rate over Rayleigh fading Channel,
Channel, BT = 4 and fc = 8kHz. BT = 4 and fc = 8kHz.

0
10
Also, Fig.5 shows that how the performances of both linear Trigonometric Cirp, K=1
Trigonometric Chirp, K=2
and trigonometric chirp are affected by different Doppler shifts −1
10
Trigonometric Chirp, K=5
Linear Chirp, K=1
in a fixed SNR and BT. We can see that the full period

Bit Error Probability

Linear Chirp, K=2
Linear Chirp, K=5
trigonometric chirp in low Doppler frequencies has better −2
10
performance due to its mathematical nature.
−3
10

0
10
−4
10

−1
10 −5
10
Bit Error Probability

−6
−2 10
10 0 5 10 15 20
Eb/N0 (dB)

−3
Fig. 7: Linear and Trigonometric chirp Bit Error Rate over Rician Channel, BT = 4.
10

As shown in Fig. 6 and Fig. 7, the full period trigonometric

−4
10
Trigonometric Chirp
chirp modulated system has better performance in comparison
LFM Chirp
with the system based on an LFM chirp modulated over
−5
fading channels.
10
0 100 200 300 400 500
Doppler Shift(Hz)

IV. E XPERIMENTAL R ESULTS

Fig. 5: Linear and Trigonometric chirp Bit Error Rate vs. Doppler Shift, BT = 4,
SNR = 2 dB and fc = 8kHz.
The set of our underwater experiments were conducted in
a water tank with 3m × 1m × 0.6m dimensions in the Sonar
laboratory at Shiraz University, using one SX147 spherical
B. Rayleigh and Rician Fading Channels transducer and one B&K hydrophone type 8105. Different
In this section, the performance of a CSS system using time duration T = 10 and 20 msec were used for transmission.
linear and trigonometric chirp signals over two fundamental We sent about 3500 bits of information with spreading
channels i.e. Rayleigh fading channel with Doppler shift and factor BT = 4, sampling frequency fs = 32 kHz and carrier
Rician fading channel, will be illustrated in Fig.6 and Fig.7. frequency fc = 8 kHz for linear and trigonometric chirp
signals. Each frame contained 100 bits/frame, a 10 msec LFM
In Rician fading channel with a basic point of view, we can signal as probe signal for synchronization at the beginning of
consider the K factor as frame and a 50 msec silence as Guard before the data and
P ower of LOS path
one another Guard after that were structured as the data packet.
K= (12)
P ower of N on LOS paths

ISBN: 978-1-941968-47-5 ©2018 SDIWC 10

Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018
TABLE I: Real UWA trials for LFM and Trigonometric when BT = 4

LFM Trigonometric 1

0.9
T (ms) BER T (ms) BER
0.8

0.7
10 0.0063 10 0
0.6

|h(t)|
20 0.0055 20 0 0.5

0.4

0.3

1 0.2

0.9 0.1

0.8 0
−2 0 2 4 6 8 10
0.7 Time Delay (ms)

0.6
(a) |h(t)|, BT = 4 and T = 20 ms
|h(t)|

0.5
1
0.4
0.9
0.3
0.8
0.2
0.7
0.1
0.6
0

|h(t)|
−2 0 2 4 6 8 10
0.5
Time Delay (ms)
0.4
(a) |h(t)|, BT = 4 and T = 20 ms
0.3

1 0.2

0.9 0.1

0.8 0
−2 0 2 4 6 8 10
0.7 Time Delay (ms)

0.6
(b) |h(t)|, BT = 4 and T = 10 ms
|h(t)|

0.5

0.4
Fig. 9: Channel Impulse Response of Water Tank in presence of Doppler effect

0.3

0.2
V. C ONCLUSION
0.1

0 In this paper, by using simulations, theoretical analysis and

−2 0 2 4 6
Time Delay (ms)
(b) |h(t)|, BT = 4 and T = 10 ms
Fig. 8: Channel Impulse Response of Water Tank
We can observe that even at a low spreading factor,
when SNR is sufficient (for first trial is between 18 -
23 dB and on the average 20.88 dB) trigonometric chirp
signal has approximately ideal performance. Also, when a
relative speed about 2 m/s is conducted to the system, the
Doppler shift is ∆fd = νc fc ∼
= 10.67 Hz and SNR is between
6 - 19 dB and on the average 11.92 dB, the results is as follows
8 10
real experiments, it has been illustrated that excluding the
Doppler effect, the performance of the CSS system based on
trigonometric chirp outperforms that of Linear chirp based
system in AWGN and other fading channels, especially
when the spreading factor is small enough. Generally, BER
of CSS systems using both LFM and trigonometric chirp
increases when Doppler shift increases as well. But, for
the trigonometric chirp it decreases in some practical lower
Doppler frequencies and it is one of important results of this
signaling. So, it is concluded that under limited bandwidth
condition, the full period trigonometric chirps can be replaced
to the linear chirps for better system performance.

TABLE II: Real UWA trials for LFM and Trigonometric when BT = 4, fc = 8kHz R EFERENCES
and ν = 2m/s
LFM Trigonometric [1] M. Stojanovic, “On the relationship between capacity and distance in
an underwater acoustic communication channel,” ACM SIGMOBILE
T (ms) BER T (ms) BER Mobile Computing and Communications Review, vol. 11, no. 4, pp. 34–
43, 2007.
[2] I. F. Akyildiz, D. Pompili, and T. Melodia, “State-of-the-art in protocol
10 0.2772 10 0.147
research for underwater acoustic sensor networks,” in Proceedings of
the 1st ACM international workshop on Underwater networks, pp. 7–
20 0.1941 20 0.0813 16, ACM, 2006.

ISBN: 978-1-941968-47-5 ©2018 SDIWC 11

Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018

[3] A. Jamshidi, “Direct sequence spread spectrum point-to-point

communication scheme in underwater acoustic sparse channels,” IET
communications, vol. 5, no. 4, pp. 456–466, 2011.
[4] M. Chitre, S. Shahabudeen, and M. Stojanovic, “Underwater acoustic
communications and networking: Recent advances and future chal-
lenges,” Marine technology society journal, vol. 42, no. 1, pp. 103–116,
2008.
[5] R. Diamant, A. Feuer, and L. Lampe, “Choosing the right signal:
Doppler shift estimation for underwater acoustic signals,” in Proceedings
of the Seventh ACM International Conference on Underwater Networks
and Systems, p. 27, ACM, 2012.
[6] R. C. Dixon, Spread spectrum systems: with commercial applications,
vol. 994. Wiley New York, 1994.
[7] M. Winkler, “Chirp signals for communication,” 1962., 1962.
[8] A. Jamshidi and S. Moezzi, “Experiential assessment of iteratively
residual interference elimination in the passive phase conjugation for
acoustic underwater communications,” Ocean Engineering, vol. 105,
pp. 287–294, 2015.
[9] A. W. Rihaczek, “Principles of high-resolution radar,” Norwood, MA:
Artech House, 1996., 1996.
[10] “Ieee standard for information technology - telecommunications and
information exchange between systems - local and metropolitan area
networks - specific requirement part 15.4: Wireless medium access con-
trol (mac) and physical layer (phy) specifications for low-rate wireless
personal area networks (wpans),” IEEE Std 802.15.4a-2007 (Amendment
to IEEE Std 802.15.4-2006), pp. 1–203, 2007.
[11] A. C. Pecci, C. Laot, and A. Bourre, “Quadratic chirp modulation for
underwater acoustic digital communications,” in OCEANS 2015-Genova,
pp. 1–7, IEEE, 2015.
[12] Q. Wang and J. Jiang, “Performances of trigonometric chirp spread
spectrum modulation in awgn & rayleigh channels,” in Proceedings of
the 8th ACM workshop on Performance monitoring and measurement of
heterogeneous wireless and wired networks, pp. 203–210, ACM, 2013.

ISBN: 978-1-941968-47-5 ©2018 SDIWC 12

Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018
Multi-keyword Search Employing Identity-Based Encryption Technique (MKS-IDE)
Regina Esi Turkson
School of Computer Science and Engineering, University of Electronic Science and
Technology of China, Chengdu, China.
Email: regina_turkson@yahoo.com or rturkson@ucc.edu.gh
Abstract— Secured storage in cloud computing has been one of
the challenging areas in research. Searching and retrieval of
outsourced encrypted data is very difficult due to the voluminous
data stored in cloud. Various strategies need to be devise to
eradicate unauthorized access, retrieval and use of such data.
Employing a secured scheme for an outsourced data is very
relevant in order to protect the confidentiality and privacy of the
data content.
In this work, we propose a Multi-Keyword Search employing
Identity-based Encryption techniques (MKS-IDE) which enables
a data user to search for and retrieve encrypted files that has
been outsourced into the cloud by a data owner. The retrieved
encrypted files are ranked based on their relevance scores and
the top-k relevant files are returned by the cloud to the data user.
The data user then obtains a decryption key from the
appropriate data owner to decrypt his selected file. The security
requirement of our scheme is provably secure and the
performance of our scheme is also more efficient as compared to
other PEKS schemes.
Keywords— Cloud Computing, Identity-Based Encryption
(IDE), Multi-Keyword Search, Privacy Preserving
1. INTRODUCTION
The rapid growth in cloud computing is due to it wide usage
and popularity. The tremendous increase in the data stored in
cloud create security and privacy issues. These challenges are
not only on how to store and manage these data but also on
how to effectively and efficiently analyze these data to gain
insightful knowledge in making smarter decision [1][2] and
also to ensure the data is secured enough.
Cloud computing is a network-based environment that centers
on sharing computations or resources. The numerous benefits
of cloud has accounted for different data owners outsourcing
their voluminous data. IT organizations have articulated
concern about critical security issues that exist with the
widespread implementation of cloud computing.
Security in cloud has been one of the most argued-about issues
in cloud computing making privacy preserving a very
important mechanism in cloud computing. Outsourcing data
into the cloud is economically attractive for the cost and
complexity of long-term large-scale data storage, however,
data integrity and availability is not guaranteed. Thus, to
efficiently verify the correctness of the outsourced cloud data
without the local copy of the data becomes a big challenge for
data storage security in cloud computing. Therefore, to fully
guarantee the security of the data and save the cloud user’s
computation resources, it is of appropriately importance to
ISBN: 978-1-941968-47-5 ©2018 SDIWC
enable public auditability for cloud data storage to allow users
to resort to a Third Party Auditor (TPA), who has expertise
and capabilities to audit the outsourced data when needed [3].
To avoid the leakage of the outsourced data to external parties,
data encryption can be adopted to alleviate this privacy
concerns. Encryption ensures data confidentiality, however
data search also becomes a challenge. Secure search over
encrypted data was first initiated by Song et al [4] by
proposing a cryptographic primitive concept called searchable
encryption which enables users to perform a keyword-based
search on an encrypted data, just as on plaintext data. Wang et
al [5] were the first to define secure search over encrypted
cloud data, however, further development made by
[6][7][8][9] and more, seeks to reduce computation and
storage cost and also enrich the category of search functions
such as fuzzy keyword search, secure ranked multi-keyword
search and similarity based search but these works are limited
to single-owner model. In 1984, a public key encryption
scheme was introduced by Shamir [10]. The notion of identity
based cryptosystem is that the public key can be an arbitrary
string. Shamir's original motivation for identity-based
encryption was to simplify certificate management in e-mail
systems. Since the problem was posed in 1984 there have been
several proposals for IBE schemes [11][12]. Boneh and
Franklin in 2001, designed the first practical identity-based
cryptosystem [13]. Zhang et al [14] defined a multi-owner
model for privacy preserving keyword search over encrypted
cloud data, however, their scheme did not use identity based
encryption. A similar system was proposed in [15] of which a
cryptographic techniques, query, response randomization and
ranking capability was used. However, IDE was not used in
their system.
In our work, we propose a multi-keyword search in cloud that
employs an identity-based encryption techniques (MKS-IDE).
The framework of our scheme and its security requirements are
defined. We prove that our proposed scheme satisfies the
ciphertext and trapdoor indistinguishability in the random
oracle [16][17]. This work adopt the ranking technique used in
[14]. Finally, we demonstrate the advantage of our MKS-IDE
scheme by comparing with previous PEKS schemes.
This paper is organized as follows: section II outlines the
system model, threat model, and design goals. Section III states
the preliminaries, Section IV outlines the algorithms and the
security requirements of our scheme. We present a concrete
description of our proposed MKS-IDE scheme in section V.
Section VI has the security analysis. Section VII, we briefly
outline the privacy preserving ranking method which we
13
Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018
adopted from [14], section VIII gives the performance analysis
of our scheme and we conclude in section IX.
2. SYSTEM AND THREAT MODEL AND DESIGN GOALS
In this section, we describe the system model, threat model,
and design goals.
2.1 System Model
In our MKS-IDE scheme, we have three entities namely; data
owner, data user and cloud server. The data owner has a
collection of files to be outsourced to the cloud. To ensure
confidentiality, these files ought to be encrypted. For efficient
and adequate search operation, the data owner builds a secure
searchable index on keywords sets extracted from the files.
The data owner then outsource the encrypted files together
with their indexes. To facilitate keyword search over these
encrypted files, a data user submits a hashed-keywords to the
data owner which is then used for trapdoor creation for the
data user. Upon receiving the trapdoor, the data user submits it
to the cloud. The cloud server then performs secure search
over the encrypted indexes from various data owners and
return the corresponding set of files that matches same
patterns as that in the trapdoor. The cloud then rank these files
based on their relevance score and return the top-k important
files to the data user. The data user can then obtain a
decryption key from the data owner to decrypt the file.
2.2 Threat Model
In our threat model, we assume that the cloud server is honest
but curious. The server follows our protocol but it keen to
know the content of the encrypted files, keywords and the
relevance scores. The data owner and data users who are
authenticated by PKI are assumed to be trusted.
2.3 Design Goals
In order to ensure privacy preserving, our MKS-IDE scheme
should satisfy some security and performance goals. The
proposed MKS-IDE should
 Enable multi-keyword search over different
encrypted files from data owners.
 Must provide data user scalability.
 Ensure that authenticated data users perform correct
and appropriate search
3. PRELIMINARIES
This section gives a brief review of the various concepts of
bilinear pairing, identity-based encryption and other related
mathematical problems used in this paper. Due to limited
page, we omit the details of these concepts.
3.1 Bilinear Pairing
3.2 Identity-Based Encryption
3.3 Other mathematically related problems and assumptions
3.3 Other mathematically related problems and assumptions.
Two mathematical hard problems are used.
 Bilinear Diffie-Hellman (BDH) problem
ISBN: 978-1-941968-47-5 ©2018 SDIWC
 Computational Diffie-Hellman (CDH) problem
4. ALGORITHMS AND SECURITY
REQUIREMENTS
We define the algorithms and security requirements for our
MKS-IDE by modifying the ones in [8] [10] [18]
4.1 Algorithms
An MKS-IDE scheme consist of the following polynomial
time randomized seven (7) algorithms:
i. Setup: A probabilistic algorithm which takes a
security parameter l as input and generates a master
secret / public key pair.
ii. Key extract: This is a deterministic algorithm which
takes as input a user’s identity, a master secret key
and system parameters and returns a user secret key
skID .
iii. MKS-IDE: This is a probabilistic algorithm which
takes a public key of the data owner, a set of
keywords in a document and system parameters and
replies with MKS-IDE searchable ciphertext index
I . The data owner encrypts each document with
Symmetric-Key Encryption method using different
keys for each document. The symmetric keys is then
encrypted with a Public Key Encryption which has
blinding capabilities.
iv. Trapdoor generation: Given multiple keywords, data
user’s secret key and system parameters, this
probabilistic algorithm generates the trapdoor Tw .
v. Test: This deterministic algorithm takes the MKS-
IDE searchable ciphertext index I , trapdoor Tw and
system parameters. Upon receiving a query request
Tw from the user, the cloud server match all
keywords stored on it against the query request. It
then extracts and return the candidate files whose
indexes have same pattern as the query request.
vi. Ranking: The cloud ranks the candidate file set
according to their relevance score and forward the
top-k relevant file to the data user. An order and
privacy preserving encoding scheme which encodes
the relevance score to obtain the top-k search result
will be adopted into our MKS-IDE scheme.
vii. Retrieval: This algorithm takes a blinded encrypted
symmetric key and returns a blinded symmetric key
for decryption of a document.
4.2 Security Requirements
Our MKS-IDE scheme must meet the following security
requirements: (1) ciphertext indistinguishability and
(2) trapdoor indistinquishability.
We define a security for Indistinguishability of Ciphertext
from Ciphertext of Chosen Keyword Attack (IND-CC-CKA)
and trapdoor indistinquishability of the MKS-IDE scheme in
Game 1 and 2 involve interactions between an adversary A
and a challenger C. Since there is limited pages, we omit the
proof of both games.
14
Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018
5. A CONCRETE MKS-IDE SCHEME
The proposed scheme is a non-interactive public key
encryption with keyword search which consist of seven
algorithms. Details of algorithms are outlined below:
- Setup: Given l as a security parameter, a trusted
Private Key Generator (PKG) can generates a master
private/public key pair. Let G1 and G2 be an additive
and multiplicative cyclic groups respectively,
generated by P with a prime orders of q . The PKG
generate a bilinear map e : G1  G1  G2 . The PKG
picks a random number x  Zq* as a master secret
key and computes the system’s public key Ppub as
Ppub  x.P . Let H 1 , H 2 and H 3 be three
cryptographic hash function defined as
H 1 :{0,1}*  G1* , H 2 :{0,1}*  G1* and
H 3 : G 2  {0,1}n where n is fixed length depending
on l. The system’s parameters
{G1, G2, e, H 1, H 2, H 3, P, Ppub, n, params} are published
and the master secret keys x is kept secret.
- Key Extract: Given a user’s identity IDi {0,1}* , the
PKG returns the user’s private key skIDi  xQIDi
where QIDi  H 1( IDi ) .
- MKS-IDE: First computes the searchable index
I  (U ,V ) for a random a  Zq* where U  aQID
and V  H 3(k ) and k is computed as
k  e(U , Ppub)e( H 2(wi), P) .
- Trapdoor Generation: Let W be the set of keywords
the data user want to search for. The data user
computes t  H 2(W ) and sends to the data owner.
The data owner then returns the trapdoor
Tw  askID  t to the data user.
- Test: Given I and Tw , the cloud server test for each
file if H 3(e(Tw, P)  V it returns the file and it id.
- Ranking: We adopt a privacy preserving ranked
search scheme implement in [14] into our MKS-IDE
scheme to facilitate the ranking of the candidate file
to determine which file is more relevant to a certain
keyword according to the encoded relevance scores.
- Retrieval: Given the IBE of the symmetric key
C  (C1, C 2) where C1  rP and
C 2  sk.(e( Ppub, QID)r ) and r  Zq* used to encrypt
the document. The data user blind C by computing
C '  (C )  (C1 , C 2 ) and sends it to the data
owner. The data owner computes
sk   C 2 .e(r Ppub, QID)1 . The secret key can be
deduce thereafter and be used to decrypt the
encrypted document [19] .
ISBN: 978-1-941968-47-5 ©2018 SDIWC
6. SECURITY ANALYSIS
We prove our MKS-IDE system is a non-interactive
searchable encryption scheme that is semantically secure in a
random oracle model[16][17][18] . The proof of security relies
on the difficulty of the BDH and CDH problem.
6.1 Ciphertext Indistinguishability
The proposed MKS-IDE scheme satisfies ciphertext
indistinguishability under an adaptive chosen plaintext attack.
For the security of this scheme to be simplified, a proof of a
lemma is demonstrated in which the adversary is assumed to
be an outside attacker.
Lemma 6.1 We assume an adversary A with a non-negligible
probability  1 can break the ciphertext indistinguishability of
the proposed MKS-IDE scheme in the random oracle under an
adaptive chosen plaintext attack, then there exist a challenger
2 1
C with a non-negligible probability   who can
e(qR  1).qH 3
solve the BDH problem where qR and qH 3 represents the
maximum numbers of making key extract and H 3 queries
respectively.
Proof: The Challenger C is given an input of the BDH
parameters as {q, G1, G 2, e} produce by G . The BDH problem
is defined as: Given (P, aP, bP, cP) = (P, P1, P2, P3) with
P being random in G1* and a, b, c  Zq* where q is the order
of G1 and G 2 as input, compute D  e( P, P)abc  G 2 . We say
that BDH is intractable if all polynomial time algorithms have
a negligible advantage in solving BDH. The Challenger C find
D by interacting with the adversary A as follows:
Setup: The challenger C runs the setup algorithm to generate
the public parameters {q, G1, G2, e, P, Ppub, H 1, H 2, H 3} by
setting QID  P2 and Ppub  P3 . H 1 , H 2 and H 3 are random
oracles controlled by C. The challenger C gives the public
parameters to A. The challenger C runs the Key Extract
algorithm to generate A’s private key skIDA with the public
key being IDA as skIDA  aQID  abP . The challenger keeps
the secret key msk and A’s private key skIDA is given to him.
H1 queries: At any time, the adversary A can query the
random oracle H 1 adaptively. To react to these queries, the
Challenger C keeps a list of tuples  IDi, Qi, bi, coini  , called
the H 1LIST . Until the adversary makes queries to the oracle,
H 1LIST is initially empty. The adversary A queries the oracle
with IDi , the Challenger C replies as follows:
a) If query IDi already exist on the H 1LIST in the tuple
 IDi, Qi, bi, coini  then C answers with
H 1( IDi )  Qi  G1*
15
Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018
b) Otherwise C generates a random coini {0,1} so that
Pr[coini  0]= for some  that will be determined
later.
c) The challenger C picks a random value bi  Zq* .
If coini  0 , C computes Qi  biP  G1* .
If coini  1 , C computes Qi  biQID  G1* .
d) The challenger C adds the tuple  IDi, Qi, bi, coini 
to H 1LIST and returns H 1( IDi )  Qi to A. Note that
Qi is uniform in G1* and is independent of A’s
current view as required.
H2 queries: Similarly, at any time the adversary A can query
the random oracle H 2 . The Challenger C reacts to a query by
maintaining a list of tuple  IDi, wi, Qwi, xi  called the
H 2 LIST . The list is initially empty until the adversary makes a
query. When the adversary A queries the oracle H 2 for
 IDi, wi  , The challenger C responds as follows:
a) If  IDi, wi  appears in the list H 2 LIST , C replies
with H 2( IDi, wi)  Qwi
b) Otherwise, the challenger C selects randomly xi  Zq*
and computes Qwi  H 2(IDi, wi)  xi.P . Finally, C
adds the tuple  IDi, wi, Qwi, xi  in the list H 2 LIST
and replies to the adversary A with
H 2( IDi, wi)  Qwi
H3 queries: The adversary A can query the random oracle H 3
at any time. The challenger C responds to the query by
maintaining a list of tuple  mi , ni  called H 3 LIST which is
initially empty. Responds to query is described as follows:
The challenger C replies to the adversary’s queries to the
oracle H 3 as indicated below:
a) If mi appears in the list H 3 LIST , the challenger C
replies with H 3(mi )  ni
b) Otherwise, the challenger C select at random
ni {0,1}n and fix H 3(mi )  ni . The challenger C
then adds the pair  mi, ni  to the H 3 LIST and return
H 3(mi )  ni to the adversary A.
Phase 1: Let IDi be a private key extraction query issued by
A. The Challenger C replies as follows:
a) C runs the algorithm above for responding to H 1-
Queries to obtain a Qi  G1* such that H 1( IDi )  Qi
and let  IDi, Qi, bi, coini  be the corresponding
tuple on the H 1LIST . If coini  1 , then C reports
failure and abort. The attack on MKS-IDE failed.
b) If coini  0 , C gets Qi  biP and defines
skIDi  biPpub  G1* . Observe that skIDi  cQi and
therefore skIDi is the private key associated to public
key IDi . The Challenger then sends skIDi to A.
ISBN: 978-1-941968-47-5 ©2018 SDIWC
Trapdoor queries: Adversary A issues a trapdoor query for
keyword  IDi, wi  , the Challenger C then access the
corresponding tuples  IDi, Qi, bi, coini  and
 IDi, wi, Qwi, xi  in the H 1LIST and H 2 LIST respectively to
computes the trapdoor Twi  xiskIDi  t where xi  Zq* and
t  H 2(wi ) for the adversary A.
a) Challenge: The adversary A sends ( IDC* ,W 0*W 1* ) to
the challenger C where W 0* and W 1* are two
challenged keywords. Upon receiving
( IDC* ,W 0*W 1* ) from A, the challenger C picks a
random value b {0,1} and access the corresponding
tuple  IDc* ,Wb* , Qwb* , x*  in H 2 LIST to generate an
MKS-IDE ciphertext I *  (U * ,V * )  (U * , R) , where
R {0,1}n is a random value. The restrictions
remains that the adversary A did not queried a private
key extraction for IDc* and also for trapdoor query
for W 0* and W 1* . The challenger C eventually sends
I * to the adversary A.
b) Phase 2: A continues with the request for key extract
queries adaptively for any identity IDC and the
trapdoor query for any keyword W from the
challenger C subject to the restriction that
IDc  IDc* and w  {W 0* or W 1*} .
c) Guess: Finally, the adversary A outputs a guess of a
value b'  {0,1} . A wins the game if b'  b .
By the assumption, the adversary A with a non-negligible
probability  1 can distinguish the MKS-IDE ciphertext I *
under an adaptive chosen plaintext attack. Now, the challenger
C picks a tuple  m* , n*  in the H 3 LIST and outputs
v*  m* / e( H 2(wb), P) as the solution for the BDH instance
(P, aP, bP, cP) for a, b, c  Zq* . It can be deduced that the
output m* / e( H 2(wb), P) is equal to e( P, P)abc . Observe that
U  aQID and m*  (e(U , Ppub)e( H 2(wb), P)) . The challenger
C accesses the corresponding tuple  IDc* ,Wb* , Qwb* , x*  in
the list H 3 LIST and compute
(e(aQID, cP)  e(abP, cP)  e( P, P) abc
By adopting the similar technique used in [18][20], we
compute the probability of our MKS-IDE scheme by
discussing the probability that the challenger C does not abort
during the challenge phase. Suppose the adversary A makes
qR queries to the key extract query. In such an instance, the
probability that the challenger C does not abort in phase 1 or 2
qR
is ( ) and the probability that the challenger C does not
terminate during the challenged step is (1   ) . Therefore, the
probability that the challenger C does not terminate during the
qR qR
simulation is ( )  (1   ) . This value can be maximize at
16
Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018

1 obtain:
 abt  1  . By using  abt , the probability that the
qR  1
1 1
challenger C does not abort is at least . Observe that
e(qR  1)
the probability analysis uses the same techniques as Coron’s
analysis of the Full Domain Hash in [21]. The Challenger C
2 1
outputs the correct D with the probability at least [21]
qH 3
where qH 3 represents the total number of requesting H 3
queries. Therefore, the Challenger C with a probability
2 1
 can solve the BDH problem. This contradict
e(qR  1).qH 3
to the BDH assumption.
Theorem 6.1: The proposed non-interactive searchable
encryption scheme (MKS-IDE) above is semantically secure
against an adaptive chosen plaintext attack in the random
oracle by satisfying the ciphertext indistinquishability under
the BDH assumption.
6.1 Trapdoor Indistinguishability
Lemma 6.2 In the random oracle model, we assume that if
there is an adversary A with a non-negligible advantage that
can break the trapdoor indistinguishability of the proposed
MKS-IDE scheme under the adaptive chosen keyword attack,
then there exist a challenger C with a non-negligible
advantage who can solve the computational CBH problem.
Proof: The challenger C receives a CDH instance of
(P, aP, bP) for a, b  Zq* where q is the order of G1 and G 2 .
By interacting with the adversary A, the challenger C returns
the CDH solution abP in Game 2. Since there is limited
pages, we omit the proof for trapdoor indistinguishability
since it’s similar as defined in lemma 6.1.
It is assumed that the adversary A with a non-negligible
advantage can differentiate the trapdoor Twb* under the
*
adaptive chosen keyword attack. Meaning the trapdoor Twb
satisfy the equation e(Twb* , P)  V ' where
V '  e(U , Ppub).e( H 2(wb), P) and Let U  xQID . We can then
1
e(Twb*  H 2(wb), P) x  e(abP, P)
Which implies that e(abP, P)  e(Twb*  H 2(wb), P) x . Hence
the Challenger can obtain abP  x1 (Twb*  H 2(wb)) , which
contradicts the CDH assumption.
Theorem 6.2 The proposed MKS-IDE scheme is semantically
secure and satisfies the trapdoor indistinquishability against
an adaptive chosen plaintext attack under the CDH
assumption in the random oracle.
7. PRIVACY PRESERVING RANKED SEARCH
Our scheme adopts the privacy preserving ranked search in
[14] . It is important to note that, after the retrieval of all the
candidate files, the cloud server cannot return all the
undifferential files to the data user due to (1) acquisition of
excessive communication cost and overhead for the system if
the cloud server decides to return all the candidate files. (2)
the data users may only be concerned with the top-k relevant
files that correspond to their queries. The scheme in [14]
illustration an additive order preserving and privacy
preserving encoding scheme. It then uses the encoded
relevance score to obtain the top-k search result.
8. PERFORMANCE ANALYSIS AND COMPARISON
Adopting similar strategy used in [18] we analyze the
performance of our MKS-IDE scheme. We compare our
MKS-IDE scheme with some previously proposed PEKS
schemes by concentrating more on some time consuming
operations defined as follows:
 TGe : The execution time of a bilinear map operation
e : G1  G1  G2
 TGmul : The execution time for scalar multiplication
operation in G1
 TGH : The execution time for map-to-point hash
function, thus H 1, H 2, H 3 :{0,1}*  G1
 Tinv : The execution time of a modular inverse
operation in Zq
The most time consuming operation is the execution time for

TABLE 1: Comparison between our IDRMKS and previously proposed dPEKS schemes

Scheme of Scheme of Scheme of Scheme of Wu Our

Hwang and Rhee et al. Hu and Liu et al. MKS-IDE
Lee [24] [25] [23] [18]
Public Key Setting Pairing-based Pairing-based Pairing-based ID-based ID-based
Certificate Management Required Required Required Not Required Not Required
Computational cost for ciphertext (2n  2)TGmul (2n  2)TGmul (2n  2)TGmul TGe  (n  2)TGmul 2TGe  TGmul
generation (conjunctive n keywords) 2nTGH 2nTGH 2nTGH (n  2)TGH (n  1)TGH

Computational cost for ciphertext 3TGmul TGe  2TGmul TGe  2TGmul TGe  3TGmul 2TGe  TGmul
generation (1 keywords) 2TGH TGH TGH 3TGH 2TGH
Computational cost for trapdoor 3TGmul  2TGH 3TGmul  2TGH 3TGmul  2TGH 2TGmul  TGH TGmul  TGH
generation Tinv Tinv Tinv

Computational cost for test 3TGe TGe  2TGmul TGe  2TGmul 2TGe  TGmul TGe  2TGmul
ISBN: 978-1-941968-47-5 ©2018 SDIWC TGH TGH Tinv Tinv17
Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018
bilinear map operation TGe as compared to the other
operations stated above. In [22], the performance simulation
results show that TGe  2.5TGmul . We therefore analyze the
performance of our MKS-IDE scheme for each phase. In the
MKS-IDE ciphertext generation phase, it required
2TGe  TGmul  (n  1)TGH to generate an MKS-IDE
ciphertext, note that n represents the total number of
keywords. TGmul  TGH is essential to generate the trapdoor
Tw in the trapdoor generation phase,
Table 1 above outlines the comparison between our MKS-IDE
scheme to other dPEKS schemes [23][24][25] ] in terms of
public key setting and performance. From the table in [18], it
can be observed that conjunctive keywords is not supported
by the schemes in [23][25] and they require
nTGe  (n  1)TGmul  nTGH for their ciphertext generation. We
can therefore conclude from table 1 above that our MKS-IDE
scheme is more efficient than the schemes in [23][24][25] with
respect to ciphertext generation especially when n is
sufficiently large and also efficient in the trapdoor generation.
Since our MKS-IDE scheme is based on ID system, the load
of certificate management associated with the other schemes
which are based on pairing-based public key system are
eradicated. Although the scheme in [25] is also based on ID-
system, however in the ciphertext and trapdoor generation, our
MKS-IDE is more efficient than [25].
9. CONCLUSION
In this paper we proposed an MKS-IDE scheme which
supports conjunctive keywords. We defined the framework
and the security requirements for our MKS-IDE scheme and
when we compared our scheme to previous dPEKS scheme,
the performance of our scheme is more efficient in both the
ciphertext and trapdoor generation phase. Our ID-based
system also has an advantage of eliminating certificate
management associated with PKI. We also demonstrated that
our MKS-IDE scheme possesses the ciphertext
indistinquishability and trapdoor indistinguishability under
BDH and CDH assumptions respectively.
Our future work will seek to implement our MKS-IDE scheme
in a hybrid cloud and also to investigate our MKS-IDE
scheme without random oracle with a pairing free algorithm to
determine how feasible it will be in cloud.
REFERENCES
[1] I. Abaker, T. Hashem, I. Yaqoob, N. Badrul, S. Mokhtar, A. Gani,
and S. Ullah, “The rise of ‘ big data ’ on cloud computing : Review
and open research issues,” Inf. Syst., vol. 47, pp. 98–115, 2015.
[2] L. Ballard, S. Kamara, and F. Monrose, “Achieving efficient
conjunctive keyword searches over encrypted data,” in Lecture
Notes in Computer Science (including subseries Lecture Notes in
Artificial Intelligence and Lecture Notes in Bioinformatics), 2005,
vol. 3783 LNCS, pp. 414–426.
ISBN: 978-1-941968-47-5 ©2018 SDIWC
[3] C. Wang, Q. Wang, K. Ren, and W. Lou, “Privacy-preserving
public auditing for data storage security in cloud computing,” Proc.
- IEEE INFOCOM, 2010.
[4] D. Wagner, A. Perrig, D. X. Song, D. Wagner, and A. Perrig,
“Practical techniques for searches on encrypted data,” in Proceeding
2000 IEEE Symposium on Security and Privacy, 2000, pp. 44–55.
[5] P. Golle, J. Staddon, and B. Waters, “Secure conjunctive keyword
search over encrypted data,” Appl. Cryptogr. Netw. Secur., vol.
3089, pp. 31–45, 2004.
[6] R. Li, Z. Xu, W. Kang, K. C. Yow, and C. Z. Xu, “Efficient multi-
keyword ranked query over encrypted data in cloud computing,”
Futur. Gener. Comput. Syst., vol. 30, no. 1, pp. 179–190, 2014.
[7] J. Li, Q. Wang, C. Wang, N. Cao, K. Ren, and W. Lou, “Fuzzy
Keyword Search over Encrypted Data in Cloud Computing,” 2010
Proc. IEEE INFOCOM, pp. 1–5, 2010.
[8] B. Wang, S. Yu, W. Lou, and Y. T. Hou, “Privacy-preserving multi-
keyword fuzzy search over encrypted data in the cloud,” in
Proceedings - IEEE INFOCOM, 2014, pp. 2112–2120.
[9] M. Chuah and W. Hu, “Privacy-aware bedtree based solution for
fuzzy multi-keyword search over encrypted data,” in Proceedings -
International Conference on Distributed Computing Systems, 2011,
pp. 273–281.
[10] A. Shamir, “Identity-Based Cryptosystems and Signature Schemes,”
Adv. Cryptol., vol. 196, pp. 47–53, 1985.
[11] U. Maurer and Y. Yacobi, “Non-interactive public-key
cryptography,” Adv. Cryptology—EUROCRYPT’91, pp. 498–507,
1991.
[12] L. Problem, “An ID-Based Cryptosystem Based on the Discrete,”
vol. 7, no. 4, 1989.
[13] D. Boneh and M. Franklin, “Identity-Based Encryption from the
Weil Pairing,” SIAM J. Comput., vol. 32, no. 3, pp. 586–615, 2003.
[14] W. Zhang, Y. Lin, S. Xiao, J. Wu, and S. Zhou, “Privacy Preserving
Ranked Multi-Keyword Search for Multiple Data Owners in Cloud
Computing,” IEEE Trans. Comput., vol. 65, no. 5, pp. 1566–1577,
2016.
[15] P. D. Mgmt, C. A. Secondary, and C. Author, “Distributed and
Parallel Databases An Efficient Privacy-Preserving Multi-Keyword
Search over Encrypted Cloud Data with.”
[16] R. Canetti, O. Goldreich, and S. Halevi, “The random oracle
methodology, revisited,” J. ACM, vol. 51, no. 4, pp. 557–594, 2004.
[17] M. Bellare, “R a n d o m Oracles are Practical : A Paradigm for
Designing Efficient P r o t o c o l s,” pp. 62–73, 1993.
[18] T.-Y. Wu, T.-T. Tsai, and Y.-M. Tseng, “Efficient searchable ID-
based encryption with a designated server,” Ann. Telecommun. -
Ann. Des Télécommunications, vol. 69, no. 7–8, pp. 391–402, 2014.
[19] C. Orencik and E. Savas, “Efficient and secure ranked multi-
keyword search on encrypted cloud data,” Proc. 2012 Jt.
EDBT/ICDT Work., pp. 186–195, 2012.
[20] Y. M. Tseng and T. T. Tsai, “Efficient revocable ID-based
encryption with a public channel,” Comput. J., vol. 55, no. 4, pp.
475–486, 2012.
[21] J.-S. Coron, “On the Exact Security of Full Domain Hash,”
CRYPTO 2000 Adv. Cryptol., pp. 229–235, 2000.
[22] T.-Y. Wu and Y.-M. Tseng, “An ID-Based Mutual Authentication
and Key Exchange Protocol for Low-Power Mobile Devices,”
Comput. J., vol. 53, no. 7, pp. 1062–1070, 2009.
[23] C. Hu and P. Liu, “An enhanced searchable public key encryption
scheme with a designated tester and its extensions,” J. Comput., vol.
7, no. 3, pp. 716–723, 2012.
[24] Y. H. Hwang and P. J. Lee, “Public key encryption with conjunctive
keyword search and its extension to a multi-user system,” Lect.
Notes Comput. Sci. (including Subser. Lect. Notes Artif. Intell. Lect.
Notes Bioinformatics), vol. 4575 LNCS, pp. 2–22, 2007.
[25] H. S. Rhee, J. H. Park, W. Susilo, and D. H. Lee, “Trapdoor security
in a searchable public-key encryption scheme with a designated
tester,” J. Syst. Softw., vol. 83, no. 5, pp. 763–771, 2010.
18
Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018
Image Processing Based Anti-Sleep Alarm System for Drowsy Drivers
Barış Ozan DİNÇER, Ahmet SEVİM, Alaa ELEYAN
Electrical & Electronics Engineering, Avrasya University, Yomra, Trabzon
aeleyan@avrasya.edu.tr
Abstract
With the predictions of World Health Organization
(WHO) that number of deaths due to traffic accidents will be
around 2 million with less than 15 years, researchers
nowadays are paying more attention in how to help in
preventing traffic accidents and lower the number of occurred
fatalities. The purpose of this study is an attempt to prevent
traffic accidents due to fatigue or sleepiness. Developed
system uses a camera and image processing techniques
embedded in a raspberry pi 3 module to detect driver’s eyes
and decide weather the driver is sleepy or not. Based on this
decision an alarm system will be activated. Alarm can be
visual, audio and even a simple vibration in the steering
wheel.
Keywords: python, opencv, image processing, camera,
drowsy driver, traffic accident
I. Introduction
Road traffic injuries and deaths have a terrible
impact on individuals, communities and countries. They
involve massive costs to often overburdened health care
systems occupy scarce hospital beds consume resources and
result in
significant losses of productivity and prosperity, with deep
social and economic repercussions [1]. According to the 2016
report of WHO, 1.24 million road traffic deaths occur every
year. This makes it the number one cause of death among
those aged 15-29 years. This number is predicted to increase
to around 1.9 million by 2030 and to become the seventh
leading cause of death if no action is to be taken [1].
According to the monthly report of traffic accidents in
Turkey, 16.785 out of the 18.965 accidents occurred in April
2018 because of the diver faults while the rest of these
accidents reasons were divided between passengers,
pedestrian, car itself and road [2].
Image processing has been used for many decades
for processing videos and images for different real-time
applications. with the invention of new processors of high
processing capabilities and high definitions cameras, it
became much easier to develop real-time application that can
perform similar than humans with much better accuracy and
less expenses. To do so, many software and libraries has been
introduced to help researches and developer in building their
systems in a faster and easier way. OpenCV (Open Source
Computer Vision) is a library of programming functions
ISBN: 978-1-941968-47-5 ©2018 SDIWC
mainly aimed at real-time computer vision, developed by Intel
Russia research center in Nizhny Novgorod. It is a free cross-
platform library. It focuses mainly on real-time image
processing. Dlib library contains a wide range of machine
learning algorithms. All designed to be highly modular, quick
to execute, and simple to use via a clean and modern C++ API.
It is used in a wide range of applications including robotics,
embedded devices, mobile phones, and large high-
performance computing environments [3].
The Haar Cascade classifier was proposed in 2001 by
Paul Viola and Michael Jones to find objects in videos or
pictures. They used a boosted cascade of simple features to
rapidly detected objects. Haar Cascade is also known as
Viola-Jones detector [4].
To prevent traffic accidents caused by sleepy drivers,
many systems have been proposed [5]-[8]. These systems use
different type of sensors to asset drivers during driving. Some
of these systems are listed below:
• A project called drowsy driver alarm system has installed an
alarm system with sensors and special equipment attached to
the driver [5].
• Detecting Driver Drowsiness Based on Sensors, the sleep
state is processed in 3 different levels and the alarm system is
activated by using the sensor readings the driver's steering and
the driver himself [6].
• Researchers in [7] developed a system called Drowsy Driver
Monitor and Warning System. A driver monitor was made
available for the driver and the images were processed by
sending light to the driver’s eyes according to the day time.
In this study, an image processing approach was
followed. Raspberry pi 3 was utilized with codes written in
python language with OpenCV and Dlib libraries. A Haar
Cascade classifier was used to detect faces in the video
sequences acquired from the camera in a real-time mode. The
system is very useful, fast and advantageous, and it worked
smoothly in during our field tests.
II. Methodology
Firstly, when scanning for the desired object, scanning with
set of frames of specific size and shape is performed. Certain
target values are generated by controlling the dark and light
reas in within the frame. This is done by comparing values the
sum of the pixel values in the black area of the frames and the
sum of the pixel values in the white zone. Examples of such
frames are shown in Figure 1.
19
Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018
Figure 1. Examples of frames used for detection
These frames are heavily applied on sample positive images
as shown in Figure 2. Because of this screening, correct and
incorrect values are compared within the classifier and object
identification is made.
Figure 2. Use of frames to detect the nose
The written code and the used libraries are expressed with 68
dots in the face area using the frames to locate the eye,
eyebrow, nose, mouth and chin lines of the driver (Figure 3).
Figure 3. The Dlib library is utilized to identify facial
Landmarks
Figure 4. Numbering facial landmarks within the code
ISBN: 978-1-941968-47-5 ©2018 SDIWC
Figure 5 68 Point representation of the facial landmarks
After detection of eye landmarks, The distance/gap between
the upper and lower points are calculated. Based on these
measures, system will be able to identify if the eye is open or
close.
Figure 6 Example of opening and closure of an eye.
Figure 7 consecutive frame and threshold values
During our experiments, this distance is set to 0.230. to
consider the eye closed and not just a blinking, the frame
value is set to 4. This means if we get a value less than the
threshold for more than 4 consecutive frames then eye will be
declared close. Be make heathier decision the values obtained
from both eyes are averaged and if still less than threshold
then eye will confidently be declared close and alarm will be
activated Otherwise, the alarm will stay de-active.
III. The Developed System
The developed system was equipped in a car and tested in
real-time. Equipment and software used in system
development are as follows:
• Raspberry pi 3 Model B: a mini computer used for
installation of the whole system. It has a 4-core 1.2 Ghz ARM
Cortex-A53 processor, a Broadcom VideoCore IV video card,
and a 1GB LPDDR2 RAM. It operates with 5V, 2.5A.
• Logitech C920 Webcam: High Definition HD Camera used
for acquiring video sequence for face detection of driver. It
works with 1.5 V.
20
Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018

• Logitech Speakers: Speakers are used to provide audio Example of the tests done in the lab are shown in Figure 9 and
alarm. It works with 5V. Figure 10. Figure 9 shows eye detection of closed eyes where
• Raspberry pi chamber/case and fan: The case is used to the opening ratio between upper and lower parts is smaller
protect the system from external environment, and a fan is than threshold. Alarm will be activated.
used to prevent the system from overheating due to its long
usage.
Software used in the study:
• Python Language: The programming language used in the
project.
• Opencv and Dlib Libraries: Independent libraries used in
this project to develop real-time computer vision applications.
• Haar Cascade Classifier: The classifier used in the project
identifies and implements the screening method of the human
face detection.
The above mentioned are the main software. Additionally;
time, imutils, video, numpy, argparse libraries are also used.

The step by step working mechanism of the developed system

Figure 9 Eye detection of closed eyes (inside the lab)
is shown in Figure 8. It can be summarized as follows:
Figure 10 shows eye detection of open eyes where the opening
• System will be activated as soon as the driver starts the
ratio between upper and lower parts is larger than threshold.
car’s engine.
Alarm will stay de-active.
• The camera continuously and simultaneously detects the
face of the driver.
• after detecting the face, facial landmarks are obtained,
and eye landmarks are acquired
• distance between the upper and lower eye landmarks is
calculated for both eye and then average distance is
generated
• If the obtained distance is less than the assigned threshold
the eye will be declared close otherwise it will be open.
• This distance will be calculated for each frame of the
video.
• If the distance falls below the threshold for 4 consecutive
frames or more the alarm will be activated.
• Alarm will be deactivated when the driver opens his eyes
• System will run and keep monitoring the driver as long Figure 10 Eye detection of open eyes (inside the lab)
as the car’s engine is running
Figure 11 shows our developed system integration inside the
vehicle for the testing stage.

Figure 11 System integration inside the car

After integration of the system inside the car, experimental

tests were conducted in different conditions and with different
drivers. Figure 12 and Figure 13 show example of these tests.

Figure 8 Block Diagram of the developed system

ISBN: 978-1-941968-47-5 ©2018 SDIWC 21

Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018
Figure 12 Eye detection of open eyes (inside the car)
Figure 13 Eye detection of close eyes (inside the car)
IV. Conclusions
Our developed system was an attempt to help in decreasing
and/or prevent traffic accidents that happen due to drivers’
drowsiness and fatigues. Based on our conducted
experiments, simply using a camera and image processing
software to build the system is an advantage as the drivers
using this system will not have to wear any sensor or put on
any gadgets. It is designed so that it can be used in all the
vehicles desired regardless of the price or model of the
vehicle. Our system still has problems in handling extreme
illumination conditions. Also, night vision camera can be used
so the system will be able to work both day and night.
ISBN: 978-1-941968-47-5 ©2018 SDIWC
References
[1] WHO. Road Safety: Basic Facts; 2016
(http://www.who.int/violence_injury_prevention/publications/
road_traffic/Road_safety_media_brief_full_document.pdf).
[2] Monthly Traffic Accidents report in Turkey. April 2018.
http://www.trafik.gov.tr/SiteAssets/istatistik/nisan18.pdf
[3] D E King. “Dlib-ml: A Machine Learning Toolkit”. Journal of
Machine Learning Research, vol 10, 2009.
[4] P Viola, M Jones, “Rapid Object Detection Using A Boosted
Cascade of Simple Features”, In proceedings of Computer
Vision and Pattern Recognition (CVPR01), pp. 511-518, 2001.
[5] V Tengshe, V Tengshe, “Drowsy Driving Alarm System”,
2005, US Patent US20060214807A1, 2005.
[6] A Sahayadhas, K Sundaraj, M Murugappan. “Detecting Driver
Drowsiness Based on Sensors: A Review”. Sensors, vol.12,
no.12, pp. 16937-16953. 2012.
[7] R Grace, S Steward, “Drowsy Driver Monitor and Warning
System”, Proceedings of the First International Driving
Symposium on Human Factors in Driver Assessment, Training
and Vehicle Design. USA, 2001, pp. 64-69.
[8] E E Galarza, FD Egas, F M Silva, P M Velasco, E D Galarza,
“Real Time Driver Drowsiness Detection Based on Driver’s
Face Image Behavior Using a System of Human Computer
Interaction Implemented in a Smartphone”, Proceedings of the
International Conference on Information Technology &
Systems (ICITS 2018), pp. 563-572, 2018.
22
Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018

Design and Implementation of Energy Management System for Buildings

SUDAD J ASHAJ
Gaziantep University
sudad_80@yahoo.com
Prof. Dr. Ergun ERÇELEBİ
Gaziantep university
ergun.ercelebi@gmail.com

ABSTRACT automation and buildings, there is an increase

Collecting data and monitoring from the related
remote sensing activates perform a significant role
for records the variants of environmental as a
function of time by using Raspberry PI in green
buildings. The main goal of this project is to
develop a data acquisition system (DAQ) for
monitoring the measurements of temperature and
the humidity in order to extend this functionality
for developing multipurpose DAQ and controlling
system. In this work we have utilizes the digital
sensors which have been connecting to I2C (Inter-
Integrated Circuit) bus circuits with programs.
The controlling units or analog to digital circuits
translate the relative the analog data of
temperature and humidity that sensed by the
sensors to be a digital signal, after that, the client
transmits these data to the server and then the data
are going to be processing by the computers. The
programmed interface system has been
implemented by used Raspberry PI rather than PC
that make it possible for users to manage all
parameters of operation at high speed and low
cost. The DAQ system can be operates either as
standalone system or to work with PC. The
monitoring and the recording system should be
pre-loaded with online software such as Linux to
be able to get the appropriate measurements.
KEYWORDS
Smart Buildings, Environment sensor,
temperature monitoring, humidity monitoring,
Raspberry PI.
1 INTRODUCTION
Due to an increase in the affordability and
availability of wireless networks for home
of interest for the environmental impact of the
of buildings in both commercial and
residential requirement. Thus, Building the
energy management system can provide a
central platform for controlling the energy
usage of the buildings, which can be detecting
and eliminating the waste and permit the
efficient use of electricity resources. These
systems should have a control unit along with
sensors that can be used to enable the
monitoring of room occupancy, lighting,
ambient temperature, and other inputs in order
to achieve useful management of lighting
systems, climate control (ventilation, heating
and air conditioning), and security [1, 2].
However, according to the drawbacks of the
typical greenhouse control system, for
example, higher cost and traditional bus mode
network problem, this study focused on
employed the low-cost component that can be
achieved the task. Last years, the
microcontroller units (MCU) and tiny
computer has been improved and large
number of MCU has been available for
everyone because of its low cost and simple
programming such as Arduino families,
Raspberry PI, etc. In this work we have used
Raspberry PI which is a low-cost, single-
board, high-performance computer [3].
In this work we have designed a Raspberry PI
communication module in order to offer a low
power consumption and high radio frequency
and we have built the user interface based on
Client/Server webserver. The web server has
been designed based on utilizing the S3C2440

ISBN: 978-1-941968-47-5 ©2018 SDIWC 23

Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018

microprocessor and Linux operation system.
The proposed system has the voice prompt,
remote control greenhouse device and remote
monitor. The test of system shows that it is
stable, and it can be developing to use as an
authorized devise to be maintains that is
appropriate for DAQ and as device control for
several kinds of agricultural sites.
2 LITERATURE SURVEY
This section discussed the in short the several
literatures presented related to Smart Home
and Energy Management System
aggregated by the PC server and appropriately
compare them for making statistical analysis
info.
Suresh et al. (2016) [7], developed an
automatic control system for lighting in order
to got an effective use of energy. They also
have presented mobility and remote command
performance to system by using Android
mobile App through Bluetooth to control the
lighting based upon the voice command
3 PROPOSED SYSTEM STRUCTURES

Sankaranarayanana et al .(2014) [4], The proposed system is aimed to repeatedly

developed a smart home monitoring prototype monitoring the realtime humidity and
by utilizing the wireless sensor systems and temperature in a cost effective way via polling
android mobile handset. This prposed system sensor at fixed time interval. The system
has the ability to monitors the usage features requires designing of hardware, embedded
of electrical power for the socket outlet in software and application programs.
realtime. It can measures the current, voltage
and temperature of the socket outlet A. Hardware design:
frequently from every room and the
The hardware component of proposed system
monitoring data has been sent from the
is including: sensors, controls unit, actuators,
system to computing.
analog to digital converts, busses cards and
other components.
Lavanya et al. (2016) [5], developed an
Automated Humidity and Temperature
1. Controls Unit: The Minicomputer type
Control and Monitoring system by using
Raspberry pi (Rev 2 model B) has been
raspberry pi. The raspberry pi has received
used as control card and gateway for
the humidity and temperature values from
remotely access. Raspberry pi has a 1.2
sensors and then sent to the internet.
GHz, 64bit quad-core ARM V8 CPU, a
However, this project has leaded to
802.11n Wireless LAN, Bluetooth Low
development of prototype of automated
Energy version 4.1, the ports includes:
humidity and temperature control with good
Display port (HDMI), Display interface
feasibility.
(DSI), composite video Camera interface
(CSI) four USB ports, Full Ethernet port
Kshirsagar and Upasani (2016) [6], developed
(LAN), 3.5mm audio jack, 40 GPIO pins,
the energy control system for smart home in
and Micro SD card slot. The video Core is
order to manage energy at the ranges of
IV 3D graphics core. Figure 1 shows the
appliances. Thus, towards this they developed
Raspberry pi board and its component
the architecture of smart home energy
figure 2 illustrated the PIN out from it [4].
management system . In this system, the
sensors have controled the energy
consumption of home devices. Furthermore,
the solar energy has been used as an
alternative source where it can be dependent
on switched based o the weather conditions.
The energy data via various home servers are

ISBN: 978-1-941968-47-5 ©2018 SDIWC 24

Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018

Fig.1. Raspberry pi board [5]

Fig.2. PIN out of Raspberry

2. Temperature/Humidity Sensors: The

KEYES SHT10 Digital Temperature
Humidity Sensor has been used to
gathering the environment temperature
degree and humidity percent.

3. Other component: Breadboard, resistors,

jumper wires, green and red LED.

A. Software design:

The software has been designed to manage

the overall operation, which includes: sending
measured temperature and humidity data from
sensors to user, getting user requests for turn
off/on lamps or other electrical appliances.

1- Raspberry Pi programing: The main

system software is the Raspbian [6], which
is a free-license Linux operating system
“Debian” which optimized for its usage
with Raspberry Pi hardware. It is
obtainable directly from the Raspberry Pi
site as an SD Card image which requires to
be copy to an SD card. Rasbian has been
loaded into the SD card to be used in
Raspberry Pi. RPi executes the Python
script to communicate with the node and
the web interface.

2- Server programing: Linux has a web

server available named Apache, it is the
most traditional web server software in the
world. It also requires PHP complements
for web-page creation and PostgreSQL for
database management.

ISBN: 978-1-941968-47-5 ©2018 SDIWC 25

Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018

In this work we have used a server has Energy Management use

windows operating system then we have Raspberry PI
converted the windows platform to linux
distribution Ubuntu14.04 based on the
Select a sensor
instruction ofered by ubuntu.com site [7].
Figure 3 shows the prposed system
expremental setup.
Make the modification for all the
code

NO YES
Accuracy >90 OR
time < 20 second

Note the result YES

Any more
codes
Fig.3. Raspberry pi connection with
sensors

As shown from figure 3, the sensor has YES Any more NO

been connected to the raspberry pi by used selected
socket
jumper wire and the computer (server) has
been connected to Raspberry pi via
ethernet. Compare the results

The raspberry pi is gathering information NO

from sensors then store and display it the Group the most noticeable features from different
real time temperature and relative groups and test them algorithms
humidity.The raspberry pi has been
programmed by used python language.
Specify the best code
Figure 4 shown the block diagram of the
proposed method.The temperature is
displayed in degree celsius or Fahrenheit Stop
as required. The user is able to know the
current temperature and relative humidity Fig.5. raspberry pi program Flow Chart
at remote location by sign in the web
browser by putting login name and
password by the client.Web application 4 RESULTS
starts after sucsess authintification and
then the output graphical representation This program opens in UBUNTU terminal to
can be obtained. The processed data will configure the code of the parameters with
be updated repeatedly on cloud server and shown GUNPLOT interface results. The steps
the user can get stored data on on an of connecting with Raspberry pi is as
hourly basis and daily basis. following steps:

1. Open the Ubuntu terminal

2. Enter the code “cd
/sys/bus/w1/devices” then enter “ls”

ISBN: 978-1-941968-47-5 ©2018 SDIWC 26

Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018
and the terminal will showing the w1
bus master.
3. Enter the w1 bus master “28-xxxx”,
4. Enter cat w1_slave
After doing this steps the server and
Raspberry pi, figure 5 shows the opration
steps in terminal.
Fig 4. Ubuntu terminal
In this paper the time taken to get the results
is not long because it tests some sensors at a
time but if there are many ligands it may take
more than one hour depending on the number
of the processor, to monitor the processors
usage install one of Linux system monitoring
tools, this thesis uses htop which is a much
advanced interactive and real time Linux
process monitoring tool to install it enter on
the terminal.
Fig.5. diagram for the temperature sensing
records
ISBN: 978-1-941968-47-5 ©2018 SDIWC
Fig.6. Diagram for the humidity sensing
records.
5 CONCLUSION
Recent advances in the LAN technology
enabled developing sensor networks, which
enable implementing robust, flexible and
reliable infrastructure by deploying high
number of sensors within the environment.
According to sensor networks can have very
high number of sensors, densely deployed,
regular change (addition, removing, or
substituting), with limited requirements
(power consumption and computational
capabilities. In this work we have designed a
monitoring management control system that
is based on employed network and Raspberry
transmission technology, and quickly presents
the fundamental, system design and
experimental results. The proposed system is
used TCP/IP mode in network, the embedded
development platform carrying out the remote
monitoring, and have the ability to analyse
and dispose the factors of the environmental
such as: CO2 concentration, temperature,
light, humidity, etc) automatically via the
display of the field environmental data. The
test provide that the proposed system has high
reliability and applicability, the remote
monitoring operation is simple, and in
agriculture area have got a good prospect. In
summary, we can conclude a main point in
this project:
27
Proceedings of the Fifth International Conference on Technological Advances in Electrical, Electronics and Computer Engineering (TAEECE2018), Turkey, 2018

1- The proposed system can process, display,
and store the data in both way (as DAQ
system or dedicate system).
2- The benefit of utilizing a Raspberry PI for
DAQ is that a Raspberry PI has the
flexibility to adjust the changing
requirements and its low cost.
3- The design and constricted of this model
can be done in highest accuracy with time
that consumed.
Instrumentation Engineering. vol.6, Issue 4, pp.
3162- 3169. April 2017.
[9] Raspberry official Site, Link:
https://www.raspberrypi.org/.
[10] J. R. C. León, R. F. Martínez-González, A. M.
Medinay and L. A. Peralta-Pelaez. “Raspberry Pi
and Arduino Uno Working Together as A Basic
Meteorological Station,” International Journal of
Computer Science & Information Technology
(IJCSIT) vol.9, no.5, pp.97-104, October 2017.

[11] ubuntu, Link:https://help.ubuntu.com/lts/

serverguide/dpkg.html.
REFERENCES

[1] S. K. Das, D. Cook, A. Bhattacharya, E. O.

Heierman and T.-Y. Lin, “The Role of Prediction
Algorithms in the MavHome Smart Home
Architecture,” IEEE Wireless Communications
Magazine, vol.9, no.6, pp.77-84, Dec. 2002.

[2] V. Mani, Abhilasha Gunasekhar, Lavanya and S.

Sankaranarayanan. “IoT Based Smart Energy
Management System,” International Journal of
Applied Engineering Research vol.12, no.16, pp.
5455-5462, 2017.

[3] I.F. Akyildiz, W. Su, and Y. Sankarasubramaniam,

“A survey on sensor networks. Communications
Magazine,” IEEE Communications Magazine,
vol.40, Issue: 8, pp. 102-114, Aug 2002.

[4] S. Sankaranarayanana , A. T. Wanb, A. H. Pusac.

“Smart Home Monitoring usingAndroid and
Wireless Sensors,” I.J. Engineering and
Manufacturing, vol.2, pp.12-30.2014.

[5] M. Lavanya , P. Muthukannan, Y. S. S. Bhargav

and V.Suresh, “IoT Based Automated Temperature
andHumidity Monitoring and Control”, Journal of
Chemical and Pharmaceutical Sciences (JCHPS)
Special Issue 5, pp.86-88, 2016.

[6] S. Kshirsagar and D. E. Upasani, “Energy

Management System for Smart Home”,
International Research Journal of Engineering and
Technology (IRJET), vol.3, Issue 6, 2016.

[7] S. Suresh, H.N.S. Anusha and T. Rajath.

“Automatic lighting and Control System For
Classroom,” 2016 International Conference on ICT
in Business Industry & Government (ICTBIG),
2016.

[8] S. B. Abhishek and J. E. Marties. “Connected

Home Using MQTT and Common Sensor
Framework,” International Journal of Advanced
Research in Electrical, Electronics and

ISBN: 978-1-941968-47-5 ©2018 SDIWC 28