Volume 7, Number 1, January – June, 2015

BIJIT - BVICAM's International Journal of Information Technology is a half yearly publication of Bharati Vidyapeeth’s
Institute of Computer Applications and Management (BVICAM), A-4, Paschim Vihar, Rohtak Road, New Delhi – 110063
(INDIA).
Editor- in-Chief Editor
Prof. M. N. Hoda Dr. Anurag Mishra
Director, BVICAM, New Delhi (INDIA) DDU College, University of Delhi
E-Mail: [email protected] Delhi (INDIA)
Associate Editors
Soft Computing
Prof. K. S. Ravichandran
Professor and Associate Dean, Dept. of Information and Communication Technologies, Sastra University Thanjavur – 613401,
Tamil Nadu (INDIA)
AI and Innovative Learning Technologies
Dr. Mohamed Hamada
Senior Associate Professor, Dept. of Computer Science, The University of Aizu, Aizu (JAPAN)
Data Mining, Analytics and Big Data
Dr. Girija Chetty
Associate Professor, Faculty of Information Technology and Engg, University of Canberra (AUSTRALIA)
Image Processing
Dr. Pradeep K. Atrey
Associate Professor, Dept. of Applied Computer Science, The University of Winnipeg (CANADA)
Information Security, High Speed Networks and Cloud Computing
Prof. D. K. Lobiyal
Associate Professor, School of Computer and Information Sciences, Jawaharlal Nehru University (JNU) New Delhi (INDIA)
Information Systems and e-Learning
Prof. Mohammad Yamin
Dept. of MIS, King Abdul Aziz University, Jeddah-21589, Saudi Arabia (KSA)
Resident Editorial Team
Dr. Anupam Baliyan Vishal Jain Ritika Wason
Associate Professor, BVICAM Asstt. Professor, BVICAM Asstt. Professor, BVICAM
New Delhi (INDIA) New Delhi (INDIA) New Delhi (INDIA)

Copy Right © BIJIT – 2015 Vol. 7 No. 1

All rights reserved. No part of the material protected by this copyright notice may be reproduced or utilized in any form or by
any means, electronic or mechanical including photocopying, recording or by any information storage and retrieval system,
without the prior written permission from the copyright owner. However, permission is not required to copy abstracts of papers
on condition that a full reference to the source is given.
ISSN 0973 – 5658
Disclaimer
The opinions expressed and figures provided in the Journal; BIJIT, are the sole responsibility of the authors. The publisher and
the editors bear no responsibility in this regard. Any and all such liabilities are disclaimed

All disputes are subject to Delhi jurisdiction only.

Address for Correspondence:

Prof. M. N. Hoda
Editor-in-Chief, BIJIT
Director, Bharati Vidyapeeth’s Institute of Computer Applications and Management (BVICAM),
A-4, Paschim Vihar, Rohtak Road, New Delhi – 110063 (INDIA). Tel.: +91 – 11 – 25275055
Fax: +91 – 11 – 25255056; E-Mail: [email protected], Visit us at www.bvicam.ac.in/bijit

Published and printed by Prof. M. N. Hoda, Editor-in-Chief, BIJIT and Director, Bharati Vidyapeeth’s Institute of Computer
Applications and Management (BVICAM), A-4, Paschim Vihar, New Delhi – 110063 (INDIA). Tel.: +91 – 11 – 25275055,
Fax: +91 – 11 – 25255056. E-Mail: [email protected]; [email protected]; Visit us at www.bvicam.ac.in/bijit
 Our Major Indexing at International Level
The INSPEC, IET (UK), formerly IEE (UK), database is an invaluable information resource
for all scientists and engineers, that contains 13 million abstracts and specialized
indexing to the world's quality research literature in the fields of physics and
engineering. For further details, click at http://www.theiet.org/resources/inspec/

Index Copernicus International (Poland) is a journal indexing, ranking and abstracting

site. This service helps a journal to grow from a local level to a global one as well as
providing complete web-based solution for small editorial teams. ICV 2012 for the BIJIT
is 4.75. For further details, click at
http://jml2012.indexcopernicus.com/BVICAMs+International+Journal+of+Information+
Technology,p4852,3.html

ProQuest (UK) connects people with vetted, reliable information. Key to serious
research, the company has forged a 70-year reputation as a gateway to the world’s
knowledge – from dissertations to governmental and cultural archives to news, in all its
forms. For further details, click at http://www.proquest.co.uk/en-UK/default.shtml

EBSCOhost Electronic Journals Service (EJS) is a gateway to thousands of

e-journals containing millions of articles from hundreds of different publishers, all at
one web site. For further details, click at http://www.ebscohost.com/titleLists/tnh-
coverage.htm

Open J-Gate is an electronic gateway to global journal literature in open access

domain. Launched in 2006, Open J-Gate is aimed to promote OAI. For further details,
click at http://informindia.co.in/education/J-Gate-Engineering/JET-List.pdf

DOAJ aims at increasing the visibility and ease of use of open access scientific and
scholarly journals, thereby promoting their increased usage and impact. For further
details, click at
http://www.doaj.org/doaj?func=issues&jId=87529&uiLanguage=en

Google Scholar provides a simple way to broadly search for scholarly literature and
repositories from across different parts of the world. For further details, click at
http://scholar.google.com/scholar?hl=en&q=BIJIT%2BBVICAM&btnG=

Cabell's Directory of Publishing Opportunities contains a wealth of information

designed to help researchers and academics, match their manuscripts with the
scholarly journals which are most likely to publish those manuscripts. For further
details, click at https://ssl.cabells.com/index.aspx

Academic Journals Database is a universal index of periodical literature covering basic

research from all fields of knowledge. For further details, click at
http://journaldatabase.org/journal/issn0973-5658

Indian Citation Index (ICI) is an abstracts and citation database, with multidisciplinary
objective information/knowledge contents from about 1000 top Indian scholarly
journals For further details, click at
http://www.indiancitationindex.com/htms/release_notes.htm

and many more..., for more details click at http://www.bvicam.ac.in/BIJIT/indexing.asp

 Editorial Board
Prof. A. K. Saini
University School of Management Studies, GGSIP University, New Delhi (INDIA)
Prof. A. K. Verma
Centre for Reliability Engineering, IIT Mumbai, Mumbai (INDIA)
Prof. A. Q. Ansari
Dept. of Electrical Engineering, Jamia Millia Islamia, New Delhi (INDIA)
Dr. Amudha Poobalan
Division of Applied Health Sciences, University of Aberdeen, Aberdeen (UK)
Prof. Anand Bhalerao
Dept. of Civil Engineering, Bharati Vidyapeeth’s College of Engineering, Pune (INDIA)
Prof. Anwar M. Mirza
Dept. of Computer Science, National University of Computer & Emerging Sciences, Islamabad
(PAKISTAN)
Prof. Ashok K. Agrawala
Dept. of Computer Science, Director, The MIND Lab and The MAXWell Lab, University of Maryland,
Maryland (USA)
Prof. B. S. Chowdhry
Dept. of Electronics Engineering, Mehran University of Engineering & Technology (PAKISTAN)
Dr. Bimlesh Wadhwa
School of Computing, National University of Singapore, Singapore (JAPAN)
Prof. Clarence Wilfred DeSilva
Dept. of Mechanical Engineering, University of British Columbia (CANADA)
Dr. D. M. Akbar Hussain
Dept. of Energy Technology, Aalborg University, Esbjerg (DENMARK)
Prof. David L Olson
Dept. of Management, University of Nebraska (USA)
Dr. Fahim Mohammad
Harvard Medical School, Harvard University, Boston (USA)
Prof Gurdeep S Hura
Dept. of Mathematics and Computer Science, University of Maryland, Maryland (USA)
Prof. Hakima Chaouchi
Telecom Sud Paris, Institute Mines Telecom (FRANCE)
Prof. Hamid R. Arabnia
Dept. of Computer Science, University of Georgia (USA)
Dr. Hasmukh Morarji
School of Software Engineering and Data Communications, Queensland University of Technology,
Brisbane (AUSTRALIA)
Dr. Javier Poncela
Dept. of Electronic Technology, University of Malaga (SPAIN)
Prof. K. K. Aggarwal
Former Vice Chancellor, Guru Gobind Singh Indraprastha University, New Delhi (INDIA)
Prof. K. Poulose Jacob
Dept. of Computer Science, University of Science and Technology, Cochin (INDIA)
Prof. Ken Surendran
Dept. of Computer Science, Southeast Missouri State University, Cape Girardeau Missouri (USA)
Dr. Ki Young Song
Dept. of Mechanical Engineering, The University of Tokyo, Tokyo (JAPAN)
Prof. Kishor Trivedi
Dept. of Electrical and Computer Engineering , Duke University (USA)
Prof. Kukjin Chun
Dept. of Electrical and Computer Engineering, Seoul National University (KOREA)
Prof. M. N. Doja
Dept. of Computer Engineering, Jamia Millia Islamia, New Delhi (INDIA)
Prof. M. P. Gupta
Dept. of Management Studies, IIT Delhi, New Delhi (INDIA)
Prof. Madan Gupta
Director, Intelligent Systems Research Laboratory, University of Saskatchewan, Saskatoon,
Saskatchewan (CANADA)
Dr. Nathalie Mitton
INRIA (FRANCE)
Dr. Nurul Fadly Bin Habidin
Engineering Business and Management, University Pendidikan Sultan Idris (MALAYSIA)
Prof. O. P. Vyas
Dept. of Information Technology, Indian Institute of Information Technology Allahabad (IIITA), Allahabad
(INDIA)
Dr. Prabhaker Mateti
Dept. of Computer Science and Engineering, Wright State University (USA)
Prof. Prasant Mohapatra
Dept. of Computer Science, University of California (USA)
Prof. Richard Chbeir
School of Computer Science, Université de Pau et des Pays de l'Adour (UPPA), Anglet (FRANCE)
Dr. S. Arockiasamy
Dept. of Information Systems, University of Nizwa, Sultanate of Oman (OMAN)
Prof. S. I. Ahson
Former Pro-Vice-Chancellor, Patna University, Patna (INDIA)
Prof. S. K. Gupta
Dept. of Computer Science and Engineering, IIT Delhi, New Delhi (INDIA)
Prof. Salim Beg
Dept. of Electronics Engineering, Aligarh Muslim University, Aligarh (INDIA)
Prof. Shiban K. Koul
Centre for Applied Research in Electronics (CARE), IIT Delhi, New Delhi (INDIA)
Prof. Shuja Ahmad Abbasi
Dept. of Electrical Engineering, King Saud University, Riyadh (KSA)
Prof. Steven Guan
Dept. of Computer Science & Software Engineering, Xi’an Jiaotong-Liverpool University (CHINA)
Prof. Subir Kumar Saha
Dept. of Mechanical Engineering, IIT Delhi, New Delhi (INDIA)
Prof. Subramaniam Ganesan
Dept. of Computer Science and Engineering, Oakland University, Rochester (USA)
Prof. Susantha Herath
School of Electrical and Computer Engineering, St. Cloud State University, Minnesota (USA)
Prof. Yogesh Singh
Director, NSIT, New Delhi & Former Vice Chancellor, MS University, Baroda (INDIA)
 Editorial

It is a matter of both honor and pleasure for us to put forth the thirteenth issue

of BIJIT; the BVICAM’s International Journal of Information Technology. It

presents a compilation of twelve papers that span a broad variety of research

topics in various emerging areas of Information Technology and Computer

Science. Some application oriented papers, having novelty in application, have

also been included in this issue, hoping that usage of these would further enrich

the knowledge base and facilitate the overall economic growth. This issue again

shows our commitment in realizing our vision “to achieve a standard

comparable to the best in the field and finally become a symbol of quality”.

As a matter of policy of the Journal, all the manuscripts received and considered

for the Journal, by the editorial board, are double blind peer reviewed

independently by at-least two referees. Our panel of expert referees posses a

sound academic background and have a rich publication record in various

prestigious journals representing Universities, Research Laboratories and other

institutions of repute, which, we intend to further augment from time to time.

Finalizing the constitution of the panel of referees, for double blind peer

review(s) of the considered manuscripts, was a painstaking process, but it helped

us to ensure that the best of the considered manuscripts are showcased and that

too after undergoing multiple cycles of review, as required.

The twelve papers, that were finally published, were chosen out of eighty nine

papers that we received from all over the world for this issue. We understand

that the confirmation of final acceptance, to the authors / contributors,

sometime is delayed, but we also hope that you concur with us in the fact that

quality review is a time taking process and is further delayed if the reviewers

are senior researchers in their respective fields and hence, are hard pressed for

time.
We further take pride in informing our authors, contributors, subscribers and

reviewers that the journal has been indexed with some of the world’s leading

indexing / bibliographic agencies like INSPEC of IET (UK) formerly IEE (UK),

Index Copernicus International (Poland) with IC Value 4.75, ProQuest (UK),

EBSCO (USA), Open J-Gate (USA), DOAJ (Sweden), Google Scholar, WorldCat

(USA), Cabell’s Directory of Computer Science and Business Information System

(USA), Academic Journals Database, Open Science Directory, Indian Citation

Index, etc. and listed in the libraries of the world’s leading Universities like

Stanford University, Florida Institute of Technology, University of South

Australia, University of Zurich, etc. Related links are available at

http://www.bvicam.ac.in/bijit/indexing.asp. Based upon the papers published in

the year 2012, its Impact Factor was found to be 0.605. These encouraging results

will certainly further increase the citations of the papers published in this

journal thereby enhancing the overall research impact.

We wish to express our sincere gratitude to our panel of experts in steering the

considered manuscripts through multiple cycles of review and bringing out the

best from the contributing authors. We thank our esteemed authors for having

shown confidence in BIJIT and considering it a platform to showcase and share

their original research work. We would also wish to thank the authors whose

papers were not published in this issue of the Journal, probably because of the

minor shortcomings. However, we would like to encourage them to actively

contribute for the forthcoming issues.

The undertaken Quality Assurance Process involved a series of well defined

activities that, we hope, went a long way in ensuring the quality of the

publication. Still, there is always a scope for improvement, and so, we request

the contributors and readers to kindly mail us their criticism, suggestions and

feedback at [email protected] and help us in further enhancing the quality of

forthcoming issues.

Editors
 CONTENTS
1. Health Management in Crowded Events: Hajj and Kumbh 791
Mohammad Yamin

2. Adopting SCRUM Agile Project Management for Managing Academic 795

Institutions
Abdullah Basahel

3. Design of a GSM Based Electronic Voting Machine with Voter Tracking 799
Vaibhav Bhatia and Rahul Gupta

4. SMITHA: Scalable Modular Interconnect for Three Dimensional High 803

Performance Applications - A New 3D Topology for NoC Based Systems
Sanju V, Niranjan Chiplunkar and Venkata Krishna

5. A Compendium Over Cloud Computing Cryptographic Algorithms and 810

Security Issues
Neha Mishra, Shahid Siddiqui and Jitesh P. Tripathi

6. Comparative Study of Cross Browser Compatibility as Design Issue in 815

Various Websites
Jatinder Manhas

7. Framework for Choosing Best Intrusion Detection System 821

Bilal Maqbool Beigh

8. A Novel Pruning Approach for Association Rule Mining 827

Lalit Mohan Goyal, M. M. Sufyan Beg and Tanvir Ahmad

9. Design of Linear-Phase Digital FIR Filter Using Differential Evolution 835

Optimization with an Improved Ripple Constraint Handling Method
Shekhar Sharma, Sumant Katiyal and L. D. Arya

10. A Novel Model for Security and Data Access for Jointly Accessing the Cloud 841
Service
Bansi Khimani and Kuntal Patel

11. Predictive Estimates of Employees’ Intelligence at Workplace with Special 845

Reference to Emotional and Spiritual Intelligence
B. K. Punia and Priyanka Yadav

12. Cloud Computing in SMEs: Case of Saudi Arabia 853

Mohammad Yamin and Ammar A. Al Makrami
 BIJIT - BVICAM’s International Journal of Information Technology
Bharati Vidyapeeth’s Institute of Computer Applications and Management (BVICAM), New Delhi (INDIA)

Health Management in Crowded Events: Hajj and Kumbh

Mohammad Yamin
Submitted in April, 2014; Accepted in December, 2014
Abstract - Reoccurrence of life threatening communicable events safer. In particular we will deal with the cases arising
diseases like EBOLA, HIV Aids, SARS, MERS in various from the communicable viruses and diseases. Our study uses
regions of the globe have made management large crowds two of the most crowded events which occur regularly, the Hajj
challenging. These viruses and the associated diseases can [7, 8 and 10] also see Pic 1 [19], an annual pilgrimage in Saudi
spread at an alarming rate and cause havoc in societies. Arabia and Kumbh [11], also see Pic 2 [20], a periodic
Crowd managers need to firstly take adequate steps to control pilgrimage in Allahabad and other cities of India. Ironically
the infiltration of infected people into the crowds and both of these events are religious in nature. Incidentally, the
secondly to have in place a management regime of effectively most crowded events of world are religious in nature. Fist we
treating the infected people, if found. Failing to manage will provide a description of the Hajj and the Kumbh from
deadly viruses and diseases in crowds can bring catastrophes organisational point of view.
of large magnitude not witnessed in the modern times. In this
article we provide an overview of managing large crowds 1.1 The Hajj
from health point of view. Our framework of health The Hajj is an annual pilgrimage to Makkah (Mecca) in Saudi
management for crowded events will be presented in the cases Arabia. Anybody belonging to the Islamic faith, irrespective of
of Hajj (an annual pilgrimage to Mecca) and Kumbh Mela (a their geographical location, can apply to perform Hajj through
periodic religious gathering in India), both of which are travel agents in the country of their residence. It takes place
known to be very complex and highly crowded events. every year during 8th-12th Dhulhijja, a month of Islamic
calendar. Being lunar, Islamic calendar is shorter than a
Index Terms – Crowd, Health, Management, Viruses, Hajj, Gregorian calendar by 10-12 days depending on the year, hence
Kumbh it completes a full cycle of all seasons in about thirty three
years. Every year about three millions out of an estimated ten
1.0 INTRODUCTION million applicants are granted permission to perform the Hajj.
Recent outbreak of EBOLA [1, 2 and 3] in some West African Main rituals of the Hajj are performed during four days.
nations has seriously challenged medical fertility in the world. However, limitation of international transportation and other
Despite many efforts by the World Health Organisation (WHO) reasons, many pilgrims spend about a month in Saudi Arabia.
and some developed nations, EBOLA continues to claim many During this extended period of their stay, they perform
human lives. During the last decade we have witnessed other pilgrimage at the prophet‘s mosque, built by Mohammed - the
kinds of dangerous outbreaks like Severe Acute Respiratory prophet of Islam in Madinah, the second most sacred mosque
Syndrome (SARC) [4]. Currently, some countries of the in the world, at a distance of about four hundred kilometres
Middle East, including Saudi Arabia, are battling with the from Makkah. One of the most complex Hajj operations is the
Middle East Respiratory Syndrome (MERS) also known as transportation of more than three million pilgrims from the tent
corona-virus [5 and 6]. There are already ongoing problems of city Mina to a short but congested journey to and from the
HIV virus in large part of the world. Valley of Mount Arafat. Some of the rituals, infrastructure,
While spread of these and other contagious viruses and management problems and other issues of the Hajj can be
diseases is not checked, events with large crowds run a risk of found in [7, 8 and 10].
catastrophic loss of human lives if they are not organised
properly. Many crowded events take place time to time. Some
are well defined and occur regularly where there are others
which are one off gatherings. Historically, crowded events have
witnessed stampedes, fires, drowning and mismanagement
resulting in the loss of thousands of human lives. An account of
the people killed in the last decade during various crowded
events in the world can be found from multiple sources
including [7, 8 and 9]. In this article we are mainly concerned
with the health issues of the participants of crowded events,
and propose solutions to deal with health risks and make the

Professor, Department of MIS, King Abdulaziz University,

Jeddah-21589, Saudi Arabia
E-mail: [email protected]
Figure 1: Hajj Rituals

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 791
 BIJIT - BVICAM’s International Journal of Information Technology

1.2 The Kumbh Mela implemented. In the last decade or so, many technological
Kumbh Mela is a pilgrimage, a festival and a fair which is the advances have produced tools and gadgets which can be
largest gathering of people in the world. In 2013 an estimated utilised to remarkably improve the crowd management’s
one hundred million pilgrims visited Kumbh (Kumbh Mela) abilities to respond to emergencies. In particular sensor and

Figure 2: Hajj Gatherings Figure 3: Kumbh Gatherings

[11]. Kumbh is organised periodically every three years in four biometric technology, tools and gadgets can be immensely
different cities of India along the banks of scared rivers helpful in improving the crowd management. Some of the
namely, Ganga (Ganges) in Haridwar and Allahabad, Yamuna intensely crowded events like the Hajj and Kumbh often
(Jamuna) and Saraswati in Allahabad, Godawari at Nasik and involve simultaneous on-foot movement of very large
Shipra in Ujjain. The largest of the Kumbh Mella, known as congregations, which creates possibilities of stampedes and
full Kumbh takes place every twelve years in Allahabad and contamination of contagious viruses and diseases. In such
the Ardha Kumbh (Half Kumbh) takes place every six years in situations, tracking, accessibility and identification of pilgrims
Haridwar. The duration of Kumbh is about six weeks during is critically important. Many of the sensor and wireless devices
which pilgrims come and go back after performing their available today, including Radio Frequency Identification
pilgrimage for a few days. One of the main rituals of Kumbh is (RFID) as shown in Pic 3, also see [12, 13, 14 and 15], can be
a dip in the scared river wherever it takes place. The pinnacle used for controlling and monitoring the movement of crowds.
of the rituals is known as Shahi Snanam on a particular day These technologies are already proving to be very beneficial in
known as Mauni Amavasya when a large number of pilgrims in securing and managing various aspects of daily life. A
dip in the river. In 2014 Kumbh at Allahabad about thirty description of ubiquitous technologies including RFID, sensor
million pilgrims made it to Shahi Snanam. Most of the Kumbh networks, biometric and scanning devices can be found in [14].
pilgrims come from within India itself. The event does,
however, welcomes and facilitates pilgrims from any country. 3.0 CROWDS AND HEALTH RISKS
Focus of the management is local population which usually Recent spread of EBOLA and other highly contagious viruses
travels with land routes including rail, road and on foot. and diseases has served an alarm to the crowd managers around
the world. Various Strands of flu viruses have already caused
2.0 AN OVERVIEW CROWD MANAGEMENT ISSUES havocs in various societies in the recent years. Re-emergence
One of the most worrying aspects of the Hajj and Kumbh of these viruses has discouraged many people from
managements is to control the number of pilgrims. The number participating in crowded events like Hajj and Kumbh. In Saudi
of permits issued every year for performing Hajj is less than Arabia, the country where the Hajj pilgrimage takes place, has
two million. However, the number of unauthorised pilgrims in been working very hard to contain the spread of the Middle
some years had nearly equalled the number of authorised Eastern Respiratory Syndrome (MERS), which has taken
pilgrims. For example, during 2012 Hajj, more than 3.65 hundreds of lives in the Middle East. Fortunately, this virus
million people performed Hajj [7]. As for the Kumbh, there are hasn’t so far affected the Hajj pilgrims. However, the Hajj and
no restrictions on the number of pilgrims as long as they Kumbh pilgrims suffer from various other health issues due to
comply with very basic health checks and undergo required unhygienic conditions as a result of crowding and
vaccinations. To minimise the chances of spread of diseases mismanagement. Usually, the lack of cleaning results in the
and other catastrophes like stampedes, drowning and fires, the growth of dangerous bacteria, which has already been
size of crowds must be contained to manageable limits. For, it witnessed in some cases [16 and 17]. Another serious health
is desirable to have some binding international standards for risk is from the inability of preventing infected illegal pilgrims
limiting the number of people which should be permitted to from entering to the crowded events like the Hajj and Kumbh.
gather in a specified space or area. Organisation of crowded Due to intense crowding and lack of infrastructure, it is not
events, in particular of those having millions of participants feasible to check and process all pilgrims within realistic
like Hajj and Kumbh, would become relatively easier to timeframe. The Hajj procedures, as can be seen in Fig 1,
administer if such standards were agreed upon and require pilgrims to undergo a medical check-up, which is a

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 792
Health Management in Crowded Events: Hajj and Kumbh

precondition to the grant of hajj permit. As for the illegal which are difficult to contain, until a satisfactory way of
pilgrims, boarders of Makkah can be sealed off to prevent the preventing their spread is available. However, people infected
entry of illegal pilgrims. The biometric scans at all the major with HIV and Hepatitis viruses are capable of being managed
air and sea ports of Saudi Arabia are capable of identifying and hence could be allowed under supervision. The
management shouldn’t neglect disabled and elderly pilgrims
and should make adequate arrangements for their well being
during their participation in the event.

5.0 DEALING WITH HEALTH RISK PILGRIMS

Despite the best efforts from crowd managers, it is probable
Figure 4: RFID Tags and Bands that some pilgrims would carry HIV, SARS, MARS and other
bacteria or virus with them into Hajj and Kumbh. Naturally
pilgrims with forged documents. The problem however still
such pilgrims wouldn’t be known to the management and
remains to check the local pilgrims from Makkah region and
hence would be living and moving with other pilgrims. These
prevent unauthorised ones from performing the Hajj. But in
pilgrims would pose a serious risk of infecting other pilgrims.
case of Kumbh, it is difficult to ensure that
To detect and manage such pilgrims, event organisers should
pilgrims do not carry dangerous viruses into the event as there
have pilgrim workflow architecture like the Hajj pilgrim
is no system in place of granting a permit on the basis of full
workflow architecture shown in Fig 2, see [7]. Event organisers
health check-up. Another aspect is that it is difficult to ensure
must also have adequate and easy access to health facilities on
that the health examinations carried out in all countries are not
the sites of the event. If any of these people become ill, they
compromised. Thus there are real possibilities of some pilgrims
should be isolated and treated promptly. If such patients were
carrying contagious viruses and diseases into the Hajj and
not identified in time and not treated accordingly, the bacteria
Kumbh.
might spread into a large number of participants. The Hajj is a
highly organised event but still needs to refine and articulate its
4.0 RECONILIATION WITH IINFACTED PILGRIMS
operations. The Kumbh is not a highly organised event because
Both Hajj and Kumbh are religious events and the aspirants of
of the huge number of participants with open borders to the rest
performing these events would want to be there with a hope of
of country. Also in the Kumbh, an overwhelming number of
washing off their sins before leaving this world. A terminally
participants come from within India. There are no visa
ill person would feel more urge to fulfil his or her religious
requirements which would otherwise provide an automatic
obligation. Socially, it is a challenging job for the governments
capture and storing data like in the case of the Hajj. It is
and hence the organisers to deny the aspirants a chance of a
recommended that the organisers of the Kumbh introduce a
lifetime pilgrimage. On the other hand, in many states, medical
permit system based on health and security. Here are some
reports can be compromised and infected people may succeed
guidelines which the event mangers could follow:
to secure a false certificate of health. So,
1. Introduce the system of ID base entry permit to the event
(1) it is almost impossible to ensure that no pilgrims with
2. Make through health check-up mandatory as a pre-
communicable diseases would enter the crowded event,
condition for an entry to the event
and
3. Manage and control health check-up centres around the
(2) it is socially not an good idea to prevent dying people from
globe with the help of established health organisations and
performing the pilgrimage
healthcare providers
That being the case, the organisers of religious events like the
4. Introduce RFID enabled wrist or waist bands, as shown in
Hajj and Kumbh should seriously revise their policies on sick
Pic 3 [21], to be worn by all pilgrims
people and seriously consider allowing them to perform the
5. Capture all personal and health data into the pilgrim tag
pilgrimage. However, for it to be accomplished, the organisers
6. Make adequate arrangements for isolation of pilgrims
must make separate arrangements for the sick, infected and
found to be carrying communicable diseases
terminally ill people to perform their pilgrimages. A huge
7. Introduce mandatory health awareness classes in all
advantage of this would be that the people with health risk to
countries with the help of local authorities to be sponsored
others will not try to gain permission by unfair or illegal means.
by tour operators of the event
Key to successful organisations of crowded events is to know
8. Punish tour operators for noncompliance of health
the kind of participants and plan for their management
procedures
accordingly. Not having basic knowledge about the pilgrims,
If the health risk pilgrims remain undetected, there is very little
in particular the health information, the organisers would not be
that event organisers could do. Crowd managers should use the
able to take precautions and hence would not be able to
latest technology to track, identify and treat ill participants. For
safeguard the health of masses in the crowded events. Spread
managing the health and wellbeing of the pilgrims, including
of a communicable disease in a crowded event may have
those infected with contagious bacteria, the RFID technology
devastating results. However, all of these considerations and
can play a very significant role [18].
concessions shouldn’t apply in cases of diseases like EBOLA

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 793
 BIJIT - BVICAM’s International Journal of Information Technology

6.0 CONCLUSIONS [5]. Wikipedia, Middle East respiratory syndrome corona-virus,

We do not want spread of life threatening diseases into masses. [Online] Available:
So far the Hajj and the Kumbh have not caused large scale http://en.wikipedia.org/wiki/Middle_East_respiratory_syndrom
infections. However, there is no assurance that this would e_coronavirus (November 7, 2014)
always be the case. Some viruses like EBOLA and SARS can [6]. WHO, Global Alert and Response (GAR) (Coronavirus
infections), [Online] Available:
http://www.who.int/csr/disease/coronavirus_infections/en
(November 7, 2014)
[7]. Yamin Mohammad and Albugami Moteb A, An Architecture
for Hajj Management, 15th IFIP WG 8.1 Proceedings
International Conference on Informatics and Semiotics in
Organisations, ICISO 2014, Shanghai, China, May 23-24,
2014, IFIP Advances in Information and Communication
Technology, Vol. 426
[8]. Yamin Mohammad, Yasser Ades, Crowd Management with
RFID & Wireless Technologies, Proceedings of First
International Conference on Networks & Communications,
IEEE Computer Society Washington, DC, USA ©2009
[9]. Kumbh Mela stampede leaves dozens of Hindu pilgrims dead
– video, [Online] Available:
http://www.guardian.co.uk/world/video/2013/feb/11/kumbh-
Figure 5: Hajj Pilgrim Workflow Architecture mela-stampede-hindu-video (November 7, 2014).
spread like a fire. The organisers of the Hajj, Kumbh and other
[10]. Hajj, [Online] Available:
crowded events must revise their procedures and action plans
http://www.saudiembassy.net/issues/hajj/ (November 7, 2014)
to prevent and contain these viruses from spreading on a larger
scale. National and international bodies should actively play [11]. Kumbh Mela, Wikipedia, [Online], Available:
http://en.wikipedia.org/wiki/Kumbh_Mela (November 7, 2014)
their role in helping to streamline the organisers to ensure the
safety and wellbeing of the participants of crowded events [12]. Klaus Finkenzeller, RFID Handbook: Fundamentals and
Applications in Contactless Smart Cards, Radio Frequency
where monitoring and controlling the participants are not easy.
Identification and Near-Field Communication, 3rd edn, Wiley
Crowd management should consider using various tracking and (2010).
identification devices like RFID as well as biometric
[13]. RFID stock photos and images, [Online] Available,
technology. Indeed the infected pilgrims must be isolated from
http://www.fotosearch.com/photos-images/rfid.html
the rest of the pilgrims. In this paper we present and analyse a (November 7, 2014).
number of problems of Hajj management, and offer some
[14]. Wen Yao, The use of RFID in healthcare: Benefits and
solutions including architecture for improving the management. barriers, Proceedings of IEEE International Conference on
These solutions can indeed be applied in many other and RFID-Technology and Applications (RFID-TA), 2010.
similar crowded events such as Kumbh.
[15]. Talevski A, Wu C, and Chang E, Wireless Sensors, SOA and
7.0 ACKNOWLEDGEMENT Web Based Approaches for Remote Operation and Control,
This work wouldn’t have been possible without the help of the 7th IEEE International Conference on Industrial Informatics,
King Abdulaziz University. The author acknowledges the help INDIN 2009.
and support from the Faculty of Economics and Management. [16]. John McConnell, Mass gatherings health Series, The Lancet
8.0 REFERENCES Infectious Diseases, Volume 12, Issue 1, Pages 8 - 9, January
[1]. Ebola Virus Disease, Word Health Organisation (WHO), 2012.
[Online] Available: [17]. Ziad A Memish, Gwen M Stephens, Robert Steff en, Qanta A
http://www.who.int/mediacentre/factsheets/fs103/en/ Ahmed, Emergence of medicine for mass gatherings: lessons
(November 7, 2014). from the Hajj, Lancet Infect Dis. 2012 Jan;12(1):56-65
[2]. About Ebola Virus Disease, Centers for Disease Control and [18]. Wen Yao, The use of RFID in healthcare: Benefits and
Prevention, [Online] Available: barriers, Proceedings of IEEE International Conference on
http://www.cdc.gov/vhf/ebola/about.html (November 7, 2014). RFID-Technology and Applications (RFID-TA), 2010.
[3]. Tracing Ebola's Breakout to an African 2-Year-Old, [Online] [19]. Hajj pictures: [Online], available:
Available: https://www.google.com.sa/search?sourceid=chrome-
http://www.nytimes.com/2014/08/10/world/africa/tracing- psyapi2&ion=1&espv=2&ie=UTF-
ebolas-breakout-to-an-african-2-year-old.html?_r=3 8&q=HPICTURES%20OF%20HAJJ%20CROWDS
(November 7, 2014) (November 7, 2014)
[4]. Wikipedia, Severe acute respiratory syndrome, [Online], [20]. Images of Kumbh, [Online], Available:
Available: https://www.google.com.sa/search?sourceid=chrome-
http://en.wikipedia.org/wiki/Severe_acute_respiratory_syndro psyapi2&ion=1&espv=2&ie=UTF-8&q=kumbh%20pictures
me (November 7, 2014) (November 7, 2014).

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 794
 BIJIT -BVICAM’s International Journal of Information Technology
Bharati Vidyapeeth’s Institute of Computer Applications and Management (BVICAM), New Delhi (INDIA)

Adopting SCRUM Agile Project Management for Managing Academic Institutions

Abdullah Basahel
Submitted in May, 2014; Accepted in December, 2014
Abstract - Managing academic institutions can be The question that arises is if SCRUM can be applied outside
challenging and different than managing other the IT domain, in different teams across different hierarchy
organizations. This is because the staffs are highly qualified levels that needs to work together in a top down approach. This
and may even be at the same and sometimes even higher paper will attempt to answer this question by illustrating how
academic level than their managers. An autocratic style of SCRUM can be used to manage academic institutions.
leadership may lead to tensions and be problematic for the
organization. Therefore, a more democratic approach is 2.0 AN OVERVIEW OF SCRUM
required. SCRUM is an agile project management method, SCRUM is an agile project management method, mainly used
mainly used in IT domain. SCRUM teams have no manager by software development teams [7], however, it can be used to
and rely on the team to control its members. This paper will manage projects and activities in different areas as well.
attempt to illustrate, how the SCRUM methodology can be SCRUM teams are self-organized and self-managed,
used to manage an academic institution. effectively there is no need for a manager [6]. The key idea
behind this flat hierarchy, is to engage the people who will
Index Terms – Academic Institution Management, Agile actually do the job, they know much better how to do it and
Project Management, SCRUM how much time it will to accomplish it rather than their
managers who may not even be relevant to the field. Like all
1.0 INTRODUCTION agile methodologies that follow the agile manifesto [5], and as
Academic Institutions differentiate in plethora ways from other it can be deducted from it, SCRUM favors activities that add
organizations, as they are dealing with educating their students value directly to the organization over other supportive
and conducting scientific research [1]. Managing such activities such as documentation. This does not mean there is
institutions can be different than managing other types of no value on supportive activities, but it means that SCRUM
organizations and it requires different management approaches. favors actions over plans. SCRUM also promotes resources
This is because, usually the human resources to be managed are sharing over dedicated use of people and equipment. In such a
highly qualified and highly respected and maybe at the same of way it has limited division of roles in a team like: the Product
even higher academic level than their managers, for example it owner, the SCRUM master and member(s). Briefly, the product
is not uncommon for a head of department to be at the owner is responsible for representing the client and clarifying
academic rank of an Assistant Professor and some of his staff the requirements. The SCRUM master is the coach of the team,
to hold the rank of Professor, with more teaching experience, organizing the team’s activities, meetings and helping the team
research contribution and even influence within the department. members to overcome various difficulties that may appear
An autocratic style of management may lead to tensions in the during a project. Neither the Product Owner nor the SCRUM
organization and possible have negative impact. In effect, a master are managers. Everybody in the team is a team member
more flat and democratic style maybe seem more appropriate and all of them work together to achieve the organization’s
[2]. goals and objectives, set by the business and introduced by the
SCRUM is an agile project management methodology mainly Product Owner. The SCRUM teams usually have six to eight
used in the IT sector [3]. SCRUM has a more flat management members. All the decisions about the team are made, reviewed
[4] and like all agile project management methods, tend to and enforced by the team in a democratic style of management.
focus more on activities that add value directly to an
organization rather that supportive activities [5]. Members of
SCRUM teams are equal and have no manager. Instead of a
manager, they use the influence of the team to motivate or push
the team members to complete their tasks [6].
This paper proposes the use of SCRUM agile project
management methodology for managing academic institutions.

Management Information System Department, Faculty of

Economics and Management, King Abdulaziz University,
Jeddah, Saudi Arabia.
E-mail: [email protected]

Figure 1: SCRUM overview

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 795
Adopting SCRUM Agile Project Management for Managing Academic Institutions

Figure 1, explains the process of SCRUM agile methodology. productivity of the team, is usually measured by the ‘Velocity’
At the beginning of the project, a draft task list is generated by (how many stories they completed/time); calculated based on
the business in collaboration with the Product Owner. These the total productivity of the team and not individual
tasks may change during the project, due to changes in the productivity. The key idea behind self-managed teams is that
business environment, or for other reasons. These tasks are that if a team member does not perform well, then the team will
prioritised and divided to releases forming the Product put pressure and provide assistance to bring this team member
Backlog. up to speed. This SCRUM dependence towards the team is also
SCRUM delivers products or services that add value in its biggest drawback, because if they majority of the team
incremental releases, called ‘Sprints’. Usually, each release has members are focused on the benefits of the organization the
duration of 2 weeks to 1 month, but this can change to fit the team will fail. Aligning all the team members that may also
project needs. Each release must produce product/services or have their personal agenda can be proven a big challenge for
complete a number of tasks that add value to the organization. their SCRUM Master. However, once a team starts, it is like a
The Sprint starts with the Sprint meeting. In this meeting, the well-oiled machine. Another, potential drawback of the method
Product Owner brings the prioritized tasks, desired to be lies within the principles of the agile manifesto [5]. Because the
included in the next release. These tasks are called Stories, and team, focuses of the tasks that have add value directly, they
are organizational requirements. During the Sprint meeting, the may pay less attention to supportive activities such as
team will review each story and estimate it, by using an agile documentation. This may lead to future problems, in terms of
estimation approach, for example, story points [8] and initially, maintainability and extensibility of the system or process. This
assign it to a team member. Later on during the Sprint, a story however, does not mean that if there is a high priority story for
may be reassigned to another member, in case that the original producing documentation or any support document, it will be
owner of the story is overloaded with tasks while other ignored. Instead, it will be treated like any other story.
members have still capacity to take more, or if another member According to the author this is the best way for inserting this
is more qualified for this task. If the team is overloaded with type of activities to a Sprint. The next section will show, how
stories, they may refuse to take on stories with lower SCRUM can be applied for managing academic institutions.
organizational value. On the other side, if they still have
capacity, they may ask for more stories. So stories original 3.0 ORGANISATIONAL STRUCTURE OF A COLLEGE
intended for a release, may be moved to another release. At the This section will illustrate how the process of SCRUM can be
end of the Sprint meeting, the Sprint Backlog will be created. adopted and optimized for managing an academic institution at
During the Sprint, the SCRUM master will call for daily Stand- a college level. Focus is given on this level because it is more
Up meetings. These meetings are five to fifteen minutes long. close to the main function of the institution but the same
During the meeting, the SCRUM master will ask every member approach can be extended to cover the whole organization.
three questions; Figure 2, shows a generic organizational structure of a college.
• What they did yesterday? The aim of this structure is aid the illustration of SCRUM,
An overview of the work completed over the last working however the same logic can be applied to colleges with small
day. variations in their organizational structure.
• If there are any problems?
If they are still any problems that impended the progress.
In case of a positive response to this question, the SCRUM
master should help the team members to solve the
problem, either directly or by associating the member with
the required resources for solving the problem.
• What they will do today?
Tasks, actions that they will do today.

During the meeting, the members are expected to provide a

brief overview, avoiding including too many technical details.
Engaging discussions during the meeting is not recommended
as this is intended to be a progress report meeting that will Figure 2: College Organizational Chart
monitor and motivate the team members and not a resolution
meeting. In case of any problems, the SCRUM master can As it can been seen from Figure 2, The Deanship is the leading
follow up with the involved parties after the meeting. the organization. Under it, there are two Vice Deanships, the
SCRUM promotes good communication and teamwork [6]. Unit of Development and the Director of Administration. The
During the project, the Product Owner, is available for Unit of Development, has responsibilities regarding to quality
clarifying related requirements to the team. The team members assurance, implementation of standards and similar activities.
work together, helping each other to complete the tasks. The In this example, its size is a team of six to eight people, but in

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 796
 BIJIT -BVICAM’s International Journal of Information Technology

some other cases it may vary, depending on the size of the structure, will have less complex but precise tasks. It is the
college. The Vice Deanship of Academic Affairs is responsible belief of the author that by using this type of recursive
for the academic issues of the college and the academic approach, SCRUM but be used to manage an academic
departments. Each academic department has a number of institution.
committees, for example committee for curriculum
development, examination committee and so on. Each 5.0 LIMITATIONS
committee has six to eight academics from the department. The Applying this approach for managing the institution, may be
Vice Deanship of Research is responsible for managing the difficult at the beginning as the Heads and Chairmen, may fell
research activities of the college. The Directory of that applying the SCRUM methodology aims to take away their
Administration, is responsible for the various units within the powers. This maybe be true in some extend, but the real aim of
college. For example maintenance and IT support. applying the approach is to create a high performance, less
Management entities are managed by council/committee that bureaucratic working environment [9]. In order to manage their
consist of the head and a secretary from the entity, and the head resistance, they can be allowed to keep their titles, for example,
of the heads of the entities that are under this entity in the Head of the IT support unit and so on, as long as they do the
organizational structure. For example the Vice Deanship of tasks required by the role of the Product Owner. The second
Academic Affairs is managed by the Vice Dean of Academic limitation is that it adds some complexity to the management
Affairs and the Head of the Departments. Each Department is process as the members will be involved with more than one
managed by the head of the committees accordingly. teams. They will be members in one team and Product Owners
in the other team, so daily they have to attend two Stand-Up
4.0 APPLYING SCRUM TO ACADEMIC INSTITUSION meetings. In general, more close to the bottom of the
MANAGMENT organizational structure, there is a need for daily meetings.
They key idea behind applying SCRUM for managing However, teams at a more high level in the organizational
academic institutions is that each management entity will be structure may not be required to have daily Stand-Up meetings.
consider as a SCRUM team. The head of the management unit Instead, they could limit them to two times per week. Also the
will be the Product Owner, and the secretary or one of the other duration of the Sprint could be adopted accordingly as long as
members the SCRUM master. Any member of the team can be it does not cause co-ordination problems between various
a SCRUM master as long as fulfils the criteria and is willing to teams.
do it. However, due to the workload involved, it is
recommended that secretary can be appointed for this role. The 6.0 CONCLUSION AND FUTURE SCOPE
Stand-Up meetings can be replaced by daily conference calls, This paper illustrated how the SCRUM methodology can be
where each member has to answer, what it was done the day used to manage academic institutions. The method was applied
before, if there were any problems and what it will be done to “generic Organizational Structure” of a college. While this
today. An electronic version of the Story Board will be structure may not be the same with some academic institutions,
required for managing the team. The members of each team the author believes that the method presented in this paper, can
will be also the Product Owners of their teams. So the Head of be easily tailored to map most academic institutions. The key
the Departments, will be members of the Vice Deanship of idea, is that each managing entity will be considered as a
Academic affairs, team but they will also be Product Owners, SCRUM team. The head of the entity will be the Product
in their Departments. The committee members of each Owner, the secretary the SCRUM master and rest of the team
department, will be members in the Department council team will be the members. Each member of a team will be the
but Product Owners in their Committees. At the beginning of Product Owner of its own team, so recursively the tasks will be
the Sprint, the Product Owner of a management entity will passed to the teams, lower in the organizational structure.
bring the tasks that need to be completed during this sprint, like Further to the main limitations of applying SCRUM to manage
any Sprint Meeting. During the meeting, the tasks are academic institutions is that people who hold manager
estimated, accepted and allocated to different members. positions may fell threaten if their title is changed to Product
However, unlike a normal SCRUM team, the tasks are not Owner. The approach allows them to keep their current titles as
going to be completed by the individual person but by the long as they also deal with the tasks of the Product Owner.
member’s team. The member as a Product Owner in his own Another limitation of this approach is that it adds some
team, will be responsible for braking down the work to be done overhead, as some members will belong to two teams and may
in prioritized stories and supply these stories to its own team, in not even be in the same locality. To solve this problem, Story
their Sprint meeting. After the end of the Sprint Meeting, the Boards can be replaced with electronic boards and Stand-Up
member will take all these tasks to its own team, where the meetings with conference calls. The frequency of the Stand-Up
member will now be the Product Owner. The same process will meetings can be also be reduced and the duration of the Sprint
be repeated until tasks reach individual persons. It is worth to can be modified to fit the organizational needs. Following this
be noted that the tasks are not going to be the same. The teams approach, this paper demonstrated how SCRUM can be used to
closer to the leadership, will have higher level goals and manage academic institutions.
objectives, while the teams at the bottom of the organizational

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 797
Adopting SCRUM Agile Project Management for Managing Academic Institutions

7.0 ACKNOWLEDGEMENT
This work would have not been possible with the support of
King Abdulaziz University and the Faculty of Economics and
Management.

8.0 REFERENCES
[1]. C.Maree. 'Defining administrators and new
professionals.' PERSPECTIVES, VOLUME 4,
NUMBER 1, 2000: pp. 4-5.
[2]. N. Martindale, (2011). "Leadership Styles: How to
handle the different personas". Strategic Communication
Management 15 (8): 32–35.
[3]. K.Schwaber, J. Sutherland, “The Scrum Guide”,
SCRUM.ORG, 2013,
http://www.scrumguides.org/docs/scrumguide/v1/Scrum
-Guide-US.pdf, last accessed 3/12/2014
[4]. SCRUMstudy, “A Guide to the SCRUM BODY OF
KNOWLEDGE (SBOK™ GUIDE), 2013 Edition,
ISBN: 978-0-9899252-0-4
[5]. The Agile Manifesto.
http://agilemanifesto.org/, last accessed 3/12/2014
[6]. N. Brede Moe, T.Dingsøyr , T.Dybå, “A teamwork
model for understanding an agile team: A case study of a
Scrum project”, Elsevier Information and Software
Technology, Volume 52, Issue 5, May 2010, Pages 480–
491
[7]. http://www.successfulprojectmanager.com/agile/scrum/,
last accessed 3/12/2014
[8]. J.Choudharia, U.Suman, “Story Points Based Effort
Estimation Model for Software Maintenance”, Elsevier
Procedia Technology, Volume 4, 2012, Pages 761–765.
[9]. T. ErlendFægria, T. Dybåa, T. Dingsøyra, “Introducing
knowledge redundancy practice in software
development: Experiences with job rotation in support
work”, ElsevierInformation and Software Technology,
Volume 52, Issue 10, October 2010, Pages 1118–1132.

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 798
 BIJIT - BVICAM’s International Journal of Information Technology
Bharati Vidyapeeth’s Institute of Computer Applications and Management (BVICAM), New Delhi (INDIA)

Design of a GSM Based Electronic Voting Machine with Voter Tracking

Vaibhav Bhatia1 and Rahul Gupta2

Submitted in April, 2014; Accepted in December, 2015

Abstract – This paper presents the design of a GSM based Number of voters at the polling place can be tracked by using
electronic voting machine with voter tracking. The design infrared sensors. The functions performed by the proposed
presented here follows a GSM based approach to send the voting machine includes:
polling results to a base station via mobile network. Infrared a. To cast votes to the candidates
sensors have been used for tracking the information b. Voter Tracking
regarding the voters. After the voting process has been over, c. Vote Rejection option available
the results are sent to the base station for various analyses d. Results are sent to monitoring station via GSM
and declaring the verdict. This system is more secured and This paper is organized as: Section 2 describes the GSM
chances of tampering the results are reduced. The simulation modem, Section 3 shows the block diagram of the system,
of the system is done on Proteus Professional Software v8.0. Section 4 describes the power supply schematic, Section 5
The design presented in this paper is more secured and shows the development tools used, Section 6 shows the circuit
appropriate according to modern day requirements. simulation, Section 7 describes the results and conclusions and
Section 8 discusses the future scope.
Index Terms – Infrared Sensor, GSM, Microcontroller,
Security, Voting. 2.0 GSM MODEM
A GSM modem is a specialized kind of modulator-demodulator
1.0 INTRODUCTION in which a SIM card is accepted and it can be operated over a
Voting is the most pivotal process which is carried out to reveal subscription to the mobile operator. GSM module is used when
the opinion of the people in selecting government or in any a communication between a computer and a GSM system is
issue that is under consideration. So the conventional voting required. In many countries it is used as architecture for mobile
systems based on paper voting are being replaced by electronic communication. GSM module consists of a GSM modem and
voting machines. Voting is a decision making mechanism in a communication interfaces like RS-232, USB along with a
society and security is indeed an essential part of voting. The power supply circuit for computer. GSM modem
term “electronic voting” represents the practice of electronic communicates over the mobile network when connected to a
means in voting to safeguard the security, reliability, and computer [11-12]. GSM modems are also used to send and
transparency. The crucial role in determining the result of an receive SMS and MMS messages. A GSM can be easily
election, electronic voting systems should be developed with interfaced with the microcontroller system and uses serial
the greatest responsibility and security. Electronic voting communication for data transfer. Fig. 1 shows a GSM module:
machines aid blind users by reading off the instructions using
headphones and also provide essential tools to help people with
disabilities. Voting machines are the combination of
mechanical and electronic equipments which are needed for
casting votes and displaying the election results. The main
proposal for using the voting machines was given in 1838
[1].There are large number of smart systems present which
employ microcontrollers for their operation [2-4] and several
other voting systems have been developed for ensuring a
secured vote casting process [5-7].The design presented in [8]
incorporates voter information facility for getting the
information about the number of voters at a place. In this paper,
GSM based design of an electronic voting machine has been
presented for sending the polling results to a monitoring station
via mobile network. This system is fully secured and chances
of digital tampering are also avoided [9-10]. Figure 1: GSM Module
1 GSM Module is used to accomplish the following functions:
Department of Electrical and Electronics Engineering,
a. To receive and send messages.
Bhagwan Parshuram Institute of Technology, New Delhi, India
2 b. To make, receive and reject a voice call.
Department of Electronics and Communication Engineering,
Maharaja Agrasen Institute of Technology, New Delhi, India
E-mail: [email protected]

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 799
 BIJIT - BVICAM’s International Journal of Information Technology

3.0 BLOCK DIAGRAM OF THE SYSTEM entering the correct password, thus making the system more
For the development of the system, PIC16F877A secured and less prone to mistakes.
microcontroller based on the modified Harvard architecture
developed by microchip technology has been used. Fig.2 shows 3.5. GSM Module
the block diagram of the system. Various sensors and devices
GSM Module is used in the system to send the polling results to
interfaced with the microcontroller are shown. The machine
design consists of: the monitoring station via mobile network. In this way a more
The machine design consists of: secured system is presented and the chances of anomalies are
3.1. Crystal Oscillator reduced.
It is an oscillator circuit that deploys the property of
mechanical resonance of the piezoelectric crystals for creating 4. 0 POWER SUPPLY
an accurate electrical signal. The frequency of the crystal by Five volts power supply w.r.t ground is required for the
keeping a track of time, provides clock signal to the operation of the microcontroller. Fig. 3 shows the circuit for the
microcontroller. Quartz crystal of frequency 4 MHz is used in power supply used in the system.
the proposed system. Quartz crystals are used in wrist watches,
calculators, counters, signal generators, and oscilloscopes.

Figure 3: Power Supply

The step down transformer steps down the ac supply from the
mains and bridge rectifier converts it into dc. This dc is passed
Figure 2: Block Diagram of the System through filter circuit to get a more smoothened waveform.
Voltage regulator produces a fixed output voltage that remains
3.2. Infrared Sensors constant irrespective of changes in its input voltage and load
Infrared (IR) sensor is an electronic device consisting of conditions. 7805 voltage regulator is used here, which gives +5
transmitter and receiver LED. The transmitter LED V output which is then given to microcontroller for its
continuously emits IR rays and when an object is close to the operation.
sensor, the rays bounce off the object and received into the
receiver LED. Infrared sensors are generally used in IR 5.0 DEVELOPMENT TOOLS
imaging devices, gas analyzers, and radiation thermometers. IR The software of the machine is written in Embedded C
sensors are used in this system to track the voter count which is language and the simulation of the system is done on Proteus
used at later stage for analysis of polling results. Professional Software v8.0. MPLAB Integrated Development
Environment has been used for programming the
3.3. Liquid Crystal Display microcontroller.
A Liquid Crystal Display is dot matrix display that displays
alphanumeric characters and symbols. Liquid crystal displays 6.0 CIRCUIT SIMULATION
are used in battery-powered devices, such as digital watches, A GSM based design of an electronic voting machine with
calculators, digital thermometers etc. 16X2 LCD has been used voter tracking is presented in this paper, Fig. 4 shows the
in the modeled system to display the candidate information and simulation schematic of the machine design and snapshot of
polling results. voter count is also included (see Fig. 5). To enable the polling
results to be displayed, an LCD having two by sixteen display
3.4. Polling Switches is used. The LCD is connected to the PORT B of the
Polling Switches are used in the system to cast votes to the microcontroller which is declared as the output port. The
candidates. By pressing the switch, vote can be given to the operation of the LCD is shown in Fig. 6.
desired candidate. Six switches have been used in modeled Polling switches are used to give votes to the candidates. The
system to cast votes and one switch is for vote rejection. The design is proposed for four candidates representing different
polling results are displayed by pressing the last switch after

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 800
Design of a GSM Based Electronic Voting Machine with Voter Tracking

parties, a switch is used to show the polling results after [3]. A. Drumea, P. Svasta, "Universal electronic module for
entering the correct password. There is rejection button industrial control based on system on chip device," 30th
available in the machine, which is pressed if voter does not International Spring Seminar on Electronics Technology,
want to choose a candidate. Infrared sensors are used for ISSE 2007, Cluj-Napoca, Romania, pp. 232-235, May
tracking the number of voters at the polling booth. After the 2007.
voting process has been over, the results are sent to the [4]. Vaibhav Bhatia, Gavish Bhatia, “Room Temperature
monitoring station via GSM using mobile network. There is an Based Fan Speed Control System using Pulse Width
added advantage of using GSM that the results data cannot be Modulation Technology”, International Journal of
tampered as it is directly sent to the base station for analysis Computer Applications, Vol. 81, no. 5, pp.35-40,
and results declaration. All the data consisting of number of November 2013
voters present at the polling booth, number of votes rejected [5]. A. Villafiorita, K. Weldemariam, and R. Tiella,
and number of votes given to the candidates is also sent to “Development, Formal Verification, and Evaluation of
monitoring station for verification. This system is appropriate an E-Voting System with VVPAT,” IEEE Transactions
and the chances of mistakes are reduced. The final results are on Information Forensics and Security, vol. 4, no. 4,
also displayed on the LCD after entering the correct password, 2009.
thus making the system highly secured. [6]. S.Lavanya."Trusted secure electronic voting machine"
International Conference on Nanoscience, Engineering
7.0 RESULTS AND CONCLUSIONS and Technology (ICONSET), pp.505 – 507, 2011
A GSM based voting machine design with voter tracking has [7]. D. Chaum , P. Y. Ryan and S. Schneider S. De Capitani
been proposed in this system which is found to be appropriate. di Vimercati , P. Syverson and D. Gollmann "A
The software of the system has been written in Embedded C practical voter-verifiable election scheme", Proc.
language and Proteus Professional Software has been used for Comput. Security (ESORICS 2005), vol. 3679, pp. 118
simulating the behavior of the machine. The simulation of the -139, 2005.
machine is working properly under normal conditions. Polling [8]. Vaibhav Bhatia, Rahul Gupta, “A Novel Electronic
switches are used to give votes to the candidates and infrared Voting Machine Design with Voter Information Facility
sensors have been used to track the voter entries. The count of using Microcontroller” International Conference on
the voter entries previously stored in the register is matched Computing for Sustainable Global Development , pp:
with the total votes casted and votes rejected to avoid any 311-313, March 5th- 7th , 2014.
mistakes thus making the system more protected. After the [9]. A. Hasti, “Study of Impact of Mobile Ad – Hoc
voting process has been over, the results are displayed on the Networking and its Future Applications”, BIJIT – 2012,
machine LCD by entering the correct password and sent to the January - June, 2012, Vol. 4 No. 1, ISSN 0973 – 5658.
monitoring station via GSM for analysis and the declaration of [10]. Kusam, P. Abrol and Devanand, “Digital Tampering
the final verdict. Table 1 shows the comparative study of Detection Techniques: A Review”, BIJIT-2009, July –
different voting systems. The design presented here is cost December, 2009, Vol. 1 No. 2, ISSN 0973 – 5658.
effective, highly secured and appropriate according to the [11]. V. Otsason, A. Varshavsky, A. LaMarca, E. de Lara,
modern day requirements. "Accurate GSM indoor localization," in Proceedings of
the 7th International Conference on Ubiquitous
8.0 FUTURE SCOPE Computing, pp.141-158, Tokyo, Japan, September 2005.
The design of the GSM based electronic voting machine with [12]. J. Cai, D.J. Goodman, “General Packet Radio Service in
voter tracking proposed in this paper is accurate and it can be GSM", IEEE Communication magazine, October 1997.
further improved in terms of power consumption using
advanced VLSI applications. Table 1: Comparative study with existing voting systems
Parameters Other Systems Our System
9.0 REFERENCES
[1]. Douglas W. Jones, “Early Requirements for Mechanical Control Unit Microcontroller Microcontroller
Voting Systems”, International Workshop on Security Less secured More Secured
Requirements Engineering for E-voting Systems, Aug.
31, 2009. Comparatively less
Accuracy More accurate
[2]. Vaibhav Bhatia, Pawan Whig, “Performance Analysis of accurate
Multi Functional Bot System Design using
Microcontroller”, International Journal of Intelligent Cost High Cost Low Cost
Systems and Applications, Vol. 6, no. 2, pp. 69-75,
January 2014.

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 801
 BIJIT - BVICAM’s International Journal of Information Technology

Figure 4: Simulation of the Machine

Figure 5: Snapshot of Counted Votes Figure 6: LCD Operation

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 802
 BIJIT - BVICAM’s International Journal of Information Technology
Bharati Vidyapeeth’s Institute of Computer Applications and Management (BVICAM), New Delhi (INDIA)

SMITHA: Scalable Modular Interconnect for Three Dimensional High Performance

Applications - A New 3D Topology for NoC Based Systems

Sanju V1, Niranjan Chiplunkar2 and Venkata Krishna3

Submitted in July, 2013; Accepted in October, 2014

Abstract - Today we are in an era of green computing Today we are in an era of mobile computing running multiple
wherein the devices are high performing, occupies smaller applications. These devices being hand held running in
chip die area, and consumes low power. These systems are batteries also had additional constraints in term of power
designed and implemented using multi core architectures. consumption also. The present popular topologies namely mesh
Network on chip is establishing itself as interconnect for this and torus when scaled poses a problem in terms of performance
high performance multi core systems. Currently the systems which in term affects the power consumption of the systems.
are realized using two dimensional topologies like mesh, torus This paper discusses a new topology for network on chip based
etc. Research outcome in fabrication technology is reducing systems SMITHA (Scalable Modular Interconnect for Three
the feature size of silicon processes which enables more logic dimensional High performance Applications). The paper also
to be implanted on silicon. This was well complemented with discusses the performance / area parameters of the same and
improvement in packaging technology which led to vertical compares with popular topologies namely mesh and torus. It is
stacking of logic to form of three dimensional structures. This observed that the proposed topology performs better with lesser
paper introduces a new three dimensional topology SMITHA area requirement.
(Scalable Modular Interconnect for Three dimensional High
performance Applications). The paper discusses the two 2.0 PROPOSED ARCHITECTURE
dimensional base topology along with routing algorithms and The discussion about the proposed architecture is done as two
performance parameters and its extension to three dimension. sections. The section below discusses the two dimension
Performance parameters for both cases are also discussed. topology along with the performance parameters followed by
its extension into three dimension.
Index Terms – Network On Chip, Design, SMITHA
2.1 Two Dimensional Base Topology
1.0 INTRODUCTION The proposed architecture is obtained by deleting the base node
One of the biggest inventions of the last century is and by interconnecting the neighboring nodes along the level of
semiconductor devices. These devices allowed design and a complete binary tree. This depicted in the figure below (Fig
implementation of systems in every domain ranging from 1).
consumer electronics to industrial or defense applications. In The topology is identified by the number of layers, numbered
the initial days these systems were implemented using common from 1 to K where K is the number of layers in the
bus architecture. The different processing modules shared a configuration. A node n in layer K is linked to its neighboring
traditional bus to communicate and pass information between nodes 2n and 2n +1 in layer K + 1. The topology becomes
them. As the complexity of the systems increased, common bus bigger with the number of layer with an increment of nodes in
posed a serious problem in terms of performance which led to power of two starting with two nodes in layer one to 2K nodes
usage of parallel or multi bus structures. These prevailed in in layer K. The number of nodes in a configuration of K layer
design for decades which always saw performance bottleneck.
To overcome these performance issues, Network On Chip
(NoC), a new paradigm in design was introduced [1,2]. The
idea was to implant the techniques of data communication
network on chip. In this the modules are interconnected using
topologies like mesh, torus and packets were exchanged for 2.2 Addressing
purpose of communication. The nodes are being addressed depending on the layer which
they are and position within it. The nodes start its address from
1
Dept. of Computer Science & Engineering, Nitte Meenakhsi zero to 2K and one to k for layers.
Instituteo of Technology, Bangalore, INDIA
2 2.3 Routing Algorithm
NMAMIT, Nitte P.O, Karkala, INDIA
3
School of Computing Science & Engineering, VIT University, This section brings out an optimal routing algorithm for routing
Vellore, INDIA packets in the same level. All the packets are routed through
E-mail: [email protected] the shortest path between any source destination pairs.
Step 0: Check destination address. Initialize current_src as
current node address and current_dest as destination address.

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 803
 BIJIT - BVICAM’s International Journal of Information Technology

Case 1: current_src and current_dest in same layer.

1. Compute the minimum hop count between current_src
and current_dest. If it is greater than 3, set current_src
and current_dest as their parent nodes respectively in
the adjacent layer below.
2. Repeat Step 1 till minimum hop count is greater than
3.
3. Move to the next node from current_src towards
current_dest. Set the current_src as the next node.
4. Repeat Step 3 until current_src is equal to
current_dest.
5. Consider a complete binary tree with top right node of
the current_src as root. If destination node lies in the
tree, mark the right node as set current_src else mark
the sibling of the left node as current_src. Move to the
current_src.
6. Repeat Step 5 until current_src is equal to destination Figure 1: Proposed Topology With Three Layers In Two
address. Dimension
Case 2: When destination node is in a layer above that of the 2.4 Performance Parameters
source node This section deals with the various performance parameters of
1. Move one step at a time from the current_dest to the the proposed topology in two dimension. The performance and
layer below until layer of the current_src is reached. area parameter are discussed below.
Set current_dest as the node obtained in the layer of
the source. 2.4.1 Maximum & Average Hop: This parameter brings out
2. Repeat the steps as in Case 1. the performance of the system. Consider a configuration of K
3. Repeat Step 5 and 6 of Case 1. layers, we tabulate the number of hop required through the
shortest route between every source and destination pair. The
Case 3: when destination node is in a layer below that of the maximum value thus obtained in the set is called as maximum
source node hop for a configuration of K layers. The mean value of the set
1. Move one step at a time from current_ src to the layer is called as average hop. Maximum hop brings out the time
below until the layer of current_dest is reached. Set taken for a packet to reach any source destination pair for a
current_src as the node obtained. given configuration. Higher the value, more will the time taken
2. Repeat the procedure as in case 1. to reach the destination thereby reduces the performance of the
system. Average hop also reflects the performance of the
The algorithm presented above can be explained as follows. system by considering a cumulative effect of all source
Case 1 represents when the source and destination are in the destination pair. The maximum hop for a configuration of n
same level. If they differ by three or less positions then the nodes is expressed as
packet is just transferred to them directly else it is routed HC (Max) (1,n) = 2log 2 (n+2) – 3 ≈ 2log 2 (n+2).
downward to the layers below and then it moves to the The table (Table 1) below depicts average hop for a single
destination. level. From the equation and table quantified above, it should
be noted that both the parameters of the proposed topology
Case 2 in the algorithm explains when the destination node is grows gradually as the topology scales.
above that of the source node. In this case, the routing logic
finds the parent node of the destination in the layer of the 2.4.2 Number of Wire Segments & Wirelength: This
source. Now both the source and destination is the same layer parameter brings out the area needed by the circuit and the
and will use Case 1 to route the packet to the intermediate node power consumption of the system to an extent. The parameter
and then the packet moves upward destination. number of wire segments brings out to the number of
interconnection wire segments used to produce the topology.
Case 3 expresses when the destination node is below than the Similarly considering unit length of wire between every pair of
source node. In this case, the packet moves to the root of the nodes, we calculate the wirelength requirement of the system.
source node in the layer of the destination. Now Case 1 hold These parameters also indicate the level of difficulty for the
goods as the source and destination is in same layer. CAD tool to generate the design and complete its process.
These parameters for a configuration of n nodes in two
dimension is expressed as

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 804
SMITHA: Scalable Modular Interconnect for Three Dimensional High Performance Applications - A New 3D Topology for NoC
Based Systems

W s (1,n) = WL s (1,n) = 2n – log 2 (n+2) – 1 ≈ 2n address (1, 2, 0) represents a node in level one layer two and
position zero. This is depicted in the figure (Fig 2)
3.0 Three Dimension Topology
The proposed three dimension topology is made by placing the 3.2 Routing Algorithm
above discussed base substrate one over the other and by This section presents a routing algorithm for the proposed
interconnecting the adjacent levels as follows. topology in three dimension. Whenever the source and
• The interconnections between odd level and an even are destination nodes are in the same level, the routing algorithm
done by interconnecting the even layers through the right presented in case of two dimension applies. The algorithm
and odd layers though the left. For example presented below routes a packet which is in two distinct levels.
interconnection between level one and two. For clarity in explanation and understandability, we divide the
• Similarly the connections between even level and an odd topology vertically into two as shown in the figure (Fig 3)
are done by connecting the odd layers though the right and
even layers through the left. For example interconnection
between level two and three.
This way makes the proposed topology more scalable by
placing one layer over the other. The interconnection discussed
above is depicted in the figure (Fig 2) below for a configuration
of three levels and each level having three layers.

Figure 3: Figure Illustrating the Routing Algorithm

Figure 2: Figure Depicting The Proposed Topology In
Three Dimension For A Configuration Of Three Levels Step 0: Check destination address.
And Each Level Having Three Layers Step 1: Check node number of destination and current node
address
3.1 Addressing Step 2: Check layer of the destination and current node address
The nodes are placed and are addressed relative to the level,
layer within that level and node position along the layer to Case 1: Current node and destination on the left
which the node belongs. For example a node who has an Step 3: Move left when current node and destination
is in the same layer

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 805
 BIJIT - BVICAM’s International Journal of Information Technology

Step 4: Move top left when destination is greater than

current node HC s(Max) (L,n) = HC s(Max) (1,n) + ( L – 1) + HC s(Max) (1,n)
Step 5: Move down when destination is less than = 2 * HC s(Max) (1,n) + ( L – 1)
current node = 2 *{ 2log 2 (n+2) – 3} + (L – 1)

Case 2: Current node on the left and destination on the right where L is the number of levels and HC s(Max) (1,n) is the
Step 3: Move right when current node and destination maximum hop in case of two dimension topology with n
is in the same layer or when the current node is greater nodes.
than the destination
Step 4: Move down when destination is less than Similarly maximum hop for all source and destination was
current node tabulated and simple arithmetic mean for recorded. The effect
was the same as observed in the case of two dimension.
Case 3: Current node and destination on the right
Step 3: Move right when current node and destination 3.3.2 Number of Wire Segments & Wire length: In case of
is in the same layer the three dimensional structure, the wires used to interconnect
Step 4: Move top right when destination is greater within layers and between the different levels contribute to the
than current node wire length and number of wire segments. Considering a unit
Step 5: Move down when destination is less than length of wire for interconnecting the node within the layer and
current node between the layers, the parameters can be expressed as

Case 4: Current node on the right and destination on the left W s (L,n) = WL s (L,n) = L ( 2n – log 2 (n+2) – 1) + ( L -1)
Step 3: Move left when current node and destination (log 2 (n+2) – 1)
is in the same layer or when the current node is greater
than the destination Where n is the number of nodes in each level and L is the
Step 4: Move down when destination is less than number of levels. The first term quantifies the parameters for
current node interconnecting nodes in single level and the second term
quantifies the same for interconnecting two successive levels. It
The current node checks the destination address in the packet. should be noted that as the number of levels increases, the
The next node of transit is decided by the routing algorithm parameters for interconnecting the successive levels increases
depending on the position of destination and current node. The only in logarithmic order. This is very advantageous in case of
routing algorithm routes the packet to the nearest node towards the proposed three dimension topology.
the periphery in the same level of the current node and then to
the level of the destination node. Now the packet is in same 4.0 COMPARISON OF PERFORMANCE PARAMETERS
level as destination node, will follows the two dimensional WITH EXISTING TOPOLOGIES
routing algorithm to reach the destination node. For example, This session compares the above discussed performance
consider a packet currently in node (1,2,1) to be transferred to parameters with two dimensional and three dimensional mesh
(2,2,1). Since both the nodes are on the right topology, the and torus, the popular existing topologies in this concept.
packet is routed towards the periphery through the right to
(1,2,0), then to the level of the destination to (2,2,0) and then to 4.1 Comparison with Two Dimension Topologies:
the destination (2,2,1). Similarly the proposed routing This session compares the performance parameters in case of
algorithm routes packet to all source destination pairs. two dimension with popular existing topologies – mesh and
torus
3.3 Performance Parameters The tables ( Table 2,3) below tabulate the parameter maximum
This session discusses the different parameters in case of the hop and average hop for different number of nodes per layer
proposed three dimension topology. The parameters discussed for levels one, two and three
are same as those of two dimension. From the tables 2 and 3, it is quite evident that the maximum
hop and average hop parameters are good for the proposed
3.3.1 Maximum & Average Hop: As discussed in the case of topology when it is compared to mesh and torus. Also it should
two dimensional, the performance of the systems will be given be noted that the performance parameters for the proposed
by this parameter. As per the routing algorithm presented topology when scaled does not grow drastically when
above, the packet moves towards the node at the periphery compared with mesh and torus This implies that the packets
which is at most HC s(Max) (1,n). Then it should travel vertically reach the destination in less time and the output generation is
up through (L – 1) nodes to reach the level of the destination faster increasing the throughput of the system in the proposed
node. Now it takes another HC s(Max) (1,n) to reach the topology and a system implemented using the proposed
destination node. Summing the values above, the parameter can topology in any dimension performs better. This has an
be bounded as implication of power consumption also.

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 806
SMITHA: Scalable Modular Interconnect for Three Dimensional High Performance Applications - A New 3D Topology for NoC
Based Systems

The following tables (Table 4,5) quantify the parameters 6.0 REFERENCES
related to area of the systems. The tables below tabulate [1]. William J. Dally , Brian Towles, Route Packets Not
number of wire segments and wire length for the same. Wires : on chip interconnection network, DAC 2001
From the above tables, it is quite clear that the area requirement June 2001.
for the proposed is almost same in lower configuration / levels [2]. Tobias Bjerregaard And Shankar Mahadevan, A Survey
but as the system scales to higher levels the wire requirement is of Research and Practices of Network_on_Chip, ACM
comparatively very high when compared to the proposed Computing Survey March 2006.
topology. This effect is not only on the area requirement but [3]. Ville Rantala, Teijo Lehtonen, Juha Plosila, Network On
also on the power dissipation of the system.Summing the above Chip Routing Algorithms, TUCS Technical Report No
two results it is evident and clear that the proposed architecture 779, August 2006.
performs better with lesser area / power requirement. Apart [4]. Ahmed Hemani, Axel Jantsch, Shashi Kumar, Adam
from the analysis above, the proposed topology was subjected Postula, Johnny berg, Mikael Millberg,Dan Lindqvist,
to real time scenarios under different buffer and load / traffic Network on a Chip: An architecture for billion transistor
conditions to test the strength of the topology which also gave era, DAC 2001.
positive results. [5]. Radu Marculescu, IEEE, Natalie Enright Jerger, Yatin
Hoskote, Umit Y. Ogras and Li Shiuan Peh, Outstanding
5.0 CONCLUSION Research Problems in NoC Design: System, Micro
This paper introduces a new three dimensional topology architecture, and Circuit Perspectives, IEEE
SMITHA (Scalable Modular Interconnect for Three Transactions on computer aided design of integrated
dimensional High performance Applications). The discussion circuits and systems, vol. 28, no. 1, January 2009.
starts with the two dimension variant and extends it to three [6]. J. Nurmi: Network on Chip: A New Paradigm for
dimension. The proposed architecture is structured and System on Chip Design. Proceedings 2005 International
scalable. The paper quantifies the different performance / area Symposium on System on Chip, 15 17 November 2005.
parameters namely maximum hop, average hop. wirelength, [7]. L. Benini, G. De Micheli. Networks on chips: A new
number of wire segments in both two / three. dimension. This SoC paradigm. IEEE Computer.35(1), 2002
is compared with those of the current popular topologies [8]. S. Kumar, A. Jantsch, J. Soininen, M. Forsell,
namely mesh and torus. The growth of these parameters is M.Millberg, J. Oberg, K. Tiensyrja, and A. Hemani, A
found to be slow in case of the proposed topology implying Network on Chip Architecture and Design Methodology,
that a system implemented performs better with smaller die Proceedings International Symposium VLSI (ISVLSI),
area / power consumption. pp. 117-124, 2002.
[9]. Sanju V, Niranjan Chiplunkar, Mohd Khalid, P Venkata
Patent Information Krishna, Performance Analysis Through Simulation For
The architecture discussed in this paper is applied for patent Three Dimension Network Of Chip Based System,
with the following details Anusandhana – Journal Of Science, Engineering And
Type : India Management, Vol No 01, Issue No 02, Dec 2012, Pg.
Number : 1598/CHE/2014 18 – 29.
Date : 26/03/2014

Table 1: Depicting Average Hop For A Single Level

Number
of 1 2 3 4 5 6 7 8 9 10
Layers
Number
of 2 6 14 30 62 126 252 510 1022 2046
Nodes
Average
.5 1.2 2.17 3.29 4.65 6.21 7.9 9.75 11.64 13.5
Hop

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 807
 BIJIT - BVICAM’s International Journal of Information Technology

Table 2: Table Quantifying The Parameter Maximum Hop

No Of Topology
Nodes Number of Levels
Mesh Torus SMITHA Mesh Torus SMITHA Mesh Torus SMITHA
1 1 1 2 2 2 3 3 3
4 2 2 1 3 3 3 4 4 5
8 4 3 3 5 4 4 6 5 6
16 6 4 5 7 5 6 8 6 8
32 10 6 7 11 7 8 12 8 10
64 14 8 9 15 9 10 16 10 12
128 22 12 11 23 13 12 24 14 14
256 30 16 13 31 17 14 32 18 16
512 46 24 15 47 25 16 48 26 18
1024 62 32 17 63 33 18 64 34 20
2048 94 48 19 95 49 20 96 49 22

Table 3: Table Quantifying The Parameter Average Hop

Topology
No Of Nodes Number of Levels
Mesh Torus SMITHA Mesh Torus SMITHA Mesh Torus SMITHA
1 1 1 2 2 2 3 3 3
4 1.00 1.00 0.50 6.00 6.00 1.25 17.00 15.00 1.94
8 1.75 1.50 1.20 9.00 8.00 2.03 23.75 19.50 2.65
16 2.50 2.00 2.17 12.00 10.00 3.06 30.50 24.00 3.71
32 3.87 3.00 3.29 17.50 14.00 4.35 42.88 33.00 5.05
64 5.25 4.00 4.65 23.00 18.00 5.87 55.25 42.00 6.61
128 7.93 6.00 6.21 33.75 26.00 7.55 79.44 60.00 8.33
256 10.63 8.00 7.90 44.50 34.00 9.35 103.63 78.00 10.16
512 15.96 12.00 9.75 65.88 50.00 11.23 151.72 114.00 12.06
1024 21.31 16.00 11.64 87.25 66.00 13.16 199.81 150.00 14.00
2048 31.98 24.00 13.50 129.94 98.00 15.12 295.86 222.00 15.96

Table 4: Table Quantifying the Parameter Number Of Wire Segments

No Of Topology
Nodes Number of Levels
Mesh Torus SMITHA Mesh Torus SMITHA Mesh Torus SMITHA
1 1 1 2 2 2 3 3 3

4 4 8 1 12 24 3 20 36 5
8 10 16 8 28 48 18 46 72 28
16 24 32 23 64 96 49 104 144 75
32 52 64 54 136 192 112 220 288 170
64 112 128 117 288 384 239 464 576 361

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 808
SMITHA: Scalable Modular Interconnect for Three Dimensional High Performance Applications - A New 3D Topology for NoC
Based Systems

No Of Topology
Nodes Number of Levels
Mesh Torus SMITHA Mesh Torus SMITHA Mesh Torus SMITHA
1 1 1 2 2 2 3 3 3
128 232 256 244 592 768 494 952 1152 744
256 480 512 499 1216 1536 1005 1952 2304 1511
512 976 1024 1010 2464 3072 2028 3952 4608 3046
1024 1984 2048 2033 4992 6144 4075 8000 9216 6117
2048 4000 4096 4080 10048 12288 8170 16096 18432 12260

Table 5: Table Quantifying the Parameter Wirelength

No Of Topology
Nodes Number of Levels
Mesh Torus SMITHA Mesh Torus SMITHA Mesh Torus SMITHA
1 1 1 2 2 2 3 3 3

4 4 8 1 12 48 3 20 72 5
8 10 20 8 28 96 18 46 144 28
16 24 48 23 64 192 49 104 288 75
32 52 104 54 136 384 112 220 576 170
64 112 224 117 288 768 239 464 1152 361
128 232 464 244 592 1536 494 952 2304 744
256 480 960 499 1216 3072 1005 1952 4608 1511
512 976 1952 1010 2464 6144 2028 3952 9216 3046
1024 1984 3968 2033 4992 12288 4075 8000 18432 6117
2048 4000 8000 4080 10048 24576 8170 16096 36864 12260

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 809
 BIJIT - BVICAM’s International Journal of Information Technology
Bharati Vidyapeeth’s Institute of Computer Applications and Management (BVICAM), New Delhi (INDIA)

A Compendium Over Cloud Computing Cryptographic Algorithms and

Security Issues

Neha Mishra1, Shahid Siddiqui2 and Jitesh P. Tripathi3

Submitted in March, 2014; Accepted in November, 2014

Abstract - Cloud computing is an emerging and revolutionary networks, servers, storage, applications, and services) that can
approach towards the computing and becoming more risk be rapidly provisioned and released with minimal management
prone than ever before. It is an evolutionary approach of effort or service provider interaction [2]. There are numerous
using resources and services on demand and as per need of benedictions to adopt cloud computing but still there are few
consumers. Cloud computing providing a platform rose on loop holes that make adoption difficult to adopt. Cloud
the Internet for usage of IT services and flexible computing providers must ensure their users for hard security
infrastructure to the consumers and business. Deployment of data and relief from various attacks.
and management of services or resources are maintained by
the third party. Whereas there are innumerable advantages to 1.1 The following Definition of Cloud computing has been
approaching the cloud computing, it also contains various developed by NIST-
issues such as confidentiality, Integrity, Authenticity and Cloud computing is a model for enable convenient on-demand
Privacy. One of the prominent barrier to adopt the cloud network access to a shared pool of configurable computing
computing is security. This paper comprises the elaborated resources that can be rapidly provisioned and released with
study on various security issues allied to cloud computing are minimal management effort or service provider inter action.
presented by consolidating literature reviews on Cloud model promotes availability and is composed of five
cryptographic algorithms used for data security. essential characteristics, four deployment models and three
service models.
Index Terms – Cloud computing, Cryptographic algorithm,
Decryption, Encryption, Security issue. 1.2 Essential Characteristics of Cloud Computing
 On-demand self-service
1.0 INTRODUCTION  Broad network access
Cloud computing proffering us the delivery of computing  Resource pooling
services over the Web. Cloud services providing the usage of  Rapid elasticity
software and hardware that are maintained and deployed by  Measured service
third party to the individuals or business from a remote
location. A study conducted by Gartner [1,2] on Cloud 1.3 Cloud Service Models
Computing is regarded as the first among the top 10 most The three fundamental classifications are often referred to as
important technologies and well acknowledged by companies the “SPI Model” where ‘SPI’ refers to the Software, Platform
and organizations. Cloud computing encapsulate various or Infrastructure (as a Service), respectively.
services such as web mail, social networking sites, online file
storage and different business application. Cloud computing 1.3.1 Cloud Software as a Service (SaaS): In this type of
enable users to access services and resources from a model complete application is provided to the cloud users. It is
configurable shared pool from anywhere where network mainly accessed through web portal and services oriented
connection is available. As each users and organizations are architecture (SOA).The Main Consistence Server (MCS) and
transmigrated their information and statistics to the cloud, Domain Consistence Server (DCS)[4].
hence it uses the storage service of cloud deployed by cloud
provider. So it is essential to secure data from any illegitimate 1.3.2 Cloud Platform as a Service (PaaS):Paas encapsulate
user access or any other attack such as denial of service, environment for the development and provisioning of cloud
modification and forgery of document etc. Cloud Computing applications. Examples: Force.com, Google App Engine and
enables ubiquitous, convenient, on-demand network access to a Microsoft Azure.[4]
shared pool of configurable computing resources (e.g.,
1.3.3 Cloud Infrastructure as a Service (IaaS): Infrastructure
1
CSE, Integral University, Lucknow, India layer is used to essential IT resources. Examples: Amazon
2
Asst. Professor CSE, Integral University, Lucknow, India Elastic Cloud, Computing (EC2), Amazon S3 and Go Grid. [4]
3
Associate Professor, S R Group of Institutions, Lucknow,
India, 1.4 Cloud Deployment Models:
E-mail: [email protected], [email protected]  Public Cloud-Microsoft Azure, Google App Engine [4]
and [email protected]  Private Cloud-Eucalyptus Systems [4]

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 810
 BIJIT - BVICAM’s International Journal of Information Technology

 Community Cloud -Face book [4] users by the cloud providers. Other concern is related to the
 Hybrid Cloud -Amazon Web Services (AWS). [4] accessing of data and resources. Cloud provider must keep eye
on who is accessing data, from where this activity is taking
place and what type of control are applied. Data must be
classified for efficient accessing of data.

2.1.2 Training of cloud users-employee or users must be

trained so that they can efficiently and properly access data.
Employee must be trained to know how to maintain data.

2.1.3 Service Level Agreement (SLA)–SLA is an agreement

between the cloud users/business and the cloud service
providers to assure which services are used by an individual
user. SLA must be unambiguous or clear.

2.2 Basic Security issues for cloud

 Availability
 Data /System Integrity
 Authentication
 Storage, Backup and Recovery of data
 Data Confidentiality and privacy
 Access control
2.3 Different Threats in Cloud Computing [9]
 Account or Service hijacking
Figure 1: NIST Visual Model of Cloud Computing  Denial of service
Definition [4]  Data Scavenging
 Data Leakage
 VM escape and hopping
 Customer data manipulation
 Sniffing/Spoofing
 Attack against Web Services
 Man-in-middle attack

3.0 CRYPTOGRAPHIC ALGORITHMS FOR DATA

SECURITY
In cloud computing data security is the main concern. For the
same different cryptographic algorithms are used. Original text
message is known as plaintext and the coded form is known as
cipher text. Conversion of plaintext to cipher text is called
Figure 2: Cloud service model encryption. Cipher text can be converted back to plain text, this
is call decryption. Cryptography comprises the study of
2.0 SECURITY ISSUES AND THREATS OF CLOUD encryption and decryption.
Write correct security controls should be implemented
according to asset, threat, and vulnerability risk assessment 3.1 Symmetric encryption is a technique to camouflage the
matrices [5]. While cloud security concerns can be grouped originality of contents of blocks or streams with message file,
into any number of dimensions (Gartner names seven[6] while encryption key and password. Single key is used to encrypt or
the Cloud Security Alliance identifies fourteen areas of decrypt data. There are two kinds of symmetric-key encryption
concern[7]) these dimensions have been aggregated into three algorithms are used to wrap-up the content in a mask i.e. Block
general areas: Security and Privacy, Compliance, and Legal or cipher and Stream cipher. In block cipher a block of plain text
Contractual Issues [8]. of fixed size is encrypted at a time using key. In stream cipher a
2.1Security concern of cloud users - bit of stream is encrypted at a time using key. e.g. DES, AES,
2.1.1 Data- Data is main entity of communication and it must triple DES, Blowfish etc. are cloud computing algorithms.
be secure enough so that it cannot be hamper by any
unauthorized user. Data security should provide to the cloud

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 811
A Compendium Over Cloud Computing Cryptographic Algorithms and Security Issues

3.1.1 DES –DES is a symmetric algorithm for data encryption Brute force attack. Because it uses 128 bit block size it more
by using 56 bit key size. It uses balanced feistal structure. It is secure than any other algorithm.[9,10]
designed by IBM in 1977.DES uses 64 bit block. Feistal
function for this are – expansion, substitution, key mixing and 3.1.4 RC-5-RC-5 encryption technique is Designed by Ronald
permutation and for the encryption process of DES there are Rivest in 1994.This symmetric algorithm uses Keysize of 0-
two permutations, one is initial and the other is final 2040bit and uses variable block size (32, 64 or 128 bits), key
permutation and sixteen Feistel rounds are used to generate the size (0 to 2040 bits) and number of rounds (0 to 255).It is
key, for each round 48-bit keyis generated from the cipher susceptible to 64-bit blocks differential attack using 244 chosen
key.[9,10]. plaintexts. [9,10]

Figure 5: Encryption with AES

3.1.5 IDEA-This block cipher uses 64 bits block of message
and 128 bit key. This encryption algorithm suffer from narrow
bicliques attack.[9,10]

3.1.6 BLOWFISH-Blowfish is one of the block cipher

algorithm for encryption. This encryption technique uses the
same secret key to both encryption and decryption of
information. Blowfish uses 64 bit block size and variable
length key, from 32 bits to 448 bits. Blowfish is appropriate
Figure 3: Encryption with DES technique for applications where the key is not changed
frequently. Over the 32-bit microprocessors it is faster and
3.1.2 3DES- This encryption algorithm is derived from DES. It efficient than other encryption techniques with huge data. It
provides an easy and efficient way of increasing the key size of uses 16-round Feistel network.[9,10]
DES to protect against brute force attack.[9,10]
3.2 Asymmetric encryption is used to encrypt small block of
data. One key is used to encrypt data or other key is used to
decrypt data. Two keys are: Private Key and Public Key. The
Public key is used by the sender for the purpose of encryption
and the private key is used for the purpose of decryption of data
by the receiver. In cloud computing these algorithms are used
Figure 4: Encryption with 3DES to generate keys. Some of the common asymmetric-key
algorithms for cloud are: RSA,DH and IKE.
3.1.3 AES – AES is a symmetric algorithm for data encryption
by using 128, 192, 256 bit key. This algorithm is affected by

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 812
 BIJIT - BVICAM’s International Journal of Information Technology

3.2.1 RSA- RSA is a public key cryptographic algorithm for Larger the block size means greater security but decreases the
data security. This is a most common encryption algorithm encryption/decryption speed. AES has larger block size among
used by people to encrypt message with two keys. RSA the entire algorithms and has greater security capability than
algorithm encryption and decryption is based on the modular other.
exponential and has two exponents, a and b, where a is used for Table 1: Comparison of Cryptographic Asymmetric
public and b is used for private. Let the plaintext is M and C is encryption algorithms
cipher text, then at encryption.[9,10]
C =Ma mod n ASYMMETRIC ALGORITHM
And at decryption side
M = Cb mod n.
DESIGN
FEATURE DIFFIE-
RSA ElGamal ECC
HELLMAN

Based on
Based on
the Vulnerable
Bases on difficulty to
problem and secure
the determine
SECURITY of against
discrete secret key k
factoring eavesdroppin
logarithm given kP
large g
and p
Numbers
Free for
all,
STANDAR FIPS186- IEEE P1363
Patented ANSI X9.42
D 3
only in
US
Used for
confident
iality and Used for Implementin
key Used for Key both g algorithm
USAGE
exchange exchange encryption such as
as well as and DSA DSA
for digital
sign.
Figure 6: RSA algorithm
NO. OF
3.2.2 Diffie-Helman Key Exchange- This is created by the KEYS
2 2 2 2
Whitfield Diffie and Martin Hellman In 1976.This algorithm
depends on the complexity of discrete logarithm. Diffi-hellman
2013,224
basically used for key exchange between two users.[9,10]. KEY 512 to bits for q and 112 bit to
ALGORITHM: 2048 bit
LENGTH 15,360 2048 bits for 512 bit
p
Timing or
Brute
Chosen simple and
forced
Denial of cipher text differential
and
ATTACKS service and power attack
oracle
attack malleabilit (side
attack
y channel 0r
etc.
fault)

Larger the key size means greater security but decreases the
encryption/decryption speed .Blowfish has larger key length
3.2.3 ElGamal- This is an asymmetric algorithm used for among other algorithm.
transmitting digital signatures as well as for key exchange. El No. of rounds, multiple rounds offers greater security. Blowfish
Gamal is based on the applicability of discrete logarithms. It is has 16 rounds which is typically a standard number of rounds.
rely on the logarithmic number’s characteristics or calculations The most important thing no attack is known to be successful
of these numbers. [9,10] against Blowfish. Hence Blowfish is superior to other
algorithms.
4.0 COMPARISON AND RESULT Different asymmetric algorithms are compared below on the
Different symmetric algorithms are compared below on the basis of design feature. Analysis and performance Comparison
basis of design feature. Analysis and performance Comparison Table of symmetric encryption algorithm-[9,10,12].
Table of Asymmetric encryption algorithm-[9,10,12]

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 813
A Compendium Over Cloud Computing Cryptographic Algorithms and Security Issues

Table 2: Comparison of cryptographic symmetric AES effective in both software and hardware.3DES and DES
encryption algorithms are slow in software. Blow fish is more efficient in software.
DESIGN FEATURE
AES is excellent in security rate and execution time than RSA.
This paper encompasses a theoretical performance analysis of
SYMM symmetric or asymmetric encryption algorithm.
ETRIC
ALGOR NO. In the future, our research will be extend by providing
POSSIB
ITHM BLOCK
SIZE
KEY SIZE
OF
ROUN
NETW
ORK
ATTACK LE implementation of algorithm.
KEYS
DS

6.0 REFERENCES
10,12,1 Non- 2128,2192, [1]. Vanya Diwan, Shubhra Malhotra, Rachna Jain, “Cloud
AES 128 128,192,256 BruteForce
4 Feistal 2256
Security Solutions: Comparison among Various
Theoretical 2168 Cryptographic Algorithms”,IJARCSSE , April 2014.
168,112 or
3DES 64 48 Feistal meet-in-the 2112
56
middle attack 256 [2]. [Gartner Inc, “Gartner identifies the Top 10 strategic
Lai- technologies for 2011”. Online Available:
Narrow-
IDEA 64 128 8.5 massey 2128 http://www.gartner.com.
biclique
scheme
32,64,12
[3]. David G Rosado2, Eduardo Fernández-Medina2 and
1-255
8
(64
0-2040
(128
(12
Feistal Differential 2128
Eduardo B Fernandez, “An analysis of security issues
RC-5
suggets
suggeste suggested)
ed) for cloud computing Keiko Hashizume1”.
d)
No attack is [4]. .M. Vijayapriya, “SECURITY ALGORITHM IN
Variable
BLOWF known but
ISH
64 length 16 Feistal
suffering from
232,2448 CLOUD COMPUTING: OVERVIEW”, International
(32-448)
weak key Journal of Computer Science & Engineering Technology
ECC have advantage over RSA i.e.: requirement of less (IJCSET),2013.
memory and computation time. Advantages of ECC compared [5]. "4 Cloud Computing Security Policies You Must
to RSA increases abruptly because of length of the key. While Know". Cloud Computing Sec. 2011.
RSA need to double its key size, ECC requires few Bit to
[6]. Gartner, “Seven cloud-computing security risks".
obtain the same level of security. The RSA currently changes
[7]. Cloud Security Alliance. 2011, "Security Guidance for
its key size to 2048 Bit and ECC only need to increase its key
Critical Areas of Focus in Cloud Computing”.
size to 192 Bit.ECC devices occupies less storage, less power,
[8]. "Cloud Security Front and Center". Forrester Research.
less memory, and less bandwidth in compare to other systems.
2009-11-18.
Thus ECC has computational advantages with shorter key size
[9]. Hashizume , “An analysis of security issues for cloud
than SA.RSA is most widely used algorithm for encryption and
computing”, Journal of Internet Services and
key exchange. ElGamal is extended and updated version of
DH. Applications 2013.
[10]. Rashmi, “A Survey of Cryptographic Algorithms for
5. 0 CONCLUSION Cloud Computing”. International Journal of Emerging
Cloud computing is an innovative computing trend and many Technologies in Computational and Applied Sciences
organizations and business are shifting towards the cloud but (IJETCAS),2013.
there are certain barrier to adopt the services. The major reason [11]. EloffM.M, Smith E., “The management of security in
to avert the usage of cloud is security. There are many Cloud computing”, Univ. of South Africa, Pretoria,
cryptographic algorithms that can be deployed over the cloud South Africa,2013.
to provide the security. [12]. Maulik P. Chaudhari and Sanjay R. Patel, “A Survey on
DH and ElGamal accept the variants of elliptic curve.RSA is Cryptography Algorithms”, IJARCSMS, 2014.
faster in encryption and slower in decryption to ElGmal and
half of DH. Hence RSA is efficient among all other asymmetric
algorithms.
RSA and Diffie-Hellman Key Exchange both are asymmetric
algorithms. In cloud computing both RSA and Diffie-Hellman
Key Exchange algorithms generate encryption keys for
symmetric algorithms.
DES and AES are frequently used symmetric algorithms.DES
algorithm is easy to implement then AES. In terms of Security
of data, Flexibility, Memory usage, and performance AES
(Rijndael) is best among all.

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 814
 BIJIT - BVICAM’s International Journal of Information Technology
Bharati Vidyapeeth’s Institute of Computer Applications and Management (BVICAM), New Delhi (INDIA)

Comparative Study of Cross Browser Compatibility as Design Issue in Various

Websites
Jatinder Manhas
Submitted in February, 2014; Accepted in November, 2014
Abstract - In this current era of information technology of the websites available these days, but the most widely
websites are very important means of communication. Lot of accessible categories of the websites include social
efforts is required by different institutions / organizations to networking, Commercial, Government, Educational and job
portray complete information on beautifully designed portals. With the advancement in technologies and
websites. Websites act as an online agent through which a completions of providing all services to the citizen online, it
user can get his work done without physically visiting the has become mandatory as well as very important for different
organizations. Website design is given with a very critical organizations (big or small) to develop website for rendering
look by the designer so that it can provide users with all the all its services online to the user at global level through
facilities of the concerned institutions / organizations online. networks. In order to provide better services online the
To make websites behavior similar in all the different websites are increasing in number in a very short span of time.
browsers employed by the different categories of the users, With increase in the number of websites its popularity and
the responsibility of the designer and the concerned cater to the need of all the different categories of the users it
institutions / organizations increases manifold. In this becomes mandatory for different organizations to put lot of
research paper author developed an online tool using .NET efforts to design websites carefully so that it can easily be
Framework using C# to study cross browser compatibility as accessible for different services. Number of different websites
Design issue in various categories of the websites like Job design issues is there which needs to be kept into designers
portals, Government, educational, Commercial and Social consideration while designing websites. Browser compatibility
networking. The automated tool developed by author is one of the most important issues among all websites design
function on the basis of the different standards prescribed in parameters. These days multiple browsers are being practiced
W3C guidelines document UAAG 2.0 [7] and act like a by the different users for accessing websites. With the advent
parser and renders the complete code of the website and of multiple browsers it becomes mandatory for a website to
produces result on basis of the behavior of the websites in behave similar when open in multiple browsers. There are
five most popular and widely used Browsers like parameters different organizations which are responsible for development
like Internet Explorer[7,8,9], Chrome, Safari, Fire fox. Each and recommendations of different website design standards.
Browser is tested on the basis of the five parameters which For evaluating website design as per the different standards so
are included in the parser are Blinking, Active X control, that it behave similar in all the different browsers available,
Website Resolution; image Formats, HTML Tag errors. The the number of different automated tools are available online.
results obtained after testing five different categories of Each available online tool test the websites on its own criteria
websites shows that educational and social networking sites and supply the result accordingly. In order to test the websites
shows least compatibility in multiple browsers where as job design (Browser compatibility) w.r.t. the standards
portals, commercial and government websites shows 100% recommended by W3C, the automated tool is design and
compliance to the website design standards recommended by developed by taking into consideration the parameters which
W3C w.r.t browser compatibility of different websites on are not tested by any automated tool available online till date.
different browsing platform. The automated tool, developed shall take website URL as
input and then parse the website complete code supplied by
Key words: Websites, Browser, Compatibility, Design, .NE, the server after submitting the desired request through its
Automated Tool. interface. The HTML code supplied by the server will be
compared with the existing standards recommended by the
1.0 INTRODUCTION W3C and embedded in the automated tool for its compliance.
Website is a combination of interlinked and related web pages The working of the automated tool design and developed by
residing on a single server and severed to the user through the author is given below in Figure 1.The parser of the
single domain. There are different categories and wide variety automated tool will take website URL as input and send same
to the server as request for fetching the websites HTML code.
Assistant Professor, Department of Computer Science, The server within no time sends back the HTML code of the
Bhaderwah Campus, University of Jammu, website to the interface of the automated tool for its
Jammu and Kashmir, INDIA. comparison with the existing standards. The algorithm of the
E-mail: [email protected] online automated tool developed by the author is given below:
Algorithm 1: To determine browser compatibility of a

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 815
 BIJIT - BVICAM’s International Journal of Information Technology

website. 2.0 LITERATURE SURVEY

Input: Website URL All though many people have worked in area of website
Output: Compatibility status of website w.r.t different design and development but still the desired qualities of a
features for various browsers. website has not been achieved. Still websites fails to deliver
the desired goals as is evident from the literature available. A
details review of literature regarding the concerned topic of
research has been done through different sources of Literature
like journals, research papers, books and article with Internet
User/Designer Interface
as the major source of this literature. Most designers feel that
they should be supporting all or at least as many browsers as
they possibly can. But honestly, this is impossible. If there are
5 major browsers, IE has over 7 versions, Fire fox has over 3,
Opera has over 9, Safari over 3, and now Chrome has come
out. So even if you try to support just the 2 latest versions of
Browser Compatibility* each common browser you're looking at 10 browsers, 8 of
(Checks compatibility on the basis of which you'll have to test on both Macintosh and Windows - so
supporting capabilities of various that's 18 tests for every page. And that's not even close to all,
features/components which are or even most, of the browsers available. [1] When there are
determined by the parser for different nearly 100 web browsers available today, [2] each browser
browsers) parses a web code in a different way. It becomes very difficult
HTML Parser for a designer to design websites which shows similar
(checks html tags for
various components of behavior in each browser employed by the user. . If we are
webpage)
using java scripts or AJAX calls for UI functionality,
performing security checks or validations then give more
stress on browser compatibility testing of our web application
HTML of the
website
[3]. The components of web pages like blinking, active X
controls, resolution, image formats and HTML tag error (like
Server video tag), Ajax, Flash, and event-handling for dynamic
HTML [4] etc. have a direct effect on the cross browser
Figure 1: Block diagram of Browser Compatibility Parameter compatibility of a website. These features are supported to
different levels by various types of popularly known web
Method: browsers or their versions. E.g. the compatibility of blink tag
Begin by various browsers [5] is shown in table 1 below.
Step I: Generates request for obtaining HTML file of the Hence to avoid loss of business and reputation it is very
website by passing url to the server. important to pay attention to cross browser issues [6]. W3C
Step II: For each feature considered for determining guidelines document UAAG 2.0 [7] states that web sites must
compatibility follow the maximum standards to make website behavior
Flag=0; similar in all browsing platforms. There are other categories of
features that hinder the compatibility of websites over the
(i) Check for the presence of the feature in the website by commonly employed browsers which include Different
a look out of the HTML file. Browsers, Different Browser Versions, Different Computer
Types, Different Screen Sizes, Different Font Sizes, and
(ii) Determine the compatibility of the feature w.r.t HTML Errors, Browser Bug, Add-ons and Third Party entities
different browsers. [8]. In near future the different government organizations are
If (feature isn’t supported by any browser) going to provide all facilities through web and they are
available once the user visited there websites through its
Set flag=1; browser [9]. Web based applications are extremely helpful in
enhancing the success rate of any organization.
Step IV: if (flag==0) The sustenance of High level Process maturity can be
Website is compatible for all browsers achieved by adopting web application technology [10]. As
else already discussed the technology also plays vital role on
Website isn’t fully compatible. making websites compatible with different browsing
End platforms. Government organizations are putting lots of efforts
to provide different services in the Directorate General of
Foreign Trade (DGFT): [11] e-licensing, e-BRC, e-tendering,

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 816
Comparative Study of Cross Browser Compatibility as Design Issue in Various Websites

e-monitoring, e-meeting e-delivery, e-PRC, e-grievance levels of compatibility conflicts to various levels when
re-addressal etc. regarding the implementation of ICT websites are opened on distinct web browsers or their
and different user’s access information from different versions. These features are the basis for categorizing websites
browsing platforms. The continuous evolution of the Internet for their cross browser compatibility capabilities.
has opened unimaginable opportunities and challenges in web
based education and learning [12]. Each and every person 3.2 Online tool for testing web pages
these days is getting all his work done by the use of web based The automated tool designed and developed by the author for
system which is accessible through one oir the other browsing testing various categories of the websites in order to study the
platforms. Only Workplace (office, school, college etc) was deviation from design standards by the designer includes five
different parameters. For testing browser compatibility
found to be the most common place for internet access. Most
parameter the features (or measures) which are considered by
offices and schools provide internet (broadband) access;
automated tool in determining the compatibility of a website
therefore people indulge in internet activities at work, only
on various commonly used browsers like Internet Explorer,
22% of the respondents did not have internet facility at work Google Chrome etc. are:
[13]. So attention needs to be paid to the design of the 1) Blinking
websites in order to make uniform behaviors of all the 2) ActiveX Controls
websites in different widely used browsers. 3) Website Resolution
4) Image Formats
S. No Web Browsers Compatibility for Blink 5) HTML Tag Error
Tag The test for various websites is carried out on the basis of the
1. Internet Not Supported above features to determine compatibility conflicts. This
Explorer[7,8,9] online web diagnostic tool provides a complete statistics of the
2. Chrome Not Supported browser compatibility based on the features which are
3. Safari Not Supported considered for the purpose.
4. Fire fox Supported
3.3 Sample Data
Table 1: Browsers compatibility with blinking tags The Sample of data undertaken to study Browser compatibility
as design and development issue in various websites is given
below in table 2. the five different categories of the websites
3.0 METHODOLOGY
3.1 Problem Identification which are widely used and accessed by the users were selected
A good web design aims to give a uniform look to the website, which includes government websites, commercial websites,
viewed from any web browser. Thus, a good website should job portals, social networking and educational websites. The
be viewable in its full functionality on any web browser. different category of the websites undertaken by the author
Since every webpage is built up of various components with helps the designer to understand about diversity in website
their own characteristics and these characteristics affect the design in a better and efficient way. Total 100 websites
performance of a webpage in different contexts. Like other belonging to different categories was taken for testing which
measures of performance evaluation the browser compatibility includes 20 websites from each categories mentioned above.
feature of websites is also affected by the various components The different websites are designed in consideration with
of a webpage either directly or indirectly. It is also dependent different design constraints and they are applying specifically
on the type of technology used in building websites like AJAX to each category undertaken for testing. Moreover, it helps a
as already discussed in literature review. Different lot to study diverse nature of the website design as far as
technologies sometimes also create the compatibility problem. browser compatibility on different browsing platform is
So during the design phase of the websites they must be tested concerned.
rigorously for its compatibility at different browsing platforms
As already discussed in literature review, different 3.4 Results and Discussions
components of web pages like blinking, active X controls, The Sample data undertaken to study cross browser
resolution, image formats and HTML tag error (like video compatibility as website design issue in various websites is
tag), Ajax, Flash, and event-handling for dynamic HTML etc. shown in below given Table 2 and the results obtained are
have a direct effect on the cross browser compatibility of a shown by framing the analysis into a graph shown in below
website. These features are supported to different levels by given figure A.
various types of popularly known web browsers or their Based on the graph shown in fig.(A) above, it can be analyzed
versions. The compatibility of blink tag by various popularly that the website categories that have maximum support in
known browsers is already in table discussed in literature percentage for various types of commonly used browsers
review. belong to e-government websites, e-commerce websites and
Similarly, the other features described above offer different job portals and the websites which shows least compliance to
the different browsing platforms belongs to social networking

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 817
 BIJIT - BVICAM’s International Journal of Information Technology

websites and educational websites. b) Different Browser Versions

c) Different Computer Types
100 d) Different Screen Sizes
90
80 Education sites e) Different Font Sizes
70
60 f) HTML Errors
50
40 g) Browser Bug
30 h) Add-ons and Third Party entities
20
10 Job Sites
0
6.0 FUTURE SCOPE
Browser In future more number of measures will be added to browser
Compatibility(in %) compatibility parameter to help the designers to create page
that are more efficient, user centric and follow the standards
Figure 2: %age for Browser Compatibility provided by concerned organizations in a more appropriate
It has been observed from the results obtained that out of the way. The parameters that are under consideration to be
five different categories of the websites taken as a sample data included in the parameter belong to categories like diversity in
for testing the e-governance, e-commerce and job portals web browser versions, other HTML tag errors, screen sizes
follows the maximum of the standards recommended for the etc. All these features are under consideration and will be
websites design w.r.t. behavior of the websites in different included in the tool to enhance its capability to check websites
browsing platforms. for more number of design issues regarding cross Browser
compatibility problem.
4.0 CONCLUSION
Designers cannot assume that their application will run fine 7.0 REFERENCES
and display and work for all the browsers without cross [1]. Ralph Moseley and M.T. Savaliya- Developing Web
browser testing. The organizations must put lots of efforts in Applications, Wiley-India, First Edition: 2011 ISBN:
designing websites in order to avoid loss of business and 978-81-265-2960-5. www.wileyindia.com.
reputation. The analysis carried out on the cross browser [2]. Wikipedia. List of web browsers.
compatibility feature of websites above using the online web http: //en.wikipedia.org/wiki/List_of_web_browsers.
diagnostic tool Website Design Evaluator clearly indicates that [3]. Sanjay Dahiya, Ved Parkash, T.R. Mudgal,
websites are neglecting criteria that otherwise must be “Comprehensive Approach for Cross Compatibility
considered for complete cross browser compatibility in spite Testing of Website”, International Journal of Computer
of the web browser employed. Applications (IJCA), 2012.
By evaluating the different categories of the websites on the [4]. Ali Mesbah, Mukul R. Prasad, “Automated Cross-
automated tool design and developed by the author it has been Browser Compatibility Testing” Electrical and
observed that websites are not meeting the different standards Computer Engineering, University of British Columbia
recommended by the different standard organization (e.g. Vancouver, BC, Canada Trusted Systems Innovation
W3C) for the design and development of various websites Group, Fujitsu Laboratories of America Sunnyvale,
such as blink tag for IE [7,8,9] and Mozilla, video tag for IE 8 and CA, USA.
earlier versions etc. The result of this study confirms that the [5]. http://stackoverflow.com/questions/10248923/how-can-
developers of the websites somehow are not able to i-detect-if-a-browser-supports-the-blink-tag
completely follow the standards recommended. It is clear from [6]. Ochin, Jugnu Gaur, “Cross Browser Incompatibility:
the results that more efforts are required to follow websites Reasons and Solutions” International Journal of
design standards in order to make websites 100% compatible Software Engineering & Applications (IJSEA), Vol.2,
in different browsing platforms. No.3, July 2011.
[7]. http://en.wikipedia.org/wiki/Web_standards
5.0 LIMITATIONS [8]. http://www.netmechanic.com/products/Browser-
Appendixes, In addition to the features considered for Tutorial.shtml
compatibility test, there are also other features for determining [9]. V. S. Rana, “E-Licensing in DGFT “A Best E-
browser compatibility that can be included in order to widen Governance Application” BIJIT - BVICAM’s
the cross compatibility check to better help International Journal of Information Technology
developers/designers to build websites that have increased Bharati Vidyapeeth’s Institute of Computer
compatibility capabilities. The features considered are not Applications and Management (BVICAM), New Delhi
quite enough to fully determine the compatibility capabilities (INDIA) Copy Right © BIJIT – 2013; January – June,
of a website. There are other categories of features that hinder 2013; Vol. 5 No. 1; ISSN 0973 – 5658 545.
the compatibility of websites over the commonly employed [10]. A. V. Sharmila, “Web Based Technologies to Support
browsers e.g. High Level Process Maturity” BIJIT - BVICAM’s
a) Different Browsers International Journal of Information Technology

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 818
Comparative Study of Cross Browser Compatibility as Design Issue in Various Websites

Bharati Vidyapeeth’s Institute of Computer BIJIT – 2011; July – December, 2011; Vol. 3 No. 2;
Applications and Management (BVICAM), New Delhi ISSN 0973
(INDIA) Copy Right © BIJIT – 2013; July-December,
2013; Vol. 5 No. 2; ISSN 0973 – 5658 589. Jatinder Manhas. The author’s educational
[11]. V. S. Rana, “An Innovative use of Information & background includes masters in computer
Communication Technology (ICT) in Trade Facilitation application from the university of Jammu,
in India” BIJIT - BVICAM’s International Journal of Jammu (J&K), India in the year 2004. The
Information Technology Bharati Vidyapeeth’s Institute author also qualified the state level
of Computer Applications and Management eligibility test for lectureship in the year
(BVICAM), New Delhi, Copy Right © BIJIT – 2012; 2005 conducted by UGC. The Authors is
July - December, 2012; Vol. 4 No. 2; ISSN 0973 – likely to obtain PhD degree in the field of
5658 492. computer sciences and IT from university of Jammu (J&K).
[12]. Rajiv and Manohar Lal , “Web 3.0 in Education & The author’s major field of study is website Design and
Research” BIJIT - BVICAM’s International Journal of Development issues. He has represented 10 research papers in
Information Technology Bharati Vidyapeeth’s Institute different national and international conferences within India
of Computer Applications and Management and abroad. He has also published 8 research papers in
(BVICAM), New Delhi, Copy Right © BIJIT – 2011; different peer reviewed and indexed journals of international
July – December, 2011; Vol. 3 No. 2; ISSN 0973 – repute. The author is working as Sr. Assistant Professor in the
5658 335. Department of Computer Sciences & IT, University of Jammu
[13]. Preeti Dubey , Jeevan Jyoti and Devanand, “A Study to (J&K), India since 2006. The author was also awarded with
Examine the Digital Divide Factors: Jammu and the UGC teacher fellowship under the faculty improvement
Kashmir Perspective” BIJIT - BVICAM’s International programme to complete his PhD in the year 2012. The author
Journal of Information Technology Bharati is also an acting and life member of Computer society of India
Vidyapeeth’s Institute of Computer Applications and and IACSIT (International Association of Computer Science
Management (BVICAM), New Delhi Copy Right © & Information Technology).

Table 2: Sample Data

S.N Govt. Educational Commercial Social Networking Job Portal Websites
o Websites Websites Websites Websites
1. www.mit.gov.in www.upsc.gov.in www.bsnl.co.in www.facebook.com www.monster.com

2. www.indianrail.go www.shiksha.com www.onlinesbi.com www.gmail.com www.careerbuilder.c

v.in o.in
3. www.moia.gov.in www.apple.com www.airtel.in www.orkut.com www.indeed.co.in

4. www.rajasthan.go www.schoolcircle.c www.india.philips.com www.yahoo.com www.simplyhired.co.

v.in om in
5. www.jkgad.nic.in www.classteacher.c www.libertyshoes.com www.rediffmail.co www.nakuri.com
om m
6. www.maharashtra. www.jammuuniver www.bata.in www.frenzo.com www.glassdoor.com
gov.in sity.in
7. www.assam.gov.i www.kashmirunive www.hp.com www.bebo.com www.usajobs.gov
n rsity.net
8. www.india.gov.in www.coeju.com www.tata.in www.goibibo.com www.reed.co.uk

9. www.aponline.go www.kashmirunive www.godrej.com www.jakpinch.com www.monesterindia.c

v.in rsity.ac.in om
10. www.nagaland.nic www.jmi.ac.in www.hul.co.in www.myspace.com www.placementindia.
.in com
11. www.upsc.gov.in www.jamiahamdar www.marutisuzuki.co www.mycantos.co www.timesjobs.com
d.edu m m
12. www.petroleum.ni www.smvdu.net.in www.britannia.co.in www.friendster.co www.beyond.com
c.in m
13. www.rbi.org.in www.islamicuniver www.pg.com www.viber.com www.findtherightjob.
sity.edu.in com

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 819
 BIJIT - BVICAM’s International Journal of Information Technology

14. www.dotindia.co www.du.ac.in www.colgatepalmolive. www.netlog.com www.bright.com

m com
15. www.coal.nic.in www.jnuonline.in www.hindwarehomes.c www.flickr.com www.ziprecruiter.co
om m
16. www.tn.gov.in www.amu.ac.in www.pepejeans.com www.watsapp.com www.theladders.com

17. www.tourismofind www.skuastkashmi www.ashokleyland.co www.plus.google.c www.monester.co.uk

ia.com r.ac.in m om
18. www.upgov.nic.in www.skuast.org www.dabur.com www.tagged.com www.career.com

19. www.darpg.gov.in www.pondiuni.edu. www.pepsico.com www.ning.com www.tweetmyjobs.co

in m
20. web.guidelines.go www.bujhansi.org www.coca-cola.com www.pinterest.com www.smartbrief.com
v.in

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 820
 BIJIT - BVICAM’s International Journal of Information Technology
Bharati Vidyapeeth’s Institute of Computer Applications and Management (BVICAM), New Delhi (INDIA)

Framework for Choosing Best Intrusion Detection System

Bilal Maqbool Beigh

Submitted in February, 2014; Accepted in November, 2014

Abstract - As there are many intrusion detection systems The Computer Emergency Response Team (CERT) reported
available in the market and yet there is not a single guideline 3734 incidents in 1998, 9859 in 1999 and 8836 in the first 6
framed by any researcher or any organization so that a months of 2000. In a recent audit of U.S. federal agencies by
company or an organization will decide which intrusion the GAO [4] investigators were able to pierce security at nearly
detection system is best suited to their company for the every system they tested. The cause of these attacks are either
purpose of security. Here in this paper, we have proposed a complexity of the system itself or increasing number of hackers
novel guideline in terms of framework for choosing right day by day or market competitors or software development
most intrusion detection system for an organization. The companies itself etc. Therefore along with these tremendous
framework needs some security expert so that they can check opportunities for sharing important information and resource
the equation to be satisfied. especially used for some critical operation like military , space,
nuclear etc. It has become very much important to protect these
Index Terms – Framework, Guideline, model, IDS, Intrusion. special and important resources and information against such
attacks [5]. For protecting the same, we have the concept called
1.0 INTRODUCTION “Information security” thus we can say that information
The story of the human life started with Stone Age, then security is such area which protects our information / resources
agriculture age and now we are in the information technology from theft or misuse. But still this field of research is in its
age, where everything depends upon information and infancy days. This research started in early 90’s and so far little
information processing systems. Information ranging from has been done in this field. This research field comprises of
personnel to commercial have been processed and exchanged many subfield such as system side security, network side
by these information systems. With the advent of Internet, the security etc. One subset of information security that has been
convergence of information & communication technologies the area of much more attention in recent years is intrusion
and today’s very complex nature of business environment detection system [5]. Therefore intrusion detection system can
resulted in myriad trust and information security concerns. The be defined as the process of monitoring events occurring in a
secure functioning of these information systems is the utmost system and signaling responsible parties when interesting
important and foremost concern. Information security is a field (suspicious) activity (compromises the confidentiality,
of security which ensures the confidentiality, integrity and integrity, availability, or to bypass the security mechanisms of
availability of information and information processing a computer or network) occurs [7]. At this instant of time, there
resources. Many security professionals think that developing a are many intrusions detection systems available in market with
completely secure system is almost an impossible task. different features and uses, but it is very difficult for a user or
According to [1] the completely secure system is one that is organization to choose best Intrusion detection system for him
disconnected from a network, encased in concrete, and lying at or for his organization [8][14]. As there is no such guideline
the bottom of the ocean. In this networked environment where provided by any agency/ organization to choose the security
there are potential number of hackers and adversaries present, policy therefore there is a need of guidelines for the purpose.
security enforcing mechanisms needs to be incorporated in the Here In this research work, we will provide a framework in
information systems to with stand with the both deliberate and terms of mathematical equations and steps for choosing best
accidental malicious intents. Hence this tremendous growth in possible intrusion detection system for you and your
communication technology brings number of good things to organization. This part will ensure that the system for intrusion
human society, but it also makes us re-lay on information detection should be made in accordance to the model prepared
systems [2].As the information is increasing in digital format in terms of equations and physical model to be described in the
day by day, the vulnerabilities are also increasing in the form of next sections of this paper.
cyber threats, attacks and mis-identification of trusted users.
There are lots of intrusion attacks in today’s digital world, 2.0 NEED FOR FRAMEWORK FOR CHOOSING IDS
according to recent survey by CERT/CC [3], the rate of The intrusion detection system allows us to make the system
intrusion attacks almost doubles every year. safe from the most attackers. Thus as described in previous and
this chapter, intrusion can be defined as a process of accessing
someone’s personal property or data or information without
P. G. Department of Computer Sciences, University of proper access or proper authentication cardinalities. As all of us
Kashmir, Jammu and Kashmir, INDIA know that today’s almost 90 percent of information is available
E-mail: [email protected] online through websites or computer programs. Although this
make very ease and very fast access to the people overall the

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 821
 BIJIT -BVICAM’s International Journal of Information Technology

globe, but it also increases the risk to the maximum. According choose the best intrusion detection. The framework consists of
to Symantec report, around 1, 00,000 websites are available logical steps and which when followed revealed the desires
online and some of them share the critical information and intrusion detection system. The steps which are involved in
valuable data. In order to steal the critical data or important and choosing best intrusion detection are as:
relevant information without having legitimate access to the • Risk Analysis.
resources, the person on longer need not to be a hacking gem, • Detection Rate
just download and run the hacking program, make some • False Alarm Rate
settings and you are done [9][16]. In order to secure the • Cost Benefit analysis
companies or individual’s data/ information, firewalls are being • Updates or patches ratio.
installed, but they alone do not serve the purpose of defending When the above mentioned steps are followed in a manner
the data from attacks or intruders. The main aim of the firewall shown in figure below (cc), yield the results based on the
is to filter the traffic but they cannot block all the traffic. Also mathematical formulas. The accepted criteria must be decided
once the traffic passed through the firewall there is no such by the security professionals to choose best intrusion detection
mechanism available that traffic will be monitored inside the system.
network for rest processing. Also firewall only detects external
traffic coming to it, but doesn’t detect the internal attacks. By Identify Risk
using intrusion detection system, we can monitor or do the
following things:
• Monitors network traffic. Pick IDS
• Continuously monitors servers/ network for misuse actions
or abuse policy.
• Attack / breach alerting, response and reporting. Detection Rate
• Countermeasures.
Thus it became very much important for an organization to
install both firewall and intrusion detection system to secure True Alarm
their assets / information for hackers / attackers. Also for
securing this particular data and information from the attackers,
there are lots of intrusion detection mechanisms currently Select another Cost Benefit
available in the market. Every intrusion detection Choice of
manufacturing organization highlights his qualities of making IDS
secure the information, but none of the organization or research Updates Available
community describes the guidelines for picking the most
appropriate and good suitable intrusion detection system for
any company or individual. Thus it became very much
important to provide some guidelines either through some
model or through some mathematical formula to suggest the If
company which intrusion detection system is most suited for Satisfied Yours
them and under which norms. We have taken a step towards the
development of guidelines for choosing the right intrusion
detection in accordance to their requirements and importance’s. Figure 1: Frame Work for choosing right intrusion
In this chapter we have presented guidelines for choosing right detection system
most intrusion detection system for company or individual
under required condition. The guidelines will be discussed in 3.1 Risk Analysis
next section of this chapter. The Risk analysis step is the first step towards the choosing of
intrusion detection and prevention system for an individual or
3.0 FRAME WORK FOR CHOOSING INTRUSION an organization. This step is most important and critical
DETECTION SYSTEM towards the picking of right most intrusion detection system.
Choosing an intrusion detection system is a delicate task, as the The risk calculation is very big thing to do because it deals
whole company security responsibility lies on the shoulders of with the overall security of the organization. The risk analysis
the intrusion detection system i.e.to detect the attack made on can be considered as a tool for risk management, which is
the organization system, to mitigate them if possible or to alert helpful for identifying security issues i.e. vulnerabilities,
the administrator about the attack happened [4]. Currently there threats and unauthorized access. Also as per the general
are many intrusion detection systems available within the definition of risk on different blogs and websites, the risk can
market but it is difficult to choose the best intrusion detection be calculated as:
system for an organization[17]. In order to choose the same, we “Risk = Threats x Vulnerabilities x Impact”
have devised a framework that will help an organization to

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 822
Framework for Choosing Best Intrusion Detection System

But we have devised the risk in accordance to our own IDs etc. Before going for any products, the company should
formulas. The formula will depend upon the following factors. consider all the available resources for basic system operation
i. Summation of threats. and maintenance. Thus should be able to pick the appropriate
ii. Value / impact of threats IDS which will meet the needs within the constraints laid down
iii. Total Impact of assets under risk by company. This task is very difficult, As there is no industry
iv. Total assets of the organization. standard against which we will compare IDS. Hence there is a
Before enforcing the formula, The RAG (Risk Analysis Group) need of providing a standard benchmark for IDS. The new
will find two important aspects used in making the formula product cycle for commercial IDSs is rapid, and information
which are [10] [11] [12] [13]: and systems quickly become obsolete. Steven Northcutt
• Identifying important information and their Values recommends the use of product guides that are updated at least
• Identifying threats and Vulnerabilities for the assets monthly. Relatively little objective third party evaluation of
Identifying important information and their Values – IDSs is available, while trade press reports are generally spotty
Identifying the values of the organization’s important and superficial. Setting up a facility to objectively compare
information is the very first step for risk analysis. In this step IDSs will be prohibitively expensive for all but the largest
the risk assessment group will point out / identify the most potential users, and some third-party or industry sponsored
important assets of organization and will estimate the cost effort is needed. Marketing literature rarely describes how well
associated and damage resulted if some intrusion/ attack a given IDS finds intruders and how much work is required to
happed on an organization or we can say the group will analyze use and maintain that system in a fully functioning network
the loss made by losing the information to some other with significant daily traffic. IDS vendors usually specify
company. While identifying the assets following things must be which prototypical attacks their systems can find, but without
kept in consideration. access to deployment environments, they cannot describe how
• Cost of assets/ information that may be lost if intrusion well their systems detect real attacks while avoiding false
happen. alarms. Edward Amoroso and Richard Kwapniewski recently
• Role and usage of assets / information. provided guidance in selecting IDS [14] by making some
questionnaires, upon the receiving the answers from the users,
Identifying threats and Vulnerabilities –After pointing out they will choose the intrusion detection system. These
the important information/assets, the responsibility of the group guidelines have impact of bias towards a particular intrusion
is to identify the vulnerabilities and threats for assets/ important detection system. This step is very important, as it is concern
information as identified in the prior step. Also they have to with the security of overall system. The step choice based and
keep an order of threats i.e. which threat may damage/ theft will not be entertained in the conditions which are going to
more information according to the percentage of damage done decide that is the picked intrusion detection system the right
by these threats and vulnerabilities. Thus in general, the RAG most intrusion detection for their organization. The decision of
will gather the information about the loss of assets / this step lies on the shoulders of the Risk analysis group. They
information in total at the initial stage, if not prevented and are the security professionals which will decide the most suited
total threats and vulnerabilities that can cause these losses. as per the threats and vulnerability of the organizations.
After acquiring the above two steps, we have derived a formula
which we are going to use to calculate the risk is as under: 3.3 Detection Rate
The step is very much important as the decision is concerned.
The detection rate for a particular intrusion detection system
will be available in the literature and papers available in
different research journals. Let us make an example, in one of
our experiment, the snort has detection rate of 99.4 % which
means that 99.4% of attacks are being detected by the snort
coming towards the system. Similarly all intrusion detection
By calculating the estimated risk, we can have idea about how has the documentation, which shows the rate of detection for
much it will affect our assets. Therefore the overall percentage that intrusion detection system. Rate of detection can be
of the risk can be calculated with the help of the following calculated as:
formula.

After getting the results of assets under risk in percentage, we

will move towards the next step of the frame work. Or we can say that the rate of detection can be calculated as:
3.2 Picking of IDS
Every organization wants to secure their confidential resources,
for that they have to make some selection in terms of firewalls,

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 823
 BIJIT -BVICAM’s International Journal of Information Technology

TP = amount of attack when it actually attack Before a company or an organization decides on exactly which
TN = amount of normal detect when it actually normal IDS that organization or company should opt, it is very
Also rate of detection can be calculated using the above important to perform cost/ benefit analysis. As it is very
formula. The detection rate is very much important as per obvious and important that cost/benefit analysis is very real and
selection is concerned. It shows the overall rate. If the detection important factor in decision making of all the process related to
rate is greater than 90 %, the system is partially accepted. an organization. There funds allocated to the security or other
Which means the system will be evaluated for the next section. solutions have to have a good reason why such funds are
i.e. True alarm Rate . allocated to the said solution. This analysis can be performed
effectively once the organizations risk analysis has been
3.4 True Alarm Rate performed. This risk analysis will give the organization a very
The step is much important as per the decision is concerned. As real sense of the costs associated company assets. The
we are calculating the rate of false positive alarm rate which estimated cost/ benefit of the company can be evaluated with
can be calculated as the ratio of in-correct classified intrusions the following formulas as shown under:
to the total number of normal records. Therefore false positive
rate (FPR) can be calculated as:

But we have to calculate the True Positive alarm rate which can
be derived as from the above formula, which can be derived as
below:
True Positive Rate = 100- false Positive Rate
Let us assume we got the false positive rate as 3.06% , then we
can easily get the true positive rate as 96.77% which means that
the system is accurately identifying 96.77 % of the intrusions
that of total available in the dataset. Upon partial accepted from The equation of cost has been designed to evaluate the total
previous sections. If the true alarm rate is less than 95 %, it is cost of the security solution for an organization. The cost
partially accepted. consists of all the man power which can be used to incorporate
the security solution in accordance to the requirement of the
3.5 Cost Benefit Analysis. organization for the purpose of securing the critical data. Also
There is a variety of approaches to cost analysis, the suitability the benefit is as important as determine the cost of security
of any of which depends upon the purpose of an assessment solution. The benefit will give us figures that whether the
and the availability of data and other resources. It is rarely solution will be beneficial to the company. It will give us the
possible or necessary to identify and quantify all costs and all impact of the benefits using the big budget for the security
benefits (or outcomes), and the units used to quantify these solution. The formula helps us to estimate the benefits from the
may differ. security solution, which is usually the cost of assets which are
Main types of cost analysis include the following. currently under threat and future assets. The formulas for
• Cost-of-illness analysis: a determination of the economic benefits are as under:
impact of an illness or condition (typically on a given
population, region, or country) e.g., of smoking, arthritis or
bedsores, including associated treatment costs
• Cost-minimization analysis: a determination of the least
costly among alternative interventions that are assumed to
produce equivalent outcomes. Where n is the number of Assets
• Cost-effectiveness analysis (CEA): a comparison of costs The Net Cost-Benefit will be retrieved from the difference of
in monetary units with outcomes in quantitative non- cost from benefits. The Net-cost benefit analysis will be
monetary units, e.g., reduced mortality or morbidity. derived as under:
• Cost-utility analysis (CUA): a form of cost-effectiveness
analysis that compares costs in monetary units with
outcomes in terms of their utility, usually to the patient, This section will be critical as far as the decision will be is
measured, e.g., in QALYs. concerned. If partially accepted from last sections, if the Net-
• Cost-consequence analysis: a form of cost-effectiveness Cost Benefits is greater than Zero (>0), it is again partially
analysis that presents costs and outcomes in discrete accepted.
categories, without aggregating or weighting them. 3.6 Decision Phase
• Cost-benefit analysis (CBA): compares costs and benefits, This is one of the most important phases of our framework, if
both of which are quantified in common monetary units. the system is already partially accepted; it will go to the next

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 824
Framework for Choosing Best Intrusion Detection System

phase of detection. The final selection results will be based on Table1: Detection Rate for evaluation
the following points:
Name of IDS Benefits Name of IDS Detection Rate
Snort 62,00,000 Snort 98.3 %
Bro 94.4 %
Bro 62,00,000
NIDS 97.3%
NIDS 62,00,000
As per the statistics available in the literature, Snort has highest
1. Highest rate of detection under consideration. detection rate while on the second number NIDS is there and
2. Lowest False alarm Rate. Bro is at the third number.
3. Highest Net-Cost Benefit. False Rate: As per the literature available, we have collected
It is considered that if the one security solution has high Net the respective false rate of the all the three intrusion detection
Cost-Benefits rate and rest two options are low and other system available mentioned above. The False rate is as under:
solution have also been partially accepted but does not have Table 2: False alarm Rate for evaluation
Net Cost-Benefits but have very good statistics high in
detection rate and true positive rate, the security solutions Name of IDS False Rate
which have high detection rate and true positive rate will be Snort 2.3%
considered for selection. Bro 7.5%
NIDS 2.1%
4.0 EVALUATION OF FRAMEWORK FOR CHOOSING
INTRUSION DETECTION SYSTEM As per the statistics available in the literature, NIDS have least
The evaluation of frame work for choosing Intrusion detection false rate, on second number Snort is there and last is Bro.
system was done on the statistics provided by the Kashmir Cost-Benefit:
university IT&SS department. The department provided the The benefit of all the three is as follows:
statistics only meant for the research meant in this thesis. The
figures provided are as : Table3: Cost Benefit for evaluation
Risk Analysis: As per the departmental report, total cost of The costs of all the three are:
assets which are under risk (attacks). The risk figures are Cost of Snort = Cost of Deployment + Cost of updating +
calculated by using the above mentioned formulas. Cost Maintenance.

Cost of Snort = 1, 00,000 (purchasing of Computer) + 0 +

30,000(rule purchasing)
Risk = Cost of Results + Cost of pay generation Software +
= 1, 30,000.
Cost of E-Governance
Risk = 10,00,000 + 2,00,000 + 50,00000
Cost of Bro = Cost of Deployment
Risk = 62,00,000 ( Approx).
+ Cost of updating
Pick IDS: We have chosen three intrusion detection systems
+ Cost Maintenance (Script writing).
which are open source. The selection of intrusion detection is
based on statistics and popularity score in literature available.
Cost of Bro = 1, 00,000(purchasing of Computer) + 10,000
The intrusion detection systems are:
per Month+ 40000(Script Writing)
1. Snort.
2. Bro
= 1, 50,000.
3. NIDS.
Detection Rate: As per the literature available, we have
Cost of NIDS= Cost of Deployment + Cost of Updating
collected the detection rate of the all the three intrusion
+ Cost of Maintenance
Name of IDS Cost
Cost of NIDS = 1, 00,000(purchasing of Computer) + 50,000
Snort 1,30,000 per Month + 20,000
Bro 1,50,000 Cost of NIDS = 1, 70,000
NIDS 1,70,000 Table 4: Cost for evaluation
detection system available. The detection rate is as under:
Table5: Net Cost Benefit for evaluation
Net Cost-Benefit = Benefit – Cost

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 825
 BIJIT -BVICAM’s International Journal of Information Technology

Name of IDS Net Cost-Benefit [7]. Beigh, B. M., & Peer, M. A. (2011). Intrusion Detection
Snort 62,00,000-1,30,000 = and Prevention System: Classification and Quick.
6070000 [8]. Kovacich, G. L. (2003). The Information Systems
Bro 62,00,000- 1,50,000 = Security Officer's Guide: Establishing and managing an
6050000 information protection program. Butterworth-
NIDS 62,00,000 -1,70,000 = Heinemann.
6030000 [9]. Huang, Y. A., & Lee, W. (2003, October). A
cooperative intrusion detection system for ad hoc
Decision: networks. In Proceedings of the 1st ACM workshop on
As the Net Cost-Benefit analysis for all are almost same, Security of ad hoc and sensor networks (pp. 135-147).
therefore the deciding factor is now detection rate and false ACM.
alarm rate. As per the calculation Snort has highest detection [10]. Cavusoglu, H., Mishra, B., &Raghunathan, S. (2004). A
rate from the three and rate of false alarm rate for snort is 2.3 model for evaluating IT security investments.
and NIDS is 2.1. Therefore after seeing the results, the Communications of the ACM, 47(7), 87-92.
detection rate of Snort is high and false rate is almost same in [11]. Banerjee, U., & Arya, K. V. (2013). Optimizing
NIDS and Snort, so we choose Snort from all the three Operating Cost of an Intrusion Detection System.
intrusion detection system. International Journal of Communications, Network and
System Sciences, 6(1).
4.0 CONCLUSION [12]. Cohen, G., Meiseles, M., &Reshef, E. (2012). U.S.
The current research is focused on choosing intrusion detection Patent No. 8,099,760. Washington, DC: U.S. Patent and
and prevention system. The selection of the Intrusion detection Trademark Office
System is a very tough job. The thesis chapter provide [13]. Amoroso, E., &Kwapniewski, R. (1998, December). A
framework for choosing best intrusion detection system for an selection criteria for intrusion detection systems. In
organization. The framework is the form of flow diagram, Computer Security Applications Conference, 1998.
when followed strictly will yield a solution for choosing best Proceedings. 14th Annual (pp. 280-288). IEEE.
intrusion detection and prevention system for an organization. [14]. Chaudhary, A., V. N. Tiwari, and A. Kumar. "Analysis
The steps mentioned in framework appears to be a simple of fuzzy logic based intrusion detection systems in
exercise but are basically important/ critical steps for getting mobile adhoc networks." BIJIT – BVICAM’s
best of ID&PS for an organization . But ultimately the choice International Journal of Information Technology, 6.1
depends upon company. The researcher had made an attempt to (2014): 690-696.
provide certain guidelines in terms of frame work for choosing [15]. Beigh, Bilal Maqbool. "One-stop: A novel hybrid model
or selecting right most intrusion detection for an organization. for intrusion detection system." INDIACom - 2014,
2014 IEEE International Conference on Computing for
5.0 ACKNOWLEDGEMENT Sustainable Global Development, Bharati Vidyapeeth’s
I would like to thank Prof. S. M. K. Quadri, Head, Department Institute of Computer Applications and Management
of Computer Sciences, University of Kashmir for helping me (BVICAM). New Delhi, 2014.
throughout the course- Thank you Sir [16]. Mitra, Sulata, and Arkadeep Goswami. "Load Balancing
in Integrated MANET, WLAN and Cellular Network."
6.0 REFERENCES BIJIT – BVICAM’s International Journal of
[1]. CONNOLLY, P. J., 2001. Security protects bottom line. Information Technology, (2011): 304.
InfoWorld, Vol. 23, No. 15, p. 47
[2]. SAKURAI, K., & Kim, T. H. (2008). A Trend in IDS
researches. (Journal of Security Engineering), 5(4), 8.
[3]. Lazarevic, A., Ertoz, L., Kumar, V., Ozgur, A., &
Srivastava, J. (2003). A comparative study of anomaly
detection schemes in network intrusion detection. Proc.
SIAM.
[4]. Mathew, D. (2002). Choosing an intrusion detection
system that best suits your organization. GSEC Practical
v1. 4b, available at: www. Sans.
org/reading_room/whitepapers/detection
[5]. Brown, D. J., Suckow, B., & Wang, T. (2002). A survey
of intrusion detection systems. Department of Computer
Science, University of California, San Diego.
[6]. Grandison, T., &Terzi, E. (2009). Intrusion Detection
Technology.

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 826
 BIJIT - BVICAM’s International Journal of Information Technology
Bharati Vidyapeeth’s Institute of Computer Applications and Management (BVICAM), New Delhi (INDIA)

A Novel Pruning Approach for Association Rule Mining

Lalit Mohan Goyal1, M. M. Sufyan Beg2 and Tanvir Ahmad3

Submitted in April, 2014; Accepted in December, 2014

Abstract – The problem of Association rule mining (ARM) transactions, where each transaction is a set
can be solved by using Apriori algorithm consisting of 3-steps ofitems such that . There is a unique identifier, TID,
-Joining, Pruning and Verification. Pruning step plays an related with each transaction. We can say that a transaction
important role in eliminating weak candidate itemsets. In this contains an itemset if . A k-itemset is said to be
paper, a new pruning step is proposed as an alternate to frequent only if all its k items are in some minimum number of
Apriori’s pruning step. This alternative is depicted as a transactions. An association rule is an implication of the
filtration step. Five experiments are carried out to claim that form , where , and . X is called
proposed pruning method also works as efficient as Apriori’s antecedent and Y is called consequent; the rule means X implies
pruning method. Y. For a given set of transactions D, the problem of mining
association rules is to generate all association rules that have
Index Terms – Data mining, ARM (Association Rule certain user-defined minimum support, called minsup, and user
Mining), Apriori algorithm, pruning. defined minimum confidence, called minconf. Support of an
association rule is defined as the ratio of the number of
1.0 INTRODUCTION transactions that contain all the items of the set to the
Data mining [8] is a method of extracting non-trivial, inherent, total number of transactions in the database D, i.e.,
unfamiliar and practical information from large repositories. . Confidence of an association rule is
Association rule mining extracts frequent patterns [1], defined as the ratio of the number of transactions that contain
correlations [19], subsequences [12], substructures [11, 26, 27] all the items of the set to the total number of transactions
or associations [1] among sets of items of the databases. An that contain all the items of the set X, i.e.,
association rule is an implication , where and are sets . The problem of mining
of items. The meaning of such expression is that transactions of
association rules is a two phase process. In first phase, all sets
a database which contain likely to contain . For example-
of items which occur with a frequency greater than or equal to
95% of the students who buy a lap-top and a book related to
the given minimum support are identified and in the second
computers can also purchase a pen-drive.
phase, all sets of rules that satisfy given minimum confidence
The rules thus discovered from the databases can be used to
are generated. All sets of items which satisfy minimum support
rearrange the related items together or can be used to make new
are known as frequent itemsets and all sets of rules which
market strategies which further increase the sales. Application
satisfy minimum confidence are known as association rules.
domain of association rule mining is not only limited to the
Here, the major cost of mining association rules is contributed
context of retail application but can also be used in the decision
by first phase only. It is for the reason that most of the
logics to the medical applications [7, 13].
researchers focused their investigations on identifying frequent
This paper is organized as follows. Section 2. A describes the
itemsets.
problem statement. Section 3 summarizes the related work in
To identify frequent itemsets Agarwal and Srikant [2] proposed
this field. In section 4, a modified Apriori algorithm is
an algorithm, called Apriori. It is a three steps process: joining,
proposed with experimental results and in the last section,
pruning and verification as shown in Table 1.1. In the first step
conclusion and future directions are stated followed by section
of this algorithm, the k-frequent candidate itemsets are
of references.
generated by using joining operation which is defined on two
2.0 PROBLEM STATEMENT frequent itemsets as if and
The following is a formal statement of the association rule are two k-1 frequent itemsets and
mining problem to be solved. Let be a set then joining operation will yield the following k-frequent
containing m items. A set of items is called an itemset candidate itemset . In the second step,
and an itemset having k numbers of items is known as k- these k-frequent candidate itemsets are pruned to generate
itemset. Let be a set containing n potential k-frequent itemsets whose all subsets containing k-1
items are frequent. In the third step, potential k-frequent
2,
Department of Computer Engineering, Aligarh Muslim itemsets are verified by scanning all transactions of database D
University (AMU), Aligarh (UP), INDIA. for a given minimum support. These steps are repeated until
1, 3
Department of Computer Engineering, Jamia Millia Islamia large k-frequent itemsets are generated. Here, large k-frequent
(A Central University), New Delhi, INDIA. itemsets mean that value of k should be as large as possible.
E-mail: [email protected], Apriori algorithm uses bottom-up approach to generate k-
2
[email protected] and [email protected] frequent itemsets, so, initially, it requires 1-frequent itemsets

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 827
 BIJIT - BVICAM’s International Journal of Information Technology

which can be generated by scanning all transactions of database methodology of Srikant and Agarwal gets fail when applied to
D against all items of set N for a given minimum support. interval data where separation between data values has some
Prerequisite for the Apriori algorithm are: A set of transactions meaning [15].
(D), minimum support (minsup) and 1-frequent itemsets (F 1 ). Table 1: Apriori Algorithm

3.0 RELATED WORK Apriori (D, minsup, F1)

The very first paper which directly addresses the problem of // D is a set of Transactions, minsup is the given
association rule mining was given by Agarwal et al [1]. Their //minimum support, F1is the collection of 1-frequent
research motivates the direction of enriching the database with itemset.
more functionality to process those queries which can increase 1.
the sales of any retail market. In this paper an algorithm known //Step 1: (Joining): is the collection of k-frequent
as AIS (Agrawal, Imielinski, Swami) was proposed to answer candidate itemsets.
those queries. In AIS algorithm potential candidate itemsets are 2.
generated and verified simultaneously during database scan. 3.
Drawback of this algorithm is that all the association rules are 4.
generated with only one item in their consequent, i.e., if a rule
is generated then . 5.
Agarwal and Srikant [2] proposed a popular algorithm, called //Z is the new itemset of size k.
Apriori, which came out as revolution in this field. In this 6.
algorithm generation and verification steps are separated in two 7.
steps one after the other. Afterwards, research has been carried //Step 2: (Pruning) - is the collection of k-frequent
out to improve or extend the Apriori algorithm. The pruning potential itemsets.
method used in this algorithm is also used by Mannila et al 8.
[14]. 9.
An algorithm known as DHP used direct hashing and pruning 10.
technique which improved the Apriori significantly [18].
Alternately, Mueller [16] introduced prefix tree instead of hash 11.
tree. //Step 3:(Verification) - is the collection of k-
A sampling approach was used by Toivonen [23]. The idea is //frequent itemset.
to pick a random sample to determine the negative boundary 12.
that separate the large frequent itemsets from the small frequent 13.
itemsets and validate the results with the rest of database. // functions returns ratio of the number of
Algorithm thus produces exact association rules in one full pass transactions that contain all the items of the set to the
over the database. But if sample misinterprets the negative total number of transactions in the database D.
border then the whole process needs to be repeated. 14.
Savasere et al [20] partitioned the database and generated all
15.
association rules by scanning the database two times only. //Value of k is increased to find large frequent itemsets.
Mueller [16] also used partition technique to generate the
16.
frequent itemsets.
17. ;
A new method of overlapping generation and verification step
is evolved by Brin et al [5]. In this method, Association rules
are known as implication rules. These rules are based upon
Agrawal and Srikant [3] also focused on mining the sequential
conviction instead of confidence. Here, conviction neither talks
association rules which is a kind of navigation or noticing a
about the co-relation nor talks about the co-existence.
generalized behavior followed. Here, rules can be like- 95% of
Conviction gives equal importance to both antecedent and
the students who are buying a laptop; followed by buying a
consequent of the rule. Moreover, it is unambiguous and
book related to computers; and followed by buying a pen drive
measures actual implication. Because of these two features,
it is observed that they can purchase an external mouse also.
implication rules are more interesting than association rules.
Continuous association rule are those kinds of rules which are
Ahmed et al [4] elaborated and extended the direction of Brin
generated online. Hidber [10] investigated continuous
et al [5].
association rules and provided flexibility to change given
Srikant and Agarwal [21] have carried out their research on
minimum support during first scan of database.
interval data to generate quantitative association rules by
Due to update in database it may be possible that the large
measuring the value of each individual attributes greater than
frequent itemsets may become small frequent itemsets and
some expected value. Quantitative association rule can be like-
vice-versa. Therefore, association rules are kept maintained
95% of the students who buy a lap-top and a book related to
instead of generating a new set of rules [6].
computers can also buy at least two pen-drives. But

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 828
A Novel Pruning Approach for Association Rule Mining

Interesting information means knowing something unknown, Table 2: Set of Transactions

covering a large portion of database, and potentially useful. Transaction TID Items Transaction TID Items
Association rules generated at high levels are not interesting Purchased Purchased
and association rules generated at low levels are not useful.
101 {1,4,6,7, 223 {3,7}
Therefore, in order to get interesting and useful associations,
8,9}
multilevel association rules are generated by [9, 17, 22].
Concept of recursive median is used by [24, 25] which is a 102 {1,3,4,5,6, 507 {1,4,7,
probabilistic approach to discover frequent itemsets. It does not 7,8,9} 8,9}
consider all candidate itemsets with equal probability to be
frequent itemsets. 103 {3,4,6, 345 {1,2,3,4,5,
Next section 4.1 and 4.2 demonstrates an example to explain 8,9} 9,10}
the work carried out in this paper. Section 4.3 proposes the 201 {1,4,6} 309 {1,4,6,7}
modified algorithm and section 4.4 shows the experimental
results. 213 {1,5,7,8, 316 {1,3,4,6,
9,10} 8,9,10}
4.0 MODIFICATION OF APRIORI ALGORITHM
The modification applied to Apriori algorithm is exposed by 123 {3,4,6, 224 {4,8,10}
taking following example. 8,9}

205 {2,4,6,9} 508 {1,5,8,9}

4.1Apriori algorithm with example:
Let be a set of ten different types of items a 234 {2,3,4,8} 346 {1,4,6,9,10}
customer can purchase. Let be a set of
301 {3,7,8} 356 {1,4,5,6,9}
twenty independent transactions. Each transaction have a
transaction identifier, TID, and list of items purchased shown in 306 {1,2,3,4, 366 {1,4,6,9}
Table 2 and let us fix user-defined minimum support, i.e., 6,7,9}
minsup= 5. Frequency of each individual item purchased in
database is shown in Table 1.3.
It is easy to generate following 1-frequent itemsets Table 3: Individual Items Frequencies
Item Frequency Item Frequency
by comparing the frequency of each item number number
of Table 1.3 with the given minimum support. These itemsets 13 12
have frequency greater than or equal to given minimum 4 8
support.
Thereafter, Apriori algorithm’s first step is executed and 9 11
following 2-frequent candidate itemsets are generated.
16 14

5 5

In third step, the frequencies of above itemsets present in

the database D, shown in Table 4, are compared with given
minimum support and following 2-frequent itemsets are
In the second step, pruning is applied on the outcome of generated.
first step and following potential 2-frequent itemsets are
generated. These itemsets

are used to generate 3-frequent candidate itemsets in next

iteration. All the above three steps are executed iteratively until
the large k-frequent itemsets are generated. It is worth
mentioning here that the following itemsets
are not 2-frequent because frequency of
these itemsets is less than the fixed minimum support as shown
in Table 1.4.

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 829
 BIJIT - BVICAM’s International Journal of Information Technology

Table 4: 2-Frequent Itemsets Frequencies Frequency of the 3-frequent potential itemsets, shown in Table
Potential 2- Frequency Potential 2- Frequency 1.5, is compared with given minimum support and following 3-
Frequent Frequent frequent itemsets are generated.
Itemsets Itemsets
It is again worth
4 8
11 12 mentioning here that following itemsets
are not 3-frequent because
5 4
frequency of these itemsets is less than minimum support.
9 2 Similarly, in third iteration, following 4-frequent candidate
6 2 itemsets ; followed by 4-
6 3
frequent potential itemsets ; and then
11 5
followed by 4-frequent itemsets are
{1,10} 4 2
generated by first, second and third step respectively. At this
7 4 point again, the significance of making the following 4-
2 5 frequent candidate itemsets bold faced
will be discussed in part 3.2 of this section. Frequency of 4-
5 9
frequent potential itemsets is shown in Table 1.6. There will be
4 2 no more 5-frequent candidate itemsets generated by Apriori
algorithm for this example because none of two itemsets from
f6 5 following 4-frequent itemsets are possible
6 5 to join using joining operation in next iteration and therefore,
Apriori algorithm terminates.
2 1
2 8 4.2 Alternate to apriori’s pruning step
In section3.1, it is observed that large k-frequent itemsets can
12 3 be generated by modifying the second step of apriori
algorithm. For this, not only k-frequent but k-infrequent
5 4 itemsets are also generated in the third step of Apriori
algorithm. Then, the output of the first step is filtered by taking
In the second iteration, following 3-frequent candidate itemsets help of all infrequent itemsets generated so far. It is obvious
and following 3-frequent potential itemsets are generated by that this modification is applicable from second iteration
first and second step respectively. onwards.
Continuing the discussion from part 3.1 of this section, in the
first step of second iteration, bold faced 3-frequent candidate
itemsets and and in the first step of third
iteration, bold faced 4-frequent candidate itemset are
useless to generate because itemset is not 2-frequent.
Similarly, are useless to
generate because , are not 2-frequent.By
eliminating
and itemsets
from 3-frequent candidate itemsets following 3-frequent
At this point, it is to be noted here that 3-frequent candidate potential itemsets are generated.
itemsets and are bold
faced. The reason for making itemsets bold faced will be
discussed in part 3.2 of this section.

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 830
A Novel Pruning Approach for Association Rule Mining

Similarly, bold faced itemsets Lemma: if any itemset Z is infrequent then none of it superset
from the 4-frequent candidate itemsets are useless to generate can be frequent.
because 3-frequent itemsets are not 3- Proof: Let Y be an itemset containing all the items of set Z and
frequent. This process will result the following set as a 4- number of items in set Y is more than number of items in set Z,
frequent potential itemsets i.e., Z ⊂ Y. As if Z is infrequent, it can be stated that:-
. This method of pruning (1)
advocates an alternative to pruning step of apriori algorithm. In
It is obvious that support of itemset Y can’t be greater than
this paper, this process of eliminating some or all itemsets from
support of any of its subset in a given database because
k-frequent candidate itemsets, using infrequent itemsets
cardinality of set Y is more than cardinality of set Z, i.e.,
generated, is expressed as a “filtration” for first step of Apriori
. It can be said that:-
algorithm.
(2)
Using property of associativity from equation “(1)” and “(2)” it
Table 5: 3-Frequent Itemsets Frequencies
can be stated that:-
Potential Frequency Potential Frequency (3)
3- 3- In other words set Y can’t be frequent if it is superset of any set
Frequent Frequent Z which is infrequent.
Itemsets Itemsets Table 7: Modified Apriori Algorithm
9 6 Apriori_Filter (D, minsup, F1)
5 4 // Dis a set of Transactions, minsup is the given //minimum
support, F1is the collection of 1-//frequent itemset.
4 5 1.
//Step 1: (Joining) - is the collection of k-//frequent
8 4
candidate itemsets.
4 5 2.
3.
3 10 4.
7 3
5.
4 4 //Z is the new itemset of size k.
6.
5 6 7.
6 5 //Step 2: (Filtration) - is the collection of k-frequent
potential itemsets and is the collection of k-infrequent
5 4 itemset
8.
5
9. ;
10.
Table 6: 4-Frequent Itemsets Frequencies )
Potential 4-Frequent Frequency 11.
Itemsets 12.
13.
5
//Step 3:(Verification) - is the collection of k-frequent
5 itemset.
14.
5
15.
16.
4.3 Modified apriori algorithm 17.
A modified Apriori algorithm is proposed in Table 7. Filtration //Value of k is increased to find large frequent itemsets.
step is an alternate to the pruning step used in Apriori 18.
algorithm. Prerequisite for modified Apriori algorithm are same 19. ;
as with the Apriori algorithm.
This new pruning method is based upon following lemma.

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 831
 BIJIT - BVICAM’s International Journal of Information Technology

5.0 Experimental Results of minimum support is taken6%, 9%, 12%, 15%, 20%, and
Five experiments are performed on synthetic datasets generated 25%. Results of this experiment are plotted in Figure5.
as described by [2]. Each experiment is executed three times 700
for both Apriori and proposed algorithm. Average time taken Apriori
(in seconds) by both algorithm in each experiment is shown 600

Time in Seconds
Proposed
below in respect to minimum support value. Following 500
parameters are set for each experiment.
1. Number of total transactions in database are 100000, i.e., 400
, numbers of items are 100, i.e., 300
,average size of transactions is 5, i.e., , average 200
size of maximal potentially large itemsets is 2,i.e., ,
100
value of correlation level is set to 0.5, number of maximal
potentially large itemsets are 200,i.e., ,and value 0
of minimum support is changed from 4% to 10%with a 4% 5% 6% 7% 8% 9% 10%
step increment of 1%. Results of this experiment are
Support
plotted in Figure 1.
Figure 1: Apriori algorithm vs.Proposedalgorithm for
2. Number of total transactions in database are 100000,
i.e., , numbers of items are 100, i.e., ,
average size of transactions is 5, i.e., , average size
of maximal potentially large itemsets is 2, i.e., , 500
value of correlation level is set to 0.5, number of maximal Apriori
400 Proposed
potentially large itemsets are 200, i.e., , and value
of minimum support is changed from 10% to 20% with a
300
step increment of 1%. Results of this experiment are
plotted in Figure2. 200
3. Number of total transactions in database are 100000,
i.e., , numbers of items are 100, i.e., , 100
average size of transactions is 5, i.e., , average size
of maximal potentially large itemsets is 4, i.e., , 0
value of correlation level is set to 0.5, number of maximal 10%11%12%13%14%15%16%17%18%19%20%
potentially large itemsets are 200, i.e., , and value
of minimum support is changed from 10% to 20% with a
step increment of 1%. Results of this experiment are Figure 2: Apriori algorithm vs. Proposed algorithm for
plotted in Figure3.
4. Number of total transactions in database are 100000,
i.e., , numbers of items are 100, i.e., , 360
average size of transactions is 5, i.e., , average size 320 Apriori
of maximal potentially large itemsets is 4, i.e., , 280
Proposed
value of correlation level is set to 0.5, number of maximal 240
potentially large itemsets are 200, i.e., , and value 200

of minimum support is changed from 15% to 30% with a 160

step increment of 3%. Results of this experiment are 120

plotted in Figure4. 80

5. Number of total transactions in database are 100000, 40

i.e., , numbers of items are 100, i.e., , 0

average size of transactions is 5, i.e., , average size 10% 11% 12% 13% 14% 15% 16% 17% 18% 19% 20%

of maximal potentially large itemsets is 6, i.e., , Figure 3 Apriori algorithm vs. Proposed algorithm for
value of correlation level is set to 0.5, number of maximal
potentially large itemsets are 200, i.e., , and value

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 832
A Novel Pruning Approach for Association Rule Mining

1200 8.0 REFERENCES

Apriori [1]. Agrawal R, Imielinski T, Swami A.,“Mining association
1000 rules between sets of items in large databases”. In
Proposed
Time in Seconds

Proceedings of the ACM-SIGMOD international

800 conference on management of data (SIGMOD’93),
Washington, DC, pp. 207–216, 1993.
600 [2]. Agrawal R, Srikant R.,“Fast algorithms for mining
association rules”. In Proceedings of the international
400 conference on very large data bases (VLDB’94),
Santiago, Chile, pp. 487–499, 1994.
200
[3]. Agrawal R, Srikant R.,“Mining sequential patterns”. In
0 Proceedings of the 1995 international conference on
data engineering (ICDE’95), Taipei, Taiwan, pp. 3–14,
15% 18% 21% 24% 27% 30% 1995.
Support [4]. Ahmed K M, El-Makky NM, Taha Y., “A note on
Figure 4: Apriori algorithm vs.Proposedalgorithm for “beyond market basket: generalizing association rules to
correlations””. SIGKDD Explorations 1: 46–48, 2000.
[5]. Brin S, Motwani R, Silverstein C.,“Beyond market
basket: generalizing association rules to correlations”, In
4500 Proceeding of the ACM-SIGMOD international
4000
Apriori conference on management of data (SIGMOD’97),
Proposed Tucson, AZ, pp. 265–276, 1997.
3500 [6]. Cheung DW, Han J, Ng V, Wong CY.,“Maintenance of
3000 discovered association rules in large an incremental
updating technique”. In: Proceeding of the 1996
2500
international conference on data engineering
2000 (ICDE’96), New Orleans, LA, pp. 106–114, 1996.
1500 [7]. Goyal, L.M., Beg, M.M.S., "An efficient filtration
approach for mining association rules," Computing for
1000 Sustainable Global Development (INDIACom - 2014),
500 2014 International Conference, pp.178-185, 5-7 March
0 2014.
[8]. Goyal, L.M., Sufyan Beg, M.M., "Improved filtration
6% 9% 12% 15% 20% 25% step for mining association rules," Data Mining and
Intelligent Computing (ICDMIC), 2014 International
Conference, pp.1-4, 5-6 Sept. 2014.
Figure 5 Apriori algorithm vs. Proposed algorithm for [9]. Goyal, L. M.,Sufyan Beg, M.M., "Evaluation of
filtration and pruning approach for Apriori
algorithm," Computer and Communication Technology
All experimental results show that proposed algorithm (ICCCT), 2014 International Conference, pp.23-28, 26-
improves Apriorialgorithm. 28 Sept. 2014.
[10]. Han J, Cheng H, Xin D, Yan X.,“Frequent pattern
6.0 CONCLUSION AND FUTURE WORK mining: current status and future directions”. Data Min.
In this paper, a new pruning method is proposed as an alternate Knowl.Discov. (DATAMINE) 15(1):55-86, 2007.
to pruning method of Apriori algorithm. This method is [11]. Han J, Kamber M. Data mining: concepts and
expressed as a filtration for joining operation of Apriori techniques.2ndedn. Morgan Kaufmann.
algorithm. By using this new method, same candidate k- [12]. Han J, Fu Y., “Discovery of multiple-level association
frequent itemsets are generated as by the Apriori’s pruning rules from large databases”, In Proceeding of the
method. It is observed that proposed approach works as international conference on very large data bases
efficient as the existing method. In future, proposed pruning (VLDB’95), Zurich, Switzerland, pp. 420–431, 1995.
approach may be helpful in FP-Growth, Eclat algorithm. [13]. Hidber C.,“Online association rule mining”. In Proc.
ACM SIGMOD Int. Conf. on Management of Data,
7.0 ACKNOWLEDGEMENT pages 145-154, Philadephia, Pennsylvania, USA, 1999.
Lalit Mohan Goyalis thankful to the management of Noida [14]. InokuchiA,Washio T, Motoda H.,“An apriori-based
Institute of Engg. & Tech., Greater Noida for providing algorithm for mining frequent substructures from graph
academic leaves. data”. In Proceeding of the European symposium on the

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 833
 BIJIT - BVICAM’s International Journal of Information Technology

principle of data mining and knowledge discovery [27]. V. Sharma, M. M. S. Beg.,“A Probabilistic Approach to
(PKDD’00), Lyon, France, pp. 13–23, 2000. AprioriAlgorithm”,International Journal of Granular
[15]. Ji X, Bailey J, Dong G.,“Mining minimal distinguishing Computing, Rough Sets and Intelligent Systems
subsequence patterns with gap constraints”. In (IJGCRSIS), Inderscience Publishers, ISSN (Online):
Proceeding of the international conference on data 1757-2711 - ISSN (Print): 1757-2703, vol. 2, no. 3,
mining (ICDM’05), Houston, TX, pp. 194–201, 2005. 2012, pp. 225-243, 2012.
[16]. M. M. S. Beg, C. P. Ravi Kumar,“Application of [28]. V. Sharma, M. M. S. Beg,“A Probabilistic Approach to
Parallel and Distributed Data Mining in e-Commerce”, Apriori Algorithm”, IEEE International Conference on
J.IETE Technical Review special issue on e-Commerce, Granular Computing (GrC 2010), Silicon Valley, USA,
vol. 17, no. 4, pp. 189-195, 2000. August 14-16, IEEE Computer Society Press, pp. 402-
[17]. Mannila H, Toivonen H, VerkamoAI.,“Efficient 408, 2010.
algorithms for discovering association rules”. In: [29]. Yan X, Zhou XJ, Han J.,“Mining closed relational
Proceeding of the AAAI’94 workshop knowledge graphs with connectivity constraints”, In Proceeding of
discovery in databases (KDD’94), Seattle,WA, pp. 181– the ACM SIGKDD international conference on
192, 1994. knowledge discovery in databases (KDD’05), Chicago,
[18]. Miller RJ, Yang Y.,“Association rules over interval IL, pp. 324–333, 2005.
data”,InProceeding of the ACMSIGMOD international [30]. Yan X, Zhu F, Han J, Yu PS., “Searching substructures
conference on management of data with superimposed distance”, In Proceeding of the
(SIGMOD’97),Tucson, AZ, pp. 452–461, 1997. international conference on data engineering
[19]. Mueller A., “Fast sequential and parallel algorithms for (ICDE’06), Atlanta, 2006.
association rule mining: A comparison”,Technical CS-
TR-3515, University of Maryland, College Park, August
1995, pp. 1-5, 1995.
[20]. N. Gangal, M. M. S. Beg, “Finding the Interesting
Rules” First.Proc. International Conference on Data
Mining (DMIN'08) - a track at the 2008 World Congress
in Computer Science, Computer Engineering and
Applied Computing (WORLDCOMP’08), Las Vegas,
USA, July 14-17, 2008, pp. 689-694, 2008.
[21]. Park J S, Chen M S, Yu P S.,“An effective hash-based
algorithm for mining association rules”, In Proceeding
of the ACM-SIGMOD international conference on
management of data (SIGMOD’95), San Jose, CA, pp.
175–186, 1995.
[22]. Piatetsky-Shapiro G.,“Notes ofAAAI’91 workshop
knowledge discovery in databases (KDD’91)”,
AAAI/MIT Press, Anaheim, CA, 1991.
[23]. Savasere A, Omiecinski E, Navathe S.,“An efficient
algorithm for mining association rules in large
databases”, In Proceeding of the 1995 international
conference on very large data bases (VLDB’95), Zurich,
Switzerland, pp. 432–443, 1995.
[24]. Srikant R, Agrawal R.,“Mining sequential patterns:
generalizations and performance improvements”,In:
Proceeding of the 5th international conference on
extending database technology (EDBT’96), Avignon,
France, pp. 3–17, 1996.
[25]. Srikant R, Agrawal R.,“Mining generalized association
rules”, In Proceeding of the international conference on
very large data bases (VLDB’95), Zurich, Switzerland,
pp. 407–419, 1995.
[26]. Toivonen H.,“Sampling large databases for association
rules”, In Proceeding of the international conference on
very large data bases (VLDB’96), Bombay, India, pp.
134–145, 1996.

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 834
 BIJIT - BVICAM’s International Journal of Information Technology
Bharati Vidyapeeth’s Institute of Computer Applications and Management (BVICAM), New Delhi (INDIA)

Design of Linear-Phase Digital FIR Filter Using Differential Evolution Optimization

with an Improved Ripple Constraint Handling Method

Shekhar Sharma1, Sumant Katiyal2 and L. D. Arya3

Submitted in April, 2014; Accepted in December, 2015

Abstract - For the optimal design of frequency-selective 1.0 INTRODUCTION
digital filters, evolutionary optimization algorithms have been Digital filters are recognized by great flexibility in design and
applied. In these design methods the goal of the optimization implementation. This makes it is easier to implement complex
process is to find the optimal filter coefficients which closely signal processing schemes utilized in digital communication
approximate the desired frequency response. In this paper, systems. The digital filters are classified according to the
an efficient alternative method for the design of linear phase length of the impulse response as Finite Impulse Response
digital FIR filter with ripple constraint is discussed. This (FIR) and Infinite Impulse Response (IIR) digital filters [1].
method of optimization uses DE algorithm with modified Special features of the FIR filters are their inherent stability
selection rule for ripple constraint handling. The results and precise linear phase. The phase response of the FIR filter
obtained using this method are compared with those obtained is linear if the coefficients of FIR are symmetric around the
for another method of ripple constraint handling based on centre coefficient. Linear phase FIR filters have many
penalty function using DE algorithm. From the simulation applications such as in telecommunications, for demultiplexing
results it is observed that ripple constraint handling method the data that have been frequency-division multiplexed,
based on the modified selection rule of DE shows better without distorting the data in this process, and in systems,
performance than that obtained using DE with ripple where it is necessary to have minimum signal distortion and
constraint method based on penalty function when number signal dispersion so as to avoid inter symbol interference.
of runs is applied. In order to seek better control over different parameters in the
design of digital filters, the design methods based on
Index Terms — Differential Evolution algorithm, FIR filters, optimization algorithms are developed. Thus, Evolutionary
Frequency response, Ripple constraint. algorithms (EA), such as Genetic algorithm (GA), Particle
swarm optimization (PSO), and Differential evolution (DE)
NOMENCLATURE and many others have been used for better individual control
H (e jω ) : The frequency response of a digital filter, over the parameters of digital filters. Generally, in these design
h(n) : The impulse response of a digital filter, methods, the unconstrained optimization is used. In this paper,
DE algorithm is used for the design of linear phase digital FIR
A(ω ) : The magnitude response,
filter with two ripple constraint handling methods. One method
θ (ω ) : The phase response, is based on penalty function [2] and the other is based on a
a(k ) : The filter coefficients, method proposed by Lampinen [3]. The simulation results
E (ω ) : The approximation error function, obtained for these two methods show that when number of
runs is applied, the ripples obtained in different frequency
W (ω ) : The weighting function,
bands using the second method have smaller ripple size and
D(ω ) : The desired frequency response, smaller value of error as compared to those obtained for the
L2 ( X ) : The discrete form of L2 norm approximation error, first method. Thus a better performance is exhibited by the
th second method.
Xk : The k solution vector ,
xkj : The jth component of the solution vector, 2.0 LITERATURE SURVEY
F : The scaling factor, For the design of the digital FIR filters two classical methods
CR : The crossover factor. are used namely, windowing method and frequency sampling
method [4]. In general, an approximation error norm is used in
these methods for designing an FIR filter. The two most
1 commonly used norms are the least-squares (L 2 ) norm and
Dept. of Electronics & Telecommunication Engg.,
S.G.S.I.T.S., Indore, Madhya Pradesh Chebyshev (L ∞ ) norm.
2 In the design method based on windowing, the decrease in the
School of Electronics, Devi Ahilya Vishvavidyalaya, Indore,
Madhya Pradesh transition bandwidth causes increase in the magnitude of the
3 side lobes and consequently an increase in the approximation
Dept. of Electrical Engg., S.G.S.I.T.S., Indore, Madhya
Pradesh error. Also this method does not offer individual control over
E-mail: [email protected], [email protected] the approximation errors in different bands with any constraint
and [email protected] criterion. The design method based on frequency sampling,

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 835
 BIJIT - BVICAM’s International Journal of Information Technology

provides good control over the transition bandwidth; however, 3.0 PROBLEM FORMULATION
the approximation error is zero exactly at the sampling The frequency response of a linear-phase FIR filter is given
frequencies. by:
Recently, the design methods based on optimization N

algorithms are developed. Genetic Algorithm (GA) is one of H (e jω ) = ∑ h( n)e

n =0
− jωn (1)
the most useful, general purpose optimization algorithm [5].
Where, h(n) is the real-valued impulse response of filter,
GA has been used to solve a wide range of engineering design
(N+1) is the length of filter and ω is the frequency of interest.
and testing optimization problems such as ATM network
The linear phase is possible if the impulse response h(n) is
design [6], optimal testing of nonlinear allocation problems in
modular softwares [7], etc. GA has also been applied for the either symmetric {i.e. h(n) = h(N − n)}, or, is antisymmetric
design of digital FIR filters by Xu and Daley [8], Cen, [9] and {h(n) = − h(N − n)} for 0 ≤ n ≤ N.
others. Although GA is a good global searching algorithm, In general, for causal linear-phase FIR filters, the
sometimes it gets trapped into the local minima, and is frequency response H (e jω ) can be expressed as:
complex in coding. Also GA has slower convergence and takes H (e jω ) = e − jNω / 2 e jβ A(ω ) (2)
more execution time. Thus, the magnitude response is:
Particle swarm optimization (PSO) algorithm developed by
H (e jω ) = A(ω ) , (3)
Kennedy and Eberhart [10] requires less parameters and is
simple. PSO has been used for various engineering & the phase response is:
optimization problems such as, in malicious node detection  ω
− N 2 + β , for A(ω ) ≥ 0, (4)
and path optimization for wireless sensor networks [11]. PSO θ (ω ) = 
ω
and its variants have also been applied for the design of digital − N + β − π , for A(ω ) < 0.
filters [12]-[15]. It gives faster convergence as compared to  2
GA as shown by Ababneh [12]. The modified PSO is applied When β = 0, h(n) is symmetrical, and when β = π/2, h(n) is
by Sharma and Arya [13] for the design of linear phase digital antisymmetrical.
FIR filter to control global exploration and local exploration. The amplitude response [1], for the case of type-I linear
Another optimization algorithm, Differential evolution (DE), phase FIR filter, by substituting N = 2M, is given by:
M
developed by Storn and Price [16] has been used in power A(ω ) = ∑ a ( k ) cos(ωk ), (5)
systems for optimization in planning, operation and k =0
distribution etc. [17]. DE has also been used for the design of where, a (0) = h( M ) and a (k ) = 2h( M − k ) , for 1 ≤ k ≤ M .
digital FIR filters by Zhao and Meng [18], Albataineh et al. For the design of low pass digital FIR filters, the objective
[19], Singh and Kaur [20], Sharma et al. [2] and others. DE of the algorithm used for computation, is to determine the
algorithms provide good global optimization if its control
parameters are adjusted properly. DE algorithm, with two vector X of coefficients a (k ) , so as to minimize the difference
ripple constraint handling methods, is used in this paper for the between the desired frequency response, D (ω ) , and the
design of linear phase digital FIR filter. One method is based realized amplitude frequency response, A(ω ) . Generally this
on penalty function [2], and another is based on the method difference is specified as a weighted error function E (ω )
proposed by Lampinen [3]. The comparison of the simulation
results obtained for these two methods shows that the given by:
maximum error magnitude between desired frequency E (ω ) = W (ω )[ A(ω ) − D (ω )] (6)
response and the designed frequency response is equal to or Where, W (ω ) is a non-negative weighting function and is
below the constraint in the specified frequency bands. accepted for the given pass band attenuation δ p and stop band
However, when number of runs is applied, the ripples obtained attenuation δ s , as:
in different frequency bands for the first method have almost (δ s / δ p ), in the passband , (7)
constant magnitude; while the ripple size and final error value W (ω ) = 
 1, in the stopband .
is less in the case of second method and thus a better And D(ω), the desired magnitude response for the low pass
performance is exhibited by this method. filter given by:
This paper is organized in six sections as follows: In Section 3,
1, in the passband , (8)
Problem formulation of linear phase digital FIR filters is D (ω ) = 
presented. Section 4 explains the DE algorithm for optimizing 0, in the stopband .
filter coefficients and describes the ripple constraint handling The least-squares, or, L 2 norm; which considers error energy,
methods. In Section 5, design of linear phase FIR filters using is defined in the integral form [2] as:
12
DE with ripple constraint methods is given. Then, in Section 6, 1 π 2

simulation results are discussed and analyzed. Finally,
ε 2
=
π
∫
0
W (ω )[ A(ω ) − D (ω )] dω 

(9)

conclusion and future scope are discussed in Section 7. In practice, the discretized version of integral scalar error used
in L 2 norm is approximated by a finite sum given by:

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 836
Design of Linear-Phase Digital FIR Filter Using Differential Evolution Optimization with an Improved Ripple Constraint
Handling Method

1 2
between two randomly selected distinct population vectors,
( )  K 
2

L2 X =  ∑ W (ωi )[ A(ωi ) − D (ωi )] 

(10) G G G
 i =1  X r1 and X r 2 , to the best vector X best as follows:
 
( )
Where, L2 X is L 2 norm approximation error determined for
G +1
Vi
G G
= X best + F ( X r1 − X r 2 )
G
(15)
G +1
a vector X and ω i is suitably chosen grid of digital angular
G
Where, V is a mutant vector, X
i is the best vector of the
best
frequencies for the range 0 ≤ ω ≤ π for 1 ≤ i ≤ K. current population which gives the lowest cost function value;
r1 and r 2 are randomly chosen integers such that
4.0 DE ALGORITHM r1, r 2 ∈ {1,2,..., NP} , r1 ≠ r 2 ; and F is a real and constant
DE algorithm introduced by Storn and Price [16] is a simple scaling factor which usually lies in the range [0, 1].
population based stochastic search algorithm for objective
function minimization. Application of DE for the design of Step 3:
digital filters has been described in [2], [18][19][20]. In basic Crossover: To increase the diversity of population, crossover
DE algorithm, the initial NP population vectors are formed operation is used. This operation causes crossover or exchange
randomly from the vectors having bounded parameter values. of parameters of mutant vector with those of the target vector
Each of these vectors has D-parameters and belongs to a D- G +1
dimensional vector space. The optimization task is to and generates trial vector T i given by:
G +1
successively improve these vectors by applying mutation, Ti = [t iG1 +1 , t iG2+1 ,..., t iD
G +1
] (16)
crossover and selection operators; similar to those used by GA.
DE generates new solution vectors in the D-dimensional vector In the binomial crossover scheme, uniform crossover is
space using mutation. To produce mutated vector the weighted performed as follows:
difference between two randomly chosen, distinct population G +1
 G +1
vij , if ( rand j ≤ CR or j = jrand ) (17)
tij = G
vectors, is added to another distinct vector. Then crossover is 
 xij , if ( rand j > CR and j ≠ jrand )
performed to produce a trial vector from target vector and G +1
Where, tijG +1 is jth component of trial vector T i , vijG +1 is jth
mutated vector. By evaluating objective function for target
G +1
vector and trial vector, either one is then selected on the basis component of mutant vector V i and xijG is jth component of
of their fitness. G
target vector X i . rand j is the jth evaluation of the random
4.1 The Steps of DE Algorithm number in the range [0, 1]. CR is the crossover constant in the
Step 1:
range [0, 1] and jrand is randomly chosen index within the
Initialization: An initial population of ‘NP’ solution vectors is
generated as follows: range [1, D]. As shown above, the trial vector component is
0 0 0 G +1
P 0 = [ X 1 , X 2 ,..., X NP ] (11) adopted from the mutant vector V i , if the random number
Where, P 0 is the initial population of solution vectors, X i ,
0 rand j is less than or equal to CR, or j is equal to index jrand .
for 1 ≤ i ≤ NP given by: Otherwise, the trial vector component is adopted from target
G
0
X = [ x , x ,..., x ]
i
0 0 0
(12) vector X i . The index jrand ensures that the trial vector
i1 i2 iD
G +1
The jth component, or parameter, x , for 1 ≤ i ≤ NP ,
0 contains at least one parameter from mutant vector
ij
Ti
G +1
1 ≤ j ≤ D , is obtained from uniform distribution as follows: V i and does not duplicate the target vector.
xij0 = x Lj + ( xUj − x Lj ) * rand j , (13)
Step 4:
Where, x Lj and xUj are lower and upper bounds on parameter G +1
Selection: In order to decide whether trial vector T i , or, the
x j and rand j is a random number in the range [0, 1]. G
target vector X i , is to be selected as the member of
population vectors in next generation G + 1, the objective
Step 2: function is evaluated for target vector and trial vector. If the
Mutation: A mutant vector in the generation (G + 1) is created trial vector gives a smaller value of objective function, then
for each population vector by mutation: this vector replaces the target vector for the next generation;
G +1
V i = [viG1 +1 , viG2+1 ,..., viD
G +1
] (14) otherwise, the old target vector is retained as follows:
In this paper DE/best/1/bin is implemented for the design of G +1 T Gi +1 ,

G +1
if f (T i ) ≤ f ( X i ),
G

Xi =  G (18)
FIR digital filter, hence a single difference of vectors is 
G +1  i
X , otherwise .
utilized. In DE/best/1/bin, a mutant vector V i is generated for The process of mutation, crossover and selection is executed
G
each target vector X by adding a weighted difference
i for all target vector index i and new population is created till

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 837
 BIJIT - BVICAM’s International Journal of Information Technology

the optimal solution is achieved. The procedure is terminated if problem into an unconstrained one as shown by Lampinen [3].
maximum number of generations has been executed. This is seen from Equation (20) where, objective
function J 2 ( X ) is used instead of L2 ( X ) .
4.2 Bounce Back Technique for Handling Bounds on In this subsection an improved version of constraint handling
Parameters of Mutant Vector: method used by Lampinen [3] is described. It allows to get rid
In the process of generating mutant vector, some of the of setting of the weight parameters for individual constraints.
components of this vector may cross the lower or upper In this method, only the selection operation of the basic
bounds. In such cases bounce back mechanism [17] is adopted Differential Evolution algorithm is modified, for handling the
to bring such elements of the mutant vector within limit. In this ripple constraints. The selection criteria of Equation (18) to
method the element, which has violated the limits, is replaced G +1 G
select either trial vector T i or, target vector X i for the next
by a new element whose value lies within the best vector value G +1

and the bound being violated. The following relations are used generation vector X i is changed as follows:
for violated mutant vector elements: • If both solution vectors satisfy all ripple constraints,
G +1

x + rand ⋅ ( x Lj − x best , j ), if v ijG +1 ≤ x Lj (19) then the one with lower objective function value is
v =  best , j
 x best , j + rand ⋅ ( x j − x best , j ),
 if v ijG +1 > x Uj selected, OR,
ij U

G +1 • If target vector satisfies all ripple constraints, while

Where, vijG +1 is jth element of mutant vector, V i , xbest , j is jth trial vector does not satisfy, and if target vector also
G
element of the best vector, X best and x Lj , xUj are lower and has lower objective function value then it is selected,
OR,
upper bounds on parameter x j respectively and rand is a • If target vector does not satisfy all ripple constraints
random number in the range [0, 1]. but provides lower or equal value for all ripple
constraints as compared to the trial vector, and also if
4.3 Ripple Constraint Handling methods: target vector has lower objective function value then
In this paper two methods are used for ripple constraint it is selected.
handling. First we have discussed the method based on penalty • Else, trial vector is selected.
function used by Sharma et al. [2]. Another method is based
on modified selection rule as proposed by Lampinen [3]. 5.0 DESIGN OF LINEAR PHASE FIR FILTERS USING
DE WITH RIPPLE CONSTRAINTS
4.3.1 Method # 1: Method Based on Penalty Function: This section is divided into two subsections. Subsection 5.1
This method is based on penalty function, which penalizes describes the specifications of the digital low pass FIR filter.
infeasible frequency response ripple values obtained in the The design parameters of DE algorithm with ripple constraints
pass band and stop band. Thus, for a vector X , the objective for Method # 1 and Method # 2 are discussed in subsection
function with ripple constraint J 2 ( X ) is developed as follows: 5.2.

J 2 ( X ) = cL L2 ( X ) + c pδ p ( X ) + csδ s ( X ), (20) 5.1 Specifications of The Digital Low Pass FIR Filter
with cL + c p + cs = 1. Designed:
Type-I linear phase FIR filter is designed with the filter length
Where, cL , c p and cs are suitable weight parameters for L2 ( X ) , taken as N+1 = 31, and the grid of digital angular frequencies
δ p ( X ) and δ s ( X ) respectively. δ p ( X ) and δ s ( X ) are the as K = 180. The cut-off frequency of the pass band is ω p =
0.3π and cut-off frequency of the stop band is ω s = 0.4π. The
maximum pass band and stop band ripples given as follows:
desired ideal frequency response D(ω) has unity gain in the
δ ( X ) = max 1 − A(ω ) (21)
p
ωi∈Passband
i pass band and zero gain in the stop band and is given by:
(22)  1, 0 ≤ ω ≤ 0.3π (23)
δs (X ) = max [ A(ωi )] D(ω ) = 
ωi∈Stopband
 0, 0.4π ≤ ω ≤ π
Where, A(ωi ) is the magnitude of the frequency response For passband attenuation δ p = 0.06 and stopband attenuation δ s
of the filter, defined earlier in (3), for the suitable set of = 0.06, the weighting function W (ω ) used, is given by:
frequencies ωi . (δ / δ p ) = 1, in the passband (24)
W (ω ) =  s
4.3.2 Method # 2: Method Based on Modified Selection  1, in the stopband
Rule of DE:
The penalty function method uses additional control 5.2 Design Parameters of Differential Evolution Algorithm
parameters, which are termed as the weight parameters. Setting with Ripple Constraint for Method # 1:
For applying DE with ripple constraint for obtaining the
the weight (or, penalty) parameters for getting their appropriate
values by trial and error method, is a laborious task. The coefficients of the filter, a (k ) ; the size of each solution
penalty function method effectively converts a constrained vector, X , is taken as D = M + 1 = 16. The scaling factor F

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 838
Design of Linear-Phase Digital FIR Filter Using Differential Evolution Optimization with an Improved Ripple Constraint
Handling Method

and crossover constant CR used are taken as, F = 0.5 and CR = however, the value of standard deviation obtained for Method
0.7. The population size is taken as NP = 50. The numbers of # 2 is larger.
generations used are 200. The weight parameters cL , c p and Table 2: Frequency-Domain Parameters
cs used in the objective function of (20) are arbitrarily set as
DE with ripple constraint DE with ripple
0.0006, 0.2894 and 0.71respectively so as to reject larger Parameter Method # 1 constraint Method# 2
values of objective function for larger size of ripples. It is
observed that the amplitudes of the ripples are larger in the
stop band, so more weight is assigned to cs as compared to c p . Pass Stop Stop
Pass Band
Band Band Band

Lower Band
5.3 Design Parameters of Differential Evolution Algorithm 0.0000 0.4000 π 0.0000 0.400 π
Edge
with Ripple Constraint for Method # 2:
There is no specific parameter used for ripple constraint
Upper Band
handling Method # 2. All other parameters of DE algorithm are 0.3500 π 1.0000 π 0.3500 π 1.000 π
Edge
kept same as mentioned in the subsection 5.2.
Desired
6.0 SIMULATION RESULTS AND ANALYSIS 1.0000 0.0000 1.0000 0.0000
Value: D(ω)
The summary of the parameters obtained for the best of 30
runs, for Type-I FIR low pass filter (LPF) design, using DE Maximum
0.0243 0.0231 0.0229 0.0226
with ripple constraint Method # 1 and Method # 2, is shown in ripple
the TABLE-1 and TABLE-2 for the time-domain and the Maximum
frequency domain respectively. In TABLE-3, the statistical 0.2087 − 32.70 0.1972 − 32.88
ripple (dB)
parameters obtained for the two cases are compared.
In Fig. 1, the error plots for the best run of the two cases are Minimum
0.0010 0.0057 0.0010 0.0022
ripple
shown. The frequency responses obtained for the filter design
using DE with ripple constraint methods are compared in Fig. Minimum
0.0094 − 44.78 0.0091 − 52.81
2. It is observed from Fig. 1, that the absolute value of error ripple (dB)
obtained for Method # 2 is lower than that obtained for Method
# 1. This fact is also exhibited in TABLE-3.
By comparing the frequency responses in Fig. 2, it is observed Table 3: Statistical Parameters of DE With Ripple
that ripple constraint handling Method # 2 shows a better Constraint Method # 1 And
frequency response as compared to Method # 1. It is also obse- Method # 2 for the Low Pass Filter
(Number of runs = 30; Number of generations = 200;
Population Size = 50.)
Table 1: Time-Domain Parameters Sr. Parameters DE with DE with
No. ripple ripple
DE with ripple constraint DE with ripple Constraint constraint constrai
Method # 1 Method # 2 Method # 1 nt
Impulse response h(n) Impulse response h(n) Method
#2
h(0) = − 0.0033 = h(30) h(0) = − 0.0058 = h(30)
h(1) = 0.0081 = h(29) h(1) = 0.0025 = h(29) 1 Best fitness value of error of all 0.2711 0.1552
h(2) = 0.0170 = h(28) h(2) = 0.0104 = h(28) runs
h(3) = 0.0090 = h(27) h(3) = 0.0071 = h(27)
2 Average value of minimum error 0.4599 0.3174
h(4) = − 0.0080 = h(26) h(4) = − 0.0051 = h(26) of all runs
h(5) = − 0.0226 = h(25) h(5) = − 0.0201 = h(25)
h(6) = − 0.0104 = h(24) h(6) = − 0.0136 = h(24) 3 Worst minimum value of all 0.9015 0.7021
h(7) = 0.0229 = h(23) h(7) = 0.0154 = h(23) runs
h(8) = 0.0356 = h(22) h(8) = 0.0367 = h(22) 4 Standard deviation of minimum 0.1389 0.1440
h(9) = 0.0120 = h(21) h(9) = 0.0195 = h(21) error from average
h(10) = − 0.0412 = h(20) h(10) = − 0.0385 = h(20)
h(11) = − 0.0708 = h(19) h(11) = − 0.0717 = h(19) 5 Frequency of convergence* 0.5000 0.6000
h(12) = − 0.0167 = h(18) h(12) = − 0.0195 = h(18) *(Frequency of convergence = number of better fitness values than
h(13) = 0.1316 = h(17) h(13) = 0.1250 = h(17) mean out of all runs / total no. of runs)
h(14) = 0.2804 = h(16) h(14) = 0.2839 = h(16)
h(15) = 0.3493 h(15) = 0.3535
-rved that the stop band attenuation is decreased further with
the increase in frequencies in the case of Method # 2. Finally,
from the TABLE-3 it is observed that the frequency of
convergence of Method # 2 is higher than that of Method # 1;

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 839
 BIJIT - BVICAM’s International Journal of Information Technology

[4]. J. G. Proakis and D. G. Manolakis, Digital Signal Processing:

Principles, Algorithms and Applications, 4th ed., Pearson
Education, 2007.
[5]. S. V. Chande1 and Dr. M. Sinha, “Genetic Algorithm: A
Versatile Optimization Tool”, BIJIT – 2009, Vol. 1 No. 1,
January – June, 2009.
[6]. S. Routray, “An Enhanced Genetic Algorithm Approach to
ATM Network Design”, BIJIT – 2011, Vol. 3 No. 1, January –
June, 2011.
[7]. A. G. Aggarwal, P. K. Kapur, G. Kaur and R. Kumar “Genetic
Algorithm Based Optimal Testing Effort Allocation Problem
for Modular Software”, BIJIT – 2012, Vol. 4 No. 1, January -
Figure 1: Error plots of Type-I FIR LPF obtained using June, 2012.
DE with ripple constraint Method # 1 and Method # 2. [8]. D.J. Xu and M.L. Daley, “Design of optimal digital filter
using a parallel genetic algorithm”, IEEE Transactions on
Circuits and Systems II: Analog and Digital Signal Processing,
Vol. 42, No. 10, pp. 673-675, October 1995.
[9]. L. Cen, “A hybrid genetic algorithm for the design of FIR
filters with SPoT coefficients”, Signal Processing, Vol. 87, pp.
528-540, March 2007.
[10]. Kennedy J. and Eberhart R., “Particle Swarm Optimization”,
In: Proc. IEEE Int. Conf. on Neural Networks, Piscataway,
pp. 1942-1948, 1995.
[11]. D. Goyal, M. R. Tripathy, “Path optimization using APSO”,
BIJIT – 2013, Vol. 5 No. 2, July – Dec., 2013.
[12]. J. I. Ababneh and M. H. Bataineh, “Linear phase FIR filter
design using particle swarm optimization and genetic
Figure 2: Comparison of Magnitude Frequency responses algorithms”, Digital Signal Processing, Vol. 18(4), pp. 657-
of Type-I FIR LPF for DE with ripple constraint Method 668, July 2008.
# 1 and Method # 2. [13]. S. Sharma and L. D. Arya, “Design of FIR digital filters using
particle swarm optimization”, IE(I) Journal-ET, Vol. 91,
7.0 CONCLUSION AND FUTURE SCOPE pp.11-19, July 2010.
In this paper an efficient alternative method for the design of [14]. S. Mandal, S. P. Ghoshal, R. Kar, D. Mandal, “ Design of
linear phase digital FIR filter with ripple constraint is optimal linear phase FIR high pass filter using craziness
discussed. This method of optimization of the filter based particle swarm optimization technique”, Journal of King
coefficients uses DE algorithm with modified selection rule for Saud University – Computer and Information Sciences,
ripple constraint handling. The results obtained using this Vol.24, pp. 83–92, 2012.
method are compared with those obtained for another method [15]. M. Shukla, G. R. Mishra, “DAPSO and PSO-VAF in Linear
of ripple constraint handling based on penalty function. From Phase Digital Low Pass FIR Filter Design”, Circuits and
Systems, Vol. 5, pp. 57-67, 2014.
the simulation results it can be concluded that ripple constraint
[16]. R. Stron and K. Price, “Differential Evolution – A simple and
handling method based on the modified selection rule of DE efficient heuristic for global optimization over continuous
shows better performance than that obtained using DE with spaces”, Journal of Global Optimization, Vol. 11, pp. 341-359,
ripple constraint method based on penalty function. Thus this 1997.
method is seen as an efficient alternative method for ripple [17]. L. D. Arya, S. C. Chaube and R. Arya, “Differential evolution
constraint handling with DE algorithm for FIR filter design. In applied for reliability optimization of radial distribution
future the population size of the DE algorithm can be varied systems”, IJEPES, Elsevier , Vol. 33, Issue 2, Feb. 2011.
and statistically better results are expected. [18]. Q. Zhao, G. Meng, “Design of Digital FIR Filters using
Differential Evolution algorithm based on Reserved Gene”,
8.0 REFERENCES International Conference of Information Science and
[1]. A. V. Oppenheim and R. W. Schafer, Discrete-Time Signal Management Engineering, pp.177-180, 2010.
Processing, 3rd ed. Prentice-Hall of India, New Delhi, 2013. [19]. Z. Albataineh, F. Salem and J. I. Ababneh, “Linear phase FIR
[2]. S. Sharma, L. D. Arya and S. Katiyal, “Design of linear-phase Low Pass Filter Design Using Hybrid Differential Evolution”,
digital FIR filter using Differential evolutional optimization International Journal of Research in Wireless Systems
with ripple constraint”, International conference on (IJRWS), vol. 1, issue 2, pp. 43-49, 2012.
Sustainable Global Development, INDIACom-2014, pp. 474- [20]. K. Singh N. Kaur, “Design of Digital FIR Filters Using
480, March 2014. Differential Evolution Algorithm Based on Reserved Genes”,
[3]. J. Lampinen, “Multi-Constrained Nonlinear Optimization by Vol. 4, Issue 7, pp. 813-819, July 2014.
the Differential Evolution Algorithm”, Journal of Soft
Computing and Industry, Springer, pp. 305-318, 2002.

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 840
 BIJIT - BVICAM’s International Journal of Information Technology
Bharati Vidyapeeth’s Institute of Computer Applications and Management (BVICAM), New Delhi (INDIA)

A Novel Model for Security and Data Access for Jointly Accessing the Cloud Service

Bansi Khimani1and Kuntal Patel2

Submitted in January, 2014; Accepted in December, 2014

Abstract - Cloud computing is a set of resources and services Applications, and services) that can be rapidly provisioned and
offered by Internet. It provides all kinds of services for end released with minimal management effort or service provider
user. One of the most important services provided by cloud interaction.
computing is an Email (Data Storage and File Sharing). This cloud model promotes availability and is composed of five
Employees or any Committee of Institution are very interested essential characteristics (On-demand self-service, Broad
in sharing documents with group members. There is network access, Resource pooling, Rapid elasticity, Measured
possibility of creating one group to share information with Service); three service models (Cloud Software as a Service
all. So, in everybody’s registered email id, they will get (SaaS), Cloud Platform as a Service (PaaS), Cloud
notification for it. Here, everybody have their own mail id. In Infrastructure as a Service (IaaS).There are four deployment
this research paper, a model is discussed which allow one models (Private cloud, Community cloud, Public cloud, Hybrid
email id and two users sharing same Email id. This cloud). Key enabling technologies include: fast wide-area
mechanism is like Joint access of single bank account networks, powerful an inexpensive server computers and high-
between two members. performance virtualization for commodity hardware.”[1]
Cloud service is any resource that is provided over the Internet
Index Terms – Cloud Computing, Cloud services, User Access [2]. According to NIST, a cloud model is composed of three
Control, Joint Access of Cloud data service models – IaaS, PaaS and SaaS.

1.0 INTRODUCTION 1.1 Infrastructure as a Service (IaaS)

Nowadays different state governments and central government This is considered as a first layer of Cloud computing. Using
have taken initiative to successfully implement E-Governance this service model, you manage your operating systems, data,
in various areas of Service applying Information and applications, middleware and runtime. IaaS allows you to
communication Technology to provide better transparency, easily scale based on your requirements and you only pay for
Accuracy and Security of its Services to the citizens [9]. The the resources which you used. This means that extra data
current commercial Systems are aimed mainly at governments processing space is available to you whenever you need it, and
and corporations with high security requirements [10].Internet when you don’t need it then don’t pay for it.
continues to grow and bulk of information is transferred
between individuals. Evolution of smart phone and tablets 1.2 Platform as a Service (PaaS)
make more usage of cloud services. All these technological This layer provides developer the flexibility to make
developments provide new business model which is known as application on the provider’s provided platform. It’s fully
cloud computing. Main idea behind a cloud is to provide on virtualized platform that includes one or more operating
demand service with high reliability, scalability and availability systems, servers and also specific applications. Main features
in distributed environment. Cloud computing entrusts remote offered by PaaS are flexibility, scalability and database. E.g.
services with user’s data, software and computation. Thus it is Google app engine, Amazon web services s3 etc...
just like using some applications or facilities by not directly
installing in devices as we normally do. This system is remote 1.3 Software as a Service (SaaS)
version of remote access [8]. This layer delivers single software to multiple clients on
National Institute of Standards and Technology (NIST) defines demand via web browser over Internet. So, Software as a
Cloud computing as: Service consists of a software running on the provider’s cloud
“Cloud computing is a model for enabling convenient, on- infrastructure. E.g. Google docs, salesforce.com etc...
demand network access to a shared pool of configurable
computing resources (e.g., networks, servers, storage, 2.0 LITERATURE SURVEY
People use cloud because it provides on-demand services with
1 high reliability, scalability and availability in distributed
R. K. University, School of Computer Science, Bhavnagar
Highway, Kasturbadham Road, Rajkot 360020, INDIA. environment. Here, in this research paper, we start with survey
2 of major cloud providers and Authentication Techniques
Ahmedabad University, School of Computer Studies,
Commerce Six Roads, Ahmedabad 380009, INDIA. adopted by various providers. After extensive literature survey
E-mail: [email protected] and related to cloud security, we finally proposed a model related to
2 “Joint Access of Cloud Data” which is explained in this paper.
[email protected]

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 841
 BIJIT - BVICAM’s International Journal of Information Technology

2.1 Examples of Cloud Service Providers choose your password unhackable. Unhackable password is
There are many features of cloud computing. Cloud storage combination of Alpha-numeric characters, special symbols and
providers like Amazon S3, Microsoft SkyDrive, and DropBox difficult to imagine by intruder too. It is also known as one
permit consumers to access data online. Second feature is, it factor authentication.
provides computation resources for users such as amazon EC2.
Third, Google apps or versioning repositories for source code 3.2 Two Factor Authentication
are examples of online collaboration tools. Username and passwords are not enough to secure your online
Cloud service providers should ensure the security of their data. Two factor authentications must have feature for any
customer’s data and should be responsible if any security risk successful and popular service to protect it against password
affects their customers’ service infrastructure. [5]Cloud phishing, hacking and account hijacking. [3] There are several
providers must ensure that the information Security Systems solutions i.e. One Time Password (OTP), Authenticator app,
they provide are responsive to customer requirements and the SMS and email codes, Security questions, Device recognition
data, both primary and secondary, must be preserved as etc. which are not costly and secure enough too.
authentic and reliable [11].
• Several vendors with cloud offerings stand to benefit from 3.3 Biometrics Verification
this trend. Many of these vendors – Amazon, IBM, Biometric Verification enables identification based on “who
Microsoft etc. have established cloud products and have you are”. Every person has distinguishing and measurable
been active in this space for number of years. physical traits. Personal recognition based on unique physical
attributes forms a powerful tool for identity management. Other
ways of verifying authorization include “what you have” (a
key, a swipe card) and “what you know” (a password, your
mother’s maiden name).Biometrics is the only mode of
authentication that uses “who you are” for verification. [4]

4.0 PROPOSED TECHNIQUE AS “JOINT ACCESS OF

CLOUD DATA”
“Ad hoc networking” is popular, which allows device to
establish communication, anytime, anywhere without the aid of
a central infrastructure [7]. Today, many people carry numerous
portable devices, such as laptops, mobile phones, PDAs and
MP3 players, for use in their professional and private lives. For
the most part, these devices are used separately i.e. their
applications do not interact.
Now imagine, however if they could interact directly.
Participants at a meeting could share documents or
presentations; all communication could automatically be routed
through the wireless corporate campus network [7].So, like
Figure 1: Major Cloud Vendor Used in 2012[6] sharing documents and all these things if happen via single
mail id with separate password mechanism then how much it
From above Figure 1, we can observe that Microsoft is likely to will be helpful to circulate information between groups of
gain most from a broader adoption of the cloud. Of all students and participants or user. So, Proposed model follow
respondents, 68% of all respondents who expect to move somewhat same concept i.e With Single Shared Mail ID user
workloads or provision new ones to the cloud environment can access shared data by their personal password mechanism.
mention Microsoft as their preferred vendor of choice [6]. Generally single user has single id and single password to
Report Published in 2014 says that in Microsoft’s Storage – access cloud service. But in proposed model, user id will be
OneDrive, If Privacy is major Concern then it should be noted single and passwords will be infinite.
that Microsoft reserve the right to scan your files to look for Following Simple Steps shows how our proposed model will
what it would deem Objectionable Content. This could be work to access jointly accessing cloud based data:
copyrighted Material or things of an explicit nature. Apple has Step 1: Input User ID and Password.
similar policy, making the two potentially more intrusive than Step2: System will compare (by password) about which user
their competitors [12]. want to access cloud data.
Step 3: System will forward Barcode image File to registered
3.0 TECHNIQUES FOR USER ACCESS CONTROL
email id of Particular user.
3.1 Username and Password
Step 4: User will provide correct barcode image file
Unique username is provided to user for accessing services. For
security purpose, password is a powerful mechanism if you Step5: System will check that verification of correct barcode

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 842
A Novel Model for Security and Data Access for Jointly Accessing the Cloud Service

file upload or not? password is of which user. E.g. If Id [email protected]

Step 6: If step 5 is correct then User can access cloud based is shared between guide and PhD scholar.
data otherwise he will perform again from step 1. We assume that Guide’ password is “guide101” and Ph.D.
scholar’s Password is “student101”. So web service will check
Above 5 steps are explained in Figure 2 which shows flow to that which password is entered by user. If User entered
retrieve data from cloud. Here, multifactor authentication is “guide101” then barcode image file will be sent to registered
applied on cloud. email id of guide. If User entered “student101” then barcode
image file will be sent to registered email id of Ph.D. scholar.
This registered email id is personal Id of each user. If password
is wrong, then web service will assume that user is not valid.
So again user will need to enter valid id and password for
accessing cloud data.
After accessing barcode image file, user will import that
barcode image file (.jpeg) and if that code is valid for that user
then that user will access their data. If uploaded file is not valid
then user will not be able to access their data.
Sometimes it may happen that “Example University” want to
share their Exam schedule between guide and student then they
will just mail on jointly access mail id which is
[email protected].
Whenever guide or student needs to access this account they
will enter joint id and their own password. If they forget their
password then they can request their admin to send reset
password link into their registered email id.

4.1 Advantages and Disadvantages of Proposed Model

Advantages
• It is not costly compare to other high security methods.
• This model reduces number of email ids. (I.e. sharing of
email id reduces numbers of email ids).
• This model is easy to implement.
• It provides high security then single factor
authentication.
• Barcode cannot be altered or predicted by human being.
So, if text based security is provided to user then code
can be altered by human being intentionally or
unintentionally.
• This model is reliable.
• Any organization or person can share data with multiple
users having same id and different passwords.
Disadvantages
• Sometimes user may feel bore to upload and download
barcode image file.
• Many times users don’t like to have same email id.
• Network overhead will be increased compare to Present
System of 2 Step Verification.

5.0 CONCLUSION AND FUTURE WORK

In this paper, three main cloud service models are described.
Figure 2:Flowchart of “Joint Access of Cloud Data” Data security is big hurdle in cloud. It is cloud provider’s duty
From Figure 2 we can see that, to access any cloud service or to keep user’s data safely. This model is helpful to those
data, user needs to enter their Id and Password. In this model, organizations where single cloud based data need to be shared
User Id will be shared between multiple users and each user between multiple users. But, sharing should be securely. These
will have their own password. So, when user will enter their users can be of same field, same region or from same caste or
user Id and password, web service will check that entered community. Here, multifactor authentication is used to add

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 843
 BIJIT - BVICAM’s International Journal of Information Technology

second layer of security while sharing data with people.

This model is one of the small steps from our side to enhance
cloud security and accessing jointly access of cloud based data.
Presently we had proposed this model, but in near future we
have plan to test this proposed model on live cloud.

6.0 ACKNOWLEDGEMENT
We would like to thank Mr. Raghu Khimani – Cyber Crime
Expert - for their valuable suggestions.

7.0 REFERENCES
[1]. http://www.nist.gov/itl/cloud/ US: Department of
Commerce.: viewed on 14-November-2014
[2]. http://searchcloudprovider.techtarget.com/definition/clo
ud-services/:viewed on 15- November-2014
[3]. http://www.secureauth.com/blog/cloud-storage-2-factor-
authentication-review/:viewed on 10-December-2014
[4]. “Benefits of Methode Biometric Verification
Technology” Article of “Methode Electronics,
Inc.”http://www.methode.com/Documents/TechnicalLib
rary/Methode_Biometrics_-_Benefits_&_FAQs.pdf
viewed on:18-November-2014
[5]. A Mohammed, P Eric, S Ben, T Hanes, “Cloud
computing security: From single to Multi – Clouds”,
2012 45th Hawaii International Conference on System
sciences, pp.5490-5499, 2012
[6]. S Hemalatha, R Manickachezian, “Present and Future of
cloud computing: A collaborated survey report”, IJITEE,
Vol 1, Issue 2, pp. 216-223, July 2012
[7]. H Ashema , “Study of Impact of Mobile Ad – hoc
networking and its future Applications”, BIJIT –
BVICAM’s International Journal of Information
Technology , Vol 4, Issue 7, pp. 439- 444, January –
June 2011
[8]. R Vaibhav, “Innovative Use of cloud computing in
Smart Phone Technology”, BIJIT – BVICAM’s
International Journal of Information Technology, Vol 5
No 2, pp. 640-648
[9]. S Sirsendu , K Sunil, “Applications of Public Key
Watermarking for Authentication of Job-Card in
MGNREGA”, BIJIT – BVICAM’s International Journal
of Information Technology, Vol 4 No 1, pp. 435 – 438 ,
January – June 2012
[10]. S Dilbag, S Ajit , “ An Effective Technique for Data
Security in Modern Cryptosystem” , BIJIT – BVICAM’s
International Journal of Information Technology, Vol 2
No 1, pp. 189- 194, January – June 2010
[11]. D Meenu, D Mikku, Y.S.Shishodia, “Cloud Computing
and Knowledge Management as a Service: A
Collaborative Approch to Harness and Manage the
Plethora of Knowledge”, BIJIT – BVICAM’s
International Journal of Information Technology, Vol 5
No2, pp. 619-622, July – December 2013.
[12]. http://www.pcadvisor.co.uk/features/internet/3506734/b
est-cloud-storage-dropbox-google-drive-onedrive-
icloud/: visited on 15/02/2015

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 844
 BIJIT - BVICAM’s International Journal of Information Technology
Bharati Vidyapeeth’s Institute of Computer Applications and Management (BVICAM), New Delhi (INDIA)

Predictive Estimates of Employees’ Intelligence at Workplace with Special Reference to

Emotional and Spiritual Intelligence
B. K. Punia1 and Priyanka Yadav2
Submitted in June, 2014; Accepted in December, 2014
Abstract - The piece of writing investigates the relationship in job. Earlier researchers have discovered that emotionally
between employees’ emotional and Spiritual intelligence. A healthy individuals are better learners, better at resolved
conversation of spirituality and emotions within the conflicts, a reduced amount of aggressive and added
workplace can be an unthinkable topic. However, emotional understanding, feel better about themselves, have better desire
intelligence and spiritual intelligence are, at present, more control, are less involve in destructive behavior also have few
widely acknowledged. Drawing a research connected with problems related to behavior. For employees, emotional
these construct we suggest that emotional intelligence within intelligence offers an essential foundation for developing the
the employees in organisations may provide employees with a skills of effective independent learning and personal
medium to better understand and mix spiritual experiences development. Spirituality survives in the hearts and minds of
within their work. Study further explain that spiritual individual everywhere, within religious traditions and
intelligence in employees also simultaneously improve the independently of traditions. Spirituality defines as the field of
expression of emotional intelligence at work because result final concern, and then everyone is spiritual because everyone
shows that emotional and spiritual intelligence are has ultimate concerns which can be understand in different
significantly and positively correlated to each other. Also by ways. The investigations into spiritual intelligence suggested
doing regression analysis, taking emotional intelligence as that it is one of several types of intelligence and that it can be
dependent variable or spiritual intelligence as independent developed relatively independently. Spiritual intelligence also
variable and vice-a-versa model revealed that spiritual helps to knowing others. It can be refined through searching,
intelligence predict approximately 80 percent of emotional questioning, and practice.
intelligence of employees and emotional intelligence predict Emotional Intelligence (EI) has come to mean many things to
approximately 51 percent of spiritual intelligence whereas many different people. For some people, it is about being a
another 49 percent due to another factors which are not nice people whereas other people see it as a dilemma and they
explained in the study. find it hard to believe that emotions can be intelligent.
Emotional intelligence is indication as a predictor of success in
Index Terms – Intelligence, Emotional Intelligence, Spiritual a number of life domains, including work Goleman1 who
Intelligence, Employee, Workplace. conceptualized emotional intelligence as the capacity for
recognizing our own feelings and those of others, for
1.0 INTRODUCTION motivating ourselves and for managing emotions well in
In any of the organization, for educators, experts, corporate ourselves and our relationship. Individuals first notice the term
instructors, recruiters, managers there are some reasons for "emotional intelligence" around 1995 with the publication of
their extra performance apart from average performance. It is Daniel Goleman’s best-selling book Emotional Intelligence
not the scientific skills, it is not intelligence either rather it is who laid out a powerful case that such factors as self-
something else that cannot be defined in words i.e. personnel awareness, self-discipline, and empathy determine personal and
skills. In literary terminology intelligence can be understood as professional success. Based on BarOn’s [2] work, “emotional
one’s capacity for learning, reasoning and understanding. Here intelligence is an array of non-cognitive capabilities,
in the similar forms of mental activity; aptitude in grasping competencies and skills that influence one’s ability to succeed
truths, relationships, facts, meanings may also be included. in cop up with environmental demands and pressures”. The
After talking a long time about personnel skills in organizations working definition of emotional intelligence as given by
our approach to examine what makes people shine at Mayer, Caruso & Salovey [3] is ‘an ability to recognize the
workplace indicates that we can replace these personal skills meanings of emotions and their relationships, and to reason and
with objective and measurable terms i.e. emotional and problem-solve on the basis of them’. An alternative model has
spiritual intelligence. There are few of emotional intelligence been suggested to encompass a range of ‘non-cognitive
and spiritual competences that have been related with success capabilities, competencies and skills which influence one’s
ability to succeed in coping with environmental demands and
1
Professor, Haryana School of Business, Guru Jambeshwar pressure’ by Bar-On [2] which includes intrapersonal qualities
University of Science & Technology, Hisar-125 001. (e.g. emotional self-awareness), interpersonal (e.g. empathy),
2
Research Scholar, Haryana School of Business, Guru adaptability (e.g. problem-solving), stress management (e.g.
Jambeshwar University of Science & Technology, Hisar- 001. tolerance to stress) and general mood. Primarily Mayer and
E-mail1 [email protected] and Salovey [4] the inventors of the theory of emotional
2 intelligence, propounded an ability model of emotional
[email protected]

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 845
 BIJIT - BVICAM’s International Journal of Information Technology

intelligence which defines it as the capability to reason with, intelligence which helps us solve the problems of life
and concerning emotions. For them, emotional intelligence significance and value the concept of our action and life. The
combines feelings with thinking, and thinking with feeling. theory of positive disintegration having relevance to spiritual
Goleman5 described the five main components of emotional intelligence illustrated it as the one’s capability to discard the
intelligence as self-awareness, self-regulation, empathy, social normal ways of thinking and behaving for want of compassion,
skills and motivation that further assented that “a learned integrity and altruism, however the theory on self-actualization
capability based on emotional intelligence those results in emphasized the values such as justice, truth & beauty,
outstanding performance at work. Our emotional intelligence wholeness, and uniqueness. On the similar lines the humanist
determines our potential for learning the practical skills based psychology emphasized the central point of the individual’s
on the five elements: self-awareness, motivation, self- instinctive drive to be a better person through values of
regulation, empathy, and adeptness in relationships. Mayer and openness, caring, and the desire for entirety of life, body, mind
Salovey[4] coined the theory of emotional intelligence which and spirit. The theory of flow i.e. when a person’s entire being
defines emotional intelligence as the capability to reason with, is stretched in full functioning of body and mind all doings of
and concerning emotions. For them, emotional intelligence the individual become worth doing for its own sake
combines feelings with thinking, and thinking with feeling. (Dabrowski’s[9];Maslow[10];Rogers[11];and Csíkszentmihályi
Accordingly emotional intelligence is the ability to perceive [12]). The researches carried out in recent years have fostered
emotions, to access and generate emotions so as to assist increasing awareness to the study of spiritual intelligence and
thought, to understand emotions and emotional knowledge, and its application in various fields specially in working places.
to reflectively regulate emotions so as to promote emotional King [13] is defined spiritual intelligence such as:” Spiritual
and intellectual growth. intelligence is as a set of intellectual capacity that contributes to
The mix model by Goleman [6] of emotional intelligence is the integration and knowledge and adaptive application of
most widely used and accepted by people. It occupies a variety spiritual and supernatural aspects of personal existence and
of competencies which are divided in two different skills sets. lead to critical existential thinking, increased meaning, identify
Figure-2 summarized Goleman’s model. of superior universe and dominated of spiritual. In this study
four dimensions as explained in Figure-3 have been used to
Self awareness Self management measure spiritual intelligence.
Critical Existential Refers to the ability to critically
Emotional self Self control, Trustworthiness, Thinking think about the reality of
awareness, Accurate Conscientiousness, Adaptability, (CET) existence, being, universe, time,
self assessment, Self Achievement orientation, death, and other beyond
confidence Initiative normal and existential issues.
Personal Meaning It is the ability to create personal
Social awareness Social skill Production purpose and the goal existed in
(PMP) all mental and physical
Empathy, Developing others, experiments, including the ability
Organizational Communication, Change catalyst, to create and to dominate over
awareness, Service Influence, Leadership, Building the goal of life
orientation bonds, Teamwork Transcendental TA is the ability to understand
Figure 2: Goleman’s Emotional Intelligence competencies Awareness transcendental aspects in
Source: Goleman (2010) (TA) ourselves, others and the physical
world, when you are alert and
Spiritual intelligence means recognizing the work life principle awake.
and the seasons of life and may be termed as the expression of Conscious State Refers the ability to enter into
innate spiritual qualities through ones thoughts, actions and Expansion the levels and states of
attitude which expands ones capacity to understand others at spirituality and beyond
(CSE)
the deepest level. Spiritual understanding allows distinguishing consciousness and also coming
both the 'true cause' of behavior without judgment, and serve out of it when you intend.
the 'true needs' of others until they themselves learn to meet
their own needs. This capacity is developed by first learning to Figure 3: Dimensions of Spiritual Intelligence
free you from attachment and neediness and being able to meet Source: King (2008)
our own inner needs. The word spirituality has its roots in spirit 2.0 RELATED STUDIES
which means “nature of mind or outlook”, “a mental The relationships of biographical information with emotional
disposition characterized by firmness or forcefulness” Covey7. intelligence in two opposite professions of social work and
Zohar and Marshal8 define spiritual intelligence as the software was examined by Singh14 which revealed that there is

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 846
Predictive Estimates of Employees’ Intelligence at Workplace with Special Reference to Emotional and Spiritual Intelligence

no significant difference in emotional intelligence of social Sarlak [26]; Modassir & Singh [27]; Chin, Anantharaman, &
work and software professionals. The study by Naghari & Tong [28]; Yunus[29]; Salarzehi, et al. [30]; and Yoghoubi et
Radzum [15] examined that boys are less emotional intelligent al. [31] (as cited in Yadav and Punia [32]) highlighted that the
in comparison to girls and also emphasis that high emotional five aspects of emotional intelligence had significant positive
intelligence in boys enhance their achievement. The research impact on employees’ organizational citizenship behaviour.
by Akintayo [16] found that managerial effectiveness was Emotional intelligence was significantly correlated with
appreciably predicted by emotional intelligence. Emotional conscientiousness, civic virtue and altruism and also mentioned
intelligence has been found linked with job performance and that emotional intelligence of leader enhance followers’
also individuals’ high emotional intelligence reported a more organizational citizenship behaviour. Results also showed that
positive experience. However the relationship of employees’ level of emotional intelligence varies from industries to
perceptions of supervisor’s emotional intelligence with a industries because of their unique characteristics of individuals.
number of important organizational outcomes pointed out those Gender plays a significant role in emotional intelligence and
organizations in the UAE which might be benefited by organizational citizenship behaviour of supervisors and their
developing emotional intelligence skills in their leaders (Gupta subordinates and three dimensions of emotional intelligence i.e.
[17]; Whiteoak and Rana [18]). The study by Vakola, et al. [19] self control, self motivation and empathy have major effect on
explored that the “big five” dimensions of personality in organizational citizenship behaviour.
association with emotional intelligence can smooth the Gardner [33] acknowledged that there was some neurological
progress of organizational change at an individual level. The evidence for “spirituality” in patients with temporal lobe
study by Singh, Punia and Kumar [20] conducted on epilepsy as far as spiritual intelligence is concerned and they
supervisory personnel proposed the cultivation of an emotional tend to become contemplative and develop and increasingly
quotient and the desired organizational climate to enhance the tending toward the study of philosophy and religion and the
employees' mental health vis-à-vis the organizational persistent evaluation of deep questions. However Persinger34
effectiveness. The findings of the study highlighted that some and Ramachandran et al. [35] established a relationship
of the dimensions of organizational climate (performance between heightened temporal lobe activity and the spirituality
standards, reward system, decision-making process etc.) and wherein the ideas on spirituality were apparent in different
emotional competency (adequate depth of feeling, belief systems of ancient civilizations, religions, and the
encouragement of positive emotions etc.) have meaningful indigenous cultures. This entire phenomenon led to exhibit the
relationships with the anxiety level of the personnel. Khatri and connectedness or the belief that everything is part of one
Duggal [21] investigated the role of transformational leadership cosmic whole which is also acknowledged by contemporary
and explored that the factors like its effectiveness, society of the inherent wisdom and spirituality of indigenous
innovativeness & quality improvement and the transactional people within the Australian Aboriginal communities. The
leadership have direct linage with each other through the connection of the society to the land and values that place
relationship was not very strong. The result further indicated collective needs above those of the individual was apparent and
that for maximum satisfaction and performance from for many indigenous cultures, adolescence is a significant time
employees, the leader must exhibit charisma and intellectually in the spiritual life cycle and rites of channel guide people in
strong motivation. Anari [22] investigated the inter-relationship the journey from childhood to adulthood (Kerr & Mcalister
of emotional intelligence, job satisfaction, and organizational [36]). Doostar et al. [37]; Sara Piroozfar [38]; Chen & Yang
commitment, among high-school English teachers. The role of [39]; and Rastgar, et al. [40] (as cited in Yadav and Punia [32])
gender and age in emotional intelligence, the study went in highlighted spirituality as a predictor of organizational
support of gender differences, with females reporting higher citizenship behaviour and showed that there are significant
emotional intelligence same as a resulted by Punia [23] who differences between different levels of workplace spatiality and
glimpsed that women have more emotional intelligence due to organizational citizenship behaviour. Values, attitudes, and
their high score on the dimensions of emotional intelligence, on behaviors of leaders have positive effects on spirituality of the
the other hand in relation to their age Punia [23] reveals a employees and relationship among workplace spirituality,
parabolic trend with emotional intelligence whereas Anari22 organizational citizenship behaviour and job performance of
results revealed no differences among the participants employees was positive and workplace spirituality had a
according to their age. Emotional intelligence is also related to significant positive influence. The studies as reviewed above
other important work outcomes namely family-work conflict, specify that emotional and spiritual intelligence is not only an
goal conflict, frustration and readiness to create and innovate. important predictor of employees’ success but also
Significant differences with regard to the perception of organizational success. The relationship and prediction of
emotional intelligence, conflict, readiness to create and employees’ emotional and spiritual intelligence is also a
innovate have been established across the background of significant factor for job performances. Hence keeping this and
participants discovered employee’s with higher levels of the added significance of emotional and spiritual intelligence in
emotional intelligence reported to lower levels of conflict and view, the current research has been carried.
higher levels of readiness to create and innovate (Hess &
Bacigalupo [24]; Suliman & Al-Shaikh [25]). Sahafi, Danaee &

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 847
 BIJIT - BVICAM’s International Journal of Information Technology

3.0 OBJECTIVES AND METHODOLOGY only emotional intelligence, can explain 51.6 % of it. That
The present study aims to predict the employees’ emotional means other 48.4 % part of spiritual intelligence can be
and spiritual intelligence and also find out the relationship explained/predicted by some other factors but approximately
between these two intelligences. The incidental objectives half of spiritual intelligence predicted by only emotional
which contributed to the prime objective have been given as intelligence. The value of f- ratio is 38.35 and significant value
under: is 0.000* which is significant at 5 % level of significance that
1. To study the relationship between employees’ emotional means our regression model result is significantly better
and spiritual intelligence vis-à-vis their dimensions. predictor of spiritual intelligence.
2. To explore the predictive estimates of emotional and The Table-3 depicts the results of regression analysis with
spiritual intelligence of employees’ while taking them spiritual intelligence as independent variable and the emotional
dependent and independent variables on reversal basis. intelligence as the dependent variable.
This research is based on sample size of 149 employees which It can be seen from the Table-3 wherein emotional intelligence
encompass 90 male and 59 female respondents from different is taken as a dependent variable and spiritual intelligence as
organizations. The present study is based on primary data independent variable, results of regression analysis revealed
collected through well-structured research instruments that that spiritual intelligence dimensions predict 80% (R square =
measures the related variables of emotional and spiritual 0.0800) of the emotional intelligence. There might be many
intelligence. The scale designed by Goleman1 has been used to factors that can explain emotional intelligence of the employees
examine the emotional intelligence of employees which consist but as per the present study only spiritual intelligence, can
of 32 statements and represent its four different dimensions. explain up to 80.0 % of it. It denotes that means 80.00 % part
The inventory on spiritual intelligence developed by King13 has of emotional intelligence can be explained/predicted by one’s
been used which consist of 24 statements representing four spiritual intelligence only and rest of the 20 percent emotional
different dimensions. The collected data was analyzed using intelligence is predicted by other factors. The value of f- ratio is
relevant statistical techniques like mean score, standard 144.32 and significant value is again 0.000* which is also
deviations (SD) and Pearson’s correlation coefficients (r) and significant at 5 % level of significance. In short, the results
regression analysis as per the requirements of the study. pointed that 80 % of emotional intelligence of employees is
explained by spiritual intelligence. While giving a summative
4.0 RESULTS AND DISCUSSION note it can be viewed that if an individual is spiritually
Objective-1: To study the relationship between employees’ intelligent his chances of being emotionally intelligent are 80
emotional and spiritual intelligence vis-à-vis their dimensions. percent. However one’s being emotionally intelligent his
To study the relationship between employees’ emotional and spiritual intelligence can be only approximately 50 percent
spiritual intelligence, correlation analysis has done and the (51.6%). Therefore, spiritual intelligence carries more weight
results have been shown as per the Table-1. As shown in Table- as compared to emotional intelligence.
1, out of four dimensions of spiritual intelligence only one When we compare the findings of the present study with
dimension i.e. Critical Existential Thinking (CET) was not previous studies of the nature some similar and interesting
found correlated with self-awareness dimension of emotional findings have emerged. The result of present study showed
intelligence. However all the dimensions of spiritual that emotional and spiritual intelligence were highly correlated
intelligence except the one mentioned above have been found which replicate the findings of pervious researches of Kaur et
positively correlated with the all dimension of emotional al. [41], Koohbanani et al. [42], King et al. [43], and Harmer
intelligence. It signifies that both these intelligences are [44] who also found the positive and significant relationship
significantly and positively correlated to each other. between emotional and spiritual intelligence. The employees
Objective-2: To explore the predictive estimates of emotional having the knowledge of their divine, inner sense, critical
and spiritual intelligence of employees’ while taking them thinking about the reality, time and death, creating goal in
dependent and independent variables on reversal basis. mental and physical experiments, ability to enter and exist in
To estimate the effect of one’s spiritual intelligence on consciousness i.e. awareness of something and responding to
emotional intelligence and vice-a-versa regression analysis has one’s surroundings and construct a life purpose is possible only
been done while taking the two intelligences dependent and when he knows about himself or herself, about his emotions or
independent variables alternatively and the results of the same other’s emotions and feelings. All emotions are important on
have been shown as per Tables-2&3. The Table-2 depicts the the job, not just fear or worry. The answer is that emotions
results of regression analysis wherein emotional intelligence express information which you need to be effective in your
has been taken as independent variable and the spiritual work. Another major finding of present study is about
intelligence as the dependent variable. predictive estimates of emotional and spiritual intelligence of
It can be witnessed from the Table-2 that the value of R square employees’ which is in consonance with the finding by David
is 0.516, which denotes that four dimensions of emotional [45] who also found that emotional intelligence is not only the
intelligence can predict 51.6 % of the spiritual intelligence. individual predictor of workplace success also spirituality that
There might be many factors that can explain spiritual means knowing the things other than emotional awareness is
intelligence of the employees but our model, which includes also important.

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 848
Predictive Estimates of Employees’ Intelligence at Workplace with Special Reference to Emotional and Spiritual Intelligence

5.0 CONCLUSION 8. Zohar, D. and Marshall, I. (1999). SQ-Spiritual

Intelligence can be understood as one’s capacity for learning, Intelligence- The Ultimate Intelligence, London:
reasoning and understanding. Here in the similar forms of Bloomsbury Publishing.
mental activity; aptitude in grasping truths, relationships, facts,
meanings may also be included. Researchers have discovered 9. Dabrowski, K. (1967). Personality Shaping Through
that emotionally healthy individuals are better learners, better Positive Disintegration, Boston: Little & Brown.
at resolved conflicts, a reduced amount of aggressive and added 10. Maslow, A. H. (1968). Toward a Psychology of
understanding, feel better about themselves, have better desire Being. New York: Harper & Row
control, are less involve in destructive behavior also have few
problems related to behavior. For employees, emotional 11. Rogers, C., (1980). A Way of Being, Boston:
intelligence offers an essential foundation for developing the Houghton Mifflin.
skills of effective independent learning and personal
12. Csíkszentmihályi, M. (1996). Creativity: Flow and the
development whereas spirituality survives in the hearts and
Psychology of Discovery and Invention, New York:
minds of individual everywhere, within religious traditions and
Harper Perennial.
independently of traditions. Spirituality in fact delineates as the
field of final concern, and then everyone is spiritual because 13. King, D. (2008). “The Spiritual Intelligence Self-
everyone has ultimate concerns which can be understand in Report Inventory (SISRI-24)”. Available at
different ways. In the present study emotional and spiritual http://www.dbking.net/spiritualintelligence (Retrieved
intelligence were significantly positively correlated and also on April 08, 2013).
found that emotional intelligence predict only approximately
half of spirituality of employees’ while spiritual intelligence 14. Singh, S.K., (2008). “Biographical Information as
predict most of things other emotional intelligence of an Correlates of Emotional Intelligence: A Study in
employees. Social Work and Software Professions”, Indian
Journal of Industrial Relations, 43(3), 385-405.
6.0 REFERENCES 15. Naghavi, F. and Redzuan, M., (2011). “The
1. Goleman, D. (1988). The Meditative Mind: The Relationship between Gender and Emotional
Varieties of Meditative Experience, Newyork: J.P. Intelligence”, World Applied Sciences Journal, 15(4),
Terchor, Inc., p. 214. 555-561.
2. Bar-On, R., (2000). “Emotional and Social 16. Aukler, A. (2010). The Influence of Personality Factor
Intelligence: Insights from the Emotional Quotient on Organisational Citizenship Behaviour,
Inventory (EQ-i)”. In R. Bar-On & J.D.A. Parker unpublished report available at
(eds.), Handbook of emotional intelligence. San www.poczynek.org/.../2010%20Judit %20Aykler%20-
Francisco: Jossey-Bass, pp. 363-388. %20Organisationa... (retrieved on March, 22, 2013),
3. Mayer, J. D., Salovey, P. and Caruso, D. R. (2000). pp. 1-67.
“Emotional Intelligence as Zeitgeist, as Personality, 17. Gupta, R. (2012). “Analytical Study on Emotional
and as a Mental Ability”. In Bar-On, R. & Parker, J. Intelligence in Organisational Research”, VSRD-
D. A. (Eds.), Handbook of Emotional Intelligence, San IJCSIT, 2(5), 433-436.
Francisco: Jossey-Bass, pp. 92-117.
18. Whiteoak, J.W. and Manning, L.R. (2012).
4. Mayer, J. D. and Salovey, P. (1997). “What is “Emotional Intelligence and Its Implications on
Emotional Intelligence?” In Salovey, P. & Sluyter, D. Individual and Group Performance: A Study
(Eds.), Emotional Development and Emotional Investigating Employee Perceptions in the United
Intelligence: Implications for Educators, New York, Arab Emirates”, The International Journal of Human
NY: Basic Books, pp. 3-34. Resource Management, 23(8), 1660-1687.
5. Goleman, D. (1995). Emotional intelligence, New 19. Maria, V., Tsaousis, I. and Nikolaou, I. (2004). “The
York: Bantam Books. role of Emotional Intelligence and Personality
6. Goleman, D. (2010). “ Emotional Intelligence A Variables on Attitudes toward Organisational
Bantam book” 10, illustrated, reprint, annotated, Change”, Journal of Managerial Psychology, 19(2),
Bantam Books. 88-110.

7. Covey, S. R. (2009). The Leader in me: How Schools 20. Singh, S., Punia, B. K. and Kumar, R. (2006).
and Parents around the World are Inspiringgreantness, “Emotional Competence and Organisational Climate
one Child at a time, New York, NY: Frankline covey in Relation to Anxiety among the Supervisory
Co. Personnel: An Inter-Correlational Analysis,” The IUP
Journal of Organisational Behaviour, V (3), 74-83.

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 849
 BIJIT - BVICAM’s International Journal of Information Technology

21. Khatri, P. & Duggal, S. (2012) “A Test of Intelligence (EI)”, Modern Applied Science, 5(2), 119-
Transactional and Transformational Leadership 123.
Behaviour of Salesman on Customer Relationship
Marketing Behaviour: A Study of the Indian Banking 32. Yadav, P. & Punia, B.K. (2012). “Organisational
Sector”, BIJIT-BVICAM’s International Journal of Citizenship Behavior: A Review of Antecedent,
Information Technology, 4(1), 427-434. Correlates, Outcomes and Future Research
22. Anari, N. N. (2012). “Teachers: Emotional Directions”, International Journal of Human
Intelligence, Job Satisfaction, and Organisational Potential Development, 2(2), 01-19.
Commitment”, Journal of Workplace Learning, 24(4), 33. Gardner, H., (1983). Frames of Mind, New York:
256-269. Basic Books.
23. Punia, B. K., (2005). “Impact of Demographic 34. Persinger, M. A. (1996). “Feelings of past lives as
Variables on Emotional Intelligence and Leadership Expected Perturbations within the Neurocognitive
Behaviour of Corporate Executives”, The IUP Journal Processes that Generate the Sense of Self:
of Organisational Behaviour, IV(2), 7-22. Contributions from limbic lability and Vectorial
24. Habibollah, S., Yaghoubi, N. M., Naroei, M. and Sin, Hemisphericity”. Perceptual and Motor Skills, 83(3),
L. G. (2011). “A Survey of Relationship between 1107-1121.
Emotional Intelligence and Organisational 35. Ramachandran, V. S. and Blakeslee, S. (1998).
Citizenship Behaviour in Iran”, International Business Phantoms in the Brain, New York: William Morrow,
and Management, 3(1), 130-135. pp. 1-328.
25. Suliman, A.M. and Al-Shaikh, F. (2007). “Emotional 36. Kerr, B. and McAlister, J. (2002). Letters to the
Intelligence at Work: Links to Conflict and Medicine Man: An Apprenticeship in Spiritual
Innovation”, Employee Relations, 29(2), 208-220. Intelligence, Cresskill, NJ: Hampton Press.
26. Sahafi, E., Danaee, H. and Sarlak, M. A. (2011), “The 37. Doostar, M., Chegini, M. G. and Pourabbasi, S.
Impact of Emotional Intelligence on Citizenship (2012). “Survey of Relationship between Spiritual
Behaviour of Physicians (With Emphasis on Infertility Intelligence and Organisational Citizenship
Specialists)”, Journal of Family and Reproductive Behaviour”, Interdisciplinary Journal Of
Health, 5(4), 109-115. Contemporary Research In Business, 3(1)1, 54-61.
27. Modassir, A. and Singh, T. (2008). “Relationship of 38. Piroozfar, S. (2013). “The Influence Spirituality in
Emotional Intelligence with Transformational Workplace on Employee’s Organisational
Leadership and Organisational Citizenship Citizenship Behaviour”, Trends in Social Science, 6
Behaviour”, International Journal of Leadership (1), 26-29.
Studies, 4(1), 3-21.
39. Chen, C. and Yang, C. (2012). “The Impact of
28. Chin, S., Anantharaman, R. N., and Tong, D. (2011). Spiritual Leadership on Organisational Citizenship
“Emotional Intelligence and Organisational Behaviour: A Multi-Sample Analysis”, Journal of
Citizenship Behaviour of Manufacturing Sector Business Ethics, 105(1), 107-114.
Employees: An Analysis”, Management, 6(2), 107-
125. 40. Rastgar, A., Zarei, A., Davoudi, S.M. and Fartash, K.
(2012). “The Link between Workplace Spirituality,
29. Yunus, Noorlaila, H. (2010). “The Influence of Organisational Citizenship Behaviour and Job
Supervisors’ Emotional Intelligence on Non Bank Performance in Iran”, A Journal of Economics and
Supervisory Employees’ Organisational Citizenship Management, 1(6), 51-67.
Behaviour”, Paper Presented in International
Conference (CSSR) on 5-7 December. 2010, 663-667. 41. Kaur, H., Singh, V. and Singh, P. (2012). “Emotional
Intelligence: Significance of Psychology and
30. Salarzehi et al., (2011). “A Survey of Relationship Between Spirituality”. Pakistan Journal of Social and Clinical
Emotional intelligence and Organizational Citizenship Psychology, 9(2), 32-36.
Behavior in Iran”, International Business and Management,
3(1),130-135. 42. Koohbanani, S. E., Dastjerdi, R., Vahidi, T. and Far,
M., H., G. (2013). “The Relationship Between
31. Yaghoubi, E., Mashinchi, S. A. and Hadi, A. (2011). Spiritual Intelligence and Emotional Intelligence with
“An Analysis of Correlation between Organisational Life Satisfaction Among Birjand Gifted Female High
Citizenship Behaviour (OCB) and Emotional School Students ”. Procedia - Social and Behavioral
Sciences 84(2013), 314-320.

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 850
Predictive Estimates of Employees’ Intelligence at Workplace with Special Reference to Emotional and Spiritual Intelligence

43. King, D., Mara, C. A. and Decicco, T., L. (2012). What’s the relationship?” Paper presented at the 7th
“Connecting the Spiritual and Emotional Industrial and Organisational Psychology Conference
Intelligences: Confirming an Intelligence Criterion (IOP) of the APS/1st Asia Pacific Congress on Work
and Assessing the Role of Empathy”. International and Organisational Psychology (APCWOP), 2007.
Journal of Transpersonal Studies, 3(1), 1
45. David, A. (1999). "A Proper Mentality for
44. Harmer, R. (2007). “Organisational Citizenship Benchmarking," Public Administration Review, 59 (2),
Behaviour, Emotional Intelligence and Spirituality: 105-109.

Figure 1: The Concept of Emotional Intelligence Source: Salovey & Mayer (1990)

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 851
 BIJIT - BVICAM’s International Journal of Information Technology

Table-1: Correlation between Emotional and Spiritual Intelligence Dimensions

EI Self Awareness Social Awareness Self Management Social Skills

SI PC Sig. PC Sig. PC Sig. PC Sig.
CET 0.150 0.068 0.220 0.007* 0.178 0.029* 0.266 0.001*
PMP 0.497 0.000* 0.425 0.000* 0.458 0.000* 0.478 0.000*
TA 0.884 0.000* 0.692 0.000* 0.791 0.000* 0.592 0.000*
CSE 0.683 0.000* 0.598 0.000* 0.632 0.000* 0.570 0.000*
*= Significant at 5 percent level of significance.
Note: EI= Emotional Intelligence, SI= Spiritual Intelligence, CET= Critical Existential Thinking, PMP= Personal Meaning
Production, TA= Transcendental Awareness, CSE= Conscious State Expansion, PC= Pearson Correlation

Table 2: Regression analysis while Emotional Intelligence is Independent Variable and Spiritual Intelligence is the
Dependent Variable
Un-standardized Standard T value P value R Adj. R f- ratio P
Variables Coefficient beta error square square value
Self Awareness
0.292 0.100 2.916 0.004*
Social
0.111 0.079 1.418 0.158
Awareness
Self
0.237 0.093 2.562 0.011* 0.516 0.502 38.35 0.000*
Management
Social
0.191 0.085 2.249 0.026*
Skills
*= Significant at 5 percent level of significance.

Table 3: Regression analysis while Spiritual Intelligence is Independent Variable and Emotional Intelligence is the
Dependent Variable
variables Un- Standard T value P value R Adj. R f- ratio P
standardized error square square value
Coefficient
beta
CET 0.004 0.024 0.163 0.870
PMP 0.057 0.028 2.056 0.042*
TA 0.550 0.040 13.710 0.000* 0.800 0.795 144.32 0.000*
CSE 0.232 0.045 5.122 0.000*
*= Significant at 5 percent level of significance.

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 852
 BIJIT -BVICAM’s International Journal of Information Technology
Bharati Vidyapeeth’s Institute of Computer Applications and Management (BVICAM), New Delhi (INDIA)

Cloud Computing in SMEs: Case of Saudi Arabia

Mohammad Yamin1 and Ammar A. Al Makrami2
Submitted in November, 2014; Accepted in February, 2015
Abstract - Cloud computing technology is now in an advance 1.1 Growth of SMEs in Saudi Arabia
stage having crossed many hurdles that came in its way to The Kingdom of Saudi Arabia (KSA) is one of the richest
broad acceptability by the business world as well individuals. countries in the world. Yet, it falls into the category of
Cloud computing not only provides storage but also provides developing countries. The main reason for it not making into
a range of services on a rental basis, which enables firms to the list of the developed countries is the slow advancement in
limit their infrastructure to a minimum and so makes it the industrial sector, in particular the SMEs. The export of
affordable by many. Cloud computing provides many petroleum has contributed significantly to the economic
entrepreneurs as well as businesses, especially the small and growth, which has earned the Kingdom a membership of G20,
medium- sized enterprises (SMEs), with low cost alternatives an alliance of countries whose economies are highly influential
to for starting and managing their businesses. In particular in the world. The G20 countries have a lion’s share or more
SMEs of many developing countries, who cannot always than three-fourth of the world trade [2]. Therefore, it would
afford to mobilise huge capitals to start new businesses are seem to be unfair to find some statistics suggesting Saudi
immensely benefiting from this technology. Saudi Arabia is a Arabia to be one of the developing counties. However, some
developing country with a good infrastructure for small to other sources like the International Statistical Institute [3],
medium sized industries. Many of the Saudi SMEs are using excludes Saudi from the list of Developing Countries.
a range of technologies and platforms including cloud Likewise, a former US diplomat [4] is convinced that Saudi
computing to sustain and manage their businesses. In this Arabia is a semi-developed country.
article we look at the extent of cloud computing applications
in SMEs in the Kingdom of Saudi Arabia. To measure the 1.2 Historical and religious Significance of Saudi Arabia
extent of the use of cloud computing, we present an analysis Makkah (Mecca) in Saudi Arabia is well known for housing the
of a comprehensive survey conducted on a large number of Kaaba, which marks the direction of prayers for Muslims all
small organisations in the West Coast of Saudi Arabia during over the world. Makkah is more than four thousand years old.
the course of this research. The city is the centre of the annual pilgrimages known as the
Hajj and Omrah, which are attended by millions of people
Index Terms – Cloud Computing, Entrepreneurs, SMEs, IT every year. Some information and details of the Hajj and
Infrastructure, Developing Countries, Saudi Arabia Omrah can be found in [5]. These pilgrimages further inject
about 16 billion dollars to the economy[6]. Currently Saudi
1.0 INTRODUCTION Arabia is undergoing a number of major projects involving
Cloud computing has now emerged as a leading technology of various arms of the Saudi industry and in particular the SMEs.
the 21st century and has been embraced by industries around These projects include a rail link between major cities and
the globe. Its progress and advancement could be compared to Makkah, huge expansion of the holy mosques in Makkah and
the advances of E-Commerce at the turn of the last century. Madinah and various other educational, health and cultural
Many entrepreneurs and businesses, in particular the small to projects which are amounting to hundreds of billions of dollars
medium sized businesses (SMEs) find it very helpful in setting in expenditures by the Saudi Arabian government, see [7] and
up new business as well as sustaining old businesses. In this [8]. Mohammad Yamin [9] and Abdullah Basahel [10]
article, authors will analyse the extent of cloud computing provided some more information about Saudi Arabia, who has
presence in Saudi SMEs. This study will be carried out with the earned the reputation of being the leading economy not only in
help of an industry survey conducted by the authors. An the Middle East but the whole Arab World.
example of a government who brought in cloud computing As has been the case with many developing countries, the
initiatives to enrich and strengthen SMEs is the UK growth of SMEs in Saudi Arabia was slow at the beginning.
government, which has initiated a cloud framework called "G- However, as the oil revenue increased and economy boomed,
Cloud framework" [1]. The difference between G-Cloud and industrial growth picked up pace. As a result, in addition to
other frameworks is that SMEs can pay for services as they use having a considerable presence of multi-national companies,
them, rather than being tied to inflexible, long-term contracts. the KSA has achieved remarkable growth in local businesses
This model of G-Clouds is very attractive to SMEs. and home grown industries. Some of the details of SMEs in
Saudi Arabia can be obtained from [11]. The aim of our study
1,2
Department of MIS, Faculty of Economics and in this paper is carry out a study on the presence and impact of
Administration, King Abdulaziz University, Jeddah, Saudi Cloud Computing in SMEs in Saudi Arabia. In particular, we
Arabia look at the case of Sanitary Ware, which represents a small
1 2
E-mail: [email protected] and [email protected] industry. To analyse the impact of Cloud computing in SMEs,

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 853
 BIJIT - BVICAM’s International Journal of Information Technology

we have surveyed eighty businesses, whose analysis will be Having realised the importance of SMEs, the Saudi Arabian
presented in this article. government has endorsed some attractive ways to finance the
SMEs, details of these ways can be found [17] and [18]. These
1.3 Role of SMEs in World Economy measures are expected to lower the unemployment rate of the
According to a major study conducted by Sweden’s Kingdom, and provide highly needed support to the
Globalisation Council [12], “One main conclusion is that development of the local economy. Once this happens, the
changes by globalization will affect the SMEs and Kingdom of Saudi Arabia may find itself a step closer to be
entrepreneurs in different type of economies both on a national classified as a developed country.
and regional perspective. The world is globalized so is the
world for entrepreneurs and innovations. All type of companies 2.0 CLOUD COMPUTING
will be affected by the changes that one nowadays could Cloud computing is a disruptive innovation and technology of
observe, not only SMEs or innovative entrepreneurship but of the 21st century. It has affected the industrial world in the same
course also the behaviour of multinationals and large firms, and manner as outsourcing, E-Business and E-Learning did in the
the relations between large and small firms”. In another major 20th century. Cloud computing has created opportunities to
study conducted by UNU World Institute for Development new and existing businesses, especially the entrepreneurs to
Economics Research [13], from a theoretical perspective, setup and realise their dream businesses without much of the
SMEs have several advantages over both SOEs and large infrastructure and capital of their own, as the technology
privatized SOEs (PSOEs). In SMEs the incentives for both provides much of these on a rental basis.
managers and workers tend to be clearer and stronger than in
larger organizations. Governance problems are less significant 2.1 Models and Characteristics
since in many cases the managers and owners are the same To be able to comprehend cloud computing, a well-organized
people. The small size of many of the SME firms allows definition is essential. As described in [19], The National
owners easily to monitor the performance of managers while Institute of Standards and Technology (NIST) has defined
the cost of monitoring labour is also lower. Table 1 shows SME cloud computing as a model for enabling ubiquitous,
employment as a percentage of total employment and convenient, on-demand network access to a shared pool of
manufacturing employment in various countries. For details, configurable computing resources (e.g., networks, servers,
see [13]. storage, applications, and services) that can be rapidly
Table 1: SMESs share in employment provisioned and released with minimal management effort or
Country Year Country Year service provider interaction. The NST definition as
EU 69 Japan 33
France 69 Croatia 30
Germany 64 Romania 27
Belgium 72 Hungary 24
Italy 80 Poland 23
Georgia 58 Slovenia 19
US 53 Russian Fed 10
Estonia 43 Belarus 6
Czech Rep 37 Kyrgyz Rep 3
SMEs in Saudi Arabia represent almost 93 percent of total
enterprises and account for about 24.7 percent of total
employment, which is fairly low as compared to many of the
developed countries listed in Table 1, see [13]. Moreover, Figure 1: Cloud Computing Definition
Saudi Arabia’s businesses contribute about 33% of the national seen in Fig 2 includes four characteristics namely, Broad
GDP whereas the figures for some developed countries are as Network Access, Rapid elasticity, Measured Service and On-
follows: US (50%), France (56%), Spain (57%) and Japan Demand Self-Service. The service models for cloud computing
(64%), for details see [14]. As can be seen, the GDP share of can be Software as a Service (SaaS), Platform as a service
SMEs of Saudi Arabia is much lower as compared to the (PaaS), and Information as a service (IaaS), and can be
countries of G20. A detailed analysis of SMEs and their socio- deployed as Public, Private, Hybrid or Community. Many other
economic contribution can be found in [15]. However, people have defined and described cloud computing. For
investment in small and medium enterprises in Saudi Arabia is example, in [20], cloud computing is referred to as both the
expected to grow to more than USD 70 billion by the end of delivery of applications as services over Internet, and the
2015, with the government and banks providing much-needed hardware and systems software in the data centres that provide
funding, according to industry data compiled by Zawya [16]. services (SaaS).

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 854
Cloud Computing in SMEs: Case of Saudi Arabia

There have been concerns about the security of data in the its model. On the other hand, many organisations, with the help
custody of the cloud providers. Initially, there were serious of local IT support, may resort to cloud computing in place of
concerns and many organisations, especially the government outsourcing; hence the outsourcing activity might decline.
owned businesses and corporations were very reluctant to make Initially the cloud providers, and hence the large corporations
use of the cloud computing platforms. Considerations of data of the developed countries and their economies may benefit in
security and privacy would assume a very high priority when the way of providing systems, infrastructure and other
trans-border data sharing was involved. As discussed in resources need by the cloud providers. However, once the
Security and Resilience in Governmental Clouds (2009), this saturation point occurs, the developing countries might make
has significant implications for the use of cloud databases to significant gains due to picking up the lost business of
manage private and sensitive data. Thus, one of the main outsourcing. It is expected that the use of cloud computing will
considerations for many corporate and government increase in Saudi Arabia, which currently has only some
organizations, especially in the industrial world where security businesses making use of it. In this paper we shall analyse the
and privacy is paramount, is whether to make the sensitive (e.g. result of a survey conducted in 2014 of several businesses in
defence, national security) personal data available to cloud Saudi Arabia to measure the usage and impact of cloud
providers. However applications used in periodic processing computing. An economic impact of cloud computing, known as
and those involving activities like database development and Cloud Economy, was presented by Mohammad Yamin in [21].
testing, data mining, database backups and off-site disaster
recovery are regarded as safe and can be managed and stored 2.3 Cloud Computing and Small Businesses
through clouds. These considerations have been quite critical With the help of cloud computing, many SMEs can start their
for many government corporations including those of the Saudi businesses with little capital. Cloud computing can be used by
government sector, where cloud computing is slowly and existing firms as well, where they can reorganize their way of
gradually taking its share in the government and privately doing business. With the implementation of cloud computing
owned businesses. For example, the webmail system of the services, many services can be achieved such as rent data
King Abdulaziz University is hosted on cloud. storage, software and systems, maintenance and services for a
From these and other numerous descriptions, cloud computing fee. Moreover, it's well known that offshore and onshore
has emerged as an easy to acquire and use infrastructure, outsourcing enables organization to do their business with less
software, systems, memory and maintenance in a variety of capital and infrastructure. As for when both cloud computing
ways to all kinds of organisations and individuals. This has and outsourcing join together to save capital for organizations,
allowed new starters of businesses to benefit from these it is expected that many existing businesses which outsource
facilities. some of their work process, would reorganize their ways of
doing business to benefit from cloud computing services, and
2.2 Impact of Cloud Computing on Economy that would maximize their efficiency. For details, see [22].
With the prospects of eliminating a need for a large capital According to a survey, conducted by Janet Kourik [23], by
investment, the economy of many countries, especially the 2020, over 70% of users will use internet-based (cloud)
developing countries, is significantly improving the financial software to perform work." Moreover, the usage of cloud
abilities of the organisations and businesses. This trend is likely computing would grow at an astonishing rate. According to the
to continue and is expected to affect the economies of a large survey, about 69% of the Americans use cloud computing
number of countries, getting deeper into the poor nations of services such as online backup, webmail, and other services.
Africa and Asia. Cloud computing is redefining methods and People in developed countries are not the only users of cloud
models, reducing the burden of having huge capital to start a computing services. Moreover, there are about one thousand
new or sustain and expand an established business. Many and five hundred Indian companies who use cloud computing.
entrepreneurs, who would otherwise find it impossible to start In [23], it was remarked that senior decision-makers in SMEs
their dream business due to lack of resources and capital for often are not aware of cloud computing.In view of [24] and
investing in IT infrastructure and systems technology, would [25], further benefits to SMEs are in the form of: friendliness,
now be in a position to fulfil their aspirations and expectations. lower cost of organisational ownership,better return on
The most attractive feature of cloud computing is its
affordability, an aspect that SMEs are expected to benefit from.
It is expected that many bonuses in poor and developing
countries would be able to enhance and strengthen their
business activity for many new IT organizations.. In the last
two decades, many leading organizations of the western world
have resorted to outsourcing, especially offshore outsourcing.
This model of business has so far benefited countries both in
the developing and the developed world. However, with the
advent of Cloud Computing, outsourcing is likely to undergo
structural changes and is likely to involve cloud computing in Figure 2: Gartner Hyper Cycle EmergingTechnologies

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 855
 BIJIT - BVICAM’s International Journal of Information Technology

investment, greater flexibility with range of plans, packages,

and add-ons to address diverse business requirements

2.4 Proliferation of Cloud Computing

In the recent decades, with the introduction of the Internet, web
2.0, technologies, and E-Business have changed the way
business is conducted in the 21st century. The political,
economic, social, and other factors have also changed the ways
in which the businesses are run. The technological factor, for
instance, has made it easy for all kind of firms to achieve more
efficiency in almost all fields. Cloud computing is now a hot
topic in the field of technology. Cloud computing has been
developing at an astonishing rate [26]. Through Gartner's Hype
Cycle methodologywhich providesa graphical view of the
maturity, adoption and business application of specific
technologies, this section aims to expose the high improvement
that cloud computing has achieved. Gartner releases an annual
update to the Hype Cycle showing how close technologies are Figure 4: Gartner 2011 Hyper Cycle for Cloud Computing
too high growth adoption [25]. Below are graphs presenting
cloud computing on the Hype Cycle from 2008 – 2014 [27].
Cloud computing's projected positioning on the Hyper Cycle
for 2014 is also included. Firstly, let us look at Fig 3, which
suggests that this year in 2015 cloud computing should be
enjoying a period of enlightenment, after going through a
technology trigger in 2008, peak of inflated expectation during
2009-11, and tough disillusionment during 2012-13.Let us now
analyse the extent of cloud computing in 2010 as shown in Fig
4. In 2010, cloud computing was only an emerging technology
in the Hyper Cloud of Information Technology of 2010, and
Gartner did consider making a separate cycle for this
technology. However, starting from 2011 onwards, Gartner has
produced separate hyper cycles for cloud computing. In 2011,
as is seen in Fig 5 Hyper Cloud for year 2012 shows the peak
of activity as a hyper activity and figure 6 of that of year 2014
shows continuation of the hyper activity of inflated
expectation, steadily rising the slope of enlightenment. These
hyper cycles are very much in line of the projections given by
Gartner. Figure 5: Gartner 2014 Hyper Cycle Cloud Computing

3.0 SURVEY: CLOUD COMPUTING AND SMEs

In this section, we shall describe our survey, which was
conducted in 2014 to measure the extent of cloud computing in

Figure 6: Gartner 2014 Hyper Cycle Cloud Computing

Figure 3: Gartner 2010 Hype Cycle Emerging Technologies

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 856
Cloud Computing in SMEs: Case of Saudi Arabia

Saudi Arabian SMEs. The main objective of the survey was to 3.2 Sample of the Survey
guide the direction of this and future research related with A suitable sample size was estimated using industry standards,
SMEs and cloud computing. The success of business firms were selected with the help their publications and public
unquestionably depends on information systems. records related to the selected demographic for the research,
SMEs and cloud computing. It is a forgone conclusion that and were contacted either by email or in person. Once an
most of the modern enterprises today depend on information organisation agreed to take part in the survey, the questionnaire
systems for their effective, efficient and smooth functioning. was emailed or handed personally to the nominated
Business intelligence is driving businesses to be increasingly representatives of the SMEs; they were asked to return the
dependent on IT solutions. However, information systems survey within a two-week period. Unfortunately, many
require an IT environment and facilities for their operation and organisations declined to receive the questionnaire despite
the cost of the IT infrastructure for operating such systems has agreeing to participate, and three organisations did not return
traditionally been very costly. With the availability and the questionnaire. One organisation was late with its responses
advancement of cloud computing paradigm, many companies and was therefore not included in the research data. Although
now can lease cloud computing services as an alternative to more than one hundred organisations were contacted as
upfront payment for setting up the needed infrastructure. As described above, only eighty of them provided the data.
discussed earlier, cloud computing provides IT infrastructure,
systems, software, tools, services and services on a rental basis 3.3 Ethical Considerations
for a specified time or by a meter system of usage. For many The aim of this survey was only to measure the extent of usage
entrepreneurs, cloud computing is an economical way to start of cloud computing SME enterprises in Saudi Arabia.
their businesses. Therefore, seeking highly personal and sensitive information
Saudi Arabia has many small enterprises that are still looking was not required. Only general questions were asked like name
for ways to reduce their cost of setting up a business and of the officer, the name of the company, contact details, name
running it efficiently - cloud computing is cost-effective and of the Cloud Computing provider used by the company and the
easy to acquire. To measure the extent of cloud computing location of the company. Moreover, they were assured that this
usage in Saudi Arabian small enterprises, a survey was information would be used purely for the use of the researcher
conducted of eighty companies in Jeddah. Due to the small size and would not be made public. In keeping up with established
of Saudi population and a relatively small number of norms, information provided by the participants was treated
companies in the Kingdom, it was concluded that a sample of confidentially and used for the sole purpose of this research.
eighty companies was sufficiently robust to enable Ethical considerations related to the research study were
generalisations to be made about SMEs in the emergent primarily based on anonymity. The data collected, as published
economy. here, does not identify the organisations or individuals who
provided the date. The respondents knew that their answers
3.1 Choice of Method would not be linked to their identity. As a result the
This research paper aims to develop a research model which respondents did not show any anxiety or fear on account of
would determine the extent of the use of cloud computing their participation. For example, if an organisation admitted to
amongst Saudi SMEs. One of the major challenges in not knowing anything about cloud computing, linking their
conducting our survey was to determine and select suitable organisation to such a response would have rendered their
SMEs to obtain the relevant data from.. Another challenge was brand or business being labelled ‘old-fashioned’ or ‘out-of-
to educate officials of some of the SMEs who had never heard date’. That would have affected their reputation, sales,
of Cloud Computing. Under the circumstances, the research investment, and partners. So, confidentiality was a key element
methodology and questionnaire, used for the survey, were kept in ensuring that the study followed an ethical code of conduct.
very simple. For relevant literature, publications and studies
were reviewed in order to get in-depth information on the use 4.0 SURVEY CONTENTAND RESULTS
of cloud computing in Saudi Arabian SMEs. The survey was The survey questionnaire contained thirteen questioners, with a
selected as a suitable tool for gathering the relevant data purpose of asking simple questions. Participants were asked to
because it was capable to collect first-hand information from provide their responses on a scale of seven indicators as
actual companies within the demographic region under follows: 1 = completely disagree, 2 = disagree, 3 = somewhat
consideration. Only the selected SMEs in Jeddah were disagree, 4 = neutral, 5 = somewhat agree, 6 = agree, 7 =
surveyed through a simple questionnaire. We acknowledge that absolutely agree. Our choice of the scale of seven choices was
there are some limitations to quantitative research methods. to allow the respondents with greater flexibility. The questioner
However, the strengths of such methods, related to the research is presented in Table 8, and the results of the survey are
aim, render a quantitative survey and the potential scope of presented in Table 9.
such a survey preferable over a qualitative alternative. While
interviews could perhaps produce more detailed responses, they
are not suited for determining widespread usage because of
their smaller size.

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 857
 BIJIT - BVICAM’s International Journal of Information Technology

Table 2: Survey Questionnaire

1 My company doesn’t use any Cloud Computing services.
Table 3: Survey Results
2 My company is planning to use hire Clood Computing
Questions Average Questions Average
services in the next 12 months.
Question 1 7 Question 8 5
3 My company is using some Cloud Computing services. Question 2 4 Question 9 5
Question 3 2 Question 10 4.4
4 My company is missing-on/taking an advantages of Cloud
Computing services. Question 4 4 Question 11 4.3
Question 5 4 Question 12 4.3
5 Cloud Computing is an economical way of employing IT Question 6 4 Question 13 4.3
facilities.
Question 7 4
6 Cloud Computing is a better way of dealing with technical
issues of enterprises is no reason for the slow growth of the usage of cloud
computing in small enterprises sector in Jeddah, Saudi Arabia.
7 Cloud Computing is helpful for establishing a new business. Performance of these SMEs could be much better. Also, more
cost reduction would be feasible on hiring cloud computing
8 Cloud Computing is way to get rid of old IT systems. services. A further research is warrented to examine the cases
of different parts in the kingdom and other developing
9 Cloud Computing eliminates the need for having technical countries of the Middle East and the region..
expertise.
4.2 Limitations
10 There are privacy issues with use of Cloud Computing This was a quantitative survey of only eight SMEs, in only one
services. region namely Jeddah, Saudi Arabia. Jeddah is the main port
city of Saudi Arabia and has a well-developed industry
11 There are data security issues associated with the use of presence. While it provides important and generalizable data
Cloud Computing.
and highlights deficiencies that need to be studied more
rigorously, it is far from exhaustive. However, the results of the
12 Use of Cloud Computing doesn’t require significant training
research study seem to be congruent with the published data
for the enterprise personnel.
related to the subject. Further research would be needed to
identify the barriers proliferation of cloud computing services
13 Cloud Computing improves performance of workers on job
in the Kingdom of Saudi Arabia and perhaps the entire Middle
East.
After carefully analysing the survey responses from our eighty
completed responses from various SMEs, we decided to base 5.0 CONCLUSIONS
our research simply on the averages of responses for each As we find from the analysis of our survey, all of the SMEs
question. These averages are provided in Table 9. who participated in the survey do not use cloud computing.
However, the majority of them said that their firm would
4.1 Implications of the Survey Responses benefit from having cloud computing and were in the process
From the results of the survey, analysts can see that it is of acquiring the technology for the future. Most of the
evident that majority of small enterprises in Jeddah in the respondents were of the view that cloud computing improves
province of Makkah of Saudi Arabia do not use cloud the performance of the workers. Many responses confirmed
computing services. It also shows that the small enterprises that cloud computing helps in getting rid of old and ailing
indicated that the SMEs have fair awareness of Clouds and systems. Hence, Saudi Arabia needs to strengthen its cloud
their benefits. In fact managers of many of these enterprises computing industry so that SMEs can benefit from this
have admitted to the fact that if they were using cloud technology. This would be a step closer to the Kingdom’s
computing, they would gain many advantages. However, small aspiration to find itself in the list of developed countries. It is
private sector companies were sceptical or uncertain whether fair to say that the Saudi Arabian government has been trying
there were any advantages of having this technology. In other to make the latest technology available to the business
words, the analysis shows that small enterprises are not against community of the country. The role of cloud computing,
hiring products or services from the cloud computing especially in the SMEs should not be ignored otherwise the
providers. It was also indicated that the small enterprises had industry will find it difficult to keep up with pace of the
very little or no concerns about privacy and data security issues developed countries. Saudi Arabia, having a good
associated with the use of cloud computing. Therefore, there technological base, is a candidate for excelling in cloud

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 858
Cloud Computing in SMEs: Case of Saudi Arabia

computing technologies. Cloud computing is out there and it arabia-starts-major-expansion-of-grand-mosque-in-

will reach the private SMEs. Therefore, in the future, we might mecca (15 Feb 2015).
expect a higher growth rate in the private as well as the public [9]. Mohammad Yamin, Health Management in Crowded
sector. The Saudi government on its part needs to continuously Events: Hajj and Kumbh, BIJIT - BVICAM’s
support and bring in fresh incentives to enrich and increase the International Journal of Information Technology,
industry, especially the SMEs. January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658
6.0 ACKNOWLEDGEMENT [10]. Abdullah Basahel, An Empirical Study of Impacts of E-
This work wouldn’t have been possible without the help of the Learning in Female Higher Education in Saudi Arabia,
King Abdulaziz University (KAU). The author acknowledges BIJIT - BVICAM’s International Journal of Information
the help and support from the Faculty of Economics and Technology, January - June, 2015; Vol. 7 No. 1; ISSN
Management of KAU. In particular, the authors are thankful for 0973 – 5658
the support and encouragement provided by Dr Abdullah [11]. Ahmed Al Saleh , Exploring Strategies for Small and
Basahel, Director of MBA Program, and Dr Nasser Kadasah, Medium Enterprises in Saudi Arabia, Strategies for
Associate Dean for Research, Faculty of Economics and SMEs in Saudi Arabia, [Online], Available:
Administration. http://www.ribm.mmu.ac.uk/symposium2012/extendeda
bstracts/AhmedAlSaleh.pdf(15 Feb 2015).
7.0 REFERENCES [12]. Anders Lundström (ed), The Role of SMEs and
[1]. Oxford Business Group, 2014. Starting small: Given the Entrepreneurship in a Globalised Economy, Expert
large number of local SMEs opening up, financing and report no. 34 to Sweden’s Globalisation Council, 2009,
opportunities is key. [online] Available at: ISSN 1654-6245 , [Online], Available:
http://www.oxfordbusinessgroup.com/analysis/starting-
small-given-large-number-local-smes-opening- [13]. http://www.regeringen.se/content/1/c6/12/58/93/d614cc
financing-and-opportunities-key (15 Feb 2015). 47.pdf
[2]. Australia 2014, G20 members, [Online], Available: [14]. Robert McIntyre, The Role of Small and Medium
https://www.g20.org/about_g20/g20_members (15 Feb Enterprises in Transition: Growth and Entrepreneurship,
2015). UNU World Institute for Development Economics
Research, Katajanokanlaituri 6 B 00160 Helsinki,
[3]. The International Statistical Institute, Developing Finland, ISBN 92-9190-095-8, [Online], Available:
Countries, [Online], available:http://www.isi- file:///C:/Users/ADMIN/Downloads/rfa49_1.pdf(15 Feb
web.org/component/content/article/5-root/root/81- 2015).
developing, (15 Feb, 2015).
[15]. Okaz, Classification of businesses in 4 levels, [Online],
[4]. John Burgess, Quora, Is Saudi Arabia a developed http://www.okaz.com.sa/new/Issues/20120126/Con2012
country? [Online], available:http://www.quora.com/Is- 0126472920.htm(15 Feb 2015)
Saudi-Arabia-a-developed-country-Why (15 Feb 2015).
[16]. Oxford Business Group, Starting Small: : Given the
[5]. Yamin Mohammad and Albugami Moteb A, An large number of local SMEs, opening up financing and
Architecture for Hajj Management, 15th IFIP WG 8.1 opportunities is key,[Online], Available:
Proceedings International Conference on Informatics http://www.oxfordbusinessgroup.com/analysis/starting-
and Semiotics in Organisations, ICISO 2014, Shanghai, small-given-large-number-local-smes-opening-
China, May 23-24, 2014, IFIP Advances in Information financing-and-opportunities-key (15 Feb 2015).
and Communication Technology, Vol. 426
[17]. Mohamed Abdulzaher, Saudi Arabia sees potential in
[6]. Ayesha Sarfraz (Posted May 30, 2013), Arabian Gazette, SME growth, Zawya, [Online], Available:
The economics of Hajj, [Online], Available: https://www.zawya.com/story/Saudi_Arabia_sees_poten
http://www.arabiangazette.com/economics-hajj- tial_in_SME_growth-ZAWYA20140223050551/(15 Feb
20130530/ (15 Feb 2015)). 2015).
[7]. Arabian Gazette (Posted by Ayesha Sarfraz / June 11, [18]. The Centennial Fund, [Online], Available:
2013), Saudi Arabia – new railways to cost $45 billion, http://www.tcf.org.sa/Dimm/TCF/Profile/Pages/Default.
[Online], Available: aspx (15 Feb 2015).
http://www.arabiangazette.com/saudi-arabia-new-
railways-45-billion-20130611/(15 Feb 2015). [19]. Kafalah program to finance SMEs, [Online], Available:
http://www.kafalah.gov.sa/Kafalah_2.5/index.php/2012-
[8]. Asma Alsharif (August 20, 2011), Saudi Arabia starts 07-04-13-47-25/2012-07-04-13-47-58(15 Feb 2015)
major expansion of grand mosque in mecca, [online],
Available: [20]. Mell, Peter, and Tim Grance. "The NIST definition of
http://blogs.reuters.com/faithworld/2011/08/20/saudi- cloud computing."National Institute of Standards and
Technology 53, no. 6 (2009): 50

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 859
 BIJIT - BVICAM’s International Journal of Information Technology

[21]. Armbrust, Michael, Armando Fox, Rean Griffith,

Anthony D. Joseph, Randy Katz, Andy Konwinski,
Gunho Lee et al. "A view of cloud
computing."Communications of the ACM 53, no. 4
(2010): 50-58.
[22]. Mohammad Yamin, Cloud Economy of Developing
Countries, World Journal of Social Sciences, Vol. 3. No.
3. May 2013, Pp. 132 – 142.
[23]. ENCIA, Cloud Computing Risk Assessment, 2011,
[Online], Available:
http://www.enisa.europa.eu/activities/riskmanagement/fi
les/deliverables/cloud-computing-risk-assessment,
Retrieved 10/04/2013 (15 Feb 2015).
[24]. Janet Kourik, For small and medium size enterprises
(SME) deliberating cloud computing: Proceedings of
European Computing Conference, ISBN: 978-960474-
297-4, 2011,
[25]. STC, "Cloud Computing services" [Online], Available:
http://new.awal.com.sa/portal.php?lang=1&p=2:14:102,
(15 Feb. 2015).
[26]. Neves, FátimaTrindade, Fernando Cruz Marta, Ana
Maria R. Correia, and Miguel de Castro Neto. The
adoption of cloud computing by SMEs: identifying and
coping with external factors, Paper presented at 11ª
Conferência da Associação Portuguesa de Sistemas de
Informação (CAPSI 2011), 19-21th Oct 2011. [Online],
available:
http://run.unl.pt/bitstream/10362/6166/1/Neves_Marta_
Correia_Neto_2011.pdf (15 Feb 2015).
[27]. Tritsiniotis, Emmanouil D. Get Ready for Clouds,
Tailoring enterprise architecture for cloud ecosystems,
Master thesis, School of Management and Governance,
University of Twente, (2013).
[28]. Gartner, "About Gartner" [Online], Available:
http://www.gartner.com/technology/about.jsp, (15 Feb.
2015).

Copy Right © BIJIT – 2015; January - June, 2015; Vol. 7 No. 1; ISSN 0973 – 5658 860
 BIJIT - BVICAM’s International Journal of Information Technology
(A Half Yearly Publication; ISSN 0973 - 5658)

Subscription Rates (Revised w.e.f. January, 2012)

1 Year 3 Years
Category India Abroad India Abroad
Companies Rs. 1000 US $ 45 Rs. 2500 US $ 120
Institution Rs. 800 US $ 40 Rs. 1600 US $ 100
Individuals Rs. 600 US $ 30 Rs. 1200 US $ 075
Students Rs. 250 US $ 25 Rs. 750 US $ 050
Single Copy Rs. 500 US $ 25 - -

Subscription Order Form

Please find attached herewith Demand Draft No._______________ dated __________________

For Rs.______________ drawn on ____________________________________________Bank
in favor of Director, “Bharati Vidyapeeth’s Institute of Computer Applications and
Management (BVICAM), New Delhi” for a period of 01 Year / 03 Years

Subscription Details
Name and Designation ________________________________________________________
Organization ________________________________________________________________
Mailing Address _____________________________________________________________
_______________________________________ PIN/ZIP ____________________________
Phone (with STD/ISD Code)_____________________________FAX___________________
E-Mail (in Capital Letters)______________________________________________________

Date: Signature
Place: (with official seal)
Filled in Subscription Order Form along with the required Demand Draft should be sent to the
following address:-

Prof. M. N. Hoda
Editor-in- Chief, BIJIT
Director, Bharati Vidyapeeth’s
Institute of Computer Applications & Management (BVICAM)
A-4, Paschim Vihar, Rohtak Road, New Delhi-110063 (INDIA).
Tel.: +91 – 11 – 25275055 Fax: +91 – 11 – 25255056 E-Mail: [email protected]
Visit us at: www.bvicam.ac.in/bijit