The Rise of Fog Computing in The Digital Era PDF

The Rise of Fog
Computing in the Digital

Era
K.G. Srinivasa
Chaudhary Brahm Prakash Government Engineering College,
India
Pankaj Lathar
Chaudhary Brahm Prakash Government Engineering College,
India
G.M. Siddesh
Ramaiah Institute of Technology, India
A volume in the Advances

in Computer and Electrical
Engineering (ACEE) Book Series
Published in the United States of America by
IGI Global
Engineering Science Reference (an imprint of IGI Global)
701 E. Chocolate Avenue
Hershey PA, USA 17033
Tel: 717-533-8845
Fax: 717-533-8661
E-mail: [email protected]
Web site: http://www.igi-global.com
Copyright © 2019 by IGI Global. All rights reserved. No part of this publication may be
reproduced, stored or distributed in any form or by any means, electronic or mechanical, including
photocopying, without written permission from the publisher.
Product or company names used in this set are for identification purposes only. Inclusion of the
names of the products or companies does not indicate a claim of ownership by IGI Global of the
trademark or registered trademark.
Library of Congress Cataloging-in-Publication Data
Names: Srinivasa, K. G., editor. | Lathar, Pankaj, 1974- editor. | Siddesh,

G. M., 1981- editor.
Title: The rise of fog computing in the digital era / K.G. Srinivasa, Pankaj
Lathar, and G.M. Siddesh, editors.
Description: Hershey, PA : Engineering Science Reference, [2018]
Identifiers: LCCN 2017061866| ISBN 9781522560708 (h/c) | ISBN 9781522560715
(eISBN)
Subjects: LCSH: Cloud computing.
Classification: LCC QA76.585 .R57 2018 | DDC 004.67/82--dc23 LC record available at https://
lccn.loc.gov/2017061866
This book is published in the IGI Global book series Advances in Computer and Electrical
Engineering (ACEE) (ISSN: 2327-039X; eISSN: 2327-0403)
British Cataloguing in Publication Data

A Cataloguing in Publication record for this book is available from the British Library.
All work contributed to this book is new, previously-unpublished material.

The views expressed in this book are those of the authors, but not necessarily of the publisher.
For electronic access to this publication, please contact: [email protected].

Advances in Computer
and Electrical Engineering
(ACEE) Book Series
ISSN:2327-039X
EISSN:2327-0403
Editor-in-Chief: Srikanta Patnaik, SOA University, India

Mission
The fields of computer engineering and electrical engineering encompass a broad
range of interdisciplinary topics allowing for expansive research developments
across multiple fields. Research in these areas continues to develop and become
increasingly important as computer and electrical systems have become an integral
part of everyday life.
The Advances in Computer and Electrical Engineering (ACEE) Book Series
aims to publish research on diverse topics pertaining to computer engineering
and electrical engineering. ACEE encourages scholarly discourse on the latest
applications, tools, and methodologies being implemented in the field for the design
and development of computer and electrical systems.
Coverage
• Programming IGI Global is currently accepting
• Sensor Technologies manuscripts for publication within this
• VLSI Design series. To submit a proposal for a volume in
• Optical Electronics this series, please contact our Acquisition
• Computer Hardware Editors at [email protected] or
• Computer science visit: http://www.igi-global.com/publish/.
• Power Electronics
• Chip Design
• Algorithms
• Computer Architecture
The Advances in Computer and Electrical Engineering (ACEE) Book Series (ISSN 2327-039X) is published by IGI
Global, 701 E. Chocolate Avenue, Hershey, PA 17033-1240, USA, www.igi-global.com. This series is composed of titles
available for purchase individually; each title is edited to be contextually exclusive from any other title within the series.
For pricing and ordering information please visit http://www.igi-global.com/book-series/advances-computer-electrical-
engineering/73675. Postmaster: Send all address changes to above address. ©© 2019 IGI Global. All rights, including
translation in other languages reserved by the publisher. No part of this series may be reproduced or used in any form
or by any means – graphics, electronic, or mechanical, including photocopying, recording, taping, or information and
retrieval systems – without written permission from the publisher, except for non commercial, educational use, including
classroom teaching purposes. The views expressed in this series are those of the authors, but not necessarily of IGI Global.
Titles in this Series
For a list of additional titles in this series, please visit:
https://www.igi-global.com/book-series/advances-computer-electrical-engineering/73675
Electronic Nose Technologies and Advances in Machine Ofaction

Yousif Albastaki (University of Bahrain, Bahrain) and Fatema Albalooshi (University of
Bahrain, Bahrain)
Engineering Science Reference • ©2018 • 318pp • H/C (ISBN: 9781522538622) • US $205.00
Quantum-Inspired Intelligent Systems for Multimedia Data Analysis

Siddhartha Bhattacharyya (RCC Institute of Information Technology, India)
Engineering Science Reference • ©2018 • 329pp • H/C (ISBN: 9781522552192) • US
$185.00
Advancements in Computer Vision and Image Processing

Jose Garcia-Rodriguez (University of Alicante, Spain)
Handbook of Research on Power and Energy System Optimization

Pawan Kumar (Thapar University, India) Surjit Singh (National Institute of Technology
Kurukshetra, India) Ikbal Ali (Jamia Millia Islamia, India) and Taha Selim Ustun (Carnegie
Mellon University, USA)
Big Data Analytics for Satellite Image Processing and Remote Sensing
P. Swarnalatha (VIT University, India) and Prabu Sevugan (VIT University, India)
Modeling and Simulations for Metamaterials Emerging Research and Opportunities

Ammar Armghan (Aljouf University, Saudi Arabia) Xinguang Hu (HuangShan University,
China) and Muhammad Younus Javed (HITEC University, Pakistan)
For an entire list of titles in this series, please visit:

https://www.igi-global.com/book-series/advances-computer-electrical-engineering/73675
701 East Chocolate Avenue, Hershey, PA 17033, USA

Tel: 717-533-8845 x100 • Fax: 717-533-8661
E-Mail: [email protected] • www.igi-global.com
Table of Contents
Foreword..............................................................................................................xii
Preface.................................................................................................................xiv
Acknowledgment..............................................................................................xviii
Chapter 1
Introduction.to.Fog.Computing. ............................................................................1
Stojan Kitanov, Mother Teresa University, Macedonia
Toni Janevski, Ss Cyril and Methodius University, Macedonia
Chapter 2
Fog.Computing.and.Its.Challenges.......................................................................36
Vighnesh Srinivasa Balaji, Ramaiah Institute of Technology, India
Chapter 3
Fog.Computing.and.Networking.Architectures....................................................53
Minal Moharir, R. V. College of Engineering, India
Bharat Rahuldhev Patil, R. V. College of Engineering, India
Chapter 4
Fog.Computing.Application.Deployment.and.Management................................68
Ranjitha G., Ramaiah Institute of Technology, India
Pankaj Lathar, Chaudhary Brahm Prakash Government Engineering
College, India
G. M. Siddesh, Ramaiah Institute of Technology, India
Chapter 5
Confidentiality.and.Safekeeping.Problems.and.Techniques.in.Fog.. Computing.
..............................................................................................................................84
Nida Kauser Khanum, Ramaiah Institute of Technology, India
College, India

Chapter 6
EdgeCloud:.A.Distributed.Management.System.for.Resource.Continuity.in.
Edge.to.Cloud.Computing.Environment.............................................................108
Jamuna S. Murthy, Ramaiah Institute of Technology, India
Chapter 7
Security.and.Privacy.Issues.in.IoT:.A.Platform.for.Fog.Computing..................129
S. R. Mani Sekhar, Ramaiah Institute of Technology, India
Sharmitha S. Bysani, Ramaiah Institute of Technology, India
Vasireddy Prabha Kiranmai, Ramaiah Institute of Technology, India
Chapter 8
Software.Engineering.in.Internet.of.Things.......................................................157
Naresh E., Ramaiah Institute of Technology, India
Vijaya Kumar B. P., Ramaiah Institute of Technology, India
Aishwarya Hampiholi, Ramaiah Institute of Technology, India
Jeevan B., Ramaiah Institute of Technology, India
Chapter 9
IOT.and.Data.Analytics.Solution.for.Reducing.Pollution,.Accidents,.and.Its.
Impact.on.Environment.......................................................................................177
Chetan Shetty, Ramaiah Institute of Technology, India
Sowmya B. J., Ramaiah Institute of Technology, India
Anemish S., Ramaiah Institute of Technology, India
Seema S., Ramaiah Institute of Technology, India
Chapter 10
IOT.and.Data.Analytics.Solution.for.Smart.Agriculture....................................210
Netravati V. Cholappagol, Ramaiah Institute of Technology, India
Compilation of References............................................................................... 238
Related References............................................................................................ 257
Index................................................................................................................... 285
Detailed Table of Contents

will.introduce.fog.computing,.its.difference.in.comparison.to.cloud.computing,.and.
issues.related.to.fog..Among.the.three.issues.(i.e..service,.structural,.and.security.
issues),. this. chapter. scrutinizes. and. comprehensively. discusses. the. service. and.
structural.issues.also.providing.the.service.level.objectives.of.the.fog..They.next.
provide.various.algorithms.for.computing.in.fog,.the.challenges.faced,.and.future.
research.directions..Among.the.various.uses.of.fog,.two.scenarios.are.put.to.use.
Chapter 3
Fog.Computing.and.Networking.Architectures....................................................53
Minal Moharir, R. V. College of Engineering, India
Bharat Rahuldhev Patil, R. V. College of Engineering, India
The. demerits. of. cloud. computing. lie. in. the. velocity,. bandwidth,. and. privacy. of.
data.. This. chapter. focuses. on. why. fog. computing. presents. an. effective. solution.
to.cloud.computing..It.first.explains.the.primary.motivation.behind.the.use.of.fog.
computing..Fog.computing,.in.essence,.extends.the.services.of.the.cloud.towards.
the.edge.of.the.network.(i.e.,.towards.the.devices.nearer.to.the.customer.or.the.end.
user)..Doing.so.offers.several.advantages..Some.of.the.discussed.advantages.are.
scalability,. low. latency,. reducing. network. traffic,. and. increasing. efficiency.. The.
chapter.then.explains.the.architecture.to.implement.a.fog.network,.followed.by.its.
applications..Some.commercial.fog.products.are.also.discussed,.and.a.use.case.for.
an.airport.security.system.is.presented.
Chapter 4
Fog.Computing.Application.Deployment.and.Management................................68
Ranjitha G., Ramaiah Institute of Technology, India
College, India
Fog.computing.enhances.cloud.computing.to.be.closer.to.the.processes.that.act.on.IOT.
devices..Fogging.was.introduced.to.overcome.the.cloud.computing.paradigm.which.
was.not.able.to.address.some.services,.applications,.and.other.limitations.of.cloud.
computing.such.as.security.aspects,.bandwidth,.and.latency..Fog.computing.provides.
the.direct.correlation.with.the.internet.of.things..IBM.and.CISCO.are.linking.their.
concepts.of.internet.of.things.with.the.help.of.fog.computing..Application.services.
are.hosted.on.the.network.edge..It.improves.the.efficiency.and.reduces.the.amount.of.
data.that.is.transferred.to.the.cloud.for.analysis,.storage,.and.processing..Developers.
write.the.fog.application.and.deploy.it.to.the.access.points..Several.applications.like.
smart.cities,.healthcare.domain,.pre-processing,.and.caching.applications.have.to.
be.deployed.and.managed.properly.

Chapter 5 Confidentiality.and.Safekeeping.Problems.and.Techniques.in.Fog..
Computing............................................................................................................84
Nida Kauser Khanum, Ramaiah Institute of Technology, India
College, India
Fog.computing.is.an.extension.of.cloud.computing,.and.it.is.one.of.the.most.important.
architypes.in.the.current.world..Fog.computing.is.like.cloud.computing.as.it.provides.
data. storage,. computation,. processing,. and. application. services. to. end-users.. In.
this.chapter,.the.authors.discuss.the.security.and.privacy.issues.concerned.with.fog.
computing..The.issues.present.in.cloud.are.also.inherited.by.fog.computing,.but.the.
same.methods.available.for.cloud.computing.are.not.applicable.to.fog.computing.
due.to.its.decentralized.nature..The.authors.also.discuss.a.few.real-time.applications.
like. healthcare. systems,. intelligent. food. traceability,. surveillance. video. stream.
processing,.collection,.and.pre-processing.of.speech.data..Finally,.the.concept.of.
decoy.technique.and.intrusion.detection.and.prevention.technique.is.covered.
Chapter 6
EdgeCloud:.A.Distributed.Management.System.for.Resource.Continuity.in.
Edge.to.Cloud.Computing.Environment.............................................................108
Jamuna S. Murthy, Ramaiah Institute of Technology, India
In.the.recent.years,.edge/fog.computing.is.gaining.greater.importance.and.has.led.to.
the.deployment.of.many.smart.devices.and.application.frameworks.which.support.
real-time.data.processing..Edge.computing.is.an.extension.to.existing.cloud.computing.
environment. and. focuses. on. improving. the. reliability,. scalability,. and. resource.
efficiency.of.cloud.by.abolishing.the.need.for.processing.all.the.data.at.one.time.
and.thus.increasing.the.bandwidth.of.a.network..Edge.computing.can.complement.
cloud.computing.in.a.way.leading.to.a.novel.architecture.which.can.benefit.from.
both. edge. and. cloud. resources.. This. kind. of. resource. architecture. may. require.
resource.continuity.provided.that.the.selection.of.resources.for.executing.a.service.
in.cloud.is.independent.of.physical.location..Hence,.this.research.work.proposes.a.
novel.architecture.called.“EdgeCloud,”.which.is.a.distributed.management.system.
for.resource.continuity.in.edge.to.cloud.computing.environment..The.performance.
of.the.system.is.evaluated.by.considering.a.traffic.management.service.example.
mapped.into.the.proposed.layered.framework.

Chapter 7
Security.and.Privacy.Issues.in.IoT:.A.Platform.for.Fog.Computing..................129
S. R. Mani Sekhar, Ramaiah Institute of Technology, India
Sharmitha S. Bysani, Ramaiah Institute of Technology, India
Vasireddy Prabha Kiranmai, Ramaiah Institute of Technology, India
Security.and.privacy.issues.are.the.challenging.areas.in.the.field.of.internet.of.things.
(IoT).and.fog.computing..IoT.and.fog.has.become.an.involving.technology.allowing.
major.changes.in.the.field.of.information.systems.and.communication.systems..This.
chapter.provides.the.introduction.of.IoT.and.fog.technology.with.a.brief.explanation.
of.how.fog.is.overcoming.the.challenges.of.cloud.computing..Thereafter,.the.authors.
discuss.the.different.security.and.privacy.issues.and.its.related.solutions..Furthermore,.
they.present.six.different.case.studies.which.will.help.the.reader.to.understand.the.
platform.of.IoT.in.fog.
Chapter 8
Software.Engineering.in.Internet.of.Things.......................................................157
Naresh E., Ramaiah Institute of Technology, India
Vijaya Kumar B. P., Ramaiah Institute of Technology, India
Aishwarya Hampiholi, Ramaiah Institute of Technology, India
Jeevan B., Ramaiah Institute of Technology, India
This. chapter. gives. an. overall. role. of. software. engineering. in. internet. of. things.
domain.. In. this. chapter,. the. following. topics. are. included:. glimpse. of. complete.
software.engineering,.main.motivation.of.IoT,.how.IoT.evolved,.usage.of.software.
engineering.concepts.in.IoT,.role.of.CBSE.in.IoT,.role.of.aspect-oriented.software.
engineering,.heterogeneous.boards.in.designing.IoT.systems,.importance.of.integration.
phase.in.IoT.systems,.comparison.of.different.IDEs.of.IoT,.testing.of.IoT.systems,.
and.a.case.study.illustrating.all.the.concepts.for.online.blood.banking.system.and.
forest.fire.detection.
Chapter 9
IOT.and.Data.Analytics.Solution.for.Reducing.Pollution,.Accidents,.and.Its.
Impact.on.Environment.......................................................................................177
Anemish S., Ramaiah Institute of Technology, India
The.goal.of.this.chapter.is.to.inspect.and.consider.the.answer.for.accidents.and.
reactions.to.the.accidents.in.the.urban.zones..Modules.have.been.made.to.manage.
the.colossal.datasets.and.to.bring.interesting.bits.of.knowledge.into.the.outcomes..
This.is.done.by.utilizing.decision.tree.analysis.

Chapter 10 IOT.and.Data.Analytics.Solution.for.Smart.Agriculture...............210
Netravati V. Cholappagol, Ramaiah Institute of Technology, India
This.chapter.gives.the.real-time.solutions.to.the.farmers.by.providing.smart.solutions.
for.irrigation,.disease.monitoring,.and.decision.supporting.systems.(which.involves.
giving.suggestions.and.solutions.to.the.farmers.by.monitoring.soil.conditions,.rain,.
weather,.and.overall.quality.of.crop.growth.and.the.effect.on.the.growth.of.the.crop.
due.to.infertile.soil.or.bad.climatic.conditions)..These.solutions.are.provided.using.
the.IOT.and.data.analytics.technology.
Compilation of References............................................................................... 238
Related References............................................................................................ 257
Index................................................................................................................... 285
xii
Foreword
The Internet of Things and current-state cloud computing have created a situation
where billions of remote sensory devices work perpetually to transmit raw data to
centralized cloud data centers for storage and eventual retrieval and analysis. This
has caused some concern over whether enterprises, developers and cloud solution
designers can build appropriate and secure systems to store the growing surplus
of data while allowing quick accessibility to recover the data for useful analytics.
Pushing computing, control, data storage and processing into the cloud has
been a key trend in the past decade. However, the cloud alone encounters growing
limitations, such as reduced latency, high mobility, high scalability and real-time
execution in order to meet the upcoming computing and intelligent networking
demands. A new paradigm called Fog Computing has emerged to overcome these
limits. Fog extends cloud computing and services to the edge of the network. It
provides data, computing, storage, and application services to end-users that can
be hosted at the network edge. It reduces service latency, and improves QoS/QoE,
that results in superior user experience.
The topics covered in this book are fairly wide and would meet the training
requirement for both applied science as well as engineering disciplines. This volume
is intended to serve as a general introduction to fog computing for a senior-level
student or graduate student. The reader is expected to have a background in basic
concepts of IoT and Cloud Computing; only topics specific to the discipline will be
discussed herein. It is not intended to be a comprehensive authority on each subject,
but rather to serve as a reference and concept review for the upper-level reader.
The authors have endeavored to present a balanced view of issues, diligently
avoiding personal biases and fashionable philosophies. It is not the purpose of this
textbook to tell you what to think. Rather, our goal is to provide access to information
and the conceptual framework needed to understand complex issues so that you can
Foreword
comprehend the nature of problems and formulate your own views. Fog Computing
is not only a very important subject, but it is extremely interesting. The extensive
use of this book by students, faculty and practicing engineers will bring satisfaction
to the authors who have put in a lot of effort in writing this book.
Manohar Lal
IGNOU, India
xiii
xiv
Preface
With the immense growth of information, we are embracing the prevalence of

ubiquitously connected smart devices, which are now becoming the main factor
of computing. The information storage, processing and sharing has moved to
sophisticated smart gadgets. “Fog Computing” deals with the platforms that provide
compute, storage, and networking services between end devices and traditional
computing data centre’s. The demand for applications for fog computing are foreseen
in areas such as manufacturing, smart cities, connected transportation, smart grids,
e-health, and oil and gas. Fog computing is considered as a key enabler for, 4G
and 5G mobile networks for providing effective ways to address a wide range of
challenges that includes securing resource-constrained endpoints or supporting local
analytics. Cloud computing and Internet of things(IoT) in the current world of smart
devices has changed the way of computing, networking and services provided to
the users. Even though cloud computing and IoT can provide a salable solution for
deploying and managing applications, it can’t meet the stringent requirements of
applications with the present generation of computing devices especially in a mobile
environment, that includes challenges of latency-sensitive, security/privacy-sensitive,
or geographically constrained applications, scheduling and power management,
heterogeneity of devices. Fog computing is proposed to enable computing directly at
the edge of the network, which can deliver new applications and services especially
for the future of Internet.
In recent times, the number of Internet of Things (IoT) devices/sensors increased
tremendously. To support the computational demand of real-time latency-sensitive
applications of largely geo-distributed IoT devices/sensors, a new computing
paradigm named ‘Fog computing’ has been introduced as the demerits of cloud
computing lie in the velocity, bandwidth, and privacy of data. Fog computing is an
extension of cloud computing, and it is one of the most important archetype in the
current world. Fog computing is like cloud computing as it provides data storage,
computation, processing and application services to end-users.
Preface
The fog computing enhances the cloud computing to be closer to the process that
act on the IOT devices. Fogging was introduced to overcome the cloud computing
paradigm which was not able to address some services, applications and other
limitations of cloud computing such as security aspects, bandwidth and latency. In
the recent years, Edge/Fog Computing is gaining greater importance and has lead to
the deployment of many smart devices and application frameworks which supports
real-time data processing. But, Security and privacy issues are the challenging areas
in the field of Internet of Things (IoT) and Fog computing. In current ages, IoT and
Fog has become an involving technology allowing major changes in the field of
information systems and communication systems.
The target audience of this book is researchers, graduate students and practitioners
in the area of Cloud Computing, Internet of Things, Smart Systems and Fog
Computing.
The book contains 10 chapters, which are briefly described as follows:
The first chapter on Introduction to Fog Computing is about introduction and
overview of fog computing, comparison between fog computing and cloud computing,
fog computing and mobile edge computing, possible fog computing architecture,
applications of fog computing and possible research directions.
In Chapter 2, the authors have introduced Fog computing, its difference in
comparison to Cloud Computing and issues related to Fog. Among the three issues
i.e. service, structural and security issues, this article, scrutinizes and comprehensively
discusses the service and structural issues also providing the service level objectives
of the Fog. They next provide various algorithms for computing in Fog, the challenges
faced and future research directions. Among the various uses of Fog, two scenarios
were the advantages of Fog computing are put to use.
Chapter 3 focuses on why fog computing presents an effective solution to cloud
computing. It first explains the primary motivation behind the use of Fog computing.
Fog computing, in essence, extends the services of the cloud towards the edge of
the network, i.e. towards the devices nearer to the customer or the end user. Doing
so offers several advantages, some of the discussed advantages are scalability, low
latency, reducing network traffic and increasing efficiency. The chapter then explains
the architecture to implement a Fog network, followed by its applications. Some
commercial Fog products are also discussed and a use case for an airport security
system is presented.
Chapter 4 provides a good insight into how fog computing applications are
managed and deployed. As Fog computing provides the direct correlation with the
internet of things, IBM and CISCO are linking their concepts of internet of things
with the help of fog computing. Application services are hosted on the network edge.
It improves the efficiency and reduces the amount of the data which is transferred to
the cloud for analysis, storage and processing of the data. Developers write the fog
xv
Preface
application and deploy it to the access points. Several applications like smart cities,
healthcare domain, pre-processing and caching applications has to be deployed and
managed properly.
In Chapter 5, authors discuss the security and privacy issues concerned with fog
computing. The issues present in cloud are also inherited by fog computing, but same
methods available for cloud computing are not applicable to fog computing due to
its decentralized nature. We also discuss few real-time applications like healthcare
systems, Intelligent food trace-ability, Surveillance video stream processing,
Collection and pre-processing of speech data. Finally, the concept of decoy technique
and intrusion detection and prevention technique is covered.
Edge Computing which is an extension to existing Cloud Computing environment
and focus on improving the reliability, scalability and resource efficiency of cloud
by abolishing the need for processing all the data at one time and thus increasing the
bandwidth of a network is elaborately discussed in Chapter 6. Edge Computing can
complement Cloud Computing in a way leading to a novel architecture which can
benefit from both edge and cloud resources. This kind of resource architecture may
require resource continuity provided that the selection of resources for executing a
service in cloud is independent of physical location. Hence this research work proposes
a novel architecture called “EdgeCloud” which is a distributed management system
for resource continuity in edge to cloud computing environment. The performance
of the system is evaluated by considering a traffic management service example
mapped into the proposed layered framework.
Chapter 7 provides the introduction of IoT and Fog Technology with a brief
explanation of how Fog is overcoming the challenges of cloud computing. Thereafter
authors discuss the different security & privacy issues and its related solutions.
Furthermore we present six different case studies which will help the reader to
understand the platform of IoT in Fog.
Chapter 8 gives an overall role of software engineering in internet of things
domain. In this chapter, the following topics are included: Glimpse of complete
software engineering, Main motivation of IoT, How IoT evolved, Usage of Software
Engineering concepts in IoT, Role of CBSE in IoT, Role of Aspect-oriented software
engineering, Heterogeneous boards in designing IoT systems, Importance of
integration phase in IoT systems, Comparison of different IDEs of IoT, Testing of
IoT systems and a case study illustrating all the concepts for on-line blood banking
system and Forest Fire Detection.
The International Traffic Safety Data & Analysis Group (IRTAD) produces
enormous amount of accident and mishap datasets. These datasets consists of detailed
information of accidents data, for example, the city names, the type of accidents,
and amount of light at the accident site, seriousness of mischance, fatality rate,
speed zone, Risk factor due to utilization of liquor and so on. The emphasis is laid
xvi
Preface
on the use of information examination to anticipate and reduce the effect of the
mishaps and reveal the significant patterns via preparing the framework with the past
information utilizing Data Analytics approach. The real causes for the mischance’s
are recognized and studied by utilizing a portion of the applicable situations like,
liquor utilization, attempt of manslaughter cases and so forth. Chapter 9 discusses
how IoT and Data analytics in this case study helps to provide a feasible solution.
Finally, Chapter 10 provides a case study on how Fog Computing can be effectively
used in Smart agriculture. IoT investigation applications can enable organizations
to comprehend the Internet of Things information available to them, with an eye
toward diminishing upkeep costs, evading hardware disappointments and enhancing
business operations. Internet of Things gadgets can open a radical new universe of
information for associations - for instance, sensors can be utilized to proactively
screen and keep up pipelines and system loads, which can help maintain a strategic
distance from mishaps and blackouts. Understanding IoT-delivered information
requires something beyond propelling a Hadoop information lake and retiring until
tomorrow, notwithstanding. Here, when we are concerning about the health care we
are mainly dealing with Activity trackers during cancer treatment. We analyze the
cancer data and give solutions in terms of its symptoms its solution and monitoring
the patients using IOT and Data analytics.
The book is a good collection of articles which provide a basic insight into Fog
Computing, its applications and different deployment scenario. We would like to
express our sincere appreciation to each contributor for his/her work and for their
patience and attention to detail during the entire production process. We sincerely
hope these eminent contributors will encourage us in the future as well, in the
greatest interest of academia. The book will be of interest to a broad audience of
Fog Computing, IoT, Cloud Computing and Analytics, or newcomers who want to
learn more about the topic.
xvii
xviii
Acknowledgment
We attribute our efforts for completing this book to all the people who have inspired
us and shaped our careers. We thank our college administration, colleagues and
students who encouraged us to work on this book.
Srinivasa would like to thank Prof. K C Tiwari, Principal of CBP Government

Engineering College, New Delhi who encouraged working on this book. His special
thanks to his esteemed colleagues Prof. Harne, Dr. Athar Hussain, and Seema Rani
for their valuable suggestions and continuous encouragement.
Both Siddesh and Srinivasa would like to thank Dr. NVR Naidu, Principal, Ramaiah
Institute of Technology for his valuable guidance and support. Special thanks
to Dr. T V Suresh Kumar, Dr. B P Vijay Kumar and Mr. Ramesh Naik for their
continuous encouragement. The authors would like to acknowledge valuable inputs
and suggestions by Dr. Anita K, Dr. Seema S, and Mr. Srinidhi H.
Pankaj would like to thank Prof. Manohar Lal, Former Director, SOCIS, IGNOU,
New Delhi for his matchless mentorship and also for writing forward for the book.
His special thanks to his esteemed senior colleague Prof. Girish Kumar Sharma,
BPIBS, Delhi who has always been a source of inspiration. He is grateful to Prof.
Yudhvir Singh, Dean Colleges, MDU for his incredible guidance and suggestions.
We are extremely grateful to our families, who graciously accepted our inability to
attend to family chores during the course of writing this book, and especially for
their extended warmth and encouragement. Without their support, we would not
have been able to venture into writing this book.
1
Chapter 1
Introduction to Fog Computing
Stojan Kitanov
Mother Teresa University, Macedonia
Toni Janevski
Ss Cyril and Methodius University, Macedonia
ABSTRACT
Pushing computing, control, data storage, and processing into the cloud has been a
key trend in the past decade. However, the cloud alone encounters growing limitations,
such as reduced latency, high mobility, high scalability, and real-time execution
in order to meet the upcoming computing and intelligent networking demands. A
new paradigm called fog computing has emerged to overcome these limits. Fog
extends cloud computing and services to the edge of the network. It provides data,
computing, storage, and application services to end-users that can be hosted at the
network edge. It reduces service latency, and improves QoS/QoE, that results in
superior user experience. This chapter is about introduction and overview of fog
computing, comparison between fog computing and cloud computing, fog computing
and mobile edge computing, possible fog computing architecture, applications of
fog computing, and possible research directions.
INTRODUCTION
The future Internet of Everything (IoE) would become the linkage between extremely
complex networked organizations (e.g. telecoms, transportation, financial, health
and government services, commodities, etc.), which would provide the basic ICT
infrastructure that supports the business processes and the activities of the whole
society in general (Brech, Jamison, Shao, & Wightwick, 2013.), (Mitchell, Villa,
DOI: 10.4018/978-1-5225-6070-8.ch001
Copyright © 2019, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
Stewart-Weeks, & Lange, 2013). Frequently, these processes and activities should be
supported by orchestrated cloud services, where a number of services work together
to achieve a business objective (Zhang, Zhang, Chen, & Huo, 2010).
However, these demands can only be partially fulfilled by existing mobile cloud
computing solutions. This is because the future Internet would exacerbate the need
for improved QoS/QoE, supported by services that are orchestrated on-demand and
are capable to adapt at runtime, depending on the contextual conditions, to allow
reduced latency, high mobility, high scalability, and real-time execution. In addition,
the emerging wave of Internet of Things (IoTs) would require seamless mobility
support and geo-distribution in addition to location awareness and low latency. Also,
the existing cloud computing security mechanisms such as sophisticated access
control and encryption have not been able to prevent unauthorized and illegitimate
access to data.
A new paradigm called Fog Computing, has emerged to meet these requirements
(Bonomi, Milito, Zhu, & Addepalli, 2012; Mouradian, Naboulsi, Yangui, Glitho,
Morrow, & Polakos, 2017). Fog Computing extends cloud computing and services
to the edge of the network. Fog computing would combine the study of mobile
communications, micro-clouds, distributed systems, and consumer big data. It
is a scenario where a huge number of heterogeneous (wireless and sometimes
autonomous) ubiquitous and decentralized devices communicate, and potentially
cooperate among them and with the network to perform storage and processing
tasks without the intervention of third parties (Vaquero & Rodero-Merino, 2014;
Liu, Fieldsend, & Min, 2017). These tasks support basic network functions, or new
services and applications that run in a sand-boxed environment. Users leasing part
of their devices to host these services get incentives for doing so. The distinguishing
fog characteristics are its proximity to end-users, its dense geographical distribution,
and its support for mobility. Therefore, fog paradigm is well positioned for real-time
big data analytics. Services are hosted at the network edge or even end devices such
as set-top-boxes, or access points (Gao, Luan, Liu, & Yu, 2017). By doing so, fog
reduces service latency, and improves QoS, resulting in superior user-experience.
It supports emerging IoE applications that demand real-time/predictable latency
(industrial automation, transportation, networks of sensors and actuators).
This chapter provides an introduction and overview of fog computing. Initially it
provides an overview of cloud computing, mobile edge computing, and fog computing.
Then it makes a comparison between fog computing and cloud computing, as well
as fog computing and mobile edge computing. After that fog computing features are
discussed. Finally, at the end fog computing open research directions and conclusion
of the chapter are provided.
2
CLOUD COMPUTING
The idea of cloud computing is based on a very fundamental principal of reusability

of IT capabilities (Zhang, Zhang, Chen, & Huo, 2010). Cloud computing is a
computing paradigm, where a large pool of systems is connected in private or public
networks, in order to provide dynamically scalable infrastructure for application,
data and file storage (Dialogic Corporation, 2010). At the same time, the shared
cloud resources (networks, servers, data warehouses, applications and services) can
be rapidly provisioned and managed with minimal interaction by service providers.
The cloud computing users may use these resources for development, hosting
and running of services and applications on demand in a flexible way at any device,
at any time and at any place in the cloud. With the advent of this technology, the
cost of computation, application hosting, content storage and delivery is reduced
significantly.
The framework for cloud computing is defined in the recommendations ITU
Y.3501 (ITU-T, Cloud Computing Framework and High Level Requirements, 2013)
and Y.3510 (ITU-T, Cloud Computing Infrastructure Requirements, 2013) as well
as the NIST standards for cloud computing (NIST, 2013).
Mobile Cloud Computing
Today the mobile devices (smartphones, tablets, etc.) became essential part of the life,
as well as, the most effective and convenient tools for communication without any
limitations for time and space. The mobile user device accumulates rich experience
of the mobile applications (iPhone applications, Google applications etc.), that are
executed either on the mobile devices, or on the distant servers through the wireless
networks. The fast development of the Mobile Computing (MC) resulted to become
a powerful trend in the information technology. However, the mobile devices face
with many challenges with their resources (battery life, memory, storage, bandwidth,
processing power), environment (heterogeneity, availability and scalability) and
security (reliability and privacy). The constrained resources additionally worsen
the improvement of the quality of the services.
With the rapid increase of the mobile applications and the support of the cloud
computing for different types of services for the mobile device users is introduced
the Mobile Cloud Computing (MCC) as an integration of the cloud computing in
the mobile environment (Dihn, Lee, Niyato, & Wang, 2011; Khan, Othman, Madani,
& Khan, 2014; Fowler, 2013). The mobile cloud computing brings new types of
services for the mobile device that would use the benefits of the cloud computing.
3
MCC is an infrastructure used by mobile applications where both the data

storage and data processing are moved away from the mobile device to powerful and
centralized computing platforms located in the clouds (Kitanov & Janevski, 2014).
The access to this platform is enabled through the wireless network by using a thin
web client, or browser. Mobile cloud computing is usually referred to the following
two perspectives (Khan, Othman, Madani, & Khan, 2014):
• Infrastructure Based: Where the hardware infrastructure is static and

provides cloud services to mobile users; and
• Ad-Hoc Mobile Cloud: Where a group of mobile devices acts as a cloud and
provides cloud services to other mobile devices.
Cloud Computing Model and Architecture
(Mobile) cloud computing is made up of complex network and relationships of and

in between infrastructure Providers, Application/Services Providers, End-Users and
Developers, all producing and/or consuming applications and/or services on web
(Qureshi, Ahmad, & Shuja-ul-islam, 2011). Such MCC model is given on Figure 1.
Figure 1. (Mobile) Cloud Computing Model

Source: Qureshi, Ahmad, and Shuja-ul-islam, 2011
4
The Infrastructure Providers provide hardware and software infrastructure, or

services and applications, and/or all the above. The Application/Services Providers
are 1st tier consumer of Cloud Computing. They are typically business consumers
of cloud computing infrastructure and providers of applications and/or services. The
Developers are 2nd tier consumer of Cloud Computing, and they develop applications
and services that are typically hosted on the Cloud. The End Users also known as
3rd tier consumer of Cloud Computing, are typical end users of applications. They
consume applications that in turn consume services on the cloud, and they care
whether the application works well when needed with the necessary availability
level and the security.
A general mobile cloud computing architecture (Dihn, Lee, Niyato, & Wang,
2011) is given on Figure 2. Mobile devices are connected to the mobile, or wireless
network (GSM, GPRS, UMTS, HSPA, LTE, LTE-Advanced, LTE-A Pro, etc.) through
a base station (BTS, UTRAN, nodeB, enodeB), satellite link or access point (WiFi
or WiMAX). The mobile, or wireless network provides internet connectivity to the
users. Therefore, the users can access cloud-based services by Internet, if they have
mobile devices that support network connectivity.
Mobile users’ requests and users’ profiles are transmitted to the central processors
that are connected to servers providing mobile and wireless network services. Mobile
and wireless network operators can provide services Authentication, Authorization
and Accounting (AAA) for mobile users based on the Home Agent (HA) and
subscribers’ data stored in databases. After that, the subscribers’ requests are
delivered to cloud through Internet. In the cloud, the cloud controllers process the
users’ requests and provide the corresponding services from the Cloud Computing
Service Provider to mobile users.
Recently mobile applications have begun to adapt to cloud computing environment.
However, these applications are often linked with server instances running in the
Figure 2. Mobile Cloud Computing Architecture

Source: Dihn, Lee, Niyato, and Wang, 2011
5
cloud. Because of this, the MCC users may face some problems such as congestion
due to the limited bandwidth, network disconnection, and the signal attenuation
caused by mobile users’ mobility. This would cause delays when MCC users want
to communicate with the cloud, so QoS and QoE are significantly reduced.
CloneClouds and Cloudlets are some of the possible solutions which would
reduce the network delay (Dihn, Lee, Niyato, & Wang, 2011). CloneCloud uses
nearby computers, or data centers to increase the speed of running smart phone
applications, by cloning the entire set of data and applications from the smartphone
onto the cloud and selectively executing some operations on the clones, reintegrating
the results back into the smartphone. Cloudlet is a trusted, resource-rich computer,
or cluster of computers that has good Internet connection and it is available for use
by nearby mobile devices. The MCC users may use a cloudlet if it is available, and
if they do not want to offload the content and the request to the cloud (maybe due
to the delay, cost, etc.).
Cloud Computing Service Oriented Architecture
Once a cloud is established, how its cloud computing services are deployed in terms
of business models can differ depending on requirements (Dihn, Lee, Niyato, & Wang,
2011; Goyal, 2014). A layered architecture of cloud computing that demonstrates
the effectiveness in terms of meeting the user’s requirements is given on Figure 3.
The layer of Data Centers provides the hardware facility and infrastructure for
clouds. A number of servers are linked with high-speed networks to provide services
for customers. Usually, data centers are built in less populated places, with high
power supply stability, and a low risk of disaster.
The Infrastructure as a Service (IaaS) layer is built on top of Data centers layer. The
consumer is not able to manage the underlying cloud infrastructure. However, IaaS
provides provision processing, storage, networks, and other fundamental computing
resources where the consumer is able to deploy and run arbitrary software, which
can include operating systems and applications.
The Platform as a Service (PaaS) layer provides an advanced integrated
environment for building, testing and deploying custom applications. The consumer
Figure 3. Cloud Computing Service Oriented Architecture
6
does not control the underlying cloud infrastructure such as network, servers,
operating systems, or storage, but has control over the deployed applications and
possibly application hosting environment configurations.
The Software as a Service (SaaS) layer provides applications running on a cloud
infrastructure. These applications can be accessed by various client devices by using
a thin client interface such as a web browser. The consumer is not able to manage,
or control the underlying cloud infrastructure including network, servers, operating
systems, storage, or even individual application capabilities. However, the consumer
may control and manage only limited user-specific application configuration settings.
Also, there are subsets of these service layers that relate to a particular industry
or market. Such models are for example Network as a Service and Communication
as a Service (Dialogic Corporation, 2010).
Network as a Service (NaaS) is a category of the cloud computing services
where the user has an opportunity to use the transport connection services, and/or
network connection services between the clouds. NaaS services include flexible
and extended VPN, bandwidth on demand, etc.
Communications as a Service (CaaS) is a category of cloud computing services
where the user has an opportunity to use real time communication and collaborative
services. CaaS services include VoIP, instant messaging service and video conference.
Cloud Computing Deployment Models
Depending on the requirements different deployment models for (mobile) cloud

computing exist. The deployment models can be: private cloud, community cloud,
public cloud, hybrid cloud and inter-cloud computing (Goyal, 2014; Sqalli, Al-
saeedi, Binbeshr, & Siddiqui, 2012).
The Private Cloud, also known as Corporate Cloud, or Internal Cloud is a
proprietary computing architecture for an organization that provides hosted services
on private networks. The deployment, the maintenance, and the operations of the
cloud infrastructure are performed by the organization itself. The operation may
be in-house, or with a third party on the premises. However, this model has its own
disadvantages, since the organizations still need to purchase, set up, and manage
their own clouds.
The Community Cloud is a private cloud infrastructure that is shared among a
number of organizations with similar interests and requirements. This may help limit
the capital expenditure costs for its establishment as the costs are shared among the
organizations. The operation may be in-house, or with a third party on the premises.
The Public Cloud, or External Cloud infrastructure is available to the public on
a commercial basis by a cloud service provider. This enables a consumer to develop
and deploy a service in the cloud with very little financial outlay compared to the
7
capital expenditure requirements normally associated with other deployment options.

The users are connected to the cloud data centers that provide the cloud services
via the public Internet.
The Hybrid Cloud can be a combination of private and public clouds that support
the requirement to retain some data within the organization, and also the need to
offer services in the cloud. The clouds in this model have the ability through their
interfaces to allow data, and/or applications to be moved from one cloud to another.
The hybrid cloud offers a suitable environment for the needs of the enterprises, but
it also introduces the complexity of determining which services and applications
should be distributed across the private, public, or both clouds. In this model the
Cloud Management System (CMS) is responsible for the administration of hybrid
clouds. The CMS should contain some functionality such as security management,
resource scheduling (e.g., immediate, on-demand, or for later use), resource allocator,
and monitoring the university activities and performance. This information can be
used to determine the required resources that need to be allocated in the future. An
overview of hybrid cloud computing model is given in Figure 4.
The Inter‐cloud computing is a model that provides оn demand cloud computing
resources, such as processing, storage and network, as well as, a distribution of the
workload through the internetworking of the clouds.
Cloud Computing Benefits
Cloud computing and mobile cloud computing would provide many benefits and
advantages for cloud computing providers, mobile network operators, and cloud
computing consumers such as:
Figure 4. Hybrid Cloud Computing Model

Source: Sqalli, Al-saeedi, Binbeshr, and Siddiqui, 2012
8
• Sharing information and applications without the need of complex and costly
hardware and software since data processing is running on the cloud;
• Enhanced features and functionalities of mobile devices through new cloud
applications;
• Easy access to mobile cloud computing through a browser with the mobile
devices;
• Possibility of one application to be shared and accessed by many mobile
device users;
• Broader reach and dissemination of mobile cloud computing applications;
• Increased battery power for mobile devices;
• Improved data storage capacity and processing power since cloud computing
will enable the users to store/access and to process the big data such as
modeling of 3D graphics visualization and animation in ecology, global
climate solutions, financial risks, healthcare and medical learning,
• Decoding genome projects, etc. through the wireless and mobile networks on
the cloud where High Performance Computing (HPC) system resides;
• Higher level of availability of the cloud computing;
• Improved reliability since data and computer applications are stored and
backed up on a number of computers; and
• Possibility for Distance Learning system.
Cloud Computing Services
Cloud computing and mobile cloud computing would provide to the consumers the
following basic services: platform services, application services and context-rich
support services (Bahl, Han, Li, & Satyanarayanan, 2012). Platform services include
the following: storage, database, memory caching, content distribution, processing
and computing. Cloud computing provider may also offer the following application
services: presence service for location-based applications, video transcoding and
streaming proxy for video streaming applications, push notifications, and speech
and image recognition.
In the near future, many applications would become more personalized, and
more context-aware. They will be able to recognize user’s identity, user’s location,
and user’s preferences. To support these applications Cloud Computing providers
should support context-rich support services such as context extraction service,
recommendation service and group privacy service. Of particular importance is
the context extraction service that performs data mining analysis of mobile data
combined with other forms of data such as social networking data, and sensor network
data in order to extract contextual clues relevant to the user. Data mining services
should be able to scale and analyze large group of people and large quantities of
9
data (big data) that that they generate in order to extract collective trends among
the population of users in real time.
Additionally, crowd actuation services, such as recommendation services based
on collective group context rather than individual context need to be created and
scaled. By using these clues, a layer of cloud recommendation services can be built
that creates output that is adjusted to a user, or set of users with those contextual
characteristics.
CLOUD COMPUTING AND TELECOM OPERATORS
In general, the big Over The Top (OTT) players have a leading role in the cloud
computing market. Dropbox, Apple, and Google are the main personal cloud
providers, while Salesforce.com, Google, and Microsoft are the leaders in SaaS
market. Amazon is mostly used for the PaaS and IaaS services.
One part of the cloud computing market would also belong to the Telecom
operators. Compared to OTT players, the telco operators have recently entered in
the cloud computing market (Brown, 2013).
Why Telecom Operators Should Use the Cloud
There are two main reasons why telecom operators should include the cloud computing.
The first reason is to use the benefits of cloud computing for IT optimization (lower
costs, bigger elasticity and speed). The second reason is to use the new business
opportunities.
Тhe telco operators have a competitive advantage on a local level, where they
can benefit from the local commercial presence, compared to many ОТТ players.
Due to their relative close proximity to the users, the telecom operators may offer
cloud services with very low latency. Additionally, they can offer services to the
users that are flexible and elastic in terms of the frequency bandwidth. Finally, the
users have more trust and confidence in telco operators rather than in many Internet
brands, or OTT players.
The Cloud in Telecom Operators
Telecom operators are investigating how they can make the best use of their assets,
and how future network investment can be aligned with a cloud model (Brown,
2013). One option is to make use of network equipment to host server modules to
create telecom-grade clouds. This could include a mix of centralized cloud and a
10
distributed cloud using access network and RAN elements known as Cloud RAN
or CRAN, as it is shown on Figure 5.
Cloud Radio Access Network (CRAN)
Since the smart mobile devices have limited capabilities for storing and processing of
data, mobile cloud computing solves this issue by moving the storing and processing
of data from the smart mobile device to the cloud computing centers. However, this
requires high bandwidth and low latency.
In order to solve this issue, the CRAN moves the cloud computing functionalities
to the radio access network (Brown, 2013; Peng, Li, Jiang, Li, & Wang, 2014).
Currently hosting content and applications in the CRAN has become popular. By
placing storage and computing resources at, or close to, the cell site, operators can
improve response times for the services requested by the users in the prevailing radio
conditions. This might be useful for congestion control, or rate adaptation for video
streams. On the network side, cell site caching can reduce demand on the backhaul
network, and potentially play a role in limiting signaling to the core network. The
two primary advantages of placing content close to the radio and close to the user
are application performance and network efficiency. One emerging application
Figure 5. Cloud Integration with Telecommunication Network

Source: Kitanov, & Janevski, 2014
11
where both application performance and network efficiency could work together in
a distributed cloud model is LTE Broadcast that uses evolved Multimedia Broadcast
Multicast (eMBMS) technology (Brown, 2013).
However, the application storing and all radio signal processing functions are still
centralized at the cloud computing server in the core part of the network. Because
billions of smart user devices need to transmit and exchange their data fast enough
with the Base Band Unit (BBU) pool, there is a requirement for high bandwidth
and low latency.
To overcome this, Heterogeneous Cloud Radio Access Networks (H-CRANs)
have been proposed in which the user and control planes are decoupled (Peng, Li,
Jiang, Li, & Wang, 2014; Peng, Li, Zhao, & Wang, 2015). as it is shown in Figure
6. The centralized control function is shifted from the BBU pool in CRANs to the
High-Power Nodes (HPNs) in HCRANs. HPNs are also used to provide seamless
coverage and execute the functions of control plane. The high-speed data packet
transmission in the user plane is enabled with the radio heads (RRHs). HPNs are
connected to the BBU pool via the backhaul links for interference coordination.
Figure 6. Mobile Network with CRAN and HCRAN
12
However, HCRANs still have its own drawbacks. The data traffic data over the
fronthaul between RRHs and the centralized BBU pool surges a lot of redundant
information, which worsens the fronthaul constraints. In addition, HCRANs do not
take fully utilize the processing and storage capabilities in edge devices, such as
RRHs and smart user devices, which is a promising approach to successfully alleviate
the burden of the fronthaul and BBU pool. Finally, operators must deploy a huge
number of fixed RRHs and HPNs in H-CRANs in order to meet the requirements
of peak capacity, which makes a serious waste when the volume of delivery traffic
is not sufficiently large.
MOBILE EDGE COMPUTING
Recently the mobile devices became an important tool for learning, entertainment,
social networking, and acquiring new information from the news and the business
(Ahmed, Gani, Sookhak, Hamid, & Xia, 2015). However, due to the constrained
resources of the mobile devices (the processing power, battery life, and the capacity
for data storing) the mobile users do not receive the same quality as the conventional
desktop users (Dihn, Lee, Niyato, & Wang, 2011). When the mobile cloud computing
appeared, the limited resources in the mobile devices of data storage and data
processing was resolved with the transfer of these resources to the cloud. Many
cloud computing services such as m-health-care (Hoang, & Chen, 2010), (Adibi,
2014), mobile learning, or m-learning (Sun, & Shen, 2014), mobile video games
(m-gaming) (Wu, Yuen, Cheung, Chen, & Chen, 2015) and m-governance (Sabarish,
& Shaji, 2014). are already directly available from the mobile devices.
However, because the data need to be transferred from the mobile devise to the
cloud computing centers, there is a need for increased bandwidth, and lower latency,
which causes a network overload to occur. It is expected the need for the frequency
bandwidth to be doubled every year (Ericsson, 2016), (Atzori, Iera, & Morabito,
2010). Moreover, the Internet of Things and the future Internet of Everything, would
provide the devices with constrained resources to be connected on Internet (Brech,
Jamison, Shao, & Wightwick, 2013.), (Atzori, Iera, & Morabito, 2010).
In order to resolve this issue, one possible proposed solution is the Cloudlet,
where the mobile devices transmit data via a WiFi network for further processing
to a server with less resources than the cloud, but is nearby to the mobile devices
(Satyanarayanan, Bahl, Caceres, & Davies, 2009). However, this approach is less
effective due to the following two reasons. Firstly, the access to the Cloudlet is only
possible through a WiFi access point. And secondly the Cloudlet possess less resources
than the cloud, which means it is not scalable in resource and service provisioning.
13
Therefore, a better proposed solution is the Mobile Edge Computing (MEC)

(Ahmed, & Ahmed, 2016; Beck, Werner, Feld, & Schimper, 2014). MEC is a model
that enables business-oriented cloud platform to be implemented within the radio
access network, close the mobile users in order to serve applications that are delay
sensitive and context aware.
This approach is initiated by ETSI, where data processing and storage happens
at the edge in the radio access network in the base station, rather than in cloud
computing centers, in order to create new application and service possibilities and
opportunities. In this way, the mobile edge computing manages to reduce latency,
alleviates the network congestion in the core, enables efficient traffic management, and
opens possibility other industrial services to be delivered for the critical applications
through the mobile network. This approach provides innovative network architecture
where the possibilities for cloud computing and IT services are converged with the
mobile network.
The MEC architecture is given on Figure 7. The mobile edge computing entity is
positioned next to the radio access network. This entity works with downlink bitstreams
from the cloud computing servers to the mobile device and uplink bitstreams from the
mobile device to the cloud computing servers. The MEC platform contains standard
IT servers and network devices inside, or outside of the base station (Yu, 2016). The
external applications are executed in the Virtual Machines (VMs) (Buyya, Broberg,
& Goscinski, 2011), connected with network devices. Also, there is a possibility
the МЕС platform to be implemented with standard IT servers, where the network
device is implemented as a software entity, such as Open Virtual Switch, or Оpen
vSwitch (OVS) (Pfaff, Pettit, & Koponen, 2015).
The MEC platform contains the following fundamental modular functionalities:
routing, exposure of network capabilities and management. The routing module is
responsible for forwarding the packets between the MEC platform and the radio
access network and the core part of the mobile network, as well as within the MEC
platform itself. The module network exposure provides authorized information to the
Radio Network Information Service (RNIS) and the Radio Resource Management
(RRM). The management module supports the ААА security function, as well as
the management of the external applications within the MEC platform itself. Here
is included the orchestration of setting up the application and the authorization for
the exposure of the network capabilities station (Yu, 2016).
МЕС enables the mobile network to evolve in an innovative way to handle the
increased traffic volume, that arise from various domains in the next 5G mobile and
wireless network, which are grouped into enhanced Mobile BroadBand - eMBB (Liu,
Hou, & Jin, 2017), massive Machine Type Communication - mMTC (Jovović, &
Forenbacher, 2015), and Ultra-Reliable and Low Latency Communication – URLLC
(Yilmaz, Wang, & Johansson, 2015).
14
Figure 7. Mobile Edge Computing Architecture
However, mobile edge computing devices and entities within the domain are
standalone, or interconnected through proprietary networks with custom security
and little interoperability. Although the modern Mobile Edge Computing attempts
to include some functions of Cloud Computing like interoperability, local security
etc., however, it does not extend to the cloud or across domains.
15
FOG COMPUTING
In order to resolve the limitations of (mobile) cloud computing and mobile edge
computing, a new paradigm known as fog computing arose. Fog Computing extends
cloud computing and services to the edge of the network. Fog computing would
combine the study of mobile communications, micro-clouds, distributed systems,
and consumer big data. It is a horizontal, system-level architecture that distributes
computing, storage, control and networking functions closer to the users along a
cloud-to-thing continuum (OpenFog Consortium, 2017).
The fog computing architecture uses one, or a collaborative multitude of end-
user clients, or near-user edge devices to carry out a substantial amount of storage,
communication and management (Chiang, 2015). It carries out a substantial amount
of storage, communication and management such as network measurement, control
and configuration, at or near the end-user (rather than stored primarily in large-scale
data centers).
Rather than concentrating data and computation in a small number of large
clouds, many fog systems will be deployed close to the end users or where computing
and intelligent networking can best meet user needs. The core idea is to take full
advantages of local radio signal processing, cooperative radio resource management,
and distributed storing capabilities in edge devices, which can decrease the heavy
burden on front haul and avoid large-scale radio signal processing in the centralized
baseband unit pool (Peng, Yan, Zhang, & Wang, 2016).
Reasons for the Rise of Fog Computing
There are four main reasons for the rise of fog computing (Chiang, 2015):
1. Real Time Processing and Cyber-Physical System Control: Edge data

analytics, as well as the actions it enables through control loops, often have
stringent time requirement in the order of few milliseconds that can be carried
out only at the edge of the network. This is particularly essential for Tactile
Internet, that enables virtual-reality-type interfaces between humans and
devices.
2. Cognition or Awareness of Client-Centric Objectives: The applications can
be enabled by knowing the requirements and the preferences of the clients.
This is particularly true when privacy and reliability cannot be trusted in the
Cloud, or when security is enhanced by shortening the extent over which the
communication is carried out.
16
3. Increased Efficiency by Pooling of Idle and Unused Local Resources: The

idle and unused gigabytes on many devices, the idle processing power, the
sensing ability and the wireless connectivity within the edge may be pooled
within a fog network.
4. Agility, or Rapid Innovation and Affordable Scaling: It is usually much
faster and cheaper to experiment with client and edge devices. Rather than
waiting for vendors of large boxes inside the network to adopt an innovation,
in the fog computing world a small team may take advantages of smart phone
Application Programming Interface (API) and Software Development Kit
(SDK), proliferation of mobile applications, and offer a networking service
through its own API.
Fog Computing Applications
Fog computing was proposed to support applications whose requirements don’t

quite match the QoS and QoE guarantees provided by the cloud computing, such
as (Gupta, Chakraborty, Ghosh, & Buyya, 2017):
• Applications having stringent latency requirements, such as mobile gaming,

video conferencing, etc.;
• Geo-distributed applications where the data collection points are distributed
over a wide area, for instance, pipeline monitoring, or sensor networks to
monitor the environment;
• Fast mobile applications involving highly mobile users (smart connected
vehicle, connected rail); and
• Large-scale distributed control systems consisting of a vast number of sensors
and actuators, such as smart grid, connected rail, smart traffic light systems,
etc., working in a coordinated manner to improve user experience.
Fog Computing Architecture
An overview of Fog Computing architecture is given in Figure 8 (Kitanov, Monteiro,

& Janevski, 2016). It consists of centralized cloud computing center, IP/MPLS core
network, RAN network with distributed fog computing intelligence and smart things
network. Each smart thing device is attached to one of fog devices in the RAN
network. The fog devices could be interconnected to each other, and each of them is
linked to the centralized cloud computing center via the IP/MPLS core the network.
The RAN network is actually an intermediate fog layer that consists of geo-
distributed intelligent fog computing servers which are deployed at the edge of
networks, e.g., parks, bus terminals, shopping centers, etc. Each fog server is a
17
Figure 8. Cloud Integration with Telecommunication Network

Source: Kitanov, Monteiro, & Janevski, 2016
18
highly virtualized computing system and is equipped with the on-board large volume
data storage, computing, and wireless communication facility (Gao, Luan, Liu, &
Yu, 2017).
The role of fog servers is to bridge the smart mobile device things and the cloud
(Vaquero & Rodero-Merino, 2014). The geo-distributed intelligent fog servers directly
communicate with the mobile users through single-hop wireless connections using
the off-the-shelf wireless interfaces, such as, LTE, WiFi, Bluetooth, etc. They can
independently provide pre-defined service applications to mobile users without
assistances from cloud, or Internet. In addition, the fog servers are connected to
the cloud in order to leverage the rich functions and application tools of the cloud.
Fog computing and networking contains both data plane and control plane that
enable different applications with different communication protocols over all layers
in the OSI system (Chiang, 2015). This is illustrated in Figure 9. Fog data plane is
focused on 5G mobile network, IoT, and the future IoE. Fog control plane is mainly
about cyber physical system control and real-time data analytics.
A Comparison Between Cloud and Fog Computing
A comparison between Fog Computing and Cloud Computing is given in (Gao,

Luan, Liu, & Yu, 2017), which is summarized in Table 1.
Fog Computing is targeted for mobile users, while cloud computing is targeted
for general internet users. Cloud Computing provides global information collected
from worldwide, while fog computing provides limited localized information services
related to specific deployment locations. Cloud services are located within the
Internet, while fog services are located at the edge of the network. There is a single
hop between the smart user devices and the fog computing server, and multiple hops
between the user devices and the cloud computing centers. Fog computing contains
very large number of server nodes, while cloud computing contains few server nodes.
The latency and the delay jitter are very low in fog computing environment, while
Figure 9. Data and Control Plane of Fog Computing

Source: Chiang, 2015
19
Table 1. A Comparison between Cloud and Fog Computing
Cloud Computing Fog Computing

Target Type General Internet Users Mobile Users
Limited localized information
Global Information
Service Type services related to specific
Collected from Worldwide
deployment locations
Service Location Within the Internet At the edge of the local network
Distance between client and server Multiple hops Single hop
Number of server nodes Few Very large
Latency High Low
Delay jitter High Very low
Geo-distribution Centralized Distributed
Security Undefined Can be defined
Ample and scalable storage, Limited storage, processing
Hardware processing and computing and computing power, wireless
power interface
Centralized and maintained Distributed in regional areas and
Deployment
by OTT service providers maintained by local businesses
Source: (Gao, Luan, Liu, & Yu, 2017)
in cloud computing provides there are very high latency and delay jitter. The cloud
computing centers are centralized, while the fog computing nodes are densely geo-
distributed. The security in cloud computing is undefined, while in fog computing it
can be defined. The hardware in cloud contains ample and scalable storage, and very
high processing and computing power. On the other hand, the hardware in the fog
contains limited storage and limited processing and computing power, and wireless
interface. Finally, the cloud computing deployment is centralized and maintained
by OTT service providers, while the fog computing is distributed in regional areas
and maintained by local business, which are close to the end users and therefore
they have bigger trust in them.
In addition, the cloud computing represents an efficient and scalable centralized
solution for information management and distribution, for the traditional desktop
users that request global information from the remote central server such as world
news, stock market in different countries, etc. However, nowadays there are mobile
Internet users using smartphones and tablets, which massively demand local
information around them (Gao, Luan, Liu, & Yu, 2017). For example, a mobile user
in a shopping center is interested about the sales, open hour, restaurants and events
inside the attended shopping center; while such information becomes useless once
he/she leaves the shopping center.
20
The conventional cloud-based Internet is inefficient in serving the local information

desired by mobile users, since the local stores need first to upload their information
to a remote cloud server over Internet, and then the mobile users to obtain the desired
information from the remote cloud server. Although the physical distance between
the mobile user and the stores is short, by using the remote cloud to provide the
necessary information, the actual communication distance, from the mobile user
to the cloud server to can be far, which may result with increased latency, and high
congestion.
Fog Computing, on the other side represents a practical and efficient solution
to resolve the mismatch between physical and communication distances. A Fog
server can be deployed inside the shopping center or store that will distribute the
local store flyers to mobile users. As such, the physical distance is equal to the
communication distance and users can acquire low-latency desirable services (Gao,
Luan, Liu, & Yu, 2017).
Cloud and fog form a mutually beneficial, inter-dependent continuum (Chiang,
2015). They are inter-dependent, e.g., coordination among devices in a Fog may
rely on the Cloud. They are also mutually beneficial: certain functions are naturally
more advantageous to carry out in Fog while others in Cloud.
A Comparison Between Mobile Edge

Computing and Fog Computing
Although the Mobile Edge Computing provides many benefits, still it has certain
constraints and limits compared to the fog computing. In order to see these constraints
and limits it is necessary to make a comparison between these two approaches.
Table 2 summarizes the main characteristics and features of mobile edge
computing and fog computing (Nebbiolo Technologies, 2015). MEC provides limited
application hosting, while fog computing fully supports it. Data service at edge and
device and application management are provided by both MEC and Fog Computing.
The security and safety are partially solved with Virtual Private Network (VPN) or
Firmware (FW) at MEC. On the other side, Fog Computing provides a complete
E2E solution for data protection on session level and hardware. Fog computing
also provides support of elastic compute and resource pooling, modular hardware,
real time control high level availability and virtualization with windows support.
The МЕС is developed by ETSI and the vendors for telecommunication equipment
in order to deliver standardized МЕC architecture and industrial standardized
application programming interface for external applications. This concept represents
a control and management of independent end user device individually, or through a
set of software functions in a cloud domain in the radio access network. The devices
21
Table 2. Main Features of Mobile Edge Computing and Fog Computing
Mobile Edge Computing Fog Computing

Application hosting Limited Yes
Data Service at Edge Yes Yes
Device and Application Management Yes Yes
A Complete E2E solution, data
Partially solved with VPN
Security and Safety protection, on session level and
or FW
hardware
Elastic compute / resource pooling No Yes
Modular Hardware No Yes
Virtualization with Windows support To be defined Yes
Real time control high level availability No Yes
Source: (Nebbiolo Technologies, 2015)
and the entities that perform computing at the edge of the network in the cloud domain
are either independent, or mutually connected through proprietary networks with
usual security and low degree of interoperability. Recently this approach attempts to
redefine the data computing range at the edge by including some functionalities for
fog computing, such as interoperability, local security and safety, etc. However, this
is not extended to the cloud or through the domains (Nebbiolo Technologies, 2015).
Оn the other hand fog computing is a concept proposed by CISCO and other
manufacturers for Internet equipment and it represents a completely distributed
network, multi-layered cloud computing architecture for data processing, where exists
billions of devices as part of IoT, multiple local clouds at the edge of the network – fog
and main central hyper scalable cloud computing data center. A single application in
the fog is distributed through the devices through the cloud components embedded
in the nodes in the different network levels, for example in the radio access network,
multi-service edge, the core of the network (in the IP/MPLS routers and switches,
the gateways of the mobile packet core, etc.). In this way the cloud is closer to the
users of mobile devices and is capable to offer ultra-low latency, quicker response,
high bandwidth, as well as real time access to the radio information that will be
used by the applications in order to offer context related services.
Compared to Mobile Edge Computing, the edge devices in the Fog Computing
may independently self-optimize among themselves and collectively to perform
measurement and management to the remaining of the network. The fog computing
moves the operational functions of the digital objects in Information Centric Networks
(ICNs) and virtualized functions in the Software Defined Networks (SDNs) at the
edge of the network (Chiang, 2015). Table 3 summarizes the main differences
between mobile edge computing and fog computing.
22
Table 3. Differences between Mobile Edge Computing and Fog Computing
No Mobile Edge Computing Fog Computing

Device aware and few services aware, Device independent, intelligent, and aware of the entire
1
unaware of the entire domain fog domain
2 Limited control in the edge domain Controls all devices in the domain
3 Not aware for the cloud Extends the cloud to a fog level in a continuum
4 Limited network scope Complete network scope
5 No IoT vertical awareness Support and enabler for multiple IoT verticals
6 No IoT vertical integration Integrates multiple verticals
Uses fog nodes that are very versatile and capable
Uses Edge Controllers that are focused on
7 of performing a variety of functions like Real time
edge device command and control
Control, application hosting and management.
8 Security scope is limited to devices End-to-End security through data ownership
9 Not designed with virtualization Enables rich virtualization
Fog Analytics enables collection, processing and
analysis of data from multiple devices in the edge for
10 Analytics scoped to a single device
analysis, machine learning, anomaly detection and
system optimization.
Edge Computing typically is embedded
Fog Computing uses the devices and the embedded
in and controls the edge. Certain devices
edge control. However, Fog
11 require hard real time control and others
Computing also enables virtual machines that host soft
require non real time control, and the Edge
PLC used in real time control.
Computing performs these functions.
Source: (Nebbiolo Technologies, 2015)
Although these two concepts are different from each other, still they share a similar
vision, which is guided by the future predictions that will be characterized with
the Internet of Things, Internet of Everything, Tactile Internet and the existence of
appropriate mobile and wireless connectivity. It can be concluded that fog computing
is a better solution than MEC and cloud computing. However, better results can be
achieved if both concepts are combined (Singh, Chiu, & Yang, 2016).
FOG COMPUTING FEATURES
The following subsections describe some of the main Fog Computing Features.
Ubiquity and Heterogeneity of Fog Nodes and Devices
The main factor which would bring fog into reality is the ubiquity of the devices,
whose increase is driven by the user devices and sensors/actuators, and enabled by
23
the presence and usage of devices everywhere around us for different services and
applications. By decreasing the size of the device, the device portability is increased.
However, the power consumption is also reduced, which may be crucial in some
context application. This can be solved with packaging and power management
technologies such as System on Chip and System in Package Technologies, 3D
Micro batteries, RF powered Computing, etc. that aim to create smaller and more
autonomous mobile devices which would run longer at a minimum price (Vaquero
& Rodero-Merino, 2014).
Fog nodes come range from high performance servers, edge routers, gateways,
access points, base stations, etc. These hardware platforms have varying levels of
computation and storage capabilities, run various kinds of operating system (OS),
and load different software applications. (Aazam, & Huh, 2016). Therefore, fog
nodes are heterogeneous.
Heterogeneity of Fog Computing Network Infrastructure
The network infrastructure of fog computing is also heterogeneous, which includes

high-speed links connecting to data center, as well as wireless access technologies
(for example, WLAN, WiFi, 3G, 4G, ZigBee, etc.) connecting to the edge devices
(Bonomi, Milito, Natarajan, & Zhu, 2014).
Support for Seamless Mobility
In fog computing scenarios, there are various mobile devices such as smart phones,
vehicles, and smart watch, as well as there are also static end devices, such as traffic
cameras. In addition, fog computing nodes in the radio access network can also
be a mobile or static computing resource platform. (Varshney & Simmhan, 2017;
Hossain & Atiquzzaman, 2013).
Therefore, it is essential for fog computing to communicate directly with mobile
devices. Moreover, various mobile devices can also communicate directly to each
other. The data does not have to be transmitted to the cloud or even the base station.
The end device itself or intermediate devices process the massive data generated by
the Internet of things, and truly realizing mobile data analysis. So, it can provide
services for more extensive nodes.
Low Latency and Real Time Interactions
Fog nodes are capable to process and store data generated by sensors and devices,
by network edge devices in local area network. This significantly reduces data
movement across the Internet and provides speedy high-quality localized services
24
supported by endpoints. Therefore, it provides low latency and meets the demand of
real time interactions, especially for latency-sensitive or time sensitive applications
(Bonomi, Milito, Zhu, & Addepalli, 2012), Tactile Internet, etc.
Effective Bandwidth Saving
Fog computing extends the computation and storage capabilities to the network
edge to perform data processing and storing between the end nodes and traditional
cloud. Some computation tasks, such as, data preprocessing, redundancy removing,
data cleaning and filtering, valuable information extraction, are performed locally.
Only part of useful data is transmitted to the cloud, and most of the data don’t need
to be transmitted over the Internet. In this way, fog computing saves the bandwidth
effectively, especially for big data processing.
Geographical Distribution and Decentralized Data Analytic
In IoT and ubiquitous computing environment, the purpose is to achieve the

interconnection and interworking among ubiquitous things, which are huge in
number, and widely distributed. The characteristic of geographical distribution and
decentralized data analytics can effectively meet the above demands.
The fog computing architecture consists of large number of widely geographically
distributed nodes, that have the ability to track and derive the locations of end
devices in order to support the mobility. Instead to process and store information in
centralized cloud computing data centers far away from end-user, the decentralized
architecture of fog computing ensures the proximity of data analytics to the customer.
This characteristic can support faster analysis of big data, better location-based
services, and more powerful capabilities of real-time decision making.
For example, fog computing can provide a wealth of Internet of Vehicles (IoV)
services (including traffic security and data analysis, urban and road conditions,
entertainment information, etc.) based on the connection and interaction of vehicle
to vehicle, vehicle to access points (Kang, Wang, & Luo, 2016).
Interoperability
Due to their heterogeneous nature, fog nodes and end devices come from different
providers and are usually deployed in the various environments. Therefore, fog
computing nodes and devices must be able to interoperate and cooperate from
different providers in order to cope with wide range of services and seamlessly
support certain services (Kang, Wang, & Luo, 2016).
25
Network Management
The configuration and the maintenance of many different types of services running
on many heterogeneous devices would only exacerbate the current management
problems. Therefore, in fog computing environment heterogeneous devices and
their running services need to be managed in a more homogeneous manner with
the following technologies: Network Function Virtualization (NFV); small edge
clouds to host services close to, or at the endpoints; and peer-to-peer (P2P)- and
sensor network-like approaches for auto-coordination of applications (Vaquero &
Rodero-Merino, 2014).
NFV is the reaction of telecom operators to their lack of agility and constant need
for reliable infrastructures. It is capable of dynamically deployment for on-demand
network services, or user-services. Software Defined Networks (SDNs) are one of
the main enablers for NFV. For example, the router can be seen as an SDN-enabled
virtual infrastructure where NFV and application services are deployed close to
the place where they are actually going to be used, which would result in cheaper
and more agile operations. However, NFV capabilities still do not reach end user
devices and sensors.
Telecom operators had already started to deploy clouds in their Long-Term
Evolution (LTE) Networks closer to the edge (to the user), because the Evolved
Packet Core (EPC) more efficiently delivers services close to users (at the edge)
and confines traffic there while reducing the traffic overload with the help of SDNs.
The fog would enable the devices to become virtual platform, that can lease some
computing/storage capacity for applications to run on them.
The peer-to-peer (P2P) and sensor network-like approaches exploit the locality
and allow endpoints to cooperate in order to achieve similar results, but can scale
better, and can be implemented in a fog. A fog application can be seen as a Content
Distribution Network (CDN) where a data is exchanged between peers. As a result,
the applications and data are no longer required to stay in centralized data centers.
A part of a network and some user devices/sensor can act as mini-clouds in a fog
computing environment. The mini clouds can be implemented by using droplets,
or small pieces of code that run in a secured manner in devices at the edge with
minimum interaction with central coordinating elements, and thus reducing the
unnecessary and undesired uploads of data to central servers in corporate data
centers. The users are able to retain control and ownership of their own data and
applications, and the scalability is improved.
26
Fog Connectivity
The presence of many mobile devices that consume and produce big data at the
edge of the network may cause a huge bottleneck in the fog (Vaquero & Rodero-
Merino, 2014). On physical level the following technologies can cope with this:
LTE-Advanced, WiFi ac, Bluetooth Low Energy, ZigBee, etc.
On the network level, each node must be able to act as a router for its neighbors
and must be resilient to node churn (nodes entering and leaving the network) and
mobility. Mobile Ad-hoc Networks (MANETs) and Wireless Mesh Networks (WMNs)
can provide these functionalities (Vaquero & Rodero-Merino, 2014). MANET would
enable the formation of densely populated networks without requiring fixed and
costly infrastructures to be available beforehand.
WMNs on the other side use mesh routers at its core. Nodes can use the mesh
routers to get connectivity, or other nodes if no direct link with the routers can be
established. Routers would grant access to other networks such as cellular, Wi-Fi, etc.
On higher levels some protocols already exist for IoT, such as Message Queue
Telemetry Transport (MQTT) and Constrained Application Protocol (CoAP), that
provide low resource consumption and resilience to failure (Vaquero & Rodero-
Merino, 2014). The network and IoT protocols can benefit from data locality, since
they no longer need to send all the data around the world all the time, except for
potential congestion problems at the edge of the network. Data locality has also a
very positive impact on privacy.
Fog Privacy and Security
The greatest concern of fog users is data ownership, i.e. data security and privacy
(Stojmenovic & Wen, 2014; Stolfo, Salem, & Keromytis, 2012). One method to
maintain the privacy is by storing encrypted sensitive data in traditional clouds.
However, the existing cloud computing data protection mechanisms such as
encryption have failed in preventing data theft attacks, especially those perpetrated
by a malicious insider.
However, the value of stolen information can be decreased. This can be achieved
through a preventive disinformation attack, by using the following additional security
features: user behavior profiling, decoys and combination of both (Stolfo, Salem,
& Keromytis, 2012).
User behavior profiling is used to model the normal user behavior, i.e. how, when,
and how much a user accesses his/her information in the cloud (Stolfo, Salem, &
27
Keromytis, 2012). Such profiles contain volumetric information, i.e. how many files
are accessed and how often. The occurrence of abnormal access to a user information
in the cloud can be determined by monitoring this normal user behavior, based
partially upon the scale and the scope of data being transferred.
Decoys are any bogus information that can be generated on demand. They are
used to:
• Validate whether data access is authorized when abnormal information access

is detected, and;
• Confuse the attacker with bogus information (Stolfo, Salem, & Keromytis,
2012).
The serving decoys would confuse the malicious attacker into believing that
he/she has ex-filtrated useful information, but he/she has not. The attacks can be
prevented by deploying decoys within the fog by the service customer and within
personal on-line social networking profiles by the individual users.
A combination of decoys with user behavior profiling would provide unprecedented
levels of security for the fog, and would improve detection accuracy. When the access
to user information is correctly identified as an unauthorized access, the fog security
system would deliver unbounded amounts of decoy information to the attacker. Thus,
the true user data is protected from unauthorized disclosure. When abnormal access
to the fog service is not recognized, decoy information may be returned by the fog
and delivered in such a way as to appear completely legitimate and normal. The
true owner of the information, would identify when decoys are returned by the fog.
Hence the legitimate user could alter the fog responses through a variety of means,
such as challenge questions, to inform the fog security system that has inaccurately
detected an unauthorized access. At the moment the existing security mechanisms
do not provide this level of security (Stolfo, Salem, & Keromytis, 2012).
Low Energy Consumption
Since the fog computing nodes are widely dispersed geographically, it will not generate
a lot of heat due to concentration, and need not additional cooling system. In addition,
short range communication mode and some optimal energy Management Policies
of mobile nodes evidently reduce communication energy consumption (Zhang,
Niyato, Wang, & Dong, 2016). This would lead to reduce power consumption, save
energy and decrease the cost. Therefore, fog computing would provide a greener
computing paradigm.
28
FOG COMPUTING RESEARCH DIRECTIONS
Many of the ideas in Fog Computing have evolved from Peer-To-Peer (P2P) networks,
Mobile Ad Hoc Network (MANET) and edge computing (Chiang, 2015). Compared
to P2P networks, fog is not just about content sharing (or data plane as a whole),
but also network measurement, control and configuration, and service definition.
Compared to MANET, today there are much more powerful and diverse off-the-shelf
edge devices and applications, together with the broadband mobile and wireless
networks. Compared to edge computing, in the edge devices in fog computing
environment may optimize among themselves, and they collectively measure and
control the rest of the network.
Fog computing also relocates the operating functions on digital objects of
Information Centric Networks and the virtualized functions in Software Defined
Networks at the network edge (Chiang, 2015). Because fog computing is recently
new emerged paradigm there are many challenging questions that need to be
resolved, such as:
• Cloud-Fog Interface and Fog-Fog Interface: Cloud would remain useful as

Fog arises. The tasks that require real-time processing, end user objectives
or low-cost leverage of idle resources should go to Fog, while the tasks that
require massive storage, heavy-duty computation, or wide area connectivity.
Therefore, the cloud-fog interface and fog-fog interface need clearly to
defined in order to make the transfer of information much easier.
• Interactions between Smart User Device Hardware and Operating
Systems: Once the operating functions are moved on smart user devices,
the interface with their OS and hardware becomes essential, particularly for
pooling idle edge resources.
• Trustworthiness and Security (Vaquero & Rodero-Merino, 2014):
Because of the proximity to end users and locality on the edge, fog computing
nodes can act as the first node of access control and encryption, provide
contextual integrity and isolation, and enable the control of aggregating
privacy-sensitive data before it leaves the edge.
• Convergence and Consistency Arising Out of Local Interactions: Typical
concerns of distributed control, divergence/oscillation and inconsistency of
global states, become more acute in a massive, under-organized, possibly
mobile crowd with diverse capabilities and virtualized pool of resources
shared unpredictably. Use cases in edge analytics and stream mining provide
additional challenges on this recurrent challenge in distributed systems.
29
• A Decision of What Kind of Information to Keep Between Distributed

(Local) and Centralized (Global) Architectures: While maintaining
resilience through redundancy.
• Distributed P2P Mobile Cloud Computing Among the Smart User
Devices: In order to guarantee certain QoS/QoE level.
CONCLUSION
This chapter provided an introduction and overview of fog computing. Initially it

provided an overview of cloud computing, mobile edge computing, and fog computing.
Then was made a comparison between fog computing and cloud computing, as well
as fog computing and mobile edge computing. After that fog computing features
were discussed. Finally, at the end fog computing open research directions and
were provided.
Fog computing is an extension of the traditional cloud-based computing model
where implementations of the architecture can reside in multiple layers of a network’s
topology. Due to its proximity to end-users, its dense geographical distribution, and
its support for mobility, fog computing reduces service latency, and improves QoS,
resulting in superior user-experience. It is well positioned for real-time big data
analytics, emerging IoE applications, 5G, Tactile Internet, etc.
However, all the benefits of cloud should be preserved with these extensions
to fog, including containerization, virtualization, orchestration, manageability, and
efficiency. Therefore, many challenging questions such as cloud-fog interface, and
fog-fog interface, security, etc. need to be resolved in order fog computing together
with cloud computing to be fully operational.
In the future, more and more virtual network functionality would be executed
in a fog computing environment, which would provide mobiquitous service to the
users. This should enable new services paradigms such as Anything as a Service
(AaaS) where devices, terminals, machines, and also smart things and robots would
become innovative tools that would produce and use applications, services and data.
REFERENCES
Aazam, M., & Huh, E. N. (2016). Fog computing: The cloud-iot/ioe middleware
paradigm. IEEE Potentials, 35(3), 40–44. doi:10.1109/MPOT.2015.2456213
30
Adibi, S. (2014). Biomedical Sensing Analyzer (BSA) for Mobile-Health (m-Health)

– LTE. IEEE Journal of Biomedical and Health Informatics, 18(1), 345–351.
doi:10.1109/JBHI.2013.2262076 PMID:24403433
Ahmed, A., & Ahmed, E. (2016). A Survey on Mobile Edge Computing. IEEE
International Conference on Itelligent System and Control (ISCO 2016). 10.1109/
ISCO.2016.7727082
Ahmed, E., Gani, A., Sookhak, M., Hamid, S. H., & Xia, F. (2015). Application
Optimization in Mobile Cloud Computing: Motivation, Taxonomies, and Open
Challenges. Journal of Network and Computer Applications, 2, 52–68. doi:10.1016/j.
jnca.2015.02.003
Atzori, L., Iera, A., & Morabito, G. (2010). The internet of things: A survey. Computer
Networks, 54(15), 2787–2805. doi:10.1016/j.comnet.2010.05.010
Bahl, P., Han, Y. R., Li, E. L., & Satyanarayanan, M. (2012). Advancing the State
of Mobile Cloud Computing. Proceedings of the 3rd ACM workshop on Mobile
cloud computing and services (MCS 2012), 21 – 28. 10.1145/2307849.2307856
Beck, M., Werner, M., Feld, S., & Schimper, S. (2014). Mobile edge computing: A
taxonomy. Sixth International Conference on Advances in Future Internet.
Bonomi, F., Milito, R., Natarajan, P., & Zhu, J. (2014). Fog Computing: A Platform
for Internet of Things and Analytics. Springer International Publishing.
Bonomi, F., Milito, R., Zhu, J., & Addepalli, S. (2012). Fog Computing and its Role
in the Internet of Things. Proceedings of the First Edition of the ACM SIGCOMM
Workshop on Mobile Cloud Computing (MCC 2012). 10.1145/2342509.2342513
Brech, B., Jamison, J., Shao, L., & Wightwick, G. (2013). The Interconnecting of
Everything. IBM Corp.
Brown, G. (2013). Converging Telecom and IT in the LTE RAN. White paper at
Heavy Reading on behalf on Samsung Cloud Computing Principles and Paradigms.
Chiang, M. (2015). Fog Networking: An Overview on Research Opportunities.
White Paper.
Dialogic Corporation. (2010). Introduction to Cloud Computing. Dialogic Corporation
White Paper.
Dihn, T. H., Lee, C., Niyato, D., & Wang, P. (2011). A Survey of Mobile Cloud
Computing: Architecture, Applications, and Approaches. Wireless Communications
and Mobile Computing, Wiley, 13(18), 1587–1611.
31
Ericsson. (2016). Ericsson Mobility Report on the Pulse of the Networked Society.
Stockholm, Sweden: Ericsson.
Fowler, S. (2013). Survey on Mobile Cloud Computing Challenges Ahead. IEEE
CommSoft E-Letters, 2(1), 13–17.
Gao, L., Luan, T. H., Liu, B. W. Z., & Yu, S. (2017). Fog Computing and Its
Applications in 5G. In 5G Mobile Communications (pp. 571-593). Springer
International Publishing Switzerland.
Goyal, S. (2014). Public vs Private vs Hybrid vs Community – Cloud Computing:
A Critial Review. International Journal of Computer Network and Information
Security, 6(3), 20–29. doi:10.5815/ijcnis.2014.03.03
Gupta, H., Chakraborty, S., Ghosh, K. S., & Buyya, R. (2017). Fog Computing in
5G Networks: An Application Perspective. In Cloud and Fog Computing in 5G
Mobile Networks: Emerging advances and applications (pp. 23-54). The Institute
of Engineering and Technology (IET), IET Telecommunications Series 70.
Hoang, D. B., & Chen, L. (2010). Mobile cloud for assistive healthcare (mocash).
Services Computing Conference (APSCC).
Hossain, M. S., & Atiquzzaman, M. (2013). Cost analysis of mobility protocols.
Telecommunication Systems, 52(4), 2271–2285. doi:10.100711235-011-9532-2
ITU-T. (2013). Cloud Computing Framework and High Level Requirements. ITU-T
Recommendation Y.3501.
ITU-T. (2013). Cloud Computing Infrastructure Requirements. ITU-T
Jovović, I., & Forenbacher, I. M. P. (2015). Massive Machine-Type Communications:
An Overview and Perspectives Towards 5G. The 3rd International Virtual Research
Conference In Technical Disciplines.
Kang, K., Wang, C., & Luo, T. (2016). Fog computing for vehicular ad-hoc networks:
Paradigms, scenarios, and issues. Journal of China Universities of Posts and
Telecommunications, 23(2), 56–96. doi:10.1016/S1005-8885(16)60021-3
Khan, A. R., Othman, M., Madani, S. A., & Khan, S. U. (2014). Survey of Mobile
Cloud Computing Application Models. IEEE.
32
Kitanov, S., & Janevski, T. (2014). State of the Art: Mobile Cloud Computing.
Proceedings of the Sixth IEEE International Conference on Computational
Intelligence, Communication Systems and Networks 2014 (CICSYN 2014), 153-158.
10.1109/CICSyN.2014.41
Kitanov, S., Monteiro, E., & Janevski, T. (2016). 5G and the Fog – Survey of Related
Technologies and Research Directions. Proceedings of the 18th Mediterranean IEEE
Electrotechnical Conference MELECON 2016. 10.1109/MELCON.2016.7495388
Liu, G., Hou, X., Jin, J., Wang, F., Wang, Q., Hao, Y., ... Deng, A. (2017). 3-D-MIMO
With Massive Antennas Paves the Way to 5G Enhanced Mobile Broadband: From
System Design to Field Trials. IEEE Journal on Selected Areas in Communications,
35(6), 1222–1233. doi:10.1109/JSAC.2017.2687998
Liu, Y. J. E., Fieldsend, E. J., & Min, G. (2017). A Framework of Fog Computing:
Architecture, Challenges and Optimization. IEEE Access: Practical Innovations,
Open Solutions, 5, 25445–25454. doi:10.1109/ACCESS.2017.2766923
Mitchell, S., Villa, N., Stewart-Weeks, M., & Lange, A. (2013). The Internet of
Everything for Cities Connecting People, Process, Data, and Things to Improve
the ‘Livability’ of Cities and Communities. CISCO, white paper.
Mouradian C., Naboulsi D., Yangui S., Glitho H. R., Morrow J. M., & Polakos A. P.
(2017). A Comprehensive Survey on Fog Computing: State-of-the-art and Research
Challenges. IEEE Communications Surveys and Tutorials, 1 – 51.
NebbioloTechnologies. (2015). Fog vs Edge Computing v1.1. Milpitas, CA: Nebbiolo
Technologies Inc.
NIST. (2013). NIST Cloud Computing Standards Roadmap. NIST Special Publication
500-291, Version 2.
OpenFog Consortium. (2017). OpenFog Reference Architecture for Fog Computing.
OpenFog Consortium Architecture Working Group.
Peng, M., Li, Y., Jiang, J., Li, J., & Wang, C. (2014). Heterogeneous cloud radio
access networks: A new perspective for enhancing spectral and energy efficiencies.
IEEE Wireless Communications, 21(6), 126–135. doi:10.1109/MWC.2014.7000980
Peng, M., Li, Y., Zhao, Z., & Wang, C. (2015). System architecture and key
technologies for 5G heterogeneous cloud radio access networks. IEEE Network,
29(2), 6–14. doi:10.1109/MNET.2015.7064897 PMID:26504265
33
Peng, M., Yan, S., Zhang, K., & Wang, C. (2016). Fog Computing based Radio
Access Networks: Issues and Challenges. IEEE Network, 30(4), 46–53. doi:10.1109/
MNET.2016.7513863
Pfaff, B. J., Pettit, J., & Koponen, T. (2015). The design and implementation of open
vSwitch. Networked Systems Design and Implementation.
Qureshi, S. S., Ahmad, T., & Shuja-ul-islam, K. R. (2011). Mobile Cloud Computing
as Future for Mobile Applications – Implementation Methods and Challenging
Issues. Proceedings of IEEE Conference on Cloud Computing and Intelligence
Systems (CCIS).
Sabarish, K., & Shaji, R. (2014). A Scalable Cloud Enabled Mobile Governance
Framework. Global Humanitarian Technology Conference - South Asia Satellite
(GHTC-SAS). 10.1109/GHTC-SAS.2014.6967554
Satyanarayanan, M., Bahl, P., Caceres, R., & Davies, N. (2009). The Case for VM-
based Cloudlets in Mobile Computing. IEEE Pervasive Computing, 8(4), 14–23.
doi:10.1109/MPRV.2009.82
Singh, S., Chiu, Y., & Yang, Y. T. (2016). Mobile Edge Fog Computing in 5G Era:
Architecture and Implementation. 2016 International Computer Symposium.
Sqalli, H. M., Al-saeedi, M., Binbeshr, F., & Siddiqui, M. (2012). UCloud: Simulated
Hybrid Cloud for a University Environment. IEEE 1st International Conf. on Cloud
Networking (CLOUDNET).
Stojmenovic, I., & Wen, S. (2014). The Fog Computing Paradigm: Scenarios and
Security Issues. Federated Conference on Computer Science and Information
Systems, 2(5).
Stolfo, S. J., Salem, M. B., & Keromytis, A. D. (2012). Fog Computing: Mitigating
Insider Data Theft Attacks in the Cloud. Proceeding of IEEE Symposium on Security
and Privacy Workshops (SPW). 10.1109/SPW.2012.19
Sun, G., & Shen, J. (2014). Facilitating Social Collaboration in Mobile Cloud-based
learning: A Teamwork as a Service (TaaS) Approach. IEEE Transactions on Learning
Technologies, 7(3), 207–220. doi:10.1109/TLT.2014.2340402
Vaquero, L. M., & Rodero-Merino, L. (2014). Finding your Way in the Fog: Towards
a Comprehensive Definition of Fog Computing. ACM SIGCOMM Computer
Communication Review Newsletter, 44(5), 27–32. doi:10.1145/2677046.2677052
Varshney, P., & Simmhan, Y. (2017). Demystifying Fog Computing: Characterizing
Architectures, Applications and Abstractions. arXiv:1702.06331
34
Wu, J., Yuen, C., Cheung, N.-M., Chen, J., & Chen, C. (2015). Enabling Adaptive
High-frame-rate Video Streaming in Mobile Cloud Gaming Applications. IEEE
Transactions on Circuits and Systems for Video Technology, 1-1.
Yilmaz, O. N., Wang, Y.-P. E., & Johansson, N. A. (2015). Analysis of ultra-
reliable and low-latency 5G communication for a factory automation use case.
IEEE International Conference on Communication Workshop (ICCW). 10.1109/
ICCW.2015.7247339
Yu, Y. (2016). Mobile Edge Computing Towards 5G: Vision, Recent Progress,
and Open Challenges. China Communications, 13(Supplement No. 2), 89–99.
doi:10.1109/CC.2016.7833463
Zhang, S., Zhang, S., Chen, X., & Huo, X. (2010). Cloud Computing, Research
and Development Trend. Proceedings of the 2nd IEEE International Conference
on Future Networks.
Zhang, Y., Niyato, D., Wang, P., & Dong, I. K. (2016). Optimal energy management
policy of mobile energy gateway. IEEE Transactions on Vehicular Technology, 65(5),
3685–3699. doi:10.1109/TVT.2015.2445833
Zhang, Z., Zhang, J., & Ying, L. (n.d.). Multimedia streaming in cooperative mobile
social networks. Academic Press.
35
36
Chapter 2
Fog Computing and
Its Challenges
Vighnesh Srinivasa Balaji
ABSTRACT
In recent times, the number of internet of things (IoT) devices/sensors increased
tremendously. To support the computational demand of real-time latency-sensitive
applications of largely geo-distributed IoT devices/sensors, a new computing
paradigm named fog computing has been introduced. In this chapter, the authors
will introduce fog computing, its difference in comparison to cloud computing, and
issues related to fog. Among the three issues (i.e. service, structural, and security
issues), this chapter scrutinizes and comprehensively discusses the service and
structural issues also providing the service level objectives of the fog. They next
provide various algorithms for computing in fog, the challenges faced, and future
research directions. Among the various uses of fog, two scenarios are put to use.
INTRODUCTION
In recent times the number of internet of things has increased rapidly in such a
way the use of “Cloud Computing” is not sufficient. Due to the increasing need for
real-time latency-sensitive applications of geo-distributed devises a new computing
paradigm called “Fog Computing” has been introduced. Generally, Fog extends
the services of the cloud-based computing, storage and networking facilities as
the fog resides closer to the Internet of Things (IOT) devices. Fog computing is a
distributed computing paradigm that acts as an intermediate layer in between Cloud
data centres and IoT devices/sensors. Fog computing was introduced by Cisco in
DOI: 10.4018/978-1-5225-6070-8.ch002
Fog Computing and Its Challenges
2012 to resolve the challenges faced by cloud computing. Since cloud computing
is geo-centric, it often fails to deal with storage and processing demands of billions
of geo-centered IoT devices. Fog datacentres are distributed at the edge of the
network along with latency-sensitive requirements. However, fog computing does
not substitute cloud computing, rather these two complement each other which
allows the users to experience a new breed of computing technology that provides
the advantages of both the paradigms (Haruna, Abdu, Manis, Francisca, Oladipo,
Ezendu, & Ariwa, 2017).
In simple words, computation done at the network’s edge is referred to as fog
computing, with its purpose to provide computing, storage and connectivity services
to the users at the network edge. It’s a supporting computer paradigm for the internet
of things and will also boost the development of IOT applications (Makwana, 2017).
This paper scrutinizes the challenges related to Fog computing, ways to overcome
them and future research directions.
DIFFERENCE BETWEEN FOG AND CLOUD
Cloud computing is usually a model for enabling convenient, on-demand network

use of a shared pool of configurable computing resources (e.g., networks, servers,
storage, applications, and services) that may be rapidly provisioned and released
with minimal management effort or vendor interaction. Cloud is located within
the network with various topologies, speeds and no central control due to which
there are a few qualities of service factors unresolved. One such issue is latency, as
many applications require real time data processing, and services provided by the
cloud cannot satisfy these requirements. Another such a problem is security and
privacy. In the internet today, the applications are located far off from the service
providers and so depending on the number of intermediate nodes the data moves
through public cloud thus compromising confidentiality and integrity of the data
as specified in Figure 1.
Fog computing was introduced by CiscoSystems as new a model to ease wireless
data transfer to distributed devices in the Internet of Things (IoT) network paradigm.
Fog Computing acts as a paradigm that extends Cloud computing and brings its
related services to the network edge. Fog, similar to Cloud, provides data, compute,
storage, and application services to end-users. The characteristics distinguishing Fog
are its dense geographical distribution, its proximity to end-users, and its support
for mobility. By doing so, it improves QoS and also reduces latency, increases
its mobility which supports the internet of everything (IoE). Thanks to its wide
geographical distribution the Fog paradigm is well positioned for real time big data
and real time analytics. Fog supports densely distributed data collection points,
37
Figure 1. Confidentiality and integrity of the data
Table 1. Differences between Fog and Cloud Computing (Maher, 2015)
Cloud Computing Fog Computing

Requires less time as processing of data and application is
Requires more time as it involves processing
done at network edge rather than working on a centralized
of data and applications in a cloud
cloud
Instead of sending every bit of data over cloud channel,
Sending every bit of data over a cloud
they were aggregated at certain points which reduces the
channel results in a bandwidth problem
demand for bandwidth
It is possible to remove the problem of response time and
Depending on remote servers, results in slow
scalability problems by placing small edge servers in the
response time and scalability problems
visibility of the users
hence adding a fourth axis to the often-mentioned Big Data dimensions (volume,
variety, and velocity).
Fog
• Fog nodes are geographically distributed over the edge each network and are
logically decentralized as different nodes are owned by different organizations
which makes them unreliable.
38
• Provide transient storage, often 1–2 hours.

• Send periodic data summaries to the cloud.
Cloud
• Cloud servers are centralized and situated in large warehouse like datacentres.
This allows for recovery, load balancing and failure recovery which makes
them more reliable.
• Receives and aggregates data summaries from many fog nodes.
• Stores large amounts data and performs analysis on IOT data. (CISCO, 2015).
SERVICE LEVEL OBJECTIVES
The fog has various services at different stages and each service pertains to a
particular functionality, each crucial and different in its own way. However, each
service needs to performed with certain objectives kept in mind. E.g. the services
must ensure there is no latency in their execution.
Here are a few objectives of the services provided by the Fog.
1. Latency Management
The most severe dis-advantage of cloud is the time latency, this has to be overcome
using Fog computing. Each service will have a maximum tolerable threshold for
service delivery which gives us the maximum tolerance of a service or a QoS
requirement and we must ensure that the service needed is given with this threshold.
This is done using Latency Management. This can be done in various ways one such
was where nodal collaboration is emphasized, so that any service can be distributed
among all the nodes so that the service is provided within the threshold (Oueis,
Strinati, Sardellitti, & Barbarossa, 2015).
Another way is to distribute tasks equally between the client and fog nodes so as
to improve the latency problem (Zeng, Gu, Guo, Cheng, & Yu, 2016). Yet another
solution is to use a low latency fog network. The main objective here is to select a
fog node among the entire network that provides the least latency.
2. Cost Management
Cost management in Fog computing is related to Capital Expenses(CAPEX) and

Operating Expenses(OPEX). The main contributor of CAPEX is the deployment
cost of the fog nodes and the network. One way to reduce to this to optimize the
39
number of nodes based on their functionality and also to optimize the places where
the nodes have to be placed (Zhang, Lin, Yin, & Zhao, 2016). Fog nodes have been
considered a platform to launch VMs. One way to reduce OPEX is to find a suitable
set of fog nodes to host the VM so as to reduce total overdue cost (Gu, Zeng, Guo,
Barnawi, & Xiang, 2015).
3. Network Management
Network management in Fog computing includes core-network congestion control,

support for Software Define Network (SDN)/ Network Function Virtualization
(NFV), assurance of seamless connectivity, etc.
This can be done in three ways:
1. Network Management: The overhead of fog nodes which includes interaction

with cloud and also with various IOT end devices and sensors greatly decrease
efficiency. This burden of overhead can be solved using a layered architecture
of fog nodes so that the computational tasks and overhead is distributed among
the various layers (Aazam, & Huh, 2014).
2. Virtualization: Virtualization is one of the best ways to improve network
management. This is done by virtualizing the conventional networking system.
One of the key enablers of virtualized network is SDN. SDN networking
technique that decouples the control plane from networking equipment and
implements in software on separate servers. An important aspect of SDN is
its support for NFV. NFV is a brilliant architectural concept that virtualizes
traditional networking functions like Network Address Translation (NAT),
Domain Name Service(DNS), caching etc.
3. Connectivity: This ensures seamless communication of end devices with other
entitieslike Cloud, Fog, Desktop computers, Mobile devices, end devices,
etc. despiteof their physical diversity. As a consequence, resource discovery,
maintenance of communication and computation capacity become easier within
the network. Several works in Fog computing have already targeted this issue
and proposed new architecture of Fog nodes e.g. IoT Hub (Cirani, Ferrari,
Iotti, & Picone, 2015) and Fog networking, Vehicular Fog Computing (Hou,
Li, Chen, Wu, Jin, & Chen, 2016) for connectivity management and resource
discovery.
4. Computational Management
In all the Fog nodes there is a lot of data needed to be processed and with the growth
of IoT, computation on data is required every second.
40
Computation, again can vary among nodes depending in the applications and
the services required. Computational management ensures all the computation of
all the nodes have to equally distributed.
The most crucial Service Level Objective is proper Computational Management.
This is done in three ways.
1. Resource Estimation: Knowing the exact resources required the resources

can be allocated according to certain policies so that certain QoS could be
achieved by which accurate service price can be imposed. Resource estimation
policies are based on user characteristics, experienced QoE, features of service
accessing devices etc. (Aazam, & Huh, 2016).
2. Workload Allocation: Ensures that there is maximum allocation of resources
and minimum delay in the provision of services. The main reason for this
introduction is to balance computation load on Fog nodes and client devices.
As a consequence, overhead on both will be reduced (Zeng, Gu, Guo, Cheng,
& Yu, 2016).
3. Coordination: Due to the distribution of fog nodes among various organizations,
the deployment of large scale applications is done in various nodes. In such
cases proper coordination among fog nodes is vital (Giang, Blackstock, Lea,
& Leung, 2015).
5. Application Management
Programming platforms are essential for efficient application management. Besides,

scalability and offloading techniques contribute to application management.
1. Programming Platforms Provide Run Time Environment, Libraries and

Packages to the Fog Nodes: Due to the dynamic nature of the fog nodes assurance
of proper programming platforms is challenging. To solve this challenge, a
new programming platform called Mobile fog is introduced (Hong, Lillethun,
Ramachandran, Ottenwälder, & Koldehofe, 2013). This provides simplified
abstraction of programming models to develop large scale applications.
2. Scaling: Refers to the adaption capability of applications even after unexpected
events. Scaling techniques also be applied inusers service access and application
scheduling. To support scalable scheduling of data stream applications,
architecture of a QoS-aware self-adaptive scheduler has been recently proposed
in Fog computing.
3. Offloading: Refers to techniques to send tasks from resource constrained
nodes to resource enriched nodes for computation (Carla, Mouradian, Diala,
Naboulsi, Sami, Yangui… Polakos, 2017).
41
ISSUES RELATED TO FOG COMPUTING
The three main issues are discussed as follows:
1. Structural Issues
a. Identification of techniques and metrics for inter nodal collaboration and
resource sharing are vital and equally difficult as in Fog computing the
nodes are distributed across the edge of the network and maybe shared
or virtualized.
b. The components used in Fog computing usually both edge and core
network components have various kinds of processors but not usually
equipped with general purpose computing. To provide general purpose
computing to them might be difficult.
c. Although the structural orientation is compatible with IOT, competency
assurance in other networks like CDN, vehicular network is a real challenge.
d. The selection of nodes, corresponding resource allocation is also crucial
in Fog computing.
2. Service Oriented Issues
a. Some fog nodes are resource enriched while others are resource constrained.
Large scale applications when developed in resource constrained nodes
makes it difficult for computations to continue. So, programming platforms
are required.
b. There must be specific policies to distribute computational tasks are
resource among Fog nodes.
c. There are many factors that affect the Service Level Agreement(SLA)
e.g. service cost, energy usage, application characteristics, data flow etc.
3. Security Oriented Issues
a. As fog nodes are distributed among various organizations, security
concerns are rising. Authentication is required for services.
b. The design of Fog nodes is based on networking, fog nodes are vulnerable
to security attacks.
c. Malicious users might also be present among the normal users, which
may cause additional threats.
ALGORITHMS FOR COMPUTING IN THE FOGSYSTEM
Here we list a few existing algorithms for each sub-heading and their associated
drawbacks. Like for any large-scale computing system, application-agnostic and
42
application-specific algorithms have been proposed for fog. Furthermore, the

application-agnostic algorithms cover the computing, storage/distribution, and
energy consumption, like in most large-scale distributed computing systems. This
section discusses the application- agnostic algorithms and the application-specific
algorithms for fog systems.
Many algorithms have been proposed for computing in the fog systems. In the
following, we review them. We first present compute resource sharing algorithms.
We then cover task scheduling algorithms and present after that offloading and load
redistribution algorithms. In each we suggest the best algorithms available.
Resource Sharing
In computing, resource sharing refers to one host making available computer

resources to other hosts on the same network. It refers to a piece of information that
can be remotely accessed for various reasons on any other computer. Inter process
communication makes resource sharing possible.
When it comes to computing in the fog systems, a first aspect that has been
investigated is the compute resource sharing and cooperation among the nodes.
These aspects have been tackled so far in the fog stratum, with the objective of
executing compute demands.
1. One way is to provide resource sharing among the fog nodes in the same fog
domain. Fog nodes involve themselves in a one-to-one paring in such a way
that the required task can be fulfilled. This algorithm defines a utility metric
for a couple of nodes that account for communication costs. Using this metric,
a node first finds a preference list of pairing nodes. The target node based on
its previous level of pairing decides whether or not to make connection with
this new node. This pairing continues to happen across the fog domain. The
evaluation of this strategy shows that it outperforms a greedy approach when
it comes to total utility. In this strategy Quality of Service criterion is not met
and the evaluation is based on a small section of nodes and scalability and
mobility of nodes criterions are also not met.
2. Another algorithm forms small cluster of nodes containing a small number of
nodes that share resources for offloading. It aims at simultaneously forming
clusters and respecting each node’s latency constraint keeping the main
objective to minimize power consumption. Even in this algorithm the mobility
and scalability criterions are not met.
43
Task Scheduling
In computing, scheduling refers to the method in which work by some means is

allocated to resources. The way in which task is scheduled by the fog nodes among
the billions of tasks available to them is task scheduling.
As the fog systems provide additional computing capabilities at the edge of the
network, a major question that they raise is how to manage task execution. More
precisely, how to decide which tasks to execute in the IoT/end-user’s stratum, the
fog stratum, and the cloud stratum? On a finer level, to which nodes a particular
task should be assigned? What metrics to consider in deriving decisions?
1. First algorithm’s task scheduling operates in two major steps. First step allocates
computational resources to each node based on a specific objective and at the
end of this step some request may not be processed due to lack of resources
and this carried out in the next step again based on a particular order and
objective. However, this algorithm fails to meet the scalability and mobility
of nodes criterion.
2. A second algorithm proposes a proactive resource sharing algorithm to all
users based on their historical data and allocates resources to users that are
loyal to the services requested. By doing this the QoS has greatly improved.
In addition to this the algorithm also has a pricing strategy that accounts for
user’s loyalty.
3. A third algorithm considers that in a fog domain in the fog stratum, a fog server
manager receives the user’s requests and is responsible for matching the fog
resources and the user demands. Upon the receipt of a request, the fog server
manager verifies whether enough computational resources are available in the
fog domain. Depending on the available resources, either it executes all tasks,
executes part of them and postpones the execution of others, or it transfers the
demand to nodes in cloud stratum, in order to run tasks over the cloud nodes.
Simulation results show that the proposed algorithm is more efficient than
other existing strategies aiming at optimizing the response time or enabling
load-balancing. It leads to a lower maximum response time and lower maximum
processing time values, at lower costs. Even this like the previous algorithms
fails to meet the scalability and mobility criterions.
Offloading and Load Redistribution
Among the various Fog nodes available the tasks that are allocated to each node
may vary for various reasons. Each task requires resources to complete and if a node
44
has too many tasks then its resources may get depleted, so load among the various
nodes have to redistributed properly.
As discussed in the previous section, several task-scheduling algorithms have been
proposed. While they allow distributing compute tasks over compute nodes across
the three strata of the system, the possibility of unbalance in terms of workload has
not been discussed. Thus, the need for the algorithms that perform offloading and
load redistribution in the system has increased.
One of the algorithms start with an atomization step that maps physical resources
into virtual resources. A graph representing the system is then built. In this graph, a
node represents a virtual resource with a certain capacity and a couple of these nodes
are linked through edges, weighted by the bandwidth of communication link between
them. Upon the arrival of new fog nodes, in order to balance the loads, redistribution
of loads in its neighbourhood is done by the algorithm while accounting for the
task distribution degree and the links among the nodes. In case of node removal, a
reverse strategy is adopted. The evaluation of the algorithm is done against a static
one from the literature. The results show the proposed strategy has led to a lower
number of moves in the graph, which implies a lower migration cost. Additionally,
it requires less time to derive results with respect to the static strategy.
ALGORITHMIC CHALLENGES AND RESEARCH DIRECTIONS
In this section, the most important algorithms to fog computing are discussed along
with the challenges faced and their future research directions.
Heterogeneity
By heterogeneity, we mean the level of difference of computational facilities and

storage capabilities of the systems, this degree of heterogeneity has huge impact on
performance. In this respect, two complementary challenges are addressed:
1. First challenge concerns existing computing and storage devices around us,
e.g., end-user mobile devices that can serve as fog nodes. These end devices
which act as fog nodes apart from their own functionality need to allocate
memory and resources to serve the requests from other fog nodes. The question
is how much of their resources can they spare to serve as fog nodes, capable
of processing other devices requests and storing their content? Actual usage
patterns are concluded from deriving the actual usages of the storage resources
45
and computing of the existing devices. When conducted over a large variety
of potential devices, assessing the capabilities that these devices can offer are
analysed. As a result, the degree of heterogeneity in terms of computing and
storage capabilities that these devices impose on the system is determined.
Adequate prediction models can then be derived accordingly to enable decision-
making over the future. Such models can be inspired from those proposed in
the context of volunteer computing systems (Sekma, Dridi, & Elleuch, 2015).
2. Second, apart from existing devices around us that can serve as fog nodes
additional computing and storage nodes can be added in the system. These
devices in turn participate in defining the heterogeneity degree of the system.
Here, the question is how to decide the dimensioning and placement of
these additional nodes. One research direction is to study the corresponding
optimization problem. It aims to full-fill the demand of IoT/end-user devices
in the fog system considered as an input. After consideration of a number of
aspects the algorithm, which include power consumption and resource costs,
would then optimize the dimensioning of nodes. (Farahani & Hekmatfar, 2009).
QoS Management
The incorporation of latency constraint addresses this problem. Along with the
latency, other crucial performance metrics such as uplink and downlink bandwidth
or resource usage costs on the user’s side are generally not taken into account –
which have to be addressed as a different challenge. Other algorithms, targeting
resource utilization or particular applications, need in turn to account for different
performance metrics and relevant costs. Instead of having these metrics as part of
constraints, they should form the algorithm’s objective. This can improve QoS,
rather than imposing a limit on a certain QoS metric.
One big challenge is that QoS metrics are covered as individual objectives, instead
of considering them together with other objectives, has not received significant
attention so far. However, in reality, several objectives that can even be contradictory
may need to be covered at the same time. One example is to minimize completion
time and minimize power consumption costs in the fog system when managing
compute resources. In this case, the need for more resources minimizing completion
time in the system, in contrast to minimizing power consumption. Thus, there is
always a trade-off to consider. Due to the complexity of such problems, so far, only
has addressed such contradictive objectives. However, the study remains limited
to a video streaming application in content delivery networks and only targets the
distribution of content from the data centres to the fog nodes.
46
Scalability
Operation of an algorithm in the fog system over a large scale is very important.
Validating an algorithm in a small-scale environment, with a few devices and nodes
does not guarantee it performs well over a large scale. Despite the importance of
this criterion, most of the proposed algorithms have been evaluated over small-scale
scenarios.
Besides the considered scale, proposed algorithms need to operate in real-
world conditions. Consider the case where a task scheduling algorithm needs to be
operational for real-world traffic patterns. Most of the proposed algorithms except
for a few experimental evaluations, have not been tested in real-world environment
but only validated in simplistic simulation environments, with no clear motivation
for system parameters. As a result, some other problems may arise when these
algorithms are tested in a real-world environment, diverging from the expected
performance. To handle these challenges, a possible research direction is to run
real-world experimental evaluations over a large scale. However, due to high costs
this may not be possible. Large-scale realistic evaluations remain the best option to
compare the different algorithms. However, to do so, two major aspects are to be
highlighted. First, as of now, we still unfortunately lack a clear knowledge of real-
world system deployments. Second, we do not yet acquire a clear understanding
of the real-world evolution of application and services consumption on the side of
IoT/end-user devices. Only through machine learning techniques, characterization
can only be achieved through thorough analysis of corresponding real-world traces.
When acquired, proper models need to be derived based on it to enable proper
evaluations through realistic simulations.
Mobility
Mobility poses significant challenges in fog systems when IoT devices are considered.
The services provided by a Fog domain must be valid even if it crossed the domain
of that fog through various other fog domains which needs proper strategies that
allow handling the mobility however, efforts in this direction still remain limited.
A migration strategy that allows moving components across fog domains according
to the movement of devices. A research direction here is to derive realistic mobility
models, to do so, real-world mobility traces need to be integrated into the analysis,
as done in the case of VM migration in mobile cloud systems. Moreover, accurate
mobility prediction methods are needed to complement algorithms operating in
real-time. Depending on the context, these schemes would either target individual
or group mobility and can be built based on collected traces of IoT/end-user devices.
47
Similarly, to manage the mobility of a fog node, we also need to ensure the
offered services are not interrupted. In fact, the fog nodes’ mobility is even more
complex to handle, as it involves the serving resources availability. For instance, as
a fog node leaves a fog domain, it implies the need to offload tasks assigned to it
to other fog nodes in the system. Instead, as a fog node joins/creates a fog domain,
additional/novel resources would be available for IoT/end-user devices that can be
connected to it.
Federation
Among all the algorithms it is seen that that the federation has been mostly ignored.
Resource sharing algorithms is considered among only among one distributor and
not among various distributors. Indeed, federation among the distributors greatly
improves the efficiency and functionality of the system. Indeed, the federation
among providers extends the capabilities of the system and has not received so far,
any attention from an algorithmic perspective.
USE CASES
A Fog System Use Case for CDN
To deliver Quality of Experience and delivering of video at a cost-efficient rate

CDN is used. It uses two types of servers i.e. origin servers and surrogate servers.
The content of the origin server is duplicated to the surrogate servers based on
conditions like physical distance, network conditions, content availability. Cloud
based CDN are recently developing, called CCDN. Some servers while providing a
video game to an end user may be located at a greater physical distance and might
therefore reduce the QoE. A potential solution is to introduce a fog stratum between
the cloud and the end users.
In Figure 2, use-case suppose Jason wants to stream some video on his phone,
the surrogate server replicates this data to access point A (action 1) and Jason can
access (Action 2) it with lesser delay as compared to when it is given directly by
the surrogate server. Now suppose Jessica also wants to access the same video can
do some with increased QoE (Action 3). And now Jason moves to a different point
further from A then point A distributes its content to Point B (Action 5) so that it
can be accessed by Jason (Action 6) with a better QoE as when compared to the
surrogate server replicating the data.
48
Figure 2. A Fog System Use Case for CDN
A Fog System Use Case for Fire Detection and Fighting
A Fire-detecting and Fire-fighting use case is provided. It dispatches firefighting

robots based on the intensity of the fire, contour, wind intensity. It is made up
of three components: Fire Detector (A), Firefighting Strategies (B), and Robots
Dispatcher (C). Usually these are located in the cloud far from the IoT devices but
placing them in Fog stratum increases the QoE and reduces latency and chances to
detect and fight fire are increased.
First the fire-detector or robot dispatcher are moved to the fog from the cloud
(Action 1) and detecting of fire can be made with very low latency (Action 2). In
case of fire the fire-detector notifies the fire-fighting strategies (Action 3) which
communicates with the robot-dispatcher (Action 4) which fights the fire and
extinguishes it (Action 5). Figure 3 shows the sample diagram of fire-fighting and
detection algorithm (Carla, Mouradian, Diala, Naboulsi, Sami, Yangui… Polakos,
2017).
49
Figure 3. A Fog System Use Case for Fire Detection and Fighting
CONCLUSION
In this chapter we have learnt the basics of Fog computing and its fundamental
difference to Cloud computing. In order to improve the QoS of Fog there are a
few service level objectives which have to be met such as latency, cost, network,
computational and application management. Although Fog computing has various
advantages over Cloud computing, this too has certain issues to it which are
classified as service, structural and service related issues. We have surveyed the
various algorithms to compute in the Fog and the corresponding challenges faced
in each. Moreover, based on these challenges the future research directions have
been mentioned along with the use cases of Fog computing in day-to-day activities.
50
REFERENCES
Aazam, M., & Huh, E. N. (2014). Fog computing and smart gateway based
communication for cloud of things. In Future Internet of Things and Cloud (FiCloud).
IEEE.
Aazam, M., St-Hilaire, M., & Lung, C. H. (2016). Mefore: Qoe based resource
estimation at fog to enhance qos in iot. 23rd International Conference on
Telecommunications (ICT). 10.1109/ICT.2016.7500362
Carla, Diala, & Sami, Roch, Glitho, Morrow, & Polakos. (2017). A Comprehensive
Survey on Fog Computing: State-of-the-art and Research Challenges. IEEE
Communications Surveys and Tutorials.
Cirani, S., Ferrari, G., Iotti, N., & Picone, M. (2015). The IoT hub: a fog node
for seamless management of heterogeneous connected smart objects. Sensing,
Communication, and Networking-Workshops (SECONWorkshops). 12th Annual
IEEE International Conference on IEEE. 10.1109/SECONW.2015.7328145
CISCO. (2015). Fog computing & the internet of things: extend the cloud to where
the things are. Retrieved from https://www.cisco.com/c/dam/en_us/solutions/trends/
iot/docs /computing-overview.pdf
Farahani, R. Z., & Hekmatfar, M. (2009). Facility Location, Concepts, Models,
Algorithms and Case Studies. Contributions to management science. Heidelberg,
Germany: Physica-Verlag.
Giang, N. K., Blackstock, M., Lea, R., & Leung, V. C. M. (2015). Developing iot
applications in the fog: A distributed dataflow approach. 5th International Conference
on the Internet of Things (IOT). 10.1109/IOT.2015.7356560
Gu, L., Zeng, D., Guo, S., Barnawi, A., Xiang, Y. (2015). Cost-efficient resource
management in fog computing supported medical cps. IEEE Transactions on
Emerging Topics in Computing.
Haruna, Manis, Oladipo, & Ariwa. (2017). User Mobility and Resource Scheduling
and Management in Fog Computing to Support IoT Devices. IEEE.
Hong, K., Lillethun, D., Ramachandran, U., Ottenwälder, B., & Koldehofe, B. (2013).
Mobile fog: A programming model for large-scale applications on the internet of
things. In Proceedings of the second ACM SIGCOMM workshop on Mobile cloud
computing. ACM. 10.1145/2491266.2491270
51
Hou, X., Li, Y., Chen, M., Wu, D., Jin, D., & Chen, S. (2016). Vehicular fog
computing: A viewpoint of vehicles as the infrastructures. IEEE Transactions on
Vehicular Technology, 65(6), 3860–3873. doi:10.1109/TVT.2016.2532863
Intharawijitr, K., Iida, K., & Koga, H. (2016). Analysis of fog model considering
computing and communication latency in 5G cellular networks. IEEE. doi:10.1109/
PERCOMW.2016.7457059
Maher. (2015). IoT, from Cloud to Fog Computing. Retrieved from https://blogs.
cisco.com/ perspectives/iot-from-cloud-to-fog-computing
Mahmud, R., Kotagiri, R., & Buyya, R. (2018). Fog Computing: A Taxonomy,
Survey and Future Directions. In B. Di Martino, K. C. Li, L. Yang, & A. Esposito
(Eds.), Internet of Everything. Internet of Things (Technology, Communications and
Computing). Singapore: Springer. doi:10.1007/978-981-10-5861-5_5
Makwana. (2016). Introduction to Fog Computing. AI-eHive.
Oueis, J., Strinati, E. C., Sardellitti, S., & Barbarossa, S. (2015). Small cell clustering
for efficient distributed fog computing: A multi-user case. In Vehicular Technology
Conference (VTC Fall). IEEE. doi:10.1109/VTCFall.2015.7391144
Sekma, N. C., Dridi, N., & Elleuch, A. (2015). Analyses toward a prediction system
for a large-scale volunteer computing system. 2015 World Congress on Information
Technology and Computer Applications (WCITCA). 10.1109/WCITCA.2015.7367017
Zeng, D., Gu, L., Guo, S., Cheng, Z., & Yu, S. (2016). Joint optimization of task
scheduling and image placement in fog computing supported software-defined
embedded system. IEEE Transactions on Computers, 65(12), 3702–3712.
doi:10.1109/TC.2016.2536019
doi:10.1109/TC.2016.2536019
Zhang, W., Lin, B., Yin, Q., & Zhao, T. (2016). Infrastructure deployment and
optimization of fog network based on microdc and lrpon integration. Peer-to-Peer
Networking and Applications Springer.
52
53
Chapter 3
Fog Computing and
Networking Architectures
Minal Moharir
R. V. College of Engineering, India
Bharat Rahuldhev Patil

R. V. College of Engineering, India
ABSTRACT
The demerits of cloud computing lie in the velocity, bandwidth, and privacy of data.
This chapter focuses on why fog computing presents an effective solution to cloud
computing. It first explains the primary motivation behind the use of fog computing.
Fog computing, in essence, extends the services of the cloud towards the edge of the
network (i.e., towards the devices nearer to the customer or the end user). Doing
so offers several advantages. Some of the discussed advantages are scalability,
low latency, reducing network traffic, and increasing efficiency. The chapter then
explains the architecture to implement a fog network, followed by its applications.
Some commercial fog products are also discussed, and a use case for an airport
security system is presented.
INTRODUCTION
With continuous innovations in the fields of Internet of Things(IoT), Artificial

Intelligence and Virtual Reality, there is an increase in the abundance of data. These
technologies are changing the way people live, commute and work. With future 5G
applications, data from homes, communities, hospitals, factories and other sources
DOI: 10.4018/978-1-5225-6070-8.ch003
Fog Computing and Networking Architectures
is expected to grow from 89 exabytes per year in 2016 to 194 exabytes per year by
2020. Current Cloud-only architectures cannot keep up with this increase in data.
Fog computing poses a promising solution to this problem. In order to enable
data-dense use cases, Fog architectures selectively move compute, communication,
storage and decision-making choices nearer to the network edge, where the source
of the data lies. In this way, Fog computing establishes the missing link in the cloud-
to-thing continuum. Is it frequently mistaken with edge computing, but it differs
from edge computing in that fact that Fog works in tandem with Cloud, whereas
edge computing excludes the Cloud. Also, Fog has a hierarchical structure whereas
edge computing is defined by a limited number of layers.
Fog computing can be defined as a horizontal, system-level architecture that
distributes computing, storage, control and networking functions closer to the users
along a cloud-to-thing continuum. It is an extension of the Cloud-based computing
model and hence, all the benefits of Cloud are preserved in these extensions including
containerization, virtualization, orchestration, manageability, and efficiency.
A Fog network can be defined as an architecture that uses one or a collaborative
multitude of end-user clients or near-user edge devices to carry out a substantial
amount of storage, communication and management. Fog architectures can be
centralized, distributed or a mixture of both. They are established using software,
hardware or both. Fog architectures differ from Cloud architectures primarily in
the following aspects:
• Fog employs a considerable amount of storage near or at the end-user whereas

Cloud uses distributed large-scale data centers.
• Fog carries out significant quantity of communication near or at the end-user
whereas Cloud communicates through a backbone network.
• Fog establishes a substantial amount of management near or at the end-user
whereas Cloud is controlled mainly by gateways.
Cloud and Fog are not two independent choices rather they are interdependent,
for example, coordination among devices in the Fog may partly depend upon the
Cloud. In this manner, they also form a mutually beneficial continuum. Fog provides
a strategic option to certain functions while others do well in the Cloud. In fact, the
interface between Fog and Cloud is a major aspect of Fog research.
MOTIVATION
Fog computing has emerged due to two main factors - abundance of data and privacy.
To explain the factor of ever growing data, consider Endomondo, a fitness and health
54
tracking application with over thirty million users. Endomondo has released certain
data statistics of their application. The statistics show that an average workout of a
user generates 170 GPS tuples, and in a month’s time this number can easily reach
to around 6.3 million tuples. This in turn calculates to around 25,000 tuples per
second. Within a single application, there is a considerable amount of data with
high velocity. Consider then the real-time data sources in a typical city - data from
health care chips, sensors in connected cars, sensors in pollution monitoring systems
and many more. It is estimated that an average person creates around 650MB of
data every day, and this amount is to double by the year 2020. Also, autonomous
driving cars will generate terabytes of data every day. Cloud architectures do not
have the capability to deal with such large amounts of data and velocity in real
time. The second factor that cannot be dealt with by Cloud servers is the privacy
of data. Many users are not comfortable releasing their data onto the Cloud even
if it provides for detailed analysis and reports. Therefore the need arose to design
computational and storage devices nearer to the source of the data and at the same
time require connectivity to the Internet. Such devices are called edge devices and
can collect and work on local data streams. On this note, Fog architectures have the
capability to work on large amounts of data with confidentiality.
CHARACTERISTICS
As discussed, Fog computing extends the services of the Cloud to the edge devices
as shown in Figure 1. It provides for distribution of computing power, networking
and storage devices between the Cloud computing resources and the edge devices.
Therefore Fog computing involves components of an application running concurrently
both in the Cloud as well as between the sensors and the Cloud as shown in Figure 1.
Following are several advantages of Fog computing:
1. Reduction of Network Traffic: Day by day there is an increase in the

number of smart devices being used by the people and in the industry. Cisco
estimates that the number of devices worldwide could go up to 50 million in
2020. The increased number of sensors used in these devices demand that the
computational capabilities be closer to the edge devices near the user rather
than sending the data over networks to the Cloud. These sensors, depending
on their configuration could send data every few seconds to every few hours.
Therefore, it is necessary to filter out the useful information before sending
a bulk of data to the Cloud. Hence, a Fog network acts as an analyzer of data
and filters the data generated by these sensors to generate local views. This in
turn decreases the volume of network traffic to be sent to the Cloud.
55
Figure 1. Cloud to the Edge of the Network
2. Suitability for IoT Applications: Certain applications do not necessarily need

to connect to the Cloud every now and then in order to function efficiently.
Instead, they need data from their immediate surroundings. Take for example,
the Endomondo application which helps users track their health care data and
provide statistics based on the data. If an Endomondo user wishes to find
company for a particular activity, the applications can locate users who have
similar interests in sports in his local area and suggest them. For this need,
it makes logical sense to use local data in a Fog infrastructure rather than a
Cloud. Another logical example would be that of a smart connected car. Self-
driving cars such as the Tesla need to capture events that in the range of a few
hundred metres ahead of it with extreme precision. Hence, here too by bringing
processing power closer to the edge, Fog computing presents a compelling
case.
3. Low Latency: Real-time data processing is needed by various mission critical
systems. Some examples include robots, drones and braking systems in vehicles.
In robots and drones, data is collected from their sensors and processed by a
56
control system. If these control systems were on the Cloud, this could lead
to communication failures and slower reaction times. Hence by bringing
computing power closer to the control system, Fog computing provides low
latency reaction times to real-time applications.
4. Scalability: The Cloud is said to have virtually infinite resources, but the caveat
here is that the Cloud might get congested if data is continuously being sent from
a number of sources. Hence, this results in loss of data and produces a burden
on the processing power of the Cloud. However, Fog computing processes the
data near the source itself, thereby reducing the burden of processing on the
Cloud. In this manner, Fog produces a solution to the increasing number of
sources in a network and addresses the scalability issue.
5. Efficiency: An important characteristic of Fog networks is pooling of resources.
For example, hundreds of gigabytes that sit idle among various household devices
such as tablets, laptops and so on can be pooled. Similar pooling of resources
can be done in a company or even a public area. Therefore, efficiency in Fog
networks is increased in this manner. Resources that can be pooled extend to
idle processing power, wireless connectivity and shared sensors.
ARCHITECTURE
A reference architecture for Fog computing is shown in Figure 2. The first layer
(bottom layer) consists of the sensors, edge devices, gateways and user applications.
This second layer provides network capabilities for communication among the
components of the first layer as well as with the Cloud. The third layer consists of
the Cloud services and resources. This layer provides processing power for compute
intensive tasks such as IoT. It also has resources that support resource management.
The next layer contains a software-defined resource management system that ensures
quality of service in Fog applications and manages the infrastructure of the Fog
network. The last layer consists of various applications and software that leverage the
Fog network to provide service to the end users. The various elements described as
present in Figure 2 are for reference only, all these elements need not be employed
to build a Fog application.
The software-defined resource management layer in Figure 3 provides a number
of services to ensure low latency, cost optimization and increase performance of
applications. Hence, it acts as a middleware to optimize Cloud and Fog services.
These services include:
1. Flow and Task Placement: This service constantly communicates with the
monitoring service from which it receives task and flow information and keeps
57
Figure 2. Reference Architecture for Fog Computing
track of free Fog and network resources. It provides the gathered information
to the Resource Provisioning service which carries out the dynamic allocation
of resources.
2. Knowledge Base: Statistics are needed on past application transactions and
demands in order to predict future needs beforehand. The knowledge base
stores past data so that it can be used by other services in their decision-making
processes.
3. Performance Prediction: This service uses the information gathered in the
Knowledge Base to predict the performance of current Cloud resources. These
predictions are used by the Resource Provisioning service to estimate the
volume of extra resources that are needed when the performance of the Fog
network is not as expected.
58
◦ There have been many recent advances in Fog assisted healthcare

applications. One such application is FAST, an analytics system to
monitor fall for stroke patients. The creators of the system have been
successful in developing a fall detection algorithm based on acceleration
measurements and time series analysis methods. Moreover, the fall
detection task is split between edge devices and the Cloud, thereby
keeping response time and energy efficiency to the current standards.
◦◦ There has also been development in the field of smart-healthcare
infrastructure. It involves adding a Fog layer to improve the existing
architecture. The proposed architecture consists of a role model, layered
Cloud architecture and a Fog computing layer. This Fog layer provides
low latency, mobility support and security measures. This model
can be used as a template for future sensor-based smart healthcare
infrastructure.
59
Figure 3. Applications of Fog computing
◦
◦ As augmented reality applications are highly latency intolerant, Fog
computing has the potential to contribute in this field. Zoe et al have
built an Augmented Brain Computer Game based on Fog computing.
The game employs the use of Electroencephalography (EEG) sensors
to obtain information on the state of the brain. This process involves
heavy signal processing tasks to be carried out in real-time. Therefore
a combination of Fog and Cloud servers enables continuous real-time
brain state classification based on EEG readings collected by the sensors.
In this way, Fog computing helps in compute intensive augmented
reality applications.
60
◦
◦ Fog computing can enhance manufacturing processes in small and large
scale industries. With Fog computing it is possible to provide real-time
monitoring and diagnostics and predictive maintenance to increase the
uptime and lifespan of equipment. Using Fog computing, factories can
predict unexpected failures and malfunctions of machines in real-time. It
can also help improve product quality by analyzing sensor readings and
identifying the root cause for defective parts. Semiconductor companies,
for example, are continuously trying to increase their yield rate and
maximize their profits. These companies can use Fog computing to
identify false positives and false negatives in real-time. Moreover, data
generated in the fabrication process are subject to confidentiality and
cannot be moved to the Cloud. Hence, Fog computing solves this issue
by bringing analytics close to the source of the data rather than moving
the data to the Cloud.
◦
◦ Oil and gas extraction operations are extremely high-stake operations
involving real-time on site intelligence processing. This is needed in
order to ensure monitoring and protection to reduce equipment failure
and environmental damage. This scenario presents a perfect stage to
employ Fog computing to ensure maximum production and safety.
An application of Fog computing in the oil and gas industry is the
use of proactive pipeline optimization systems. There are more than
2.5 million miles of oil and gas pipelines in the United States. Many
sensors have been installed in these pipelines to monitor flow, pressure
and compressor conditions. These sensors generate enormous volumes
of data. An average 100-mile long pipeline itself produces 50 gigabytes
of data, then the total data across the country would be unmanageable.
Therefore, Fog computing relieves this data problem by analyzing on
the data by the sensors at the source itself and reporting any issue and
acting automatically. For example, a valve can be closed automatically
in case of a leak and a report be sent via an alert to a mobile phone.
With Fog computing, pipeline operators can lower risk and costs which
is not possible with a Cloud only solution. Fog computing also helps in
anticipating machine failures. The Electric Submersible Pump (ESP)
extracts oil from the bottom of the well to pump it to the surface. Failure
of this machine could result in loss of huge amounts of money and
61
◦ ◦The primary motivation to employ Fog computing in mining industry is

that mining faces extreme environmental conditions in remote locations
that do not have access to the Internet. Fog computing can be employed
in this case since it does not need frequent access to the Cloud.
◦◦ The mining industry faces several challenges due to shortage of skilled
labor and falling commodity prices. Hence, mining companies are
moving towards the use of automated trucks, tunneling and boring
machine to increase productivity and worker safety. Sensors are installed
on these vehicles to enable real-time asset monitoring and to gather
data for predictive maintenance. In addition, health monitoring devices
and sensors can monitor the physical health of mine workers and send
alerts in case of any accidents. Therefore, a Fog computing system can
provide resources for real-time responses with low latency.
◦◦ In addition, mining machines also generate terabytes of data everyday
which the mining companies use to optimize processes, reduce energy
consumption and increase production. Since mining occurs in remote
locations, data cannot be uploaded to a Cloud network as the connectivity
is too low for such a large volume of data. Hence, Fog computing solves
this issue as it analyzes the data and provides real-time insight to the
mining companies and therefore reducing the bandwidth required as it
sends only a subset of the data to the Cloud which is needed for future
analysis.
FOG PRODUCTS
CISCO IOx
Cisco was in fact the first to introduce the term Fog computing. Cisco IOx is a
product that aims to integrate essential IoT elements and Fog computing. It provides
62
for distributed computing within the Fog, secure communications and rapid system
integration and application management. Cisco IOx has four components - Cisco
IOx, Fog Director, SDK and development tools and Fog applications. Cisco IOx
brings together Cisco IOS software, the industry-leading networking operating
system, and Linux, the leading open-source platform. The Cisco Fog Director
establishes a managing system to monitor and troubleshoot Fog applications running
in the IOx environment. Cisco’s SDK allows the users to develop applications in a
programming language of their choice, it supports Java and Python as well. Finally,
the Fog applications ready for execution are deployed onto the IOx infrastructure.
Some features of Cisco IOx include:
• Flexible application development and deployment approaches.

• Ability to process high volumes of data and deliver closed loop system
control in real time
• Built-in security for application deployment and management.
• Integration with PaaS (Platform as a Service) capabilities associated with
application programming environments.
LocalGrid Fog Computing Platform
LocalGrid Fog Computing Platform is a framework that enables users to build Fog
applications without worrying much about the complexity of the Fog network.
Therefore it allows for quick development of products with reduced costs. Some of
its features include:
• Rapid Onboarding of Heterogeneous Edge Devices: This allows sensors

and devices of different protocols to interact with each other with minimum
effort by the application developer.
• Peer to Peer Communication: Direct communication allows for real time
control and coordination without having to rely on a centralized server.
• Supports Development: In many programming languages on most platforms
(C, C++, C#, Java, Python and Labview).
• It Enables the Use of the Field Message Bus: Thereby eliminating any
dependency on a central command and thus there is no single point of failure.
• Reduced Bandwidth: Allows for faster real-time decisions
• Security: Data is stored locally on each device and not sent to the Cloud, and
accessed by devices when needed only with successful authentication.
63
Vortex Fog by Adlink Technologies IST
Vortex Fog is a platform that allows for secure forwarding of data between Fog
subsystems in a Vortex Cloud. These Fog subsystems contain edge node applications
communicating with each other on a Local Area Network (LAN) or other subsystems
over a Wide Area Network (WAN). Similar to the property of a Fog network, Vortex
Fog allows only selected data to be forwarded to the WAN. It supports secure data
sharing between the subsystems via encrypted communications and authentication
mechanisms. The Intelligent Vortex Data Sharing Platform supports automatic
discovery independent of underlying topology. In order to provide low latency,
quality of service Vortex provides implementations that are optimized for various
device platforms. Some features of Vortex Fog include:
• Plug and play integration and data sharing between different parts of a Vortex
system deployed on either a LAN or a WAN.
• Boundary Security with TLS based encryption, certificate–based
authentication and access control rules between subsystems.
• Uses UDP multicast to enable efficient, low latency device-to-device data
sharing.
• Adaptation to underlying network - for example it will dynamically switch to
TCP if UDP is not available.
USE CASE FOR FOG COMPUTING
Fog computing presents a solution for many compute intensive applications, one
such use case is security at airports. Security at airports still involves some manual
tasks which have the potential to be automated. Cameras at the airport play the most
significant role in providing security. The videos from these cameras are processed
to determine a possible threat. Each camera can generate up to 1TB of video data in
a day, which make it a suitable candidate for Fog computing. Several applications of
these cameras connected to computational units (or Fog nodes) are discussed next.
First and foremost, cameras are used to capture license plate numbers as they enter
and leave the airport, Fog nodes employed here can be used for optical character
recognition. Similarly, Fog nodes can be deployed for other visual analytics purposes
such as screening, observing parking lots, monitoring arrival and departure areas
and terminals. A convolutional neural network can be employed in the Fog node in
these cases as they achieve excellent image classification.
64
From the time a passenger enters the airport, up until he reaches his destination,
Fog computing can be employed in various scenarios to ensure monitoring a person
for security purposes. When the passenger enters the airport, the car he arrives in
is monitored by Fog nodes scanning the license number plates. The Fog node also
helps the airport authorities to scan the RFIDs if the vehicle is equipped with one
and also check if a suspicious car enters the airport.
Next, when the passenger makes his way to the screening area and checks in his
luggage, by this time the Fog nodes would have acquired the passenger’s information.
This includes the information including the car he arrived in, his destination and
flight information. The information about a particular passenger is aggregated by
using a Fog network among the Fog nodes. Then the passenger is monitored up until
he leaves the destination airport. With machine learning techniques, this acquired
information can be used to detect suspicious behavior. The several components that
are essential to the functioning of this procedure are as follows:
• Vehicle Capture: As discussed, this converts the images or video to gain

information on the car, such as the license plate, car model and other vital
data.
• Facial Capture: This component is used to track the passenger throughout
the entire process. It uses the appropriate API to tag a person’s face and track
his movements.
• Data Fusion: This component ensures that that all the gathered data is
matched accurately. For example, the passenger to his car number, the
passenger to his ticket number and so on.
• Behavior Monitor: This component uses the video feed to suspect abnormal
behavior. The Fog node associated with this component can use data from
previous cases and machine learning techniques to learn and detect anomalies.
• Baggage Capture: This component contains the API needed to detect
luggage, identify its owner and continuously track it.
• Alerter: This component sends out an alert to the respective authorities in
case of abnormal behavior so that immediate action can be taken.
Using these components, results have to be computed in real time within

milliseconds which is possible only with Fog nodes. A similar Cloud model
would not be able to achieve the same results with this large amount of data and
requirements of fast processing speed with low latency. In this way, a Fog network
forms an efficient airport security system.
65
REFERENCES
Bonomi, F. (2014). Fog Computing: A Platform for Internet of Things and Analytics.
In N. Bessis & C. Dobre (Eds.), Big Data and Internet of Things: A Roadmap for
Smart Environments (pp. 169–186). Springer. doi:10.1007/978-3-319-05029-4_7
for Internet of Things and Analytics. In N. Bessis & C. Dobre (Eds.), Big Data and
Internet of Things: A Roadmap for Smart Environments. Studies in Computational
Intelligence (Vol. 546, pp. 169–186). Springer. doi:10.1007/978-3-319-05029-4_7
Cao, Y. (2015). FAST: A Fog Computing Assisted Distributed Analytics System
to Monitor Fall for Stroke Mitigation. Proc. 10th IEEE Int’l Conf. Networking,
Architecture and Storage (NAS 15), 2–11. 10.1109/NAS.2015.7255196
Chiang, M. (2015, December). White Paper on Fog Networking: An Overview on
Research Opportunities. Academic Press.
Cisco IOx and Fog Applications. (2017). Retrieved from https://www.cisco.com/c/
en_in/ solutions/internet-of-things/iot-fog-applications.html
Cristea, V., Dobre, C., & Pop, F. (2013). Context-aware Environments for the Internet
of Things. Internet of Things and Inter-cooperative Computational Technologies for
Collective Intelligence Studies in Computational Intelligence, 460, 25–49.
Dastjerdi, Gupta, Calheiros, Ghosh, & Buyya. (2016). Fog Computing: Principles,
Architectures, and Applications. Internet of Things Principles and Paradigms, 61–75.
Gia, T. N., Jiang, M., Rahmani, A.-M., Westerlund, T., Liljeberg, P., & Tenhunen,
H. (2015). Fog Computing in Healthcare Internet of Things: A Case Study on ECG
Feature Extraction. In Computer and Information Technology; Ubiquitous Computing
and Communications; Dependable, Autonomic and Secure Computing; Pervasive
Intelligence and Computing. CIT/IUCC/DASC/PICOM.
Li, J. (2015) EHOPES: Data-Centered Fog Platform for Smart Living. Int’l
Telecommunication Networks and Applications Conf. (ITNAC 15), 308–313. 10.1109/
ATNAC.2015.7366831
OpenFog Reference Architecture for Fog Computing, Produced by the OpenFog
Consortium Architecture Working Group. (2017, February). Retrieved from http://
www.OpenFogConsortium.org
66
Stantchev, V. (2015). Smart Items, Fog and Cloud Computing as Enablers of

Servitization in Healthcare. J. Sensors & Transducers, 185(2), 121–128.
Zao, J. (2014) Augmented Brain Computer Interaction Based on Fog Computing
and Linked Data. Proc. 10th IEEE Int’l Conf. Intelligent Environments (IE 14),
374–377. 10.1109/IE.2014.54
67
68
Chapter 4
Fog Computing Application
Deployment and Management
Ranjitha G.
Pankaj Lathar
Chaudhary Brahm Prakash Government Engineering College, India
G. M. Siddesh
ABSTRACT
Fog computing enhances cloud computing to be closer to the processes that act on
IOT devices. Fogging was introduced to overcome the cloud computing paradigm
which was not able to address some services, applications, and other limitations of
cloud computing such as security aspects, bandwidth, and latency. Fog computing
provides the direct correlation with the internet of things. IBM and CISCO are
linking their concepts of internet of things with the help of fog computing. Application
services are hosted on the network edge. It improves the efficiency and reduces the
amount of data that is transferred to the cloud for analysis, storage, and processing.
Developers write the fog application and deploy it to the access points. Several
applications like smart cities, healthcare domain, pre-processing, and caching
applications have to be deployed and managed properly.
DOI: 10.4018/978-1-5225-6070-8.ch004
Fog Computing Application Deployment and Management
INTRODUCTION
The fog computing enhances the cloud computing to be closer to the process that
act on the IOT devices. It is also known as fogging. Fog computing is invented by
CISCO. Fog computing provides services and storage to the client. Fog computing
also provides the services between the cloud data centers and enterprise end devices.
It is wireless information transferring across the distributed devices in the Internet
of things (IOT) network. Fogging was introduced to overcome the cloud computing
paradigm which was not able to address some services, applications and other
limitations of cloud computing such as security aspects, bandwidth and latency.
The protection mechanism which was used in cloud for encryption was failed in
providing the security for the data from the attackers. Fog computing address these
type of issues.
Fog computing applications are considered as the edge network computing.
Edge network devices provide an entry point to the provider’s network. In fog
computing communication takes place as peer to peer. In Internet of thing the
sensor will generate extremely large amount of data. With the network connection
the fog devices which are known as fog nodes can be deployed anywhere. It can be
deployed on power pool, floor or the factory, vehicle, side of a railway track, etc.
The fog nodes are routers, switches, controllers in industries, surveillance camera’s
etc. This type of communication is used to provide storage and efficient sharing
and to take the decisions. FC is a novel idea which is expected to solve problems
and provides solution for the latency, sensitive information computing problems.
The local computing resources are used by the fog computing instead of real cloud
for processing data. The transmission latency is reduced because of the proximity
between the processors and data sources.
Figure 1. Fog computing
69
FEATURES OF FOG COMPUTING APPLICATIONS
1. Local Data Processing: It is implemented in a way that provides intelligence

at the gateway which helps to analyze the streaming of data locally (Bonomi,
Milito, Zhu, & Addepalli, 2012). It helps in handling large amount of data in
less time span. It responses to the various conditions.
2. Adaptive Applications: The fog layer data transmission need to be adaptive.
It includes data request and also transmission rate from fog to cloud layer.
3. Local Storage: Gateways process the incoming data locally. It contains an
operating system that handles the repository. In non-volatile memory it stores
the data. It compresses and encrypts the data and stores it locally.
4. Security: Security is the most major requirements of application. If the system
is not secured it leads to serious vulnerabilities. Iptables are used for configuring
the IP packets and it also contains set of rules.
5. Interoperability: It plays an important role for the success of an application.
IOT based system contains the mixture of protocols, technologies and platforms.
Integrating these application is major challenge. Fog computing provides easy
integration of these network and technologies.
6. Sensor Node Energy Efficiency: Sensor nodes has to be processed properly.
Fog computing provides a signal processing tasks from the nodes to gateways.
It uses low power transmission protocol to save the energy during transmission.
Figure 2. Features of fog computing applications
70
APPLICATIONS OF FOG COMPUTING
Fog computing provides the direct correlation with the internet of things. IBM
and CISCO are linking their concepts of internet of things with the help of fog
computing. The important area where the fog computing plays an important role
are shown in Figure 3.
Application services are hosted on the network edge. It improves the efficiency
and reduces the amount of the data which is transferred to the cloud for analysis,
storage and processing of the data. IOT is system of interconnected computing
devices, machine to machine interaction takes place. The amount of IOT devices
estimation is billions with the growing number of sensors, storage and computing
is difficult. In traditional data is just pushed into the cloud. Filtering and analysis
has to be computed properly.
Figure 3. Applications of Fog computing
71
Developers write the fog application and deploy it to the access points (Bonomi,
Milito, Zhu, & Addepalli, 2012). Then fog nodes will receive the data from IOT.
Then the data is distributed on basis of how soon the action is required. Fog nodes
will receive the data from IOT devices in real time and it will process the data that
the IOT device can make decision.
DIFFERENT TYPES OF DEPLOYMENT MODELS:
Indie fog infrastructure makes use of the customer network devices such as router
to provide fog computing environment (Chang & Satish Narayana Srirama, n.d.).
Mainly it reduces the IOT service providers to deploy devices throughout a fogging
system. It is more flexible and cost effective.
The two types of Indie fog deployment models are:
1. Integrated form.
2. Collaborative form.
1. Integrated Form: The gateway is responsible for providing functionality of

integrating it into indie server which in turn connected to user devices and
different data sources. The router itself provides virtual server which intern
acts as fog node.
2. Collaborative Form: The workstation are responsible for providing the
functionality in the subnet as the source devices through the internet gateway.
It connects through the internet gateway. Cloud service providers provides
equipment which is designed for the particular application.
Indie fog provides facilities for deploying four types of models:
1. Indie Fog Cluster: The group of fog nodes in a close proximity within a
subnet or which resides in the same building are responsible for establishing
the software defined cluster (Chang & Satish Narayana Srirama, n.d.). It helps
in performing the preprocessing of the information which is collected by the
sensors device and sources. It will reduce the time and bandwidth which are
required to send data to cloud.
2. Infrastructural Indie Fog: The deployment of fog devices in cameras provide
a ubiquitous application infrastructure. It requires temporary storage of data
which has to be processed in short time. Example: The scientist invest for indie
fog server with the sensor functions which are pre- installed to collect the data
which are required for research.
72
Figure 4. Integrated form
Figure 5. Collaborative form
3. Vehicular Indie Fog: Vehicles can also host their nodes. The system can
dynamically deploy vehicular software defined for sensing network which
collects the information of traffic, transportation and metropolitan information
for public. It enables the sequential data streams and data on the changing
environment.
73
4. Smart Phone Indie Fog: The smart phone users move regularly devices will
not be able to form stable cluster environment. The deployment of fog will
help in these areas. The server nodes collect the information of sensor data.
The phone will be able to process.
MANAGING THE FOG APPLICATION
1. Dashboard: Dashboard is the first page which is presented to the user after
the login. It gives an overview to the end users what all is happening.
2. System Overview: The system overview is the box that contains the current
end users, IP, server address and how users have been logged. It also provides
the information of how many users logged into the system and how long the
system is running.
3. System Activity: It contains the unicast queue. It shows the number of unicast
deployed and which are in progress. Based on the storage groups the queue
would differ. It also provides the information of queued task, running and
updates information. It provides bandwidth graph. The editing can also be
performed using fog configuration.
4. Service Management: It allows to control the client service functions.
5. User Management: This information is provided by menu bar. It contains the
information of administrators of individual fog resources.
6. Monitoring: It provides functionality of operational locally. It allows to save
the data and processed results in local and synchronizes it with cloud.
7. Host Management: It gives the information of host which are imaged and to
extract images.
8. Disk Information: Disk information displays the storage services. It also
provides the information of the available storage on the storage server. The
user is provided by drop-down box to change to storage nodes for monitoring
the disk information.
9. Image Management: It allows to manage the image files stored on the fog
servers.
10. Snap-in Management: It provides the imaging of the post tasks.
11. Printer Management: Allows the user to manage the printers to create printer
objects which can be assigned later to the hosts or group.
12. Fog Used Task: It provides the task id and type. It provides an information
of deploy, captured, debug and test disk depending on the current state. Type
ID contains the job.
13. Report Management: Are the information from the fog database which can
be of pdf, csv or html.
74
14. Storage Management: It allows the customer to add or remove the fogging
system.
15. Host: It contains the description information, Ip address, operating system,
kernel, arguments of kernel and primary disk information.
CASE STUDIES
Smart Cities
The challenges which are faced by large cities are traffic, safety of public, energy
usage is high and municipal services. These challenges can be overcome by single
IOT network by installing fog nodes. The connectivity and broadband bandwidth is
the major problem in establishing the smart cities. Most of the modern cities have
one or more cellular network which provides adequate coverage this network will
often have bandwidth limitation and capacity issues. The smart cities also face the
problem of security and safety where it requires real time performance. The network
may carry traffic and sensitive data.
Figure 6. Smart city
75
The deploying of fog computing will allow the fog nodes to provide storage and
local processing which in turn optimizes the network usage. Fogging will address
the issues such as security, encryption of data and distributed analytics.
Health Care Domain
The resource constrained devices are connected to the network. The majority of these
devices are not capable of storing data which are generated by large scale (Zhu, &
Jiang, 2013). This data was transferred directly to cloud for processing. It provides
local data processing which is very important in case of medical field which in turn
enables the system to react faster for medical emergencies. It provides local data
processing which is very important in cases of medical which in turn enables the
system to reach faster for medical emergencies. Fog computing provides greater
flexibility in meeting the requirement of healthcare. It filters the data and preserve
the privacy in order to reduce load at the network.
Deploying Healthcare Application
The healthcare application includes users, devices and connectivity.
Figure 7. Healthcare application
76
1. Mobiles: It acts as a hub between the cloud and the sensor devices. The base
unit are mobile phone which collect data from various devices and process
the data and returns it back to the back-end servers. The major purpose of fog
computing is to provide battery life for the wearable sensor device. Example:
COPD (chronic obstructive pulmonary disease).
2. Hospital: The devices are owned and maintained by the hospital itself. These
systems are more complex and requires more qualified professionals. The smart
shirts will allow monitoring data and the patient location. The data acquisition
and processing board will process, collect and merged. This information are
obtained from sensors. The wireless transmission board collects the information
from the data acquisition and processing board and sends it to the management
subsystem in single packed which is located at LAN level. The subsystem uses
this information to monitor the patient medical parameters and it also verifies
for alarm is activated or not.
3. Premises of Non-Hospital: It includes less staff and infrastructure facility
like clinic and nursing homes. The core devices are maintained and owned by
the clinic. It reduces the traffic between the detection system which is located
at the cloud and LAN (Zhu, & Jiang, 2013).
4. Transport: In emergency situation the data is collected from the patients
wearing the medial device.
Managing Application
1. Data Collection: The data as to be collected properly which is examined by

the doctor for reference.
2. Data Analysis: The critical analysis has to be taken care properly. The
monitoring of cardiac through ECG should alert in case of critical situations.
The control of the critical condition as to be detected when the alarm is obtained.
Example: Proper oxygen has to be provided to the patients. Fog computing will
assist the system for detecting and predicting the emergency situations. Fog
computing offers fault detection which allows faster for emergency situation.
Fog computing implements real time responses.
3. Context Management: The context management helps in taking proper
decisions and plans.
4. Filtering of Data: Data received from various sensors for pre-processing at
the edge and data analysis is performed. The Bio-signals are collected from
patient body which are the primary source of information. Information contains
77
amplitude, frequencies and complex shapes. Noise which is accumulated

distorts the quality of signals. Fog computing will address these types of issues
at interface sensors. It receives data from sensors by various communication
protocols.
5. Data Compression: It is used to reduce latency and energy which is consumed
during the transaction. For ECG monitoring of real time application and it
provides signal with high precision.
6. Fog Computing Data Fusion: It enables to decrease large volume of data and
reduces the energy consumption. The data fusion is categorized into three classes
they are: complementary, cooperative and competitive. The complementary
contains global knowledge which in turn contains the information of difference in
body temperature of patients and environment provided by sensors. Competitive
improves accuracy and consistency of the results in case of failure of sensors.
Cooperative provides the vital signals.
Figure 8. Smart grid
78
Smart Grid
The major problem in traditional electric generation was fossil fuel. The fuel
consumption was threatening the global warming. Smart grid integrates green power
resources controlling usage of power and it also helps in balancing the energy load.
Smart grid employs meters by which two-way electricity flows information to manage
and monitoring the electricity consumption. Smart meters produce huge amount
of data which is difficult to process, store and analyze using cloud computing. Fog
computing is responsible for processing, storing and collecting information before
transmitting it to cloud. It acts as a communicating bridge between smart grid and
cloud.
Fog computing for smart grid can be deployed to address the issues of moving the
data to the cloud. It enables the user to access the data in secure and faster manner.
Advanced metering infrastructure provides secure, reliable and cost-effective services.
The data which are collected by meter includes some private data about users. As
the size of smart applications increases the data collected by meter also increases.
Deploying Smart Grid Applications
1. The smart meters are used to measure the electricity usage of every applications.
To provide data privacy and authentication smart meter encrypts the data using
smart keys and shares among different fog server.
2. Upon receiving the encrypted data the fog server will decrypts and store it for
a period as required by users.
3. The fog server calculates the total usage of the smart meter and sends it to the
cloud. This work would help in reducing the data storage in the cloud centers.
4. Fog computing is responsible for storing private data of the customer locally
and sends public information to the cloud. These provides privacy of the
information.
5. Using fog computing in smart grid application reduces the access time and
searching as data are stored locally.
Managing Smart Grid Application
Smart grid offers users detailed usage information of electricity. The customer can
analyze and monitor the electricity consumption of daily, week and monthly. The
detailed information may contain some private data of the users. Fog computing will
treat it as private data. Customer can charge the device using ID information. The
smart meter sends the information to the owners. It enables the active participation
of the customers.
79
The Preprocessing and Cache Application
The edge servers are used for improvement of the website performance. Through
the fog boxes users will be able to connect to the internet Fog nodes provides an
optimization and reduces the amount of time where the user needs to wait.
Deploying and Managing Preprocessed and Cache Application
Other than the generic optimizations which includes reorganization, caching of the
html component, composition reduces the size of the object, fog computing will
also perform the optimization of user behavior and the condition of the network.
Example: In network congestion it provides low resolution graphics to the
customer to reach out the acceptable time of response. The edge deices is used for
monitoring of the client machine performance and it is also depends on the rendering
time of the browser.
The major thing which has to be considered is data trimming. It is nothing but
pre-processing of the data has to be done before sending it to the cloud. Forwarding
this large amount of information is difficult as it leads to data center congestion
problem. Fog computing provides gateway based smart communication which is
used to integrate cloud with the IOT. Fog computing will handle the pre-processing
and trimming of data before sending it to cloud directly. The smart gateway for the
operation of the data which is generated by IOT devices in context aware and sensitive
data latency manner. This approach will provide greater usage of IOT applications.
Smart Agriculture
Agriculture is most important as it provides food supply chain and provides many
ways of communication concentrates in cities. Fog computing helps in collecting
various information. The heat event may have an effect on crops. Flowering time is
major event for crops. The sensors has to collect the information of crops, climate,
weather, water and season information.
Deploying Agriculture Application
1. It contains sensors network to collect various information. The field, crop,

climate conditions are collected.
2. The sensor states are used to monitor the crops in the land. Air balloons are
used to monitor the crops in the land. Fog computing plays an important role
in sensing adaptively and efficiently analyzing the information obtained.
80
Figure 9. Smart architecture using fog
3. Fog gateway devices can be used in three different types according to phenonet
project. Phenonet has a group of sensors which helps for sensing devices. The
first gateway device is phenomobiles which acts as an edge devices.
4. The blimps will help in combining the sensors and gateway devices. The different
data frames help in reducing data storage. Sensor data has to be deployed in
such a way that it senses location and time sensitive data.
5. Each sensor has to be deployed by considering the context information. The
heat and frost can damage the mechanism of the crop. The sampling rate vary
based on the season. Example: The high sampling ideal thing may vary for
summer season. The lower sampling is required for winter season.
6. The sensor should continuously sense the crop. The frost event cannot be
detected without temperature. The configuring of humidity sensor is the ideal
idea where the temperature is replaced and starts sensing immediately.
7. These type of reconfiguration will help in eliminating the ineffectual sensing
and improves the network communication.
Managing the Smart Agriculture
Monitoring: Plants are monitored using different types of sensing devices and
techniques. Context information plays a major role. It provides an efficient sensing
technique in smart agriculture related problems.
81
The main reason for collecting data gives information of plants by applying
reasoning and fusing technique.
CONCLUSION
Fog computing is most attribute solution for the problem of IOT and data processing.
The devices relies on the network edge and reduces latency of application. The review
shows the different deployment scenarios for variety of application with benefits
of fog computing. It mainly filters the data to reduce traffic and to provide privacy.
Flexibility of fog computing is therefore suitable for meeting the requirement of
various application.
REFERENCES
Akella & Xiong. (2014). Quality of service (qos)-guaranteed network resource

allocation via software defined networking (sdn). In Dependable, Autonomic and
Secure Computing (DASC), 2014 IEEE 12th International Conference on (pp.
7–13). IEEE.
Bonomi, F. (2011). Connected vehicles, the internet of things, and fog computing.
The Eighth ACM International Workshop on Vehicular Inter-Networking (VANET).
Bonomi, F., Milito, R., Zhu, J., & Addepalli, S. (2012). Fog computing and its role
in the internet of things. In Proceedings of The First Edition of The MCC Workshop
on Mobile Cloud Computing (pp. 13–16). ACM. 10.1145/2342509.2342513
Bonomi, Milito, Zhu, & Addepalli. (2012). Fog computing andits role in the internet
of things. In Proceedings of The First Edition of The Mcc Workshop on Mobile
Cloud Computing. ACM.
Chang & Srirama. (n.d.). Indie Fog: An Efficient Fog-Computing Infrastructure for
the Internet of Things. University of Melbourne and Manjrasoft Pty Ltd.
Cisco, Tech. Rep. (2014). Cisco Delivers Vision Of Fog Computing To Accelerate
Value From Billions Of Connected Devices. Cisco.
Cisco. (n.d.). New cisco internet of things (iot) system provides a foundation for the
transformation of industries. Retrieved from https://newsroom.cisco.com
82
Rohila & Singla. (2016). An effective review of fog computing using virtualization.
International Journal of Innovation Research of Computer and Communication
Engineering, 4(4).
Zhu, J. (2013). Improving web sites performance using edge servers in fog computing
architecture. In Service Oriented System Engineering (SOSE), 2013 IEEE 7th
International Symposium on. IEEE.
83
84
Chapter 5
Confidentiality and
Safekeeping Problems and
Techniques in Fog Computing
Nida Kauser Khanum
Pankaj Lathar
Chaudhary Brahm Prakash Government Engineering College, India
G. M. Siddesh
ABSTRACT
Fog computing is an extension of cloud computing, and it is one of the most important
architypes in the current world. Fog computing is like cloud computing as it provides
data storage, computation, processing, and application services to end-users. In
this chapter, the authors discuss the security and privacy issues concerned with fog
computing. The issues present in cloud are also inherited by fog computing, but the
same methods available for cloud computing are not applicable to fog computing
due to its decentralized nature. The authors also discuss a few real-time applications
like healthcare systems, intelligent food traceability, surveillance video stream
processing, collection, and pre-processing of speech data. Finally, the concept of
decoy technique and intrusion detection and prevention technique is covered.
DOI: 10.4018/978-1-5225-6070-8.ch005
Confidentiality and Safekeeping Problems and Techniques in Fog Computing
INTRODUCTION
The Internet of things (IoT) will be the “Internet of future” due to its growth rate in
massive domains like wearable technology, smart city, smart transportation, smart
grid. These smart applications require certain amount of resources like storage,
battery, computation power and bandwidth. The IoT devices are not configured with
so much resources. Therefore, they are generally assisted by robust server ends. Cloud
is considered as an auspicious solution to deliver services to end users, and for its
adaptable resources at very low cost. The server ends used by IoT are deployed in
Cloud for its fair benefits provided to the users.
Even though cloud has several advantages, it cannot solve all problems due to
its own drawbacks. Latency is one of the major issue in cloud because the data
centers are located near the core network. The real-time applications like gaming,
augmented reality and real-time streaming are subtle to latency and cloud is not
a feasible choice for the application deployment that require very quick response
and avoid round-trip latency during transmission of data from/to terminal nodes
to/from cloud servers for processing. The data is sent through multiple gateways
during transmission. In addition, there are also unresolved glitches that regularly
need agility support, geo-distribution and location-alertness in IoT applications
(Shanhe et al., 2015).
However, the state-of-the-art technology in computing standard is to impulse
computation and storage resources to the control of networks. This results in growth
of an auspicious computing architype called fog computing that provides services
to the edge devices instead of relying on cloud services. Fog computing eradicates
many issues arising in cloud computing service. It retains computation and data
local to end users by providing low latency, location-awareness and high bandwidth.
It gets the title as fog because it is a cloud present close to the ground. The devices
that provide fog services to the edge devices are named as fog nodes. Fog nodes can
not only be resource-rich devices but they can also be resource-poor devices like
end devices, smart TVs/set-top-boxes and gateways. Usually cloud computing is
collaborated with fog computing by forming a three-layered architecture comprising
of end users, fog and cloud service as shown in Figure 1.
Cloud computing and fog computing share many similar characteristics like
adaptable resources (computation, storage and networking). There resources are
building blocks for both cloud and fog computing, signifying that utmost cloud
computing technologies can be pragmatic to fog computing. Nevertheless, the new
paradigm fog has quite a few exclusive features that makes it diverse from other
present computing architectures. One of the imperative feature is maintaining close
distance to end users. To support applications that are latency-sensitive, it is vigorous
to preserve computing resources at the core of the network. The other stimulating
85
feature is that the fog nodes which are geographically distributed can infer its own
position and keep a track of all the end devices by providing mobility. It also provides
reduced delay and bandwidth by dropping data volume to great extent at primary
stage by preprocessing the data at fog nodes before sending it to cloud for further
analytics. This property yields many benefits for stream mining and edge analytics.
This has a significant role in the era of big data (Bonomi et al., 2012).
There are many features that are enhanced in fog computing when compared to
cloud. It provides mobility, large IoT device support, extensibility, reduced delay,
decentralization and many more. Along with vast benefits, fog computing welcomes
many issues in security and privacy of devices and information stored. The main
concern of this chapter is understanding the security and privacy issues in fog
computing. It also includes the architecture of fog computing and its interaction
with other counterparts (cloud and IoT) that gives an overview of differences among
the different layers. The present techniques that are employed to make fog secure is
also discussed. Real-time applications are discussed that face security and privacy
issues. At the end of the chapter an intrusion detection and prevention technique
is proposed.
ARCHITECTURE OF FOG COMPUTING
Fog computing has a very basic and simple architecture that consists of three tiers
(Sarkar, Chatterjee & Misra, forthcoming). Figure 1 shows the architecture and
components included in fog computing. This architecture is widely used and each
of the tiers are discussed below:
• Tier 1 – End Devices: This tier consists of service patrons and IoT enabled
devices. Some of the common end devices that are likely to be a part of
fog computing are sensor nodes, Smart hand-held devices like smartphones,
smartwatches and tablets and many more. Almost all the devices are fortified
with Global Positioning System (GPS) to provide location awareness to fog
nodes. These devices are generally termed as Terminal Nodes. These devices
request services from cloud or fog nodes. Therefore fog-terminal nodes
interface must be designed to enrich secure resource exploitation.
• Tier 2 – Fog: This tier consists of devices that has minimal resource
availability like storage, computation etc. This rank is also labelled as fog
computing layer. Most of the network devices like router, switches, gateways
and Access Points (APs) are part of this tier. The devices collaborating their
services like storage and computation are often called fog nodes. These fog
nodes provide storage and computation facilities to terminal nodes whenever
86
Figure 1. Architecture of Fog computing with three tiers
requested. Fog nodes collaborate with each other to provide efficient service
by sharing storage and computation tasks. Hence it is one of the important
task to design fog-fog interface and protocol that empowers fog nodes to
collaborate with each other when required.
• Tier 3 – Cloud: The devices with appropriate storage and computing
resources are part of this layer. Traditional cloud servers, cloud gateways
and Data Centers reside on topmost layer in this architecture. Cloud servers
are responsible to distribute the tasks to fog nodes and maintain them. Thus,
cloud-fog interface (Chiang & Zhang, 2016) must be able to handle all fog
nodes and schedule the tasks among fog computing.
INTERACTION BETWEEN CLOUD, FOG

AND INTERNET OF THINGS
The advancement in IoT has led to development in many use cases that produce
substantial amount of data (Diaz, Martin & Rubio, 2016). There are huge number
of devices that are geographically distributed and capable of producing vast amount
of data, dealing with big data is another imperative challenge. Fog computing was
proposed to efficiently manage and analyze time sensitive data (Bonomi et al.,
2014). As researchers know that cloud computing models are incapable of handling
certain services requested by IoT devices that require high speed consciousness and
87
response to events. Fog computing discards this disadvantage from cloud services.
Interaction between all the devices must be without any trouble (Cisco, 2017).
The Table 1 summarizes the features of fog computing, cloud computing and
Internet of things. It describes the differences in each domain based on those features.
FEATURES OF FOG COMPUTING
Fog computing is distributed in nature where it offers services to IoT application at the
edge of the network by utilizing edge resources. One of the major feature provided by
fog computing is to confront the IoT applications by exploiting the fog nodes placed
near users to deliver convenient services like storage, computation, transmission,
control and management of data locally. Cloud computing and fog computing are
similar in many ways but they vary in many aspects as listed below and shown in
Figure 2. They constitute as features of fog computing that are discussed below.
• Decentralization: Unlike cloud computing, fog computing does not

have centralized server to manage its resources and services. Fog nodes
Table 1. Features of Fog, clod and IoT and differences among them
Features Fog Computing Cloud Computing Internet of Things

Stationery and mobile
Main Users Mobile users General Internet users
devices
Node count Large Few Large
Architecture Distributed Centralized Dense and distributed
Outdoors (fields, streets,
Indoor with immense
Working Environment tracks) or Indoor (home, Outdoor and indoor
planetary and ventilation
malls, restaurants)
Location alertness Yes No Yes
Real-time connections Supported Supported Supported
Mobility Supported Limited Support Supported
Localized data service,
Global data collected Data precise to end
Service type limited to precise
worldwide devices
deployment location
Short duration as it Months and years
Temporary because it is
Duration of data storage transmits big data to because it manages big
the source of big data
cloud data
Major service provider Cisco IOx Amazon, IBM, Microsoft Bosch, Atmel, ARM
88
Figure 2. Features of fog computing
communicate among themselves by self-organizing, scheduling and co-

operating among them to form a decentralized system. This enhances the
real-times services and IoT requests to users (Vaquero & Rodero-Merino,
2014).
• Geographic Distribution: The main purpose of fog computing is providing
minimal delay in services and mobility to terminal nodes. For this purpose,
the fog nodes are geographically distributed and end users can take benefit of
any fog node irrespective of its location. Fog nodes can be located anywhere
like highways and roadways, on a museum floor, on cellular base stations
and any other place of interest. Fog nodes receive high quality data from the
terminal nodes due this feature (Ahmed & Rehmani, 2017).
• Low Latency: In this era of real-time application services, there are certain
applications and IoT devices that require hasty analysis of data generated by
them. This is not feasible with cloud computing as the servers are present at
remote location. Fog nodes with resources play a major role by providing
services for computation and services for terminal nodes. As these fog nodes
are adjacent to terminal nodes the latency to provide amenity is highly reduced
89
when compared to cloud computing. Hence fog nodes can make decisions
based on local data without assistance of cloud (Ahmed & Rehmani, 2017).
• Location Awareness: Terminal nodes can access the services from fog nodes
based on the location of fog node. This requires tracing of devices location
actively and passively to provide high quality service to IoT applications.
In fog paradigm, the IoT application accesses the fog nodes that are present
nearest to end devices. This delivers rich service by providing the information
about the location of nodes and hence enabling superior quality of service
and resource (Ahmed & Rehmani, 2017).
• Large-Scale IoT Application Support: An evolving wave of internet
organizations, most particularly Internet of Things that require various
necessities like location awareness, mobility support, low latency in addition
to geographic distribution. All these assets are provided by fog computing.
This enable it to support large scale IoT applications to utilize the resources
offered by fog nodes. There are many applications that exploit the services
of fog nodes and manages large scale IoT applications like climate change
monitoring, smart grid management, environment monitoring etc. (Bonomi
et al., 2012).
There are many more features offered by fog computing to enhance the computation
power and quality of service to IoT applications. The other features are mobility
that enables users to move freely with same benefit of service irrespective of their
location. It also provides local preprocessing of data before sending it to the cloud
by reducing the strain on cloud servers. It also assists in local real-time online data
analytics and interplay with the cloud (Ni, n.d.; Bonomi et al., n.d.).
SECURITY AND PRIVACY ISSUES IN FOG COMPUTING
Cloud computing has huge storage and computing resources, it assists clients in
many ways by affording the services. But it also has many drawbacks, one among
them is its decentralized architecture to manage and store data. This makes cloud
computing vulnerable to external attacks on centralized data storage and computing
framework. There are high possibilities of data leakage from cloud servers, cloud
computing merchants like Google, Yahoo and Amazon have experienced data
leakage due to uncontrolled external attacks. Security is a major concern in cloud
computing that restricts its growth in many fields. As fog computing is an extension
to cloud computing, many of its drawbacks are withdrawn from fog computing. It
is considered that fog computing has more secure architecture than that of cloud
90
computing. Some of the reasons to make fog more securely dependent than cloud
are (Roman, Lopez & Manbo, 2018):
• End devices transmit the data to nearest local fog node where data is stored
and analyzed transiently, this decreases the dependency of the devices on
the Internet connection. As the data is stored locally, it’s quite difficult for a
hacker to access the data stored on fog nodes. This reduces probability of an
unauthorized access of information.
• Cloud and end devices do not exchange information in real-time as fog nodes
act as intermediate nodes and provides real-time service to end devices. This
makes harder for eavesdropper to discern sensitive data of a precise user.
Nevertheless, fog computing is not considered as completely secure as it

inherits many security and privacy issues from cloud computing. Fog and cloud
service providers are only concerned with their benefits by providing services to
end devices/end users. They may not deviate from the agreement signed between
them but they might snoop the content of the data that was entrusted by users to
store at their resource premises. They can also acquire any user sensitive data, this
might cause privacy leakage for users. There is a possibility where fog and cloud
are targets by hackers that can use various methods to reach their goal. There can
be many attacks on fog nodes and some of them are mentioned below (Roman,
Lopez & Manbo, 2018):
• Forgery: Malicious attackers can act as fake fog nodes by forging identities
and contours. They can also mislead other nodes and end devices by
generating fake information. This might also lead to excess consumption of
resources like storage, bandwidth and computation to process unnecessary
fake information.
• Tampering Attack: The fog nodes and other terminal nodes are connected
through wireless channel this may result to tampering attack by dropping,
delaying or modifying the information transmitted between the source and
fog nodes. This reduces the efficiency of fog computing due to the delay
caused to deliver the service. As fog computing provides mobility support to
users, it might lead to transmission failure and delay.
• Spam Attack: The attackers produce unwanted data, redundant information
by some means. This information is spread across different fog nodes
misleading other fog nodes and end devices. The attacker can also access
user specific information leading privacy leakage. By consuming large
amount of unwanted or redundant data will cause needless network resource
consumption.
91
• Jamming Attack: As mentioned above, the attackers can generate large

amount of unwanted information. Fog nodes get busy to compute this
fake information. Processing large amount unnecessary data will block
the fog nodes. Other legitimate user cannot communicate with fog node
to get serviced. The attacker jams fog nodes and disrupting the normal
communication between end devices and fog nodes.
• Eavesdropping: Malicious attackers can listen when there is a communication
channel established between fog nodes and terminal nodes. They can capture
the packet that are being transmitted and read the content in the packet if there
is lack of encryption on actual data. This type of attack is difficult to detect.
Hence the fog nodes and terminal nodes that are transmitting packets are
often unaware of third party reading their information in packets transmitted.
• Denial-of-Service (DoS): Fog nodes are quite vulnerable to Denial-of-
Service when compared to cloud computing as they have minimal amount
of resources for storage, bandwidth and computation. The attacker floods
unnecessary requests to fog nodes by disrupting the normal services provided
to terminal nodes by making them unavailable to their intended users. This
kind of attack consumes network resources to prevent fog nodes serve their
legitimate users.
• Collusion Attack: In general, collusion attackers are combination two or more
parties colluding together to increase the power of an attack. They mislead,
deceive or defraud the legal authorities or attain an unfair advantage. In fog
computing, many attacking parties can collude to increase their capability
to defraud or imbalance the normal communication. Several fake fog nodes
or Terminal nodes can lead to this attack. The attack can occur between IoT
devices and cloud, or Fog nodes with IoT devices.
• Man-in-the-Middle Attack: This type of attack is quite difficult to detect as
the users think they are in sync with each other but are unaware of presence
of malicious attacker that has an ability to manipulate or relay the information
transmitting between two devices. Fog computing has its application many
real-time environments like smart power grid, healthcare, transportation.
The data transmitting between these nodes in fog environment is critical
information, modification of such data can sometimes to lead to accidents or
loss of life in case healthcare environment.
• Impersonation Attack: A malicious attacker enacts as a legitimate terminal
node by spoofing the right identity and enjoying the benefits of services
provided by fog nodes. In another case, fog computing nodes can be
duplicated. They attackers can provide false data or services to users and
misleading other fog nodes as well.
92
There are twelve more threats that are identified by organization named Cloud
Security Alliance (Alliance CS, 2016). The following shows twelve security issues
recognized by some researchers (Stojmenovic & Wen, 2014; Stojmenovic et al.,
2015; Yi, Qin & Li, 2015) to formulate a systematic review:
• Advance Persistent Threats (APT): The main purpose of this attack is to

steal data and intellectual property. This can be achieved by compromising
the company’s infrastructure and devices used in fog computing.
• Access Control Issues (ACI): As mentioned above, this attack leads to
getting the access to unauthorized data and permissions to install software
and change configurations. When the organization has poor management, the
tendency of this attack increases gradually.
• Account Hijacking (AH): The name describes the purpose of the attack,
hijacking one or more fog nodes by any means to fetch the services offered to
that account and utilizing the data and permissions available for the hijacked
device. The most popular technique for account hijacking is phishing.
• Denial-of-Service (DoS): it is an attacking technique where legitimate users
are prevented or denied for their services. The attacker overloads fog nodes by
fake requests and makes fog node unavailable to other legitimate end users.
• Data Breaches (DB): When user specific data, private or confidential data
is stolen or released by an attacker it leads to severe sensitive data breaches.
• Data Loss (DL): Fog nodes are geographically distributed, this might cause
any damage to fog nodes due to natural disaster resulting to data loss stored
at fog nodes. This can be accidental or deliberately by an attacker.
• Insecure APIs (IA): Many fog service providers utilize Application
Programming Interface (APIs) to provide an interface for users to contract the
services provided by fog nodes. The security of these APIs is an important
task before implementation of any application.
• System and Application Vulnerabilities (SAV): The attacker can penetrate
and compromise any system by exploiting bugs arising in the system.
• Malicious Insider (MI): It is an authorized person with an intention of
stealing data or performing an attack to mend the architecture or system
of an organization. A fog node can become rogue and disrupt the normal
communication among other fog nodes and devices.
• Insufficient Due Diligence (IDD): This frequently occurs when an
organization urges development, design and implementation of any system
without considering all the scenarios and exceptions that can arise after
implementation.
93
• Abuse and Nefarious Use (ANU): This arises when attackers utilize the fog
resources when its freely available, they can misuse the resources and block
the fog nodes to only serve them.
• Shared Technology Issues (STI): Sharing technology, infrastructure,
hardware, applications or platforms might lead to some overload on the
system. There might be devices that cannot handle serving too many users.
All the above attacks are shown in Figure 3 along with other privacy issues in
fog computing. Users sensitive data is involved in the collection, transmission,
processing and sharing. Most of the users would like to keep their information in
private, but this is not possible in many cases. Users sensitive data can be leaked by
many reasons. Some of the attacks to get the private information are listed below:
• Identity Privacy: In most cases users submit their identity information to fog
nodes for authentication purpose. A user can have various identity information
like name, address, phone number, license number, public-key certificate,
visa number. All this information is linked to specific user. Disclosure of such
private is vulnerable to user’s identity.
• Data Privacy: Users data stored on fog nodes is on high risk of privacy
leakage to third party. The data can be exposed to an untrusted party when
the data is upheld on fog nodes and transmitting the data between two parties.
Exposure of this data enables an attacker to analyze this data and obtain user
specific details like users address, preferences, location, occupation and
political disposition. For example, leaking data related to health status of a
user can open the information regarding the health issues with the user.
• Usage Privacy: The attackers analyze the usage patterns of users and predict
their behavior and information. The pattern of any device using the service
of a fog node can be obtained and analysis of this information can constrain
usage privacy. For example, examining and analyzing reading of smart meter
can disclose the information regarding power consumption of a house and
disclose the living habits of a family like the time when family goes to sleep
or when they are not at home. This violates residents’ privacy and may lead
to robbery planning for a house using this information.
• Location Privacy: Nowadays enormous applications on mobile devices
collect users’ location information. To enjoy some online facility, users need
to sacrifice our location information. Navigation and location-based service
require users’ location to provide service to them. Though location data is
commonly shared among the application, its preservation is critical indeed.
The attacker can traverse the users travelling route by collecting location
information from fog nodes. As devices are mobile, they take services from
94
the nearest for node, hence a conclusion can be drawn that device is present
close to this fog node rather than other fog nodes. Moreover, users path
trajectory can be disclosed to fog nodes by analyzing the request sequence
to different fog nodes at different locations. As users are mobile and accesses
services from fog nodes deployed at different locations, this information can
violate the location privacy of the user by exposing the users route.
IoT devices in Fog Computing architecture are main sources for information and
security threat. With increase in count of linked IoT devices the vulnerability for
security and privacy threat also increases. IoT devices are more exposed to hacking,
breaking or getting stolen due to its fewer security features. When these devices are
hacked they can interrupt normal execution of services and tasks. Due to security
and privacy issues shown in Figure 3, it is very critical to plan, design an efficient
security and privacy-conserving mechanisms in fog computing. With lack of security
Figure 3. Security and privacy issues in fog computing
95
measures, the users might not be willing to participate in IoT applications and
impeding the growth of fog computing. The next topic discusses real-time scenarios
and security threats that in the applications.
REAL-TIME SCENARIOS OF FOG COMPUTING

AND SECURITY ISSUES CONFRONTED
This part of the chapter explains some the application of fog computing. It also
explains some of the security issues faced by those applications.
• Improvising Healthcare Systems and Their Performance: Healthcare

is a vast domain and consists of elderly care systems, where huge number
of sensors send large amount data to cloud for processing and storage. Fog
computing can be applied to healthcare system, where the sensor nodes send
data to fog nodes instead of cloud directly. Using a small organization of
healthcare with large number of sensors, the data that is sent to fog nodes is
preprocessed by tagging and classification. This reduces load on cloud servers
when the data is ready for analysis as it enters the cloud environment. Fog
layer delivers sophisticated data to cloud system for further processing and
analytics (Prieto Gonzalez et al., 2016). Combining cloud computing and fog
computing makes a better infrastructure with better Quality of Service (QoS)
and domination (Stantchev et al., 2015). One of the real-time application of
the above described methodology is OpSIT Healthcare project in Germany.
The healthcare facility takes advantage of fog computing by providing
services from nearest located fog nodes, store heterogenous information.
The facility consists of smart low power devices and has an ability to shift
among different communication protocols by providing flexibility. The entire
architecture facilitates distributed computing. The tasks are distributed
among cloud and fog nodes efficiently to provide scalable solution allowing
the system to swift detection of a patient abnormalities and development (Shi
et al., 2015).
Patient health records comprises sensitive data that describes the health conditions
of the patient. There is large probability of compromising this sensitive data. The
data can be lost or negotiated due to some external attack by an unauthorized entry
into the system. It might be also caused by any malicious insider that can leak the
sensitive and violate the privacy issues. Sensor terminal nodes keep sending data
to fog nodes continuously either through wireless or wired connection. Data can
be compromised while transmission by any third party (Li et al., 2010). There are
96
high chances of DoS, description disruption, and forwarding attacks due to the
ease-of-access provided by the system. The above security and privacy issues can
be managed and prevented by implementing strong policies to maintain high-level
control over the system. Unauthorized access can be avoided by implementing
multifactor authentication. Private network and strong encryption techniques for
communication and transmission of data can avoid privacy issues (Ren, Lou &
Zhang, 2008).
• Intelligent Food Traceability: Tracing food management can benefit for

the stakeholders in multiple ways to maintain food quality and track the
transportation of the food items. Fog computing can be applied for food
traceability management. This method utilizes value-based processing
to remove poor quality products from food chain supply. There are many
attributes like location, transportation devices and processing devices that
helps in physical tracing of the food item. Fuzzy rules are implemented to
make decisions on the quality of food based on distributed food traceability
though Cyber Physical System (CPS). Fog nodes maintain and collect
information regarding the traceability and quality, where the complete food
supply chain is observable. Now, the entire information about all traced food
objects is stored at fog nodes and the information about the quality can be
sent forward to cloud for analysis. This information from cloud can be viewed
by many other stakeholders using internet (Chen, 2017).
The security issues with this application can arise when an attacker can hinder
the normal execution of food supply chain processes. Suppose any fog node is
compromised by means of hijacking or exploiting systems and application liabilities,
this can lead to data falsification or corruption by the attacker. When food quality data
is fabricated, which eventually fallouts in the sale of substandard and low-quality food
products. Large number of wireless devices in a network and Machine-to-Machine
(M2M) communication prompts many security concerns. In one such scenario, a
resonance attack can be activated where the sensor devices are forced to function at
diverse frequency and causing incorrect data to be transmitted. This attack effects
the convenience and operation of network and data transmission. These security
concerns can be prevented by implementing integrity check, redundancy to avoid
single-point failure and detecting dishonesty attacks (Saqib et al., 2015).
• Surveillance Video Stream Processing: In an application where well-

organized processing and prompt decision-making is required, fog computing
is best solution for the application to adopt. For example, consider tracking of
multiple targets in a drone video stream as mentioned in Chen et al. (2016).
97
To bring about fog computing in this application, live video feeds can be sent
to fog nodes instead of cloud-based applications. Fog nodes can be any device
with minimal resources like mobile phones, laptops or tablets. These devices
can execute tracking algorithms and process raw video stream frames. With
this in the system architecture, the latency to deliver raw video stream frames
to cloud can be reduced to great extent. The real-time implementation results
of fog computing in this scenario reduced total processing time to average
of 13%. “Proximal Algorithm” can be implemented in fog nodes to support
large-scale video streaming application. It can also provide solution to joint
resource allocation issue. The potential in finding missing children or people
becomes strong. It can also be used to locate any person with criminal records
(Shi & Dustdar, 2016).
Fog nodes store and process the data generated from respective camera sensors.
The data can be video and audio that is stored in fog nodes, therefore the stream
must be secured and privacy of the data must be maintained as this information will
be shared among heterogenous users. All the devices involved in the transmission
must be secured. Providing security to fog nodes is not sufficient, terminal nodes
and cloud servers should also contain security management strategies especially
against APTs. The data can be leaked, tampered or even deleted if the fog platform
consists of any bugs due to negligence during design. Fine grained access control,
careful encryption mechanism and restrictive number of connections are some of
the security measures that can be considered in this real-time application of fog
computing (Do et al., 2015; Varalakshmi, Sudha & Jaikishan, 2014).
• Collection and Pre-Processing of Speech Data: The next real-time

application of fog computing is for patients with Parkinson’s disease. This
application enables android smart-watches coupled with a smart tablet that
has ability to collect, store and process speech information generated by the
patients. The fog computing interface (FIT) (Monteiro et al., 2016) reduces
the load of data that is supposed to be delivered to cloud. FIT extracts the
required information like short-time energy, spectral centroid, volume and
zero-crossing rate from the speech data collected by the fog nodes. Bandwidth
resource can be minimized by only transmitting only main factors that are
extracted by FIT instead of transmitting the complete audio speech data.
Preprocessing of data is performed before transmitting data to cloud server.
The security and quality of the service can be compromised due to large amount
applications hosted on smartphones and tablets. The data should be accessed by only
legitimate applications which is often decided by the user installation. Violating
98
the access control over the speech data can cause many cyber-attacks (Heuser et
al., 2016). Usually the fog platforms are constituted and built on mobile phones or
tablets as mentioned above. The fog platform must be secured and protected as they
are deployed on mobile operating system. When the fog platform is configured to be
open-source, there is a high risk of malicious application negotiating fog processes
and the connected network. User’s personal data is also on jeopardy due to this.
These security issues can be resolved by adopting the following methods: anti-virus,
software patching, firewall, constant data backups, Intrusion Prevention System,
creating system restore points and running behavior analysis methodologies through
dynamic monitoring (Wei et al., 2012).
Fog computing delivers and receives high quality data to cloud and from
terminal nodes. With many advantageous features provided by fog computing, it
also introduces some drawbacks related to security and privacy issues. The security
and privacy issues that can arise were discussed in previous section of the chapter,
ant this section explains the issues in real-time application. Table 1 summarizes the
possible attacks that can occur in fog computing applications.
EXISTING SECURITY SOLUTIONS FOR FOG COMPUTING
Applications that require low-latency services can adopt fog computing to enhance its
capability to serve mobile end devices. Many users do not opt for fog computing due
the security and privacy issues that come along with its implementation. To increase
the acceptance of fog computing paradigm, one must guarantee safe communication
and privacy for user’s data. Therefore, building real-time applications by considering
all the above issues is critical task. Researchers discuss some techniques that are used
to enhance privacy preservation and prevent attack on fog applications (Ni, n.d.).
Table 2. Possible security issues that can occur in real-time applications among
the twelve categories of security
Application APT ACI AH DoS DB DL IA SAV MI IDD ANU STI

Healthcare
✓ ✓ ✓ ✓ ✓ ✓ ✓
System
Food
✓ ✓ ✓
Traceability
Surveillance
Video ✓ ✓ ✓
Processing
Speech data ✓ ✓
99
• Identity Authentication: Fog nodes, cloud and terminal nodes must be

authenticated to provide authenticity and creditability to access the data and
services. Improper authentication system might lead to external attacks on
the fog nodes. Providing upright authentication service is critical task.
There many methods employed to provide authentication to all the devices on

fog computing. But these mechanisms do not support mobility of devices. In fog
computing, the services are delivered to users that are mobile in nature, these devices
move out of coverage area from one fog node to another. Providing authentication
to all the devices that are in movement can delay the operation of fog nodes when
the count of users increases. Hence, authentication service with minimal latency
and cooperative authentication schemes must designed and employed for better
authentication scheme in fog computing. These schemes must avoid redundant
authentication exertions when different set of users are communicating the same
message.
In some applications, users are not willing to share their identities to fog nodes
during validation. When a user is travelling, he/she might be reluctant to share their
identity information to avoid leakage of private data. If an attacker gets hold of
such information, then the attacker can track the users’ location. Many devices like
smart watch, smart glasses and smart jewels like bracelets reveal the location and
identity evidence to fog nodes. Researchers can solve this issue by using Anonymous
authentication mechanism, where user does not have to reveal its identification or
location. There are many anonymous authentication algorithms, some of them are
pseudonyms, group signatures and k-anonymity. With these techniques, fog nodes
cannot distinguish other users based on the identification key. A hybrid algorithm
that can have properties of both mechanism is more suitable.
• Policy-Driven Secure Management of Resources: This type of security

mechanism enhances safe communication, interoperability and sharing by
implementing preliminary policy management agenda for the capitals of fog
computing (Ahmed & Rehmani, 2017). As described by the policy there are
five major modules:
a. Policy Decision Engine (PDE) for acting based on pre-defined policy
rules.
b. Application Administrator (AA) to manage fog multi-tenancy.
c. Policy Resolver (PR) for attribute-based authentication
d. Policy Repository (Prep) holding rules and policies
e. Policy Enforcer (PE) to detect any discrepancies in policy implementation.
100
The responsibility to define rules and policies by considering multi-tenancy,

communication services, data sharing and applications is engaged by AA. All the
requests that are made from users are forwarded to PR whose responsibility is to
identify user based on a detailed set of attributes that were registered by the user
during authentication phase. The user attributes and their agreements are stored in
a database that is maintained by PR. PR examining the attributes and agreements
and then provides access to the user against requested resource. The role of PDE is
taking user specific information from the PR, it also extracts rules from Prep. After
extraction of rules, PDE analyzes them and enforces them through the PE. The best
method to create is “eXtensible Access Control Markup Language (XACML)” and
the context for building PDE is OpenAZ framework. This kind of policy management
framework is suitable for distributed computing which is enforced in fog computing.
It provides strong control for access, resource management and identification in
fog computing. Nevertheless, this security mechanism is only suitable for systems
that consists of devices which distribute dedicated resources for large number of
computations required by different modules to execute the framework within fog
platforms. This mechanism might cause some delay to execute the operations for
authentication, and hence by causing some latency to time-sensitive applications.
This mechanism is also vulnerable for DoS due to compound authentication methods
in PR and PDE. The response to devices might slow down when malicious user
sends large number of requests, the system repeats validation process for the same
connection by delaying other services. Hence a better policy management framework
is required to handle large number of requests and efficiently maintain resources
of the devices.
• Encryption and Decryption Mechanisms: To make secure communication

among the fog nodes and terminal nodes or fog nodes and cloud, light weight
encryption-decryption mechanism is implemented. The data is encrypted
based on private key that is shared by fog nodes and terminal nodes. When
data is sent from terminal node, it is encrypted in a specific format. When the
fog node receives data from terminal node, the information can be extracted
by decrypting in same format based on the encryption key shared with
devices. A malicious third party cannot get information about the information
that is being transmitted. This method prevents Man-in-Middle attack. There
are some drawbacks with this mechanism, terminal nodes lack resources for
storage and computation, therefore encrypting data before sending it to fog
nodes requires large amount of computation resources. The terminal nodes
might not be able handle the load of such heavy computations. This also
101
causes small delay in processing data as it consumes to apply encryption

and decryption at both the ends of connection. Applications that are time-
sensitive are not favorable for this mechanism. If the encryption is leaked to
any other third party then it’s no use of this technique as the attacker can get
information with key. Therefore, a secure communication protocol must be
designed to safeguard the resources and data on fog nodes.
PROPOSED SECURITY TECHNIQUE
The main reasons for an attack is unauthorized access to resources and data, when
authorization to legitimate users is given then risk of any data loss or any attack
can be reduced. Detection of malicious user is one of the main tasks. Researchers
propose an intrusion detection and prevention technique that can handle and avoid
different attacks.
If each user is provided with an authentication key, and fog node provides access
based on this authentication key. A terminal node must get the authentication key from
the cloud service or any fog node before sending the request. Cloud and fog nodes
distributes the key based on the user details and policies. The same authentication
key can be used every time the user sends a request to fog nodes. A database can
be maintained for authentication key provided to users. This database information
can be shared by other fog nodes. When the user is mobile and sends a request to
fog node, fog node checks its database for the authentication purpose, if the user
is an authorized user then service is offered to the user through fog node. The fog
node also keeps track of user location, it sends the user authentication details to
other fog nodes where the user can request for service. The database is updated
by the fog nodes time to time and the information is also maintained at cloud for
backup purpose.
When a malicious user is detected that is trying to access some data, then fog
node sends a notification to legitimate user about the action. If the legitimate user
declines the notification, then user is marked as an attacker and actions can be taken
against the malicious user by plotting honeypots. Fog node can send fake data to
the malicious user, when the user downloads the data, an .exe file can be run that
extracts IP and mac address of the user. By this way the malicious attacker can be
identified. The IP address can be changed but mac address remains same for the
device, therefore extracting mac address of the attacker device should be focused
and this information must be transmitted to the respective fog node. Fog node can
send this information to other fog nodes and cloud to prevent other attacks on other
102
fog nodes. Hence each fog node maintains two kinds of information. That is details
of authorized users and details of unauthorized malicious users. Before permitting
the user for any service or resource, fog node must check both the databases for safe
and secure protection. Figure 4 and Figure 5, shows the workflow of the proposed
mechanism to tackle security and privacy issues in fog computing.
Figure 4. Proposed Intrusion Detection and Prevention Technique
Figure 5. Proposed Intrusion Detection and Prevention Algorithm
103
CONCLUSION
Fog computing is latest technology that contributes an extension to cloud by providing

services at the edge of the network. The users rely on fog nodes for computation
and storage resources instead of cloud because of the enhanced features provided
by fog computing. The features of fog computing were discussed in the chapter are
the new benefits extended by fog. Fog computing eradicates many issues arising
in cloud computing service. It retains computation and data local to end users by
providing low latency, location-awareness and high bandwidth. Fog inherits some
issues concerned with security and privacy of data and devices in fog architecture.
Making data and other nodes secure is a vital task. Researches proposed a system
that detects an intrusion into the fog application and planned measures to be taken
after the intruder is detected. We need a more secure system that can avoid data loss
due reasons like natural disasters and attacks. The real-time service of fog computing
should not be interrupted due to privacy or security issues. Researches must focus
on developing light weight secure system for fog nodes with minimal resources.
REFERENCES
Ahmed, E., & Rehmani, M. H. (2017). Mobile Edge Computing: Opportunities,

Solutions and Challenges. Future Generation Computer Systems, 70, 59–63.
doi:10.1016/j.future.2016.09.015
Alliance, C. S. (2016). The Treacherous 12 Cloud Computing Top Threats in 2016.
Retrieved from https://downloads.cloudsecurityalliance.org/assets/ research/top-
threats/Treacherous-12_Cloud-Computing_Top-Threats. pdf
Bonomi, F., Milito, R., Natarajan, P., & Zhu, J. (2014). Fog computing: A platform
for Internet of Things and analytics. In Big Data and Internet of Things”: A Roadmap
for Smart Environments (Studies in Computational Intelligence) (pp. 169–186). New
York: Springer. doi:10.1007/978-3-319-05029-4_7
Bonomi, F., Milito, R., Zhu, J., & Addepalli, S. (2012). Fog computing and its role
in the Internet of Things. Proc. 1st Ed. MCC Workshop Mobile Cloud Comput.
(MCC), 13–16. 10.1145/2342509.2342513
Bonomi, Milito, Zhu, & Addepalli. (n.d.). Fog Computing and Its Role in the Internet
of Things. Cisco Systems Inc.
104
Chen, N., Chen, Y., You, Y., Ling, H., Liang, P., & Zimmermann, R. (2016). Dynamic
urban surveillance video stream processing using fog computing. In Multimedia Big
Data (BigMM), 2016 IEEE Second International Conference on. IEEE. 10.1109/
BigMM.2016.53
Chen, R. Y. (2017). An intelligent value stream-based approach to collaboration
of food traceability cyber physical system by fog computing. Food Control, 71,
124–136. doi:10.1016/j.foodcont.2016.06.042
Chiang, M., & Zhang, T. (2016). FogandIoT:Anoverviewofresearchopportunities.
IEEE Internet Things J., 3(6), 1–11.
Cisco. (2017). Fog Computing and Internet of Things: Extend the Cloud to Where
the Things Are. Available: http://www.cisco.com/c/dam/en_us/solutions/trends/iot/
docs/computing-overview.pdf
Díaz, M., Martín, C., & Rubio, B. (2016, May). State-of-the-art,challenges,andopen
issues in the integration of Internet of Things and cloud computing. Journal of
Network and Computer Applications, 67, 99–117. doi:10.1016/j.jnca.2016.01.010
Do, C. T., Tran, N. H., Pham, C., Alam, M. G. R., Son, J. H., & Hong, C. S. (2015).
A proximal algorithm for joint resource allocation and minimizing carbon footprint
in geo-distributed fog computing. In 2015 International Conference on Information
Networking (ICOIN). IEEE. 10.1109/ICOIN.2015.7057905
Heuser, S., Negro, M., Pendyala, P. K., & Sadeghi, A. R. (2016). Droidauditor:
“Forensic analysis of application-layer privilege escalation attacks on android.”
Technical report. TU Darmstadt.
Li, M., Yu, S., Ren, K., & Lou, W. (2010). Securing personal health records in cloud
computing: Patient-centric and fine-grained data access control in multi-owner
settings. In International Conference on Security and Privacy in Communication
Systems. Springer. 10.1007/978-3-642-16161-2_6
Monteiro, A., Dubey, H., Mahler, L., Yang, Q., & Mankodiya, K. (2016). Fit a fog
computing device for speech tele-treatments. arXiv preprint arXiv:1605.06236
Ni, Zhang, Lin, & Shen. (n.d.). Securing Fog Computing for Internet of Things
Applications: Challenges and Solutions. IEEE Communications Surveys & Tutorials.
DOI 10.1109/COMST.2017.2762345
105
Prieto González, L., Jaedicke, C., Schubert, J., & Stantchev, V. (2016). “Fog computing
architectures for healthcare”: Wireless performance and semantic opportunities. J
Inf Commun Ethics Soc, 14(4), 334–349. doi:10.1108/JICES-05-2016-0014
Ren, K., Lou, W., & Zhang, Y. (2008). Leds: Providing location-aware end-to-end
data security in wireless sensor networks. IEEE Transactions on Mobile Computing,
7(5), 585–598. doi:10.1109/TMC.2007.70753
Roman, R., Lopez, J., & Manbo, M. (2018). Mobile Edge Computing, Fog et al.: A
Survey and Analysis of Security, Threats and Challenges (Vol. 78). Futur. Gener.
Comp. Syst.
Saqib, A., Anwar, R. W., Hussain, O. K., Ahmad, M., Ngadi, M. A., Mohamad, M.
M., & ... . (2015). Cyber security for cyber physcial systems: A trust-based approach.
J Theor Appl Inf Technol, 71(2), 144–152.
Sarkar, S., Chatterjee, S., & Misra, S. (Forthcoming). Assessment of the suitability
of fog computing in the context of Internet of things. IEEE Trans. Cloud Comput.
Shi, W., & Dustdar, S. (2016). The promise of edge computing. Computer, 49(5),
78–81. doi:10.1109/MC.2016.145
Shi, Y., Ding, G., Wang, H., Roman, H. E., & Lu, S. (2015). The fog computing
service for healthcare. In Future Information and Communication Technologies for
Ubiquitous HealthCare (Ubi-HealthTech), 2015 2nd International Symposium on.
IEEE. 10.1109/Ubi-HealthTech.2015.7203325
Stantchev, V., Barnawi, A., Ghulam, S., Schubert, J., & Tamm, G. (2015). Smart
items, fog and cloud computing as enablers of servitization in healthcare. Sensors
Transducers, 185(2), 121.
Stojmenovic, I., & Wen, S. (2014). The fog computing paradigm: Scenarios and
security issues. In Computer Science and Information Systems (FedCSIS), 2014
Federated Conference On. IEEE. 10.15439/2014F503
Stojmenovic, I., Wen, S., Huang, X., & Luan, H. (2015). An overview of fog
computing and its security issues. Concurrency and Computation.
Vaquero, L. M., & Rodero-Merino, L. (2014). Finding Your Way in the Fog: Towards
A Comprehensive Definition of Fog Computing. ACM SIGCOMM Comp. Commun.
Rev., 44(5), 27–32. doi:10.1145/2677046.2677052
106
Varalakshmi, L., Sudha, G. F., & Jaikishan, G. (2014). A selective encryption and
energy efficient clustering scheme for video streaming in wireless sensor networks.
Wei, X., Gomez, L., Neamtiu, I., & Faloutsos, M. (2012). Malicious android
applications in the enterprise: What do they do and how do we fix it? In Data
Engineering Workshops (ICDEW), 2012 IEEE 28th International Conference on.
IEEE. 10.1109/ICDEW.2012.81
Yi, S., Hao, Z., Qin, Z., & Li, Q. (2015). Fog Computing- Platform and Applications.
IEEE Workshop on Hot Topics in Web Systems and Technologies.
Yi, S., Qin, Z., & Li, Q. (2015). Security and privacy issues of fog computing:
A survey. In International Conference on Wireless Algorithms, Systems, and
Applications. Springer. 10.1007/978-3-319-21837-3_67
107
108
Chapter 6
EdgeCloud:
A Distributed Management System
for Resource Continuity in Edge to
Cloud Computing Environment
Jamuna S. Murthy
ABSTRACT
In the recent years, edge/fog computing is gaining greater importance and has led
to the deployment of many smart devices and application frameworks which support
real-time data processing. Edge computing is an extension to existing cloud computing
environment and focuses on improving the reliability, scalability, and resource
efficiency of cloud by abolishing the need for processing all the data at one time
and thus increasing the bandwidth of a network. Edge computing can complement
cloud computing in a way leading to a novel architecture which can benefit from
both edge and cloud resources. This kind of resource architecture may require
resource continuity provided that the selection of resources for executing a service
in cloud is independent of physical location. Hence, this research work proposes a
novel architecture called “EdgeCloud,” which is a distributed management system
for resource continuity in edge to cloud computing environment. The performance
of the system is evaluated by considering a traffic management service example
mapped into the proposed layered framework.
DOI: 10.4018/978-1-5225-6070-8.ch006
EdgeCloud
INTRODUCTION
In the past few years we notice that there is a tremendous increase in the number
of devices getting connected to the network. The raise in number of devices are
concerned with two main resources i.e. the user devices and the sensors/actuators. The
researchers from Cisco Company extensively report that there will be approximately
50 billion devices that will be connected to the network by the year 2020 (Evans,
2011). Today in the developing countries there is a major growth in the number of
devices used by the people in terms of mobile phones, tablets etc. But very soon the
usage of these devices will be over passed by the myriad of sensing/acting devices
placed virtually everywhere (the so called Internet of Things, IoT, and pervasive
sensor networks).
The new innovations such as the concept of Smart Cities(Nam & Pardo, 2011),
Wearable Computing Devices such as smart watches and glasses, the Smart Metering
Devices for monitoring energy consumption at homes, Visual Sensor Networks,
Self Driving Vehicles with smart meters are such applications which are driving the
ubiquitous computing to the next level of usage witnessing the presence of smart
devices everywhere around us. This kind of technical achievements (i.e. invention
of Smart devices) by the researchers is made possible by the usage of widely used
technology called the Edge Computing or Fog Computing.
In the recent times Cloud Computing is slowly migrating towards the edge of
network facilitating the routers to form an efficient virtualization infrastructure to
support real-time data processing. This evolution of cloud to edge is labelled as
Edge Computing or Fog Computing. Edge Computing is a scenario where in large
number of heterogeneous devices (may be wireless or autonomous), ubiquitous
devices and decentralized devices combine together in a cooperative manner to
perform tasks such as storage and processing without any intervention of third party
devices. The tasks performed are in nature of support for some network functioning
or application interference or to provide new services in the sandboxed environment.
They are generally fast since the operations take place at the edge of the network.
The main focus of Fog Computing is on improving the reliability, scalability and
resource efficiency of cloud by abolishing the need for processing all the data at one
time and thus increasing the bandwidth of a network (Vaquero & Rodero-Merino,
2014). Edge or Fog Computing provides wide range of benefits which is listed below:
• Network Traffic Reduction: There are billions of devices connected to the

network world wide today. Amongst which the smart phones and tablets
are in greater numbers being used by the people. These devices generally
send, receive and generate data in such a way that the computing capabilities
are related to the physical location nearest to the devices rather than being
109
EdgeCloud
communicating with the data centres directly. Based on the frequency range
configured for the devices, the sensor embedded in them collect the necessary
data every few seconds. Therefore, it is neither necessary nor advisable to send
all of the raw data to the cloud. Hence Edge Computing is very beneficial here
by providing a platform for data collection, data filtering and data analysis at
the edge of the cloud without being sent all the data at one time but perhaps
sends only the absolutely necessary data for providing services (i.e. local data
view). This in turn reduces the network traffic to greater extent.
• Best Suitable for IoT Queries and Tasking: As we know that the usage
of smart devices and applications are increasing day by day. These smart
devices are usually designed to collect the surrounding information and
retrieve the information to the end user based on their service request. Most
of the applications today use Edge Computing where in they serve the
request without communicating with the global data present in the cloud. For
Example, the applications such as Google Map is used for keeping track of
surrounding information.
• Low Latency Requirement: Today most of the applications require real-time
processing in order to quicken the tasks. One of the best examples is Cloud
Robotics, which is a mission critical application where in we need to control
the motion of the robot. Here the motion control depends on the data collected
from the sensors and the processing is done based on the feedback control
system designed. If in case the data necessary for the processing is present
in the cloud then most of the times the sense-process-actuate loop processes
become slow due to the continuous requests. This leads to communication
failures and breakdown of application. This is where the Edge computing
becomes beneficial where the controlling of the robot is done at the edge of
cloud for every motion control request and helps in real-time processing of
requests.
• Scalability: Most of the times we observe that the cloud with infinite virtual
resources can also become the bottleneck when huge data generated by the
end devices are sent to the cloud continuously. At this point Edge Computing
becomes very essential where the processing of the incoming data is done
close to the data source itself rather than communicating with the cloud data
centre. This in turn increases the scalability of the endpoint devices to greater
extent.
Edge Computing can complement Cloud Computing in a way leading to a novel

architecture which can benefit from both edge and cloud resources. This kind of
resource architecture may require resource continuity provided that the selection
of resources for executing a service in cloud is independent of physical location.
110
EdgeCloud
Table 1. Resource continuity possibilities in the layered architecture
Resource Continuity from Edge to Cloud

Edge Cloud
Basic/
Intermediate
Edge Devices Aggregation Cloud
Nodes
Nodes
Smart
Sensors,
Car, phones, building,
Device actuators Datacenter
computer cluster of
wearables
devices
Sub seconds, Seconds, Minutes, days,
Response Time Milliseconds
seconds minutes weeks
M2M Dependable Visualization Bigdata
Application
communication services Simple analytics
Examples
Features haptics (eHealth) analytics Statics
How long IoT
Transient Minutes, Hours Days, weeks Months, years
data is stored
Geographic Connected
Device Area, cluster Global
Coverage devices
Let us consider few possibilities of resource continuity in layered architecture listed

in Table 1. Table briefly lists out the characteristics of different computational
layers from edge to cloud with different layers performing tasks associated with
different devices. The different resources which are selected and categorized in
the table address the optimization of service execution criteria wherein paving a
way for providing solution to the problems such as privacy and security, resource
efficiency, network overloading etc (Cisco. Inc, 2015). Even though there are many
contributions (i.e. Business Models, Applications, Resource Models etc.) in the
field of both Edge computing and Cloud Computing most of them lack Scalability,
Resource efficiency and networking issues even today which are listed in detail in
Literature Review Section.
By considering all the drawbacks from the existing edge computing architectures
this research work proposed a new architecture called “EdgeCloud” which is a
distributed management system for resource continuity in edge to cloud computing
environment. The proposed architecture forms a foundational block to both edge
and cloud computing resources being packed as different layers with various
functionalities in order to support resource continuity while executing a particular
service. A new service execution strategy is designed as a part of research work
where different resources are assigned to different computational layers of the
architecture based on their actual resource capabilities (i.e. storage, processing,
computing and networking). The proposed “EdgeCloud” architecture is an open
111
EdgeCloud
architecture and can be applied to any context or scenario at present. It is basically

agnostic and support application extendibility to any real-time applications such
as eHealth, visual sensors, military and hostile environments, language and speech
processing etc to greater extent. However in case of extreme computing capabilities
needed for any applications the requirements on demand may be added to provide
better service. New algorithm, strategies and policies may be added based on the
user requirements necessarily. A performance study is carried out at the end where
in the scenario of Traffic Management is mapped on to the different layers of the
proposed architecture to discuss the results in terms of scalability and resource
efficiency. The database characteristics in relation to the proposed layer architecture
is also discussed in detail which pave a way to further research studies in the field.
The Chapter in organized as follows, Section 2 discuss the Literature review where
existing architecture its drawbacks and proposed solutions are discussed in detail.
Section 3 discusses the proposed “EdgeCloud” architecture with different layers
and functionalities. Section 4 discussed the performance valuation of the proposed
architecture with traffic management service deployed on a smart city and Finally
Section 5 concludes the chapter.
LITERATURE REVIEW
In the past few years we witness that the concept of Edge Computing has turned
out as a major research topic where the researchers are continuously involved in
analyzing the benefits of edge or fog computing by applying it on the traditional
applications (Baktir, Ozgovde, & Ersoy, 2017). The recent contributions are in the
form of Cognitive Body Area Networks (BANs) (Quwaider & Jararweh, 2013),
Smart Grid (Stojmenovic, 2014.), Augmented Reality (Zao, Gan, You, Méndez,
Chung, Wang, ... & Jung, 2014), Assistance, Language and Speech Processing
(Dubey, Yang, Constant, Amiri, Yang, & Makodiya, 2015), Military and Hostile
Environments (Yaseen, AlBalas, Jararweh, & Al-Ayyoub, 2016), IoT and Wireless
Sensor Networks (Aazam & Huh, 2016), Video Streaming and Analysis (European
Telecommunications Standards Institute Industry Specifications Group, Mobile-
Edge Computing – Service Scenarios, 2017), etc. Also “OpenFog Consortiums”
are organized by the industrial sectors to discuss the new ideas and opportunities
in the Fog Computing field.
Kirak Hong et al. (2013) designed a framework called “Mobile Fog” which is
a programming model for large scale applications on the Internet of Things which
are basically geospatially distributed and are latency sensitive and large scale. The
Mobile Fog model is analysed to find out the performance issues by considering
the use case of vehicular network and the simulation results were tabulated. But the
112
EdgeCloud
results showed large differences in execution time and the resource were allocated
only using the distributed layered approach. But the proposed EdgeCloud model
supports resource continuity from edge to cloud environment by combining both
cloud and edge resources efficiently and hence increases the performance of the
system in terms of bandwidth and network traffic reduction to greater extent.
Clinton Dsouza et al. (2014) designed a framework for management of resources
in fog computing environment using some policy. The architecture extended the
existing fog computing architecture to support privacy and security management plane
which is collaborative and interoperable to the user requests in Fog environment.
The framework was applied on Smart Transportation scenario to evaluate the
performance. The performance was evaluated based on the network traffic load
with high and lower load points. But the results were satisfactory. Proposed system
showed best results by reducing network traffic load to 2-3Kbps.
Bo Tang et al. (2015) designed a distributed, hierarchical fog computing framework
for Bigdata analysis in smart-cities. The framework consists of different components
and services that pave a way for a multi-ownership infrastructure deployment in
smart cities scenario with Bigdata analytics. The main motive of the framework
was to optimize the response time and identify the anomalies and hazardous events.
Fibre optic sensor technology was use to monitor distinct events in smart cities. A
performance analysis was carried out to check the response time and bandwidth
of three fog layer. The result showed bandwidth reduction of 0.02% only but the
proposed framework reduced the bandwidth to 2-3% hence proposed system can
be more applicable to smart city scenario.
Xueshi Hou et al. (2016) designed a Vehicular Fog Computing (VFC) infrastructure
which uses the collaborative approach of edge devices and end user clients for
communication and computing of resources of each vehicle. The aim of VFC is to
provide better QoS and application service by aggregating resources of each vehicle.
Four types of scenarios are discussed for moving and parked vehicle to carry out
the quantitative analysis in terms of resource capability, connectivity, and mobility
of vehicles. But if we compare the VFC architecture to proposed EdgeCloud model
the main features such as Resource management, security and privacy and the
reduction of bandwidth and network traffic are missing which are very important
for any Fog/Edge computing scenario. Thus VFC is less efficient and not secure
compared to proposed system.
Arslan Munir et al. (2017) designed an Integrated Fog Cloud architecture called
IFCIoT that aims at reduction of network traffic, quick response time, reduce
latency and to increase scalability so as to facilitate better performance for future
IoT applications. IFCIoT consist of different fog nodes that are used as edge servers,
smart routers, base stations that receive computational requests from the user. Finally,
a layered approach of five different layers is used here to increase the efficiency and
113
EdgeCloud
enhance communication between each Fog layers and edge servers are deployed
for collecting the information regarding user request. The framework is applied on
transportation management service in smart city to analyse the results. Even though
the framework is very similar to proposed layered approach the performance of
IFCIoT shows large difference since in proposed system uses resource continuity
which showed best results in network traffic reduction and also proposed layered
architecture is simple compared to IFCIoT which is very complex to understand.
PROPOSED SYSTEM
From the Section 1 study on benefits of Resource Continuity model we infer that
the combination of fog and cloud resources deliver very useful services when
managed efficiently. Hence this section describes the main architectural features of
the proposed system in two major subsections. The first architecture is lined out as a
Layered, hierarchical and distributed management architecture and the second one
as “Functional block design”. The two architectural approaches describe the proposed
system in such a way that they lead to one main objective of designing a distributed
management system for resource continuity in edge to cloud environment called
the “EdgeCloud” model. The proposed architecture is illustrated by considering the
concepts of smart city scenario for better understanding purposes.
Architecture of Proposed “EdgeCloud” Model
Figure 1 shows the envisioned layered, hierarchical and distributed architecture of

the proposed system. The system describes the allocation of fog and cloud resources
to different layers based on their functional capabilities such as response time,
storage, processing, geographical coverage etc. listed in Table 1 of Section 1. The
coordinated Edge and Cloud architecture in Figure 1, illustrates four different layers
with Cloud being the top most layer and the other three being the Edge Devices
layer, Basic or Aggregation Layer and Smart nodes layer. The design follows the
bottom up approach where the functional capabilities of the layers increase from
lower to higher layer (i.e. Cloud layer). Cloud layer is the most efficient and the layer
with highest capability. We imply on the “Edge Devices” layer that if the number of
devices is considered to be less then it leads to much higher security and privacy.
The main aim behind the proposed resource continuity model is to abstract the
resources allocated to each layer in such a way that the performance of the system
turns out to be efficient in delivering services at request time. The categorization of
the physical resources remain to be similar as mentioned in Table 1, provided the
categorization of the resources from edge to cloud coordinate in such a way that the
114
EdgeCloud
abstraction between the layers facilitate resource continuity with abstracted entity
being distributed within one unique layer. Clearly the above scenario of abstraction
is described in Figure 1 as two different views such as Physical view and Logical
view. The Physical view is more of detailed layered view where the four different
layers are clearly visible as Edge Layer 1(EL1), Edge Layer 2(EL2), Edge Layer
3(EL1) and Cloud Layer (CL). EL1 consist of set of Edge devices namely Sensors,
actuators, wearables etc. EL2 consist of Basic/Aggregate nodes such as cars, mobile
phones, tablets, PCs etc. EL3 consist of Intermediate nodes such as Smart buildings,
Cluster of devices etc. and finally CL later consist of Data center. These layers are
mapped on to a single abstract view in order to facilitate resource continuity known
as the Logical View. Each of the abstracted resources is envisioned to be managed
using resource management functioning uniquely designed in the proposed system.
The abstraction of resources from edge to cloud environment can be implemented
using virtualization technology. But here the important factor is; how to provide the
coherent view of all the abstracted resources. In case if we consider virtual machines
to manage the data centers in the cloud layer, then it is also necessary to consider
virtual machines for managing the abstracted resources from edge devices. Or else
Figure 1. Architecture of Proposed “EdgeCloud” Model
115
EdgeCloud
we must use different abstraction and management strategies for different layer
such that they are adaptive to provide essential services at the request time such as
the container and virtual machines. Any system that guarantees resource continuity
must meet the following requirements:
• Firstly it must possess the coherent view of all the set of resources present in
the system so that optimization of service request can be possible.
• Secondly there should be a “Control and Management” plane specially
designed only for managing resources such that it independent of physical
location and ownership.
• Thirdly there should be a proper and efficient strategy for Selection of
resources, classifying them and allocating them in such a way that they
deliver essential services on demand.
• Fourthly the resource selection, classification and allocation strategies should
be decided mainly by considering the functional capabilities of resources
listed in Table 1.
• Lastly the set of resources used in the system must be efficiently in a manner
so that the performance of the system remains to be high.
Application of “EdgeCloud” Model on Smart-City
To study the proposed “EdgeCloud” Model in detail lets apply it on one of the
well-know examples of today i.e. Smart-City. Figure 2, shows the “EdgeCloud”
Model mapped Smart-City architecture which consist of three main layers i.e. two
Edge layers (Layer A and Layer B) and one Cloud layer (Layer C) with all the
edge devices, computing resources, storage resources, processing resources and
networking resources to be combined. If we consider the two Edge layers Layer
A and B, in the lowest layer simple edge devices such as sensors, actuators and
wearables are implemented and the upper layer consists of cars, smart phones and
PCs. Let us assume that from now on the devices are grouped into Edge domain
referred to as “Edges” based on the some policy laid down based on factors such as
capability, proximity, real-time connection and business benefits. Next the Cloud
layer, i.e. Layer C is implemented as three different clouds i.e. public, private and
hybrid clouds based on the requirements. In Layer A there are four different Edges
(Edge domains) i.e. Edge A1, Edge A2, Edge A3 and Edge A4. Also in Layer B
there are two different Edges such as Edge B1 and Edge B2. The Figure 2 consists
of other two important coordinated management artefacts such as Zone and Base
elements. A Zone is an area which consists of an Edge orgroup of Edges present in
different layers to facilitate vertical coordination of resource among the different
layers. Configuration of different zones and their control mechanisms is out of scope
116
EdgeCloud
of to research work and the details such as definition is open to the research and
can be found in related research articles. Base elements are the control elements
required for facilitating the real-time functioning of the system. Two Base elements
are proposed in the system here, they are Access Points (AP) and Zone Supervision
Units (ZSU). Access Point is a control element which helps in delivering different
services that are required for specific Edge. It also consist of a Control Block (CB)
to facilities the network functioning over real-time efficiently. In Figure 2, each of
the Edge layer consist of two different Zones, Zone 1 and Zone 2. In particular the
Zone 1 of “Edge layer B” consists of a Access Point embedded with Metro Station
of Edge A1 which in-turn is implemented with a vehicular network and edge premise
in building (i.e. a building with smart camera, face recognition systems etc.) . A
Zone Supervision Unit (ZSU) is a control element which helps in management and
supervision of different zones designed in the system based on factors such as location
identification, security and privacy etc. There are two ZSUs implemented in the
system i.e. Zone Supervision Unit-A and Zone Supervision Unit-B. The functioning of
ZSU in the proposed system is; firstly the ZSUs are connected with different Access
Points of the nearest zones such that the vertical coordination among the resources
allocated in different layers happen efficiently based on their capabilities. Secondly
the ZSUs are connected as “Inter-Zone Control Communication” by enlarging the
distinct resources available from different ZSUs such that to facilitate horizontal
coordination efficiently in the system. Finally the ZSUs are connected even with
the Cloud Layer consisting of cloud resources through “Control Communication”
channel thus enabling a concept of deploying a distributed system for resource
continuity in edge to cloud environment.
Management and Coordination Module

System for Resource Continuity
In an earlier section, we studied a concept of application of proposed “EdgeCloud”

model on Smart City. The architecture diagram was brief and clearly paved a way
for deploying a distributed framework for resource continuity in edge to cloud
environment. But the main idea behind the research work is how exactly we manage
resource continuity in edge to cloud environment. Thus in this section new set of
modules are designed for proper management functionalities. Figure 3 illustrates the
coordination and management of resource continuity in edge to cloud environment
based on three major functional blocks. The three modules are Instantiation,
Processing and Brokering.
The major part of the module system includes the Processing Component. It
consists of three major blocks namely User/Context Side Block (USB), Service Side
Block (SSB) and Resource Side Block (RSB). Each block differs in their own way
117
EdgeCloud
Figure 2. Application of EdgeCloud model on Smart-City
118
EdgeCloud
Figure 3. Distributed management architecture for Resource Continuity
119
EdgeCloud
based on their functionalities. The third component is brokering which is used at

the times of multi-ownership infrastructure. If the Edge or the Cloud layers in the
system belong to different parties then brokering helps in joint deployment. The
three major functional blocks in the proposed module system are explained in detail.
Functionalities of User Side Block(USB)
From Figure 3, we notice that the objective of USB is to facilitate user interaction
in the system and to contextualize how the user service requests are made and how
exactly those service requests are handled and executed in the system. In USB first
and the most important step for resource continuity in edge to cloud environment
follows “Profiling” which means, a user must first register or provide required
contextual information so as to deliver the necessary services on-time based on
the demands. Second step is any user who registers can benefit the “Sharing” of
resource because of the presence of Edge/Fog service in the system. Third step is any
user who stores and shares the information must follow Security and Privacy rules
such Authentication, Authorization and Encryption. That is any user who wishes
to share his resources for example, details of his car and vehicles at his house may
strictly follow rules and regulations of privacy since other users may not be loyal
and can’t be trusted. Also, the fourth step says a while deploying the corresponding
functionalities in the real world they all the functioning must be under business
policies i.e. Service Level Agreements (SLA).
Functionalities of Service Side Block(SSB)
The main aim of designing SSB is to handle the user requests effectively in the
system. As soon as the user makes any request it goes to SSB and the service made
by the user is categorized based on dynamic service taxonomy, where-in a service
can be neither pre-configured nor created instantly. User request on the other hand
can be decomposed into atomic service i.e. subservices for handling them easily. The
main goal behind this decomposition is atomic services may generally require less
computing resources, and facilitates parallel execution. This can reduce the processing
time so that in-turn can increase the performance of the system. The Repository
is a database which generally stores the of set of atomic services which are pre-
configured based on predictions from previous user requests and also adds the new
services made by the user. One of the challenges that arise here is how to maintain
the performance of the system (i.e. taking into consideration the computing resources
and service execution time) while allocating a place for subservice in repository
during decomposition. Also a very interesting issue arise here while defining the
extent of service that can be associated with the Access Points in the system such as
120
EdgeCloud
subservices and their relationships, defining graph rules and dependencies, decision
making for giving preference to search the available resources from Layer-Zone or
Zone-Layer, find out the missing subservices in the Zone or Edge Layers and planning
strategies at time where the user may not to be satisfied from the services provided
by the SSB etc. Also to handle security and privacy issues within the system such
as anomaly detection and privacy management.
Functionalities of Resource Side Block(RSB)
The objective of RSB is to facilitate coordination and resource management

functionalities in the system. The first and the fore most steps are Monitoring and
Discovery. These steps help to acquire the domain specific knowledge of resources
for each device and also the accurate representation of each device based on factors
such as tablet or mobile or PC, virtual device or Physical device, shared resource or
exclusive etc. After performing the above steps, the information or knowledge gathered
is categorized and classified (i.e. same strategy as used in SSB for categorization)
stored in an interactive database called the Distributed Repository. The distributed
repository consists of two main components. First is pre-layer for storage and
computing of resource. Second one is network connectivity feature which enables
topology information gathering. Each service is executed based on Provisioning
which include whether to decide the service is free and given by the system or it is
on-demand service by the users. After the decision is made the resources are selected
based on some runtime policies Routing techniques Resources are assigned to the
service to be executed, depending on the service demands, runtime model or policies
and are finally allocated by initiating any routing. Usually representation of resources
to each layer is a hideous task due to the hidden complexities and relationships
within the layer, the heterogeneity of the edge devices and also at extreme cases the
business policies when we have multi-ownership infrastructure. Hence in the proposed
framework resource continuity is envisioned as open source where the services and
the users benefit from the open source resources for performance optimization.
PERFORMANCE STUDY
In this section we study about the potential benefits obtained when the proposed
resource continuity management architecture is deployed on Smart City application.
Here we highlight the performance issues concerning to the traffic management
service example discussed. The performance study or analysis made here is purely
121
EdgeCloud
envision and doesn’t guarantee any profits or benchmarking. They state the potential
performance benefits that can be obtained when proposed resource continuity
concept (i.e. EdgeCloud model) is deployed using layered approach and in fact to
encourage the further research in this area. With the above objective performance
study starts with description of the Traffic Management Service which is deployed
on smart city application using proposed EdgeCloud model. Different Edge layers
and its functionalities are clearly mentioned including the topologies which are
configured for three different cities in real-world i.e. Bangalore, Mumbai and Pune.
The preliminary results are shown on proposed traffic management database.
The illustration of the Traffic Management Service is shown in Figure 4. The
figure consists of three different Edge Layers i.e. Edge Layer A, Edge Layer B and
Edge Layer C. Each layer collectively function for providing traffic monitoring
service and differ based on the resource capacity i.e. from Layer A to C the resource
capabilities keep on increasing. Edge Layer A consist of three different sub layers
i.e. Edge Layer A1, A2 and A3 where A1 and A2 function for traffic monitoring and
A3 functions for Car monitoring in Zone. Edge Layer B possesses more computing
capabilities such as storage, processing and networking than Layer A and hence
it is deployed to serve as Bus stop monitoring system. The Zone Control Unit is
assumed to be deployed in Edge Layer B. The Edge Layer C has highest resource
computing capability and hence it is deployed for resource sharing in parking lot.
All the layers are assumed to be present in one single zone and the resources are
published to the Brokering module present in the system which takes care of global
policies of multi-ownership.
J.J. Fernndez Lozano et al. (2015) discussed a unique approach for collecting
the data regarding the vehicles travelling in certain routes based on its source
and destination city using a data matrix. This is based on the sensor technology
where sensors are present at distinguished locations and the locality information is
collectively processed and reported eventually to the centralized data centre. Thus
to manage and coordinate traffic management service here a well-defined approach
discussed above is incorporated. But the difference is in proposed approach single
specific city zone is used instead of multiple cities. Thus here the approach is much
simpler and need on make decision regarding geographical shifts of distinct data
centers. Different Edge layers are used to collect the required vehicular information
and are forwarding to single data centre located in the same city zone. A traffic
management sub-matrix is defined for each city i.e. Bangalore, Mumbai and Pune
which consist of required information for local decision making in order to facilitate
the traffic management service smoothly. The potential benefits such as latency,
bandwidth reduction and execution time are however present in the system since
122
EdgeCloud
edge computing for traffic management service here. In addiction the impact of
analysis of database size, traffic load and lookup time for updating proposed traffic
management database is also shown here. Since the information gathered here is
using the real-time sensor technology the database size and lookup time will be
comparatively higher. The database results are discussed for two different scenarios
such as cloud management and distributed layered approach for three real-world
cities based on the following assumptions.
• Edge A1, Edge A2 and Edge A3 are defined as Traffic Sensor Nodes (TSN)
which detect the Car IDs or numbers with MAC address.
• TSN in Edge Layer A are separated by 100m linear distance in the city streets.
The baseline city will have 10,000m street area in the city.
• A car is assumed to occupy 4m alone and distance between two cars can be
1m and distance between a car and lane is an average of 1 meter.
• The data collected by the sensors of same city zone will be used for providing
the traffic management service of that particular zone.
Bangalore city has 1,319,125 street meters and here let us assume that the TNS
are 100m away from each other. Thus there are approximately 13k TNS that register
eventually to the data centre about the number of cars and its MAC address. Again
let us assume that there are two lanes in each street separated by 100m under each
Edge layer of zone which may have 20 cars separated by 4 meters each and 1m
distance between a car and a lane. This turns out as single TNS that stores 40 car ids
(MAC address of 48 bits). Thus in cloud approach the centralized database stores
40*48*13k = 3Mbps of car data and updates it for every few seconds. Whereas in
the distributed layered approach only the city specific information or data is stored.
Here the size of each area within the city is considered for defining the trend of the
city. Here it is assumed that each city area is considered as a layout.
In Bangalore city there are hundreds of areas but for an instance around 10
areas are considered with each of 1.3k TNS (i.e. all present in Edge Layer A). Thus
information stored in database is 40*48*1.3k= 300Kbps. But on the other hand
if we consider large set of 130 areas with each of 100 TNS then the information
stored in database is around 40*48*100=23Kbps (i.e. updated every few seconds)
as shown in Table 2. The time complexity here is assumed to be O (log(S)), where
S is the size of the database. Table 2 clearly differentiates the cloud and distributed
approach for traffic management service considering other cities such as Mumbai
and Pune as well.
123
EdgeCloud
Figure 4. Traffic Management with resource continuity concept
124
EdgeCloud
Table 2. Proposed database approach for updating network traffic
City Database Information
Bangalore City Edge Computing and Management
Mean length between

Total no. of Traffic Size of the table in Average Lookup
Traffic sensor No. of City Zones
sensor Nodes in Zone each zone(KB) Time
Nodes(m)
100 1300 10 (layouts) 300 1,21
100 100 130 23 1
Bangalore City Cloud Computing and Management
Mean length between

Traffic sensor
sensor Nodes in Zone all city zones(MB) Time
Nodes(m
100 13,000 3 1,40
Mumbai City Edge Computing and Management
Mean length between Size of the table in

Total no. of Traffic Lookup
Traffic sensor No. of City Zones each
sensor Nodes in Zone Time
Nodes(m) Zone(KB)
100 708 6 (layouts) 165 1,17
100 100 43 23 1
Mumbai City Cloud Computing and Management
Mean length between

Total no. of Traffic Size of the table in all Average Lookup
Traffic sensor
sensor Nodes in Zone city zones(MB) Time
Nodes(m
100 4,250 1 1,31
Pune City Edge Computing and Management
Mean length between

Traffic sensor No. of City Zones
sensor Nodes in Zone each zone(KB) Time
Nodes(m
100 3,680 15 (districts) 860 1,29
100 100 550 23 1
Pune City Cloud Computing and Management
Mean length between

Traffic sensor
sensor Nodes in Zone all city zones(MB) Time
Nodes(m
100 55,000 12 1,52
Database Updating Traffic
Management 1 sec 5 sec 10 sec

City
Approach (Mbps) (Mbps) (Mbps)
Bangalore 3 0,6 0,3
Cloud Mumbai 1 0,2 0,1
Pune 12 2,4 1,2
Layered Any city 23 4,6 2,3
125
EdgeCloud
The results from Table 2 clearly show the potential benefits of proposed distributed
layered approach concerned to the size and lookup time of the database. We can
observe the trade-off between the number of city areas and the services that can be
provided in each city. This gives us a clear picture that the lookup time and size of
the database increases with the increase in the area of the city. This in turn increases
the services that have to be provided as well. The database also defined the amount
of traffic load that was handled by the proposed database system and its impact
creates a motivational factor for future research on extension of database. Database
characteristics of Pune city gives us a motive that for every 5 seconds around GBs
of data is sent over the network thus provisioning the impact of other factors such
as network dimensioning, energy consumed and optimization of infrastructure only
on the updation of proposed database.
CONCLUSION
Edge Computing can complement Cloud Computing in a way leading to a novel

architecture which can benefit from both edge and cloud resources. This kind of
resource architecture may require resource continuity provided that the selection of
resources for executing a service in cloud is independent of physical location. Hence
this research work proposed a new resource continuity model called “EdgeCloud”
which consist of three different Edge layers operating in a unique way paving a
way for reducing network traffic at greater extent. Initially the proposed EdgeCloud
model is applied on a smart city scenario by implementing two important control
elements called AP and ZSU. Later detailed distributed management architecture
with three functional blocks was deployed to facilitate resource continuity in edge
to cloud environment. The architecture clearly described the security and privacy
issues, selection of strategy for resource computations and also how each edge layers
communicate to allow resource continuity to provide better services to user. Finally
the proposed approach was evaluated for performance using traffic management
service example by considering three real-world cities Bangalore, Mumbai and Pune.
The network traffic was calculated for each city using cloud and layered architecture
approaches and was tabulated using proposed database structure. Clearly it indicated
that proposed EdgeCloud model reduces the network traffic to greater extent than
the cloud computing approach alone. Hence it draws to a conclusion that proposed
model is well suitable for implementation on any smart city scenario which include
resource pooling. Future research objectives many include simulation of proposed
approach or to implement the proposed resource continuity architecture using
visualization technology.
126
EdgeCloud
REFERENCES
Aazam, M., & Huh, E. N. (2016). Fog computing: The cloud-iot\/ioe middleware
Baktir, A. C., Ozgovde, A., & Ersoy, C. (2017). How Can Edge Computing Benefit
from Software-Defined Networking: A Survey, Use Cases & Future Directions.
IEEE Communications Surveys and Tutorials, 19(4), 2359–2391. doi:10.1109/
COMST.2017.2717482
Dsouza, C., Ahn, G. J., & Taguinod, M. (2014, August). Policy-driven security
management for fog computing: Preliminary framework and a case study. In
Information Reuse and Integration (IRI), 2014 IEEE 15th International Conference
on (pp. 16-23). IEEE.
Dubey, H., Yang, J., Constant, N., Amiri, A. M., Yang, Q., & Makodiya, K. (2015,
October). Fog data: Enhancing telehealth big data through fog computing. In
Proceedings of the ASE BigData & SocialInformatics 2015 (p. 14). ACM.
European Telecommunications Standards Institute Industry Specifications Group,
Mobile-Edge Computing – Service Scenarios. (2017). Retrieved from http://
www.etsi.org/deliver/etsi_gs/MEC-IEG/001_099/004/01.01.01_60/gs_MEC-
IEG004v010101p.pdf
Evans, D. (2011). The internet of things: How the next evolution of the internet is
changing everything. CISCO. Int. J. Internet, 3(2), 123–132.
Fernández-Lozano, J. J., Martín-Guzmán, M., Martín-Ávila, J., & García-Cerezo,
A. (2015). A wireless sensor network for urban traffic characterization and trend
monitoring. Sensors (Basel), 15(10), 26143–26169. doi:10.3390151026143
PMID:26501278
Hong, K., Lillethun, D., Ramachandran, U., Ottenwälder, B., & Koldehofe, B.
(2013, August). Mobile fog: A programming model for large-scale applications on
the internet of things. In Proceedings of the second ACM SIGCOMM workshop on
Mobile cloud computing (pp. 15-20). ACM. 10.1145/2491266.2491270
Munir, A., Kansakar, P., & Khan, S. U. (2017). IFCIoT: Integrated Fog Cloud IoT:
A novel architectural paradigm for the future Internet of Things. IEEE Consumer
Electronics Magazine, 6(3), 74–82. doi:10.1109/MCE.2017.2684981
127
EdgeCloud
Nam, T., & Pardo, T. A. (2011, September). Smart city as urban innovation: Focusing
on management, policy, and context. In Proceedings of the 5th international
conference on theory and practice of electronic governance (pp. 185-194). ACM.
10.1145/2072069.2072100
Quwaider, M., & Jararweh, Y. (2013, December). Cloudlet-based for big data
collection in body area networks. In Internet Technology and Secured Transactions
(ICITST), 2013 8th International Conference for (pp. 137-141). IEEE.
Satyanarayanan, M., Schuster, R., Ebling, M., Fettweis, G., Flinck, H., Joshi, K.,
& Sabnani, K. (2015). An open ecosystem for mobile-cloud convergence. IEEE
Communications Magazine, 53(3), 63–70. doi:10.1109/MCOM.2015.7060484
Stojmenovic, I. (2014, November). Fog computing: A cloud to the ground support
for smart things and machine-to-machine networks. In Telecommunication Networks
and Applications Conference (ATNAC), 2014 Australasian (pp. 117-122). IEEE.
10.1109/ATNAC.2014.7020884
Tang, B., Chen, Z., Hefferman, G., Wei, T., He, H., & Yang, Q. (2015, October).
A hierarchical distributed fog computing architecture for big data analysis in smart
cities. In Proceedings of the ASE BigData & SocialInformatics 2015 (p. 28). ACM.
Vaquero, L. M., & Rodero-Merino, L. (2014). Finding your way in the fog: Towards
a comprehensive definition of fog computing. Computer Communication Review,
44(5), 27–32. doi:10.1145/2677046.2677052
Yaseen, Q., AlBalas, F., Jararweh, Y., & Al-Ayyoub, M. (2016, September). A fog
computing based system for selective forwarding detection in mobile wireless sensor
networks. In Foundations and Applications of Self* Systems, IEEE International
Workshops on (pp. 256-262). IEEE. doi:10.1109/FAS-W.2016.60
Zao, J. K., Gan, T. T., You, C. K., Méndez, S. J. R., Chung, C. E., Te Wang, Y.,
. . . Jung, T. P. (2014, June). Augmented brain computer interaction based on fog
computing and linked data. In Intelligent Environments (IE), 2014 International
Conference on (pp. 374-377). IEEE. 10.1109/IE.2014.54
128
129
Chapter 7
Security and Privacy
Issues in IoT:
A Platform for Fog Computing
S. R. Mani Sekhar
Sharmitha S. Bysani
Vasireddy Prabha Kiranmai

ABSTRACT
Security and privacy issues are the challenging areas in the field of internet of things
(IoT) and fog computing. IoT and fog has become an involving technology allowing
major changes in the field of information systems and communication systems. This
chapter provides the introduction of IoT and fog technology with a brief explanation
of how fog is overcoming the challenges of cloud computing. Thereafter, the authors
discuss the different security and privacy issues and its related solutions. Furthermore,
they present six different case studies which will help the reader to understand the
platform of IoT in fog.
INTRODUCTION
Internet of Things (IoT) is on the cutting edge of Technology, connecting humans,

devices and systems intelligently. It is a network of connected physical objects
that are reachable through the Internet. The ‘Thing’ in the IoT could range from a
“Smart Home” to a “Connected Inhaler”, that is assigned a unique address and has
DOI: 10.4018/978-1-5225-6070-8.ch007
Security and Privacy Issues in IoT
the potential to fetch and handover the data collected, over a network without any
assistance, making our lives more easier. IoT enables a device to represent itself
digitally over the internet and gets connected to its surrounding devices triggering
interaction among various such connected devices.
The word “Internet of Things (IoT)” was coined by Kevin Ashton in 1999 and
is defined as the network of physical nodes with other items embedded with it like
software, sensors, actuators, and network connectivity which enable these objects
to collect and exchange data. IoT integrates the real world with the computer-based
systems precisely and efficiently, reducing the human intervention.
Though the term “Internet of Things” is two decades old, it is in limelight since
2010 as there has been an exponential growth in the number of devices connected to
the internet. Starting from a ‘Connected Home’ having remotes, smart refrigerator,
security keypads to the present-day mobile-controlled devices, smart-temperature
learning device (thermostats), connected fitness tracker we have seen our lives
evolving around the internet. Medical field is tremendously improving by the
application of IoT as it aims to empower people to lead a healthier life by wearing
connected devices. The impact of IoT is challengingly growing in the Agriculture,
Poultry and Animal Husbandry fields.
There is an exciting future in the field of IoT. Growth of Artificial Intelligence
and Machine Learning will result in a new range of connecting devices in the coming
decade with improvised technologies. Concepts like “Monitoring and Reporting”
will ensure clean and safe surroundings with smart traffic systems resulting in
lessening the accident numbers, ensuring security. Applications of IoT to Plants
and Animals sector will lead to a smarter way of growing, processing and storing
food. Smart Home would automatically manage our most standard house activities
and its energy consumption.
The evolution of IoT has started with remote computing infrastructures like
data centers and moved on to the recent Cloud Computing environment to meet
the demands of enormous data that the devices generate by making use of virtual
storage applications in turn it will take off to “fog computing” in the coming days.
Fog Computing features a cloud on the edge of the device where data is generated
resulting in reducing the access time to the cloud.
Boon of IoT would be Machine to Machine (M2M) Communication leading to an
automation of daily tasks which in turn provide a controlled, efficient, timely, and
a better quality of life. Setback of IoT would be technology taking over our lives,
lesser employment of menial staff, risk of losing privacy, any failure or bug in the
software will result in serious consequences.
130
HOW IS IoT LINKED WITH FOG?
The Internet of things is developing gradually and is leaving a mark in almost all
domains. High-speed data processing, analytics and with less response times has
become mandatory in all applications and software’s and it is turning out to be a
tough task to meet these requirements in the current centralized cloud computing
technique used by the present IoT systems. Organizations storing large amounts
of data on the cloud find it difficult to cope with the swift changes in the cloud
technologies, making them dependent on the service providers. Security, Privacy
and Complexity in building their private cloud still seems challenging for the Cloud
Computing Sector (Lebied, 2017).
To overcome the challenges faced by the Cloud Computing Technology,
researchers have come up with an idea of bringing Cloud service features closer to
physical IoT devices. This computing infrastructure where the cloud is placed in
vicinity of data being generated featuring as “mini-cloud” is referred to as Fogging
or Edge Computing or Fog Computing, often associated with Cisco. The metaphor
“fog” is coined since it refers to the cloud near the ground. Substantially, the fog
layer acts as a junction between the produced data and cloud capabilities. It aims
to increase the efficiency by reducing the movement of data to and fro the Cloud.
Here, the processing and storage of computing resources and its applicant services
are closer to the edge of the network. Any device with computing, storage and
network connectivity is referred to as a fog node which can be installed anywhere
in a network connection. Fog nodes are the building blocks of the Fog Computing
Infrastructure allowing short-term analytics at the network edge.
IoT Applications which require mobility support such as geo-distribution and
location-awareness can be implemented by Fog Computing. Besides these, real
time gaming augmented reality and real time streaming which require high speed
processing and low latency can use Fog over Cloud. Ultimately, Fog computing can
be used in Big Data sector as it supports edge analytics and stream mining, helps in
reducing data volume at a very early stage, cutting down delay and saving bandwidth.
Fog computing brings forward some evidential benefits (Bonomi, Milito, Zhu
& Addepalli, 2012) such as:
• Low Latency: The proximity to end users makes possible the support of real-
time services (e.g. gaming, video streaming).
• Geographical Distribution of Network: The Fog provides distributed
computing and storage resources to, for example, large and widely distributed
sensor networks.
131
• Mobility: It delivers rich services analogous to location for moving users and
location constrained devices.
• Flexibility and Heterogeneity: This technology allows the interaction of
different physical environments and infrastructures among multiple services.
• Scalability: The closeness to the edge of the network enables scaling the
amount of connected devices and services.
• Delay Jitter: High in Cloud Computing, very low in Fog Computing.
Below section explains the different applications related to IoT and Fog
Computing:
• (Yi, Hao, Qin & Li,2015) has proposed a Smart Home concept by connecting
different devices and sensors which are capable of computing and storing
data. Fog computing can be used to integrate all the individual components
into a single system with elastic resources.
• Health-Care Information (Yi, Hao, Qin & Li, 2015) has to be handled
with great security and privacy thereby paving way for the Fog Computing
technology to be used in the Medical Sector. It can be used to store a patient’s
information in separate fogs and can be updated on regular intervals directly
into the patient-owned fog
• A smart traffic light (Bonomi, Milito, Zhu & Addepalli,2012) system
illustrates the role of fog computing in IoT in which a smart traffic light
node interacts locally with a number of sensors, which detects the presence
of pedestrians and bikers and also measures the distance and speed of the
approaching vehicles. Based on this information the smart light notifies the
approaching vehicles or may modify its own cycle to prevent accidents.
SECURITY IN IoT
By the end of 2020, around 38.5 billion devices are expected to be connected to the
Internet which has already crossed 13.4 billion in 2015 (Raval, 2016). The count
of the connected devices makes it obvious that the amount of data which will be
flown across the network will be massive and this would completely challenge the
capture, storage, and analysis of data, something that will transform the types of
database technologies we have.
132
DATA SECURITY ISSUE
Due to the predominance of Cyber Security malfunctions, people are finding it

difficult to secure their personal gadgets like Mobiles and laptops from security
threats, whereas in IoT where there are billions of devices, securing the devices
will definitely be a tedious task. Ensuring cyber security of all these devices does
not seem to be an easy task. The volume of data is also so huge that tracking and
identifying malicious traffic over the network is quite challenging. With the rise of
IoT, phishing attacks are definitely going to increase. The systems, appliances and
the embedded systems which are a part of the IoT network have the power to control
the utility grids and communication systems of the world. An interruption in the
functionality of these devices by a cyber-attack may have dreadful consequences.
The Internet of Things may range from a thousand to millions of identical devices,
so if a hacker is able to sneak an attack into one of these devices, then a similar
attack can be reproduced across all devices.
THREATS TO IoT AND FOG COMPUTING
In these sections we discuss the different threats related to smart meter, smart watch
and device kidnapping.
THREAT TO ELECTRICITY SMART METER
The users of the connected devices which are part of the Internet of Things network
must try to understand what kind of data their connected devices collect and how
much of that data is being shared with other devices and what sort of modes are
used for the transmission and retrieval of the data, where the data is being stored
and how strongly this data has been secured. These connected devices must run
the latest version of any software so that the devices remain bug-free else it may
fall prey to unauthorized access. For example, if there is a smart meter which sends
energy-usage data to the utility services companies for billing, it must be ensured
that this data is safely secured. Else if on hacking this information reaches the hands
of thieves, they can burgle houses based on their power usage information, that is, if
the power usage of a house is less, it would mean that there is no one in the house.
Hacking of a smart-meter of a house would peril on various personal information like
our email-id, phone number, bank account details or Government-Id details as they
are linked with smart meters. Thus if we have good security mechanisms to protect
all our smart devices which are connected, then we will be free from such threats.
133
THREAT DUE TO DEVICE KIDNAPPING
Another threat to the security of data in the field of IoT would be Device Kidnapping
(Suthar, 2016). Some device owners claim that their device was hacked and the
hackers were demanding certain amount as ransom, to get it unlocked. Suppose a
home security system or a webcam or any other optical device gets hacked, a hacker
can gets access to the surveillance system. It would become easier for someone to
keep an eye on that person, in their house, without them even having the slightest
of idea about it.
Due to the possibility of information getting stolen from the smart devices,
consumers are very concerned about the security of their data and thus they would
hesitate to purchase IoT enabled smart connected devices. Researchers (Meola,
2016) have been able to hack into real, on-the-market devices with enough time and
energy, which means hackers would likely be able to replicate their efforts.
THREAT TO SMARTWATCH AND FITNESS TRACKER
Wearable devices like smart watch and fitness tracker carry a lot of information on
the health data of a person. Data sent between smart watch and an android phone can
be intercepted by hackers as data exchange between these devices through Bluetooth
communication which relies on a six digit pin which can be easily cracked using a
brute force technique in turn providing easy access to the health data of the user and
manipulate it thereby leading to fatal consequences on patients’ health. It is very
crucial to ensure that motion sensors embedded in such devices do not disclose any
such information to the neighboring fog nodes.
SOLUTIONS TO SECURITY ISSUES IN IOT AND FOG
Below section provide the overview of present solution for Malware protection,
authentication, data encryption and access control etc. in the field of IoT and Fog
computing
MALWARE PROTECTION
The fog nodes in the Fog Computing Platform which are affected by malware like
virus Trojans, ransom are worms, spyware rootkits etc. can reduce the performance
134
of the fog network and also damage the data permanently. The above stated can
be resolved by installing anti-malware software programs in the fog nodes with a
deployment of certain intrusion detection systems at vulnerable points in the fog
network and stringent regular data backups can be performed.
Instead of carrying out computing and device control in the remote data centers
as in Cloud Computing, with fog it can be carried out at the end-user devices thereby
making it difficult for threats and attacks to pass through all the fog nodes as these
nodes can quickly identify strange activity and tranquilize them before the malware
enters the core system. When it comes to data storage, the data collected from the
IoT devices are managed by the distributed fog nodes, then these data will be more
easily available and can well protected in comparison to the data stored in the user
remote data centers.
AUTHENTICATION
Authentication is considered as an important issue in fog computing as these services

are made available to a huge mass of end-users by the front fog nodes. When a device
is connected to a network, authentication of the device is required prior to transmitting
and receiving data. Balfanz has put forward a cheap and user-friendly solution using
biometric authentication through fingerprint or face which can be used to simplify
the authentication procedure in a fog network thus providing security of user data.
Outsourcing of data to a fog node leads to difficulty in ensuring data integrity as this
data may be lost or modified and in turn can be manipulated by some unauthorized
alliances. In order to provide integrity, confidentiality and verifiability for a cloud
storage system, the client can verify its data stored on untrusted services, auditable
data storage services, by providing the different approaches such as combination
of homomorphic encryption and searchable encryption.
DATA ENCRYPTION
Data encryption is an extensively used approach to secure data confidentiality.

Encryption may lead to increased resource allocation issues, so only subtle and crucial
information like user’s identity in vehicular networks, patient data in healthcare
systems, cached data should be encrypted. Efficient data integrity scrutinization
must be performed to ratify the received and sent information both before and after
communication.
135
Due to the lack of available dedicated and contiguous memory in the fog platform,
there is an absence of malware protection in the fog devices. One solution for this
issue would be using a physical malware device such that it does not use any fog
resources. Tools like Bare Cloud (Kirat, Vigna, & Kruegel, 2014) can be used to
automatically recognize dangerous malware and zero day threats can be identified
using some machine learning algorithms.
ACCESS CONTROL
In order to restrict the privileges of the device components and applications, access
control measures must be built into the operating system of the fog network devices
so that only the required resources can be accessed. As access to all resources is not
essential for the devices to perform a function, the principle to minimum privilege
would minimize the intervention of security. A good practice would be to keep the
devices in the fog network relatively secluded thereby allowing only few delegated
persons to have access, making these devices tamper-proof or tamper-evident might
be favorable. This type of endpoint hardening can help restrict intruders from reaching
the data. The endpoints can be secured by using (Corser, Fink & Aledhari, 2017)
small simple plastic devices, port locks and camera covers, which would shut out
the USB and Ethernet ports and camera webcam apertures. Port locks also help in
avoiding undesirable malware from entering. When a device gets tampered, some of
the tamper-resistant techniques paralyses the device. This security countermeasure of
endpoint toughening creates a sort of a layered path that would require the attackers
to bypass a variety of difficulties planned to protect the device and its information
from illegal access.
NEED FOR PRIVACY
Digitalization of the physical world through sensors and connectivity using Internet
of Things (IoT) is widely extended to various fields making Environment Monitoring,
HealthCare, Industry Automation, and Precision Agriculture and so on possible.
These applications of IoT tend to produce enormous amount of data and is estimated
by the International Data Corporation (IDC) that the data stored in the “cloud”
would go up by two folds by 2020. Increase in the data produced by the connected
devices makes the data more vulnerable to Hackers and Cyber Criminals, urging
for the need of Security and Privacy for Data.
136
Most of the IoT applications measure variables related to consumption, velocity,

temperature, pressure, heartbeat or control some physical systems, power circuits,
automated lights, valves and brakes. Also, financial, criminal, political, medical
records maybe collected by the IoT devices. Sometimes, this may result in the
gathering of very private and sensitive data related to an individual giving rise to
some Data Privacy issues.
DATA PRIVACY ISSUES
Some of the issues related to Data Privacy in the IoT domain (Meola, 2016) are:
• Tremendous Data: The data generated by devices is massive as per Federal

Trade Commission there are more than ten thousand household devices
which can produce more than 150 million data points every day. This makes
more access points for Hackers and Cyber Criminals.
• Unwanted Public Profile: Some personal information of an individual which
might be very crucial to a company may be handed over to the company
by that individual consciously, but later the company might misuse the data
for acquiring benefits. For instance, an insurance company may collect data
related to one’s driving habits through a connected car.
• Eavesdropping: Manufacturers and Hackers can use a connected device to
invade a person’s activities by collecting private data about the individual.
• Customer Confidence: All these factors create void in the day-to-day usage
of IoT by Customers as the customers are concerned regarding the privacy
of their data.
PRIVACY IN IOT AND FOG
Privacy of Data in IoT devices is accomplished using Authentication, Encryption and

Data Masking (Madakam & Date, 2016) ensuring that the access to data is authorized.
Various devices that are connected to each other to communicate should ensure that
they are reliable and credible. Some security standards are to be set up and followed
by the devices which gather sensitive data. The mobile nodes in IoT which often
move from one cluster to another should be embedded with cryptography-based
protocols that provide quick identification authentication and privacy protection.
137
THREATS TO IOT AND FOG
Looking on to Fog Computing, fog devices face similar Security and Privacy threats
as the conventional Data Centers used in Cloud Computing. Security and Privacy is
to be tended into every layer in constructing the Fog Computing System resulting
in a better Privacy mechanism.Though the Privacy of data and information in Fog
Computing is better when compared to the conventional Cloud Computing and
other technologies as the data is collected, inspected and stored locally in the fog
located over the device, there are still some problems related to the Privacy of Data
in Fog Computing.
MALICIOUS FOG NODE PROBLEM
Malicious Fog Node Problem is a threat to the Data Privacy in Fog Computing in
which when a fog node receives heavy workload from one of the connected devices,
if there exists a malicious user amongst those nodes, then there might be a difficulty
in maintaining the integrity of data resulting in some privacy threats to the data, as
the fog nodes should trust each other before splitting the computational work amongst
themselves. In order to handle this issue, only the fog nodes which are authenticated
by the cloud should be located in that particular fog domain.
DATA ENCRYPTION AND DECRYPTION
When there is a large volume of data to be processed by the IoT device which cannot
be computed by itself then the IoT device sends messages to its nearby fog nodes.
In such cases, the data is split and sent to the nearby fog nodes to process it where
the data is expected to be processed without exposing it to the other nodes. Later,
on completing the process, integration of the processed data from each associated
node should take place ensuring the integrity of data. This brings in the need for the
encryption and decryption algorithms for the data on IoT device. This gives rise to
some concerns as the fog nodes have limited resources and a lightweight encryption
algorithms and masking techniques have to be used to hide the original data.
MAN-IN-THE-MIDDLE ATTACK
Man-in-the-Middle attack is another Privacy and Security threat faced by Fog

Computing. Here, some of the fog nodes in the Fog Computing Environment are
138
replaced by some fake fog nodes which cannot be identified by the traditional
Detection Mechanism. This brings the need for the Encryption algorithms to protect
the communication between the IoT devices and the Fog node, but these methods
take away a large amount of space on the IoT enabled devices.
VARIOUS TYPES OF PRIVACY ISSUES

RELATED TO FOG COMPUTING
Here we discuss the different privacy issues which are related to a Fog computing
such as Usage, and location privacy.
USAGE PRIVACY
Another privacy issue which comes into picture regarding the utilization of the Fog
Computing Services by the IoT devices is referred to as Usage Privacy (Yi, Qin &
Li, 2015). The IoT devices store all the sensitive information regarding the activities
and statistics of its user in the fog nodes violating the Data Privacy. One proposed
solution to this problem would be that the IoT device stores the actual task along
with some created dummy tasks into various fog nodes, resulting in the masking
of the actual task. However, this will increase the usage of resources and storage
which may result in some wastage but it helps in securing the Data.
LOCATION PRIVACY
The IoT devices (fog client) generally store the data in their nearby available fog
nodes leading to some location privacy issues (Yi,Qin & Li, 2015) of the Data.
Sometimes the fog client uses multiple fog nodes at multiple locations to store its
data, it might reveal its path trajectory to the other fog nodes assuming that the fog
node would be intrigue, causing some Location Privacy Issues to the IoT device. One
solution to this issue would be through identity obfuscation of the IoT device(fog
client) such that even if the fog node knows its fog client it wouldn’t be able to
detect and identify it, but it might know the approximate location of the fog client
but not its precise location. Identity Obfuscation can be done by trusted third party
applications generating some fake IDs for the fog client(IoT device).
139
SOLUTIONS TO PRIVACY ISSUES IN IOT AND FOG
Various solutions that have been proposed to secure the Privacy of Data are Verifiable
Computing and Data Searching and Access Control.
VERIFIABLE COMPUTING
According to Verifiable Computing (Yi, Qin & Li, 2015), an IoT device can shift
its process of computation to some insecure server or to another fog node, where
the incoming results are verified. The server computes the given task and returns
the result to the node along with a proof. To infuse confidence into the computation
offloaded onto the fog node, the fog user can verify the correction of the results.
(Gennaro, Gentry & Parno, 2010) have proposed a Verifiable Computing Protocol
that allows the server to return a computationally impressive and a unreciprocated
proof to the client(fog node) that can be verified by it. The protocol makes sure that
the server does not keep any information regarding input and output.
DATA SEARCHING
Data Searching (Yi, Qin & Li, 2015) using keywords is another technique to secure
Data Privacy. Generally, sensitive data from the IoT devices is encrypted and then
stored in the fog node. This makes Data Usage Services quite challenging. One of the
important usage services is the Keyword search which is used to search a particular
keyword over the encrypted data. Researchers have come up with various searchable
Encryption schemes that can be used by the user to search over encrypted data
without decrypting it. Using the Data Searching technique, the data can be secured
as there is ample secrecy for encryption, controlled searching and also there is a
support for the hidden query. As the data is not decrypted at any point of the search
this ensures privacy for the data.
ACCESS CONTROL
The most reliable technique for seclusion of Data is through keeping a control on
who should be accessing the data, resulting in safeguarding and ensuring the privacy
of user. The Access control mechanism in the traditional Cloud Computing is done
140
through the cryptographic methods for its expandable data (Yu, Wang, Ren & Lou,
2010) have come up with the technique of attribute based encryption (ABE) which is
finer than the other techniques like symmetric key management and couple open key
course of action. In case of Fog computing, authors of (D’souza, 2014) have come
up with the policy-based resource access control in order to support interoperability
between resources and assure joint exertion (Fakeeh,2016). However, arranging
the access control on the fog cloud as well as meeting the goals with the limited
resource would be a tedious job.
CASE STUDIES
Below section illustrates the different case studies such as smart grid system,
Surveillance Video Stream, Agriculture and Food Traceability System, Disaster
Management and Web Optimization system.
ELECTRICITY CONSUMPTION MONITORING

SYSTEM- SMART GRID
Introduction
Production of Electricity by utilization of fossil fuels has adverse effects on the

environment causing climate changes, global warming and increase in the carbon
and greenhouse gases in the atmosphere bring in the need of alternative sources
of Energy like Solar and Wind Energy. The Conventional Electricity Distribution
System (Okay & Ozdemir, 2016) which transmits electricity from power stations to
the customers does not integrate these eco-friendly alternative sources of energy to
the its transmission grid and further it does not provide any chance to the customer to
monitor and control their energy consumption. These drawbacks have been overcome
by the Smart Grid which has incorporated a two-way flow of both electricity and
data creating a more advanced and dynamic Energy Distribution System.
Smart Grid employ Smart meters to collect information regarding the electricity
consumption of a user and also to gather sensitive data like energy exhausted by
various devices. This data is often stored in Cloud based Data Centers, making this
private data prone to hackers, service providers and other intruders. Smart Grid must
ensure Security, Scalability and Flexibility while handling and storing the data. As
data managed by smart grid increases, it becomes difficult to handle these issues.
141
ISSUES FACED BY CLOUD COMPUTING IN

SMART GRID IMPLEMENTATION
Fog Computing provides a resolution to these problems as the data storage and
computation could be done locally near the edge of the network securing privacy
of data and processing it with low-latency. In smart grids, Fog Computing is not the
replacement of Cloud Computing and it provides some additional features as follows:
• Scalable Real-Time Services: This allows the users to keep a check on their
energy consumption.
• Private Data: Since Cloud Data Center is a shared environment it is affected
by the privacy issues. Fog Computing secures the private data of the users.
• Proximity: The fog nodes are close to the users allowing direct access to the
source can also simplifies the extraction of key information from the big data.
• Latency: Latency is highly interrelated with Proximity as the more the closer
the fog node to the user the faster is the reaction time and further reduces
congestion issues.
• Geo-Distribution: Power generators, distribution network and transformers
that form Smart Grid are naturally distributed where fog services can be
located at the edges of network making fog computing is advantageous.
PROPOSED SOLUTION
A proposed method for implementing Smart Grid (Okay & Ozdemir, 2016) by the
application of fog computing is a three-tier architecture where the first tier consists
of Smart Meters, second tier consists of the Fog Servers and the third tier composes
the Cloud Data Center, overall facilitating four types of communications - device
to device, device to fog server, fog server to fog server, fog server to cloud server.
First-Tier
The first tier is accountable for communication between the smart grid devices
like electrical vehicles, mobile phones, smart meters and various other electrical
appliances. This communication in first tier between the appliances could be related
to business purposes, billing reasons or location-based services. Data from the
electric appliance is fetched and computed by smart meter and then transferred to
the fog server. Smart Meters measure the consumption of energy by various electric
142
devices so that the consumers can keep a check on their usage. This fetching of
sensitive information lead to certain privacy issues which is taken care by the smart
meter as it classifies the data into public and private data. Private data generally
consists of electricity consumed by each appliance, and this data is encrypted using
an encryption function authenticating them with a key and then transmitted to the
corresponding fog server in an interval of fixed time. Further, a certain amount of
renewable energy (solar, wind) that is being produced in the house is sent to the
smart grid. The sum of the energy consumption by various devices and also the sum
of the energy produced by the house is termed as Public data.
Second-Tier
In second-tier, there is communication between the fog servers and the smart meters
and the fog server connects the smart meters with cloud. The fog servers process
the gathered public and private data coming from the smart meters separately. To
secure the private data, the fog servers collect the public data as unencrypted data
and private data as encrypted data. Fog servers aren’t capable of decrypting the
private data as the key of the encryption function is only shared with the consumer
and cloud. The fog servers aggregate public and private data without violating
consumers privacy and then it sends this data to the cloud, in order to reduce the
amount of data storage in the cloud. Fog servers store the detailed information
regarding the consumption of energy of each appliance temporarily so as to respond
to the customer requests quickly.
Third-Tier
The third tier consists of the cloud data center storing all the information (public
data) coming from various fog nodes and processing it. The data from the cloud can
be easily accessed by the customer by using a mobile application or a web interface.
A smart home can be taken as a use case for the above proposed solution where
the smart meter produces a report on the energy consumed by various appliances
and classifies them as public and private and stores it on the fog servers, further
connecting itself with the cloud via fog server.
In the proposed model the fog server stores private data of the customer which
can be periodically erased from it. Further, only the aggregated consumption and
production values are stored on the cloud thereby increasing the privacy of data. The
model is geometrically distributed providing reliability and extending the capabilities
of cloud in terms of privacy and latency for smart grid. Thereby, Fog Computing
increases the efficiency of the Cloud Computing in Smart Grids.
143
HEALTHCARE MONITORING SYSTEM-

ECG FEATURE EXTRACTION
Introduction
Internet of Things provides an adequate and organized approach to enhance the

health and prosperity of mankind. Fog computing which is an extension to cloud
computing is implemented in the healthcare and elderly care systems (Khan,
Parkinson & Qin,2017) where self-automated wireless sensors send data to the fog
nodes instead of sending them directly to the cloud.
CLOUD AND FOG COMPUTING IN HEALTHCARE SYSTEM
A smart healthcare infrastructure can be created using many sensors. In the fog
layer, semantic tagging and classification of data is performed and this polished data
can be later sent to the cloud systems for further processing and analysis. Human
health can be monitored in real-time by retrieving bio-signals from sensor nodes
and transmitting them to the gateways using wireless communication protocols.
This data is then sent to a cloud server for real time processing, visualization and
diagnosis. Fog computing is used in such health monitoring systems by facilitating
services like embedded data mining, distributed storage and notification storage at
the edge of the network.
ECG FEATURE EXTRACTION USING FOG COMPUTING
Electrocardiogram(ECG) feature extraction plays an important role in the diagnosis of

many cardiac diseases. In smart gateways (i.e, the edge of the network), the ECG signals
are examined and features (Gia, Jiang, Rahmani, Westerlund, Liljeberg & Tenhunen,
2015) like the heart rate, P wave and T wave are extracted. One of the elementary
technologies in healthcare IoT is the Wireless Body Area Network (WBAN),
which is used for retrieving real-time signals such as the Electrocardiogram(ECG),
Electromyography(EMG), body temperature and blood pressure a systematic manner.
In most of the healthcare examination systems, the data collected from the sensors
implanted in the wearable devices is sent to the cloud server for analysis, but during
this process of transmission, the data might get corrupted or lost, thereby leading to
errors in the analyzed data of the patient. Even a minute error in the health report
of a patient may lead to faulty treatment decisions for a health problem. Thus there
144
is a need for reducing the length of path travelled by the data and also guarantee
good quality of service. This can be fulfilled by the fog layer where the fog nodes
can process the data at the edge of the healthcare monitoring systems and also
diminishes the network bandwidth used. For the convenient utilization of the network
bandwidth between the health monitoring system’s gateway and a cloud server, a
flexible arrangement has been put forth where in light-weight algorithms are used
to extract the heart rate, P and T waves which can be cited at the end-user’s browser.
• WBAN Systems: When data needs to be aggregated from a patient’s body,

many devices and sensors need to be attached to their body using many wires
and cables. For diabetes and cardiovascular patients, their health needs to be
monitored for a long duration like for an entire day, in such cases the patient
may require to spend an entire day with many cables and sensors attached to
the body. Due to the body movement of the patient, the data which may be
recorded in the sensors may be incorrect. These incorrect results may lead to
inaccurate report of the patient’s health. A health monitoring system which
works on WBAN can overcome this drawback because of its characteristics
of mobility and wireless transmission. The WBAN systems have three main
components- a sensor node, a gateway and a backend, fulfilling the following
functions.
• Sensor Node: A medical sensor node consists of devices and wearables
which accumulate the health data and then transmit the data to the gateway
by means of communication protocols like Bluetooth, WiFi or ZigBee. The
gateway or the fog node is a medium for the health data to flow from the
sensors to the cloud server. It is considered as the most important part of the
WBAN architecture as if any problem occurs in the gateway the entire system
will fail.
• Gateway: The smart gateway engineering incorporates physical and
organizational structures. The physical structure consists of an embedded
router which endorses communication protocols like Bluetooth, Wifi and
Ethernet but this router does not support communication with sensor nodes
which require low power communication protocols like 6LoWPAN. This
requirement is fulfilled by another component of the physical structure called
sink nodes which incorporates 6LoWPAN protocol into the gateway. On the
other hand, the operational structure consists of the embedded operating
system, a hardware layer and a fog computing service layer.
A flexible template at the smart gateway is used to extract the ECG feature which
consists of three main parts.
145
1. First part is the ECG preprocessing which contains notch filters which is used
for removal of artifact movement.
2. Fast computation techniques and some wavelet algorithms are included in the
second part of the template called the Wavelet Transformation.
3. The third part is the ECG extraction which deals with implementing algorithms
for extracting data such as P-R interval, Q-T interval, S-T segment, QRS area
and QRS energy.
At the smart gateway, Graphic User Interfaces can be used by users such as
caregivers, medical doctors and system administrators for apprehending the ECG
and bio-data. These interfaces can be accessed by logging in.
The smart gateway provides some real time notification services to update about
anomalous situations. Real-time signals are sent to the cloud which then sends
notification signals to the users when one of the three situations occurs, either:
1. When data is not received by the gateway from the sensor nodes.
2. An embedded gateway or a sink node temperature is higher than a predefined
threshold.
3. When the heart rate is not in the normal heart rate range as per the American
Heart Association.
• Back - End Part: Backend consists of the cloud server and the other backend
services. The cloud server is responsible for the storing and processing of
data whereas backend services are liable for the dynamic data analysis and
visualization. High portability and 90% increase in bandwidth efficiency is
witnessed in the propounded system.
SURVEILLANCE VIDEO STREAM PROCESSING

USING FOG COMPUTING
Introduction
When coherent processing and immediate decision-making is necessary, fog

computing comes into picture as it has a vital role. Urban surveillance which is an
important component of circumstantial cognizance deals with heterogeneous data
from a stratified sensors environment which is useful for the improvement of urban
planning and management.
146
CLOUD AND FOG COMPUTING IN VIDEO

STREAM PROCESSING USING DRONES
When tracing of targets objects is being carried out using a drone video stream,
instead of sending the dynamic real-time video content to the cloud platform, it
can be sent to the fog node so that transmission and processing time can be cut
down and the possible network congestions can be shunned. By fog nodes, we
mean any smartphones or laptops or tablets which support the tracking algorithms
and can process the live streaming feeds. On an average, around thirteen percent
(Khan, Parkinson & Qin, 2017) of the processing time has been reduced due to the
introduction of an intermediate fog platform between the surveillance area and the
Cloud. Sending video feeds from every camera sensor to the cloud is a very tedious
task to perform, but due to the high processing services provided by the distributed
Edge computing, each video can be processed separately at the fog nodes and then
sent to the Cloud where all the data can be accumulated and a proper final result
can be obtained. Thus, Fog Computing proves to be an optimistic solution for this
operation.
With the aid of huge trajectory data gathered from various sensors that are installed,
Urban Traffic can be monitored. The city administration and law enforcement
department can use this data collected from sensors and provide services efficiently.
For example, pedestrians can be victims to over-speeding vehicles. Such fatal
accidents can be minimized by a dynamic traffic surveillance system. The sensor
on a drone records the videos of vehicles commuting on the road and sends these
videos to the controller on the ground.
The urban surveillance system has three layers (Chen, Chen, You, Ling, Liang
& Zimmermann, 2016)- a surveillance application layer, a Fog Computing layer
and a Cloud Computing layer.
• Surveillance Application Layer: The surveillance application layer contains

the drone devices which are used to monitor the regions of interest and capture
the video. The controller station at the ground level also is a part of this layer
where the video data generated from the drone is displayed on a monitor
screen. Upon finding any sceptical vehicle speeding beyond a limit, only that
particular region of the video can be locked and transmitted to the next layer
for further processing.
• Fog Computing Layer: The fog layer contains fog nodes which are usually
computer systems on which tracking algorithms can be run. One key issue
in the fog nodes is the video processing time. If the output frame rate is
made equal to or greater than the input frame rate, then the requirement of
real-time video streaming can be satisfied. This issue can be solved in two
147
ways. Decreasing the video frame rate or discarding a few frames without
altering performance could be one method. As in this method some frames
are dropped, it might result in a big gap between any two frames thereby
leading to loss of suspicious targets as the target might have moved a large
distance in this interval. Another method would be decreasing the resolution
of the surveillance video which would reduce the data size but some crucial
information in the video might be lost. If in safety or security related
applications, information is lost, then the consequences might be disastrous.
For better situational awareness and decision making, a high resolution video
is required.
The controller or operators on the ground are the police officers who on
receiving the video data sent from the drone sensors, identify over-speeding
vehicles and capture or lock that vehicle in the real-time frames for further
inspection. Using the divide-and-conquer technique, only the sub-area containing
the suspicious vehicle is obtained from the original video and is transmitted
to the fog computing units. As only a sub-area is extracted from the original
frame, the data size is toned down and thus transmission time also is reduced.
• Cloud Computing Layer: The fog computing units stores this data for a
short duration and then sends this data to the Cloud server for further analysis.
Precise speed information of the vehicle of interest can be deliberated from
the frames using powerful and efficient algorithms such as L1 tracker using
accelerated proximal gradient approach. The mostly likely time during which
over-speeding can happen can be a concern so that during those times extra
care can be taken by the police officers to avoid accidents. This scheme on
implementing using two DJI drones and a laptop server as a fog computing
device gave results which were encouraging as this urban surveillance system
had the ability to track only one target object, but even multiple targets using
the two simple algorithms mentioned above.
AGRICULTURE AND FOOD TRACEABILITY SYSTEM
Introduction
Population inflation (Goel, Bhagwan, Chaturvedi, Rai & Pandey, 2015) over the years
has led to the ever-increasing demand for food, cloth, etc. which has led to researchers
and scientists trying to innovate new technologies to fulfill this requirement. Farmers
grow crops in the fields, harvest them and vend them to the markets by seeking
knowledge from other co-farmers on various ways to maximize profits. This insight
148
of the farmers influences the outcome of their yield as they repeatedly growing the
same crops again and again without paying attention to the weather conditions, soil
fertility or pests and diseases. Due to this, a need for information on better farming
techniques has risen. Fog Computing can be extensively used in the improvement
of the agricultural sector.
PROPOSED SOLUTION
A Fog Computing model initiated by the ICT has two major parts-Cloud Agro
System and Food Traceability Management.
• Cloud Agro System: This System is used to observe and meet the obligations
with a user-friendly and quicker approach by providing services like
e-knowledge sharing, demand-supply, communication devices and conducting
research. The other part is the e-Data Bank which stores all necessary
analyzed data like the crop-related or weather and soil related propaganda,
crop growth progress data provided by the farmers, etc. in a consolidated
location. Information Technology enabled agricultural farming techniques
(Chavali, 2014) like precision farming which is information profound affects
the economy of the farmers extensively. Therefore, technologies like cloud
computing and fog computing are adopted to comply with the requirements
of the farmers in terms of affordability and convenience.
• Food Traceability Management: Another domain where Fog Computing
is applied is that of the Food Traceability Management (Khan, Parkinson &
Qin,2017) where impoverished kind of items are removed from the supply
chain using value-based processing. Using the traits such as processing and
transportation gadgets and location, the food products can be easily traced.
The Cyber Physical System(CPS) which gives judgments based on Fuzzy
rules is responsible for determining the quality of the food component. The
complete food supply chain is made track-able in the Fog Network when it
receives the food traceability and quality information. The undocked data on
all the food items which were pursued are stored in the Fog Network which
finally conveys this information to the Cloud System which can be accessed
by the allies of the Internet.
This Food Traceability System prevents harmful substances from entering the food
Chain and also removes ambiguous products from the food supply conglomerate.
149
DISASTER MANAGEMENT
Introduction
These days every country is witnessing sudden natural calamities like earthquakes,
floods, hurricane, typhoon and artificial disasters like bombing, fire accidents,
traffic jam and so on. Government and Safety Organizations have to take up instant
actions when these occur as it involves millions of lives. IoT technology can be very
useful during those situations.
DRAWBACKS OF CLOUD COMPUTING
Implementation using Cloud Computing has some drawbacks like low-latency

resulting in delay and jitter, as it involves real-time events. These downsides of
Cloud Computing can be overcome by using Fog Computing in such situations.
Any devices facilitating processing, storage and connectivity features can act as a
fog node.
PROPOSED SOLUTION
Crowdsourcing-based Disaster Management model (CDMFC)(Rauniyar, Engelstad,

Feng & Thanh, 2016) is proposed to manage disasters using Fog Computing. This
model makes sure that the real-time disaster related IoT data reaches the fog nodes.
The structure of CDMFC consists of four layers - sensing, crowdsourcing, CDMFC
and cloud computing layer.
• Sensing Layer: The sensing node facilitates the sensing of any disaster event
like earthquake or typhoon. Massive amount of Data is generated by humans
or any IoT device during a calamity, this data may or maybe not be related to
the calamity. This layer is dedicated to sensing all types of data transmitted
during the disaster.
• Crowdsourcing Layer: The data from the sensing layer comes to
crowdsourcing layer where all the data is further transferred to Cloud Layer
where various algorithms and techniques of Data Mining is applied for
further analysis of data. Since a large number of devices generate data during
a calamity a huge amount of data gets piled up at the cloud which may reduce
the processing speed of the data by increasing the latency. To avoid such
situation first the crowdsourcing layer sends only data related to emergency
150
events to the CDMFC layer where a filtration technique based on emergency

keyword search is applied to only allow such data to enter it. In addition to
this there might be situations where a link between a fog node and CDMFC
layer is broken due to poor network infrastructure at the disaster site, in such
cases that fog node send information to another fog node using CDMFC layer.
• CDMFC Layer: Data generated at the disaster site consists of Time and
Location details which can be used by CDMFC layer to precisely identify the
disaster location. This layer also consists of some emergency phone numbers
of safely organizations who can rescue people from disaster site. Data related
to the disaster like photographs, videos are stored in the fog and are accessible
to the people near the affected regions so that they are aware of the present
situation at the site. CDMFC layer facilitates machine to machine, device to
device and also human to machine interactions.
• Cloud Layer: The data which has to be stored for long duration can be sent
to the Cloud. Finally the cloud layer can be used to store certain data coming
from various other layers in the model can here data mining algorithms and
visualization techniques are applied to derive various conclusions from the
disaster related data.
CDMFC model implemented using Fog Computing results in easier planning

and reacting to disasters. CDMFC model conserves bandwidth as only data related
to disaster is computed at the fog nodes and the other data which has to be used
to draw conclusion and for long-term storage is transmitted to the Cloud. Further
CDMFC model can operate on IoT data inside the fog nodes using certain security
and privacy algorithms.
WEB OPTIMIZATION
Introduction
The speed in which web pages are displayed on the user’s browsers is referred to as
Web Performance and the techniques adopted to increase the web performance are
Web Optimization. Researchers have been focusing on various methods on optimizing
the speed with which the web page load. Web Optimization (Zhu, Chan, Prabhu,
Natarajan, Hu & Bonomi, 2013) improves the user’s experience as the web pages
load faster. It is estimated that about 80-90 percent of the load-time happens at the
end-user’s browser(client side) and this can be optimized by reducing the number
of HTTP requests, intelligently using style sheets on the web page and maximizing
the cache.
151
PRESENT SOLUTION AND ITS DRAWBACK
Various optimization tools like Google Page Speed and Yahoo! YSlow measure the
performance of the website when the developer physically submits their website that
is to be optimized. There is no idea of automatic optimization of a webpage based
on the data of the client which is known or which can be measured by devices at
the client’s network like the clients’ feedback getting worn out when it passes more
than the predefined number of hops or the data traffic could be measured by the
gateways at the client side network. Here, Fog computing can be used to optimize
the Web Performance.
CLOUD COMPUTING AND FOG COMPUTING

IN WEB OPTIMIZATION
In Cloud Computing the web applications are computed on the Centralized Data
Cloud resulting in low latency and also raise issues for applications requiring
mobility support, location awareness and geo-distribution which could be overcome
by integrating fog computing to the system. Computation and storage, Data
management and analysis become easier on using fog computing as an extension
of Cloud Computing.
There are certain informal rules laid down by researchers that can be used to speed
up the web performance by using Fog Computing. Optimization tools use these rules
to check the performance of a particular website and optimize them accordingly. One
factor that could help in the optimization of website is the reduction in the number
of HTTP requests. More the number of HTTP requests, slower are the execution of
the webpage as there are more round trips between the server and the client.
METHODS TO INCREASE OPTIMALITY OF WEBSITE

USING CLOUD AND FOG COMPUTING
• Styled Websites: A website consists of several styling elements (CSS),

images and scripting data (JavaScript) that would increase the number of
HTTP requests. Resolution to this problem would be to combine all the
style sheets into a single file and scripting sheets into another file and then
broadcast it to the user. If the website has multiple background images then
all can be combined into one and then be transmitted to the user. Further,
the HTML file has to be examined and if there are links for redirection, then
that HTTP link should be replaced by the complete URL. All the embedded
152
objects on the web page should have an expiration head with time that is in
accordance with Fog server caches’ expiration time so that the browser would
reduce its requests to objects.
• Minimization of Web Objects: Another factor that could improve
optimization is by minimizing the size of web objects. HTML, Style Sheets,
Scripting Files can be zipped before sending it to the user and mentioning
‘giz’ in the HTTP response. Minification should be applied to remove
inessential characters from the CSS and JS code. Munge Procedure should be
applied to JS file to reduce the length of the symbol names. Graphics and GIF
animations should be optimized before transmitting to the user.
• Website Requested for First Time: While if the webpage is been requested
for the first time by a client on that network or requests for a webpage that was
present in the fog server cache but later has been expired then, the incoming
parts of the web page that comes from the data center can be inspected by fog
server and then sent to the user. Although, optimization of the complete web
page is not possible however, the small incoming portions of the webpage can
be optimized by applying relevant optimization techniques. The optimization
of web pages should be acceptable by both the end user and also by the
website.
In Wide Area Networks(WAN) fog nodes are advantageous as they are closer to
the user and the fog servers were aware of the network condition near the
user, like congestion, data rate in the wireless channel. This information
can be utilized by the fog nodes for optimizing the web pages.
• Browsing Behavior of User: Web Pages can be optimized better if the
fog node knows about the browsing behavior of the user. If a user views
a particular website many times then the performance can be improved by
using external approach where the web pages are cached and stored in the
fog server and can be utilized by other clients. Identification of user can be
done by cookies or by track by using the IP address of the user but this is not
possible if the user belongs to a network behind a firewall. Fog Server can
identify the user in such cases as the fog server belongs to the same network
and it can track the users via their MAC address.
Rendering speed of the webpage on the client’s browser can be measured by

adding code fragments to the HTML file. This measurement is accurate only if the
client and web server are present in the same network, as the fog nodes can assist
the flow of data to the client browser based of this speed thus making it efficient
and reducing computation and cost. Thus, Fog Computing features dynamic and
automatic optimization of websites only based on the Client Network Conditions.
153
CONCLUSION
Security and privacy are the challenging areas in the field of IoT and Fog computing.
It provides lots of observation from academic, researcher and manufacturing peoples.
This chapter has illustrated the overview of different security and privacy issues
in IoT and Fog Technologies. Each of these issues has been discussed with present
available solutions. By the end of these chapter we have discuss the different case
studies which covers the different domain of computing technologies.
REFERENCES
Bonomi, F., Milito, R., Zhu, J., & Addepalli, S. (2012).Fog Computing and Its Role
in the Internet of Things. Proceedings of the first edition of the MCC workshop on
Mobile cloud computing, 13-16. 10.1145/2342509.2342513
Chavali, L. N. (2014). Cloud Computing in Agriculture. In Agricultural Bioinformatics.
Springer. doi:10.1007/978-81-322-1880-7_12
Chen, N., Chen, Y., You, Y., Ling, H., Liang, P., & Zimmermann, R. (2016).
Dynamic Urban Surveillance Video Stream Processing Using Fog Computing.
Second International Conference on Multimedia Big Data, 105-112. 10.1109/
BigMM.2016.53
Corser & Aledhari. (2017). Internet Of Things(IOT)Security Best Practices. IEEE
Internet Technology Policy Community White Paper.
Dsouza, A., & Taguinod, M. (2014). Policy-driven security management for fog
computing: Preliminary framework and a case study. Information Reuse and
Integration (IRI), 2014 IEEE 15th International Conference, 16-23.
Fakeeh. (2016). Privacy and Security Problems in Fog Computing. Communications
on Applied Electronics. Foundation of Computer Science FCS.
Gennaro, R., Gentry, C., & Parno, B. (2010). Non-interactive verifiable computing:
Outsourcing computation to untrusted workers. In Advances in Cryptology--CRYPTO
2010. Springer.
H. (2015). Fog Computing in Healthcare Internet-of-Things: A Case Study on
ECG Feature Extraction. 15th IEEE International Conference on Computer and
Information Technology, 356-363. 10.1109/CIT/IUCC/DASC/PICOM.2015.51
154
Goel, Bhagwan, Chaturvedi, Rai, & Pandey. (2015). Application of cloud computing
in agricultural sector. 4th International Conference on Agriculture & Horticulture.
Agrotechnol
Khan, Yongrui, & Qin. (2017). Fog computing security: a review of current
applications and security solutions. Journal of Cloud Computing.
Kirat, D., Vigna, G., & Kruegel, C. (2014), Barecloud: Bare-metal analysis-based
evasive malware detection. USENIX Security Symposium, 287–301.
Lebied, M. (2017). 6 Cloud Computing Challenges Businesses are facing in these days.
Retrieved from: https://www.datapine.com/blog/top-6-cloud-computing-challenges
Madakam, S., & Date, H. (2016). Security Mechanisms for Connectivity of Smart
Devices in the Internet of Things. Springer. doi:10.1007/978-3-319-33124-9_2
Meola, A. (2016). How the Internet of Things will affect the security & privacy?
Retrieved from http://www.businessinsider.com/internet-of-things-security-privacy-
2016-8?IR=T
Okay, F. Y., & Ozdemir, S. (2016), A Fog Computing based Smart Grid Model,
Networks, Computers and Communications (ISNCC). 2016 International Symposium
IEEE, 1-6.
Rauniyar, A., Engelstad, P., Feng, B., & Thanh, D. V. (2016). Crowdsourcing
based Disaster Management using Fog Computing in Internet of Things Paradigm.
Collaboration and Internet Computing (CIC), 2016 IEEE 2nd International
Conference, 490-494. 10.1109/CIC.2016.074
Raval, K. (2016). The Big Data of IoT – Understanding Privacy and Security
Threats. Retrieved from: https://www.letsnurture.com/blog/the-big-data-of-iot-
understanding-privacy-and-security-threats.html
Suthar, S. (2016). How is Security important to IoT devices? Retrieved from: http://
www.cxotoday.com/story/how-is-security-important-to-iot-devices/
Yi, S., Qin, Z., & Li, Q. (2015). Security and privacy issues of Fog Computing:
A Survey. In International Conference on Wireless Algorithms, Systems, and
Yi, S., Zi, J. H., Qin, Z., & Li, Q. (2015). Fog Computing: Platform and Applications.
Workshop on Hot Topics in Web Systems and Technologies. IEEE.
Yu, Ren, & Wenjing. (2010). Achieving secure,scalable,and fine grained data access
control in Cloud Computing, Infocom. 2010 Proceedings IEEE, 1-9.
155
Zhu, J., Chan, D. S., Prabhu, M. S., Natarajan, P., Hu, H., & Bonomi, F. (2013).
Improving Web Sites Performance Using Edge Servers in Fog Computing
Architecture. Service Oriented System Engineering (SOSE), IEEE 7th International
Symposium, 320-323.
KEY TERMS AND DEFINITIONS
Cloud Computing: Network of servers hosted over the internet to store and
process data.
ECG Extraction: Technique to monitor the performance of human heart.
Fog Computing: Infrastructure in which storage and computing is distributed
between cloud and data source.
IoT: Network of physical devices that can connect and exchange data.
Privacy: Various techniques used to protect private data.
Security: Protection of data from threats and damage.
Smart Meter: Electrical device which records electricity usage over various
intervals.
Web Optimization: Knowledge about increasing web performance.
156
157
Chapter 8
Software Engineering in
Internet of Things
Naresh E.
Vijaya Kumar B. P.
Aishwarya Hampiholi
Jeevan B.
ABSTRACT
This chapter gives an overall role of software engineering in internet of things
domain. In this chapter, the following topics are included: glimpse of complete
software engineering, main motivation of IoT, how IoT evolved, usage of software
engineering concepts in IoT, role of CBSE in IoT, role of aspect-oriented software
engineering, heterogeneous boards in designing IoT systems, importance of integration
phase in IoT systems, comparison of different IDEs of IoT, testing of IoT systems,
and a case study illustrating all the concepts for online blood banking system and
forest fire detection.
DOI: 10.4018/978-1-5225-6070-8.ch008
Software Engineering in Internet of Things
SOFTWARE ENGINEERING: THE SCIENCE

OF CREATING EFFICIENT SOFTWARE
Suppose we want to build a new house or any infrastructure in common, what do

we do? Well, we approach an engineer who has great experience and good name
in the industry. For a great result, it is important for the engineer to plan out the
construction process. It is also important for the customer to brief the engineer about
what exactly is he looking for in the building. If the customer is not sure about what
he wants, it becomes difficult for the engineer to carry out the further process. Once
the requirements are clearly defined, the engineer plans out the rest of the process
based on his level of expertise.
The same theory applies for the software and computer systems of today. We
are surrounded by enormous number of computer systems. To create each of these
systems, there is a certain process that needs to be followed. This process needs
to be planned very well and needs to be executed in the same order. The study of
Software Engineering teaches the aspect of planning and executing the process of
building a software system.
Software Engineering involves identifying the best practice to plan and execute
the creation of new software. Be it, the traditional method like waterfall model or
adaptive models like agile or spiral models. Every software needs to be developed in
a different and unique way. The software needs to be developed using the waterfall
model when it has all the requirements fixed in the initial stage of development.
Whereas, in a few cases the requirements keep changing every now and then. In such
cases, we go for adaptive models like agile or spiral. If the software development plan
is not made in a proper way, it does not turn out to be the way we expect it to behave.
Software Engineering also provides the methods to test the final software product
such as verification and validation. Not just that, it also includes the techniques
to manage the whole process be it in estimating of cost and effort or in managing
the staffing issue. It also throws light on the various types of software services/
development techniques (Sommerville, 2011).
• Component Based Software Engineering: This technique mainly involves the

idea of reusing already existing components instead of creating new modules
everytime from scratch while creating a new software. Components are the pre-
existing modules that have a definite functionality and can be adapted while
building a new software. More about CBSE is explained in the part 1.5 of this
chapter (Niekamp, n.d.).
• Service-Oriented Software Engineering: This technique mainly promotes
reusability of code. It is similar to component based software engineering
but, here, instead of smaller modules for individual functions, we use smaller
158
modules that provide services. For example, webservices. This includes third
party service providers as well (Cervantes, Humberto & Hall, 2004).
• Aspect-Oriented Software Engineering: This is one of the software engineering
techniques that is a little more advanced than the component-based software
engineering. In this technique, we can not only reuse the modules for a particular
functionality, we have the liberty to add many more functionalities without
changing the existing code(Kiczales, Lamping, Mendhekar, Maeda, Lopes,
Loingtier, & Irwin, 1997).
• Model-Based Software Engineering: While component-based software
engineering, gives the ability to reuse functional modules, model-based software
engineering gives the abstract models that can be reused for building any new
system (Schmidt, 2006).
Even though Software Engineering has been the backbone of software development
process, the new trend in software and technology called the Internet of Things
(IoT) has proven to be a challenge to this subject. IoT is essentially the emerging
trend in technology that involves connecting various physical systems to easily
accessible network like the internet so that they could be operated from anywhere
in the world. Its development process is totally different from that of the traditional
and well-established rules and regulations that are a part of the present Software
Engineering philosophy.
Before we get into the comparison between IoT and Software Engineering, lets
have a small look at what exactly IoT is, and why is it necessary to implement it in
the present day.
MOTIVATION BEHIND IOT SYSTEMS
As we know in the past few years, the usage of internet has become very popular.
Information about anything and everything in the world is available on our finger
tips with the help of internet. In the world driven by technology like artificial
intelligence, internet, cyber space, our lives have become super easy to lead. To
make our lives easier and more comfortable, the concept of integrating physical
things to internet came into picture. This integration is now popularly called as the
Internet of Things (IoT).
Initially, machines were made to reduce the human effort. But now, we are
finding out ways to make the machines intelligent and interactive with the human
beings. Using any machinery becomes a lot easier when you can communicate
well with the machine itself just by the press of a button or just by talking to it just
like talking to a friend. In an attempt to make the machine more efficient and user
159
friendly, we have now come up with ways to directly talk to the machines in order
to get our work done.
Due to the growth in technologies like Artificial Intelligence, Big data, Embedded
Systems and Wireless connectivity, etc., the concept of Internet of Things came into
picture. Not just these, but there are many other aspects that motivated the growth
and usage of IoT systems. They are:
• Real-time production and usage of data for various other purposes

• Optimization of process
• Adding Analysis-as-a-service(AaaS) as an important service in a cloud setup
• Self-aware and self-acknowledging intelligent systems
• All devices connected through a single network
HISTORY AND EVOLUTION OF IOT
The concept of Internet of Things came into picture in the year 1999. The term
was coined by Kevin Ashton. It was in this year that the concept of IoT came into
existence. Although this term was modified a little from its actual name: “Internet
for Things”, it seemed appropriate for the actual process that took place (Mattern,
Friedemann, Floerkemeier, & Christian, n.d.).
It all started with a coke vending machine that was connected to the internet.
People used to check the availability of coke cans and if the cans were cold or not,
using the information given by the vending machine. This gained popularity very
quickly and the idea of connecting many other such machines to the internet came
up. Then in the early 2000’s the research on this concept started. But, this concept
mainly gained most prominence from the year 2010 and onwards.
Initially, IoT was known to help only those “things” that had the option of on/
off. And also, it was physically connected through wires and cables to the local
network. As the research began, the local network was replaced by the internet that
was accessible worldwide. This was the major turning point in the evolution of IoT.
Slowly, the connectivity with the internet got rid of the physical wires. Instead of
the wired setup, the researchers proposed the option of wireless connectivity to the
“things” in Internet of Things (Cervantes, Humberto & Hall, 2004).
Now, as the advancement in sensors and embedded systems has increased,
the “thing” in IoT need not be a machine that could be in on/off state. Infact, the
technology has gone a step ahead in identifying and controlling even the smallest
of the small things that do not have any electric connection like pens, gates, doors,
chairs, plants, etc.
160
The evolution of IoT led to the development of new technologies like Big Data,
Data Analytics, Smart “things” concept, etc. As we know, in IoT, any “thing” could
be connected to the internet and its information could be collected instantaneously.
As the data flow is instantaneous, the growth of data flow raised drastically. This
gave rise to the concept of Big Data. And, as Big Data came into picture, we needed
better way to use this huge amount of data that was collected. Hence, the Data
Analytics part also grew drastically.
After the introduction of Big Data and Data Analytics, IoT spread across various
domains like (Borne, 2014):
• Agriculture
• Health
• Smart cities
• Weather monitoring system
• Energy (Smart Grids)
• Retail
• Logistics
• Industry
• Home Automation
SOFTWARE ENGINEERING AND INTERNET OF THINGS
As we have seen earlier, Process of software engineering is developing a product

on software. It has been our practice to follow the Software Engineering concepts
while developing any software. And it has been in the market since the time of start
of software development. But, IoT is a relatively new concept and has been in the
market for just a few years.
IoT takes a totally different approach to produce software systems. It does not
implement the age old methods of SE like the waterfall model. But, it could use the
approach of agile methodology for the development of systems. Ultimately, there is
a huge difference in the development of an IoT system and Software Engineering.
A few differences are listed below:
• Standardization of the System: In usual product development using

Software Engineering methodology, the developer need to keep in mind that
the product must obey certain standards so that it could be released to the
customer. But, in case of IoT systems, there are no predefined standards that
the system needs to meet in order to get released to the customer. As there’s
161
still a huge number of researchers working towards finding a suitable standard

for the IoT systems, it seems impossible to do so, because, IoT systems are
very easily prone to frequent changes.
• Component-Based Software Engineering: IoT makes use of the newer
version of Software Engineering called the component-based software
Engineering in which the system is built using the existing modules called
components. It saves a lot of time and also has a large number of pre-defined
functionality. In case of traditional SE, each system needs to be started from
the scratch. This is more time consuming. But, this approach helps in having
flexible functionalities which is unavailable in CBSE.
• Code Re-Usability: As IoT uses CBSE, the system is completely made
of reusable code. It has a set of predefined functionalities. Whereas in the
traditional SE, the concept of code re-usability does not exist. The code is
unique for each system.
ROLE OF SOFTWARE ENGINEERING

BASED ON COMPONENT IN IOT
Component Based Software Engineering is the part of Software Engineering in

which a few functionalities that are most widely used are combined to form small
modules called “packages”. These packages are mainly used in adding generic
functionalities while developing a software. This process helps in saving a lot of time
to the developer. CBSE also helps in saving a lot of amount spent on development
of a new software. CBSE mainly comprises of two entities:
1. Component Interface
2. Component Models
Component Interface
Component interface mainly has two activities:
1. Providing interface i.e., the services that are provided by the component to
other components.
2. Requires interface: i.e., the services that specifies what services must be made
available for the component to execute as specified.
162
Figure 1. Component Diagram
Component Models
A component model is a model that gives a brief introduction about the standards
for component implementation, documentation and deployment as shown in Figure
1. The main objective of a component model (Figure 2.) is to specify how interfaces
should be defined and the elements that should be included in an interface definition.
It is mainly used to integrate loosely coupled lesser functionality components to
form one huge product.
A few examples of component models are:
• EJB model (Enterprise Java Beans)

• COM+ model (.NET model)
• Corba Component Model
The Process of Component Based Software Engineering
We need to be very careful while making use of the components while developing
any new software product. We must know what the ideal requirements are and
Figure 2. Component Model
163
how to achieve it using the pre-existing modules or packages and also know which
functionalities are given by the packages if they are adapted in the software. This
mainly involves as follows and as shown in Figure 3.
• Developing abstract but clear requirements.

• Searching for the existing but most suiting components and then modifying
the software requirements according to available functionality.
• Searching again to find if there are better components that meet the revised
requirements.
During the CBSE process, the processes of requirements engineering and system
design are interleaved. Combining of components together or “Wiring” of components
in order to create a whole new system is called as Component composition. When
composing reusable components, you normally have to write adaptors to reconcile
different component interfaces.
When choosing compositions, one needs to be aware of what functional and
nonfunctional requirements are covered if that package is used and also about how the
system requirements may undergo change as the time passes. Hence the components
must also be ready to undergo slight changes in the meantime.
ASPECT ORIENTED SOFTWARE DEVELOPMENT
AOSD is technology which keeps the aspects in mind and implements the new
developmental modularization which helps in improving the business logic let’s
consider a scenario of smart environment where by considering the various
aspects related to the protocols of for the given feature IoT derives some various
challenges collected from the Internet in which we explain the context of scalability,
heterogeneity, undefined topology and point information regarding data, user
preferences has incomplete metadata and conflicts. The non-availability of data
point information is a major consequences in an unknown topology details in which
no more suitable device can be available at the geographical site in order to provide
required information or a device may not have collected the data points from the
various site. So there is a strong need in collecting the various points of data without
any human intervention. Similarly, the information related to the services has been
manually entered by the human operator at the installation time. As device strength
grows, the manual information given can be incomplete or inaccurate. SO this reason
makes the system to get all the data properly. So there is a need of a method which
can use the IoT devices such as sensors and actuators to gather data and provide the
proper information required by the system.
164
The savvy condition is given gadgets that can shift as for memory, preparing
capacities and network. This requires the middleware part itself ought to be
reconfigured powerfully in the part of runtime sending and reconfiguration of
administrations subsequently making it conceivable to include, evacuate, and supplant
the administrations amid running. In addition the administrations are disseminated in
nature, which empowers the non-utilitarian prerequisites like security and adaptability
to be considered for execution. The proposed middleware is intended to join the
dynamic reconfiguration ability keeping in mind the end goal to appropriately adjust
to the ceaseless changes in the earth and circumstance mindful administrations to
give client bolster.
Recommending a five layered design for the IoT. Those layers incorporate Edge
Technology layer which comprises of RFID labels, sensors, actuators and equipment
gadgets constituting things in the IoT worldview, Access Gateway Layer managing
distributing, subscribing administrations gave by the things, message steering and
correspondence, Internet layer for associating source and goal, Middleware layer
including functionalities like administrations revelation, benefit choice and security,
and application layer to convey the administrations. Our imagined angle arranged
IoT middleware is delineated in Figure 3. The commitments of our proposed work
are triple as takes after:
• Design and execution of angle arranged middleware.

• Multi-specialists to present the usefulness of checking the running condition
and adjusts in light of changing client and ecological changes by running new
design.
• Deploying practical substitution and on-request estimation benefits as angles
in middleware and empowering impairing at whatever point vital.
The basic parts of the proposed middleware system incorporate shrewd condition
observing administration (SeMS), reconfiguration benefit for adaptability(RSA)
and angle administration benefit (AMS). RSA contains the required rationale for
reconfiguration in light of the setting information provided by the SeMS. SeMS
comprises of portable operators and insightful specialists for keen condition
checking and extricating the data. Specialists are fit for moving and dispersing
themselves to play out their observing undertaking lead. They are utilized to take
care of issues by utilizing a decentralized approach where a few operators add to
the arrangement by participating each other. The essential favorable position of
programming operators is the insight that can be epitomized into them as per some
aggregate counterfeit consciousness approach that necessities participation among
a few specialists that can keep running on a parallel or dispersed PC to accomplish
the required superior for taking care of substantial complex issues keeping execution
165
Figure 3. Aspect Oriented Software Development
time low. AMS helps in woven and unwoven of usefulness demonstrated as angles.
When we have built up the fundamental administrations of the middleware, test
application situations reasonable for IoT condition will be planned and assessed.
The application administrations are reconfigured in light of the setting data gave by
the checking specialists and suitable useful segments are substituted in view of the
setting information. In light of the setting information and administration depiction
International Journal of Intelligent Engineering and Systems, Vol.8, No.4, 2015 24
the design storehouse is populated. The RSA module will choose proper setup in
view of the setting information.
TESTING
Testing Internet of Things (IoT)
Testing is something the we check weather the system that we build is working
properly and also is it fulfilling the necessary features that are asked for so in testing
IoT devices we check for the various factors of the IoT which is discussed below.
166
Now we have to decide on how to approach the testing mechanism we have to

put ourself few questions as:
• How do we do it now? How is it so different?

• We need to understand these first after which we can learn how to test them.
Approaches Used to Test the IOT Devices
Usability
• Each device usability has to be verified so that it’s used properly.

• Medical tracking devices has to be portable enough so that its transferable to
various segments.
• Every equipment has to be smart enough to display notifications along with
the error messages and warnings.
• In order to provide clarity to the end users the system has to be having the
option to view log events and those system who don’t have the option to store
the log should store in the database.
• Notification display has to be handled properly in the system.
• Testing of data in terms of display, processing and job scheduling should be
done thoroughly.
IoT Security
• Security challenge of IoT could be the data availability where data is stored in
data center and only available if the data is present and connected.
• There is chance to steel or sneak the data when the data is being transferred
from source to destination.
• Before testing we have to make sure the nature of the data weather it is
protected or encrypted when they are transferred.
• We have make sure to protect the UI based application with a password to
avoid unauthorized access.
Connectivity
• Connectivity plays a very importation roles as without connection we cannot

achieve anything.
• There should be a seamless connection to the server so that it could provide
the services to the stakeholder 24x7.
167
◦
◦ Data transfer should be up and running when every there is connectivity.
◦◦ When the connectivity is offline there should be method to inform the
users that the system is off-line and should try to fix the problem or
establish the signal soon.
Performance
• We have make sure the system is scalable.

• As a tester we have to make sure that the system performance is even for all
the users at every time.
• We have make sure the system consumes less power and provide high data
efficiency.
Compatibility Testing
• Testing is a must as the architecture of the IoT system are very complex.
• Testing has to done for all the various kind of system environment such as
OS, machine etc.
Pilot Testing
• Pilot testing is a must as far as the IoT is concerned.

• Pilot testing should be done on a real-time basis taking a handful of the users
and getting the feedbacks from them.
• The feedbacks got from the testing can be a backup and makes the system
robust to solve the issues.
Upgrade Testing
• Combination of various OS, devices and firmware etc. makes a IoT machine/
device.
• When we upgrade the system we have to make sure the legacy features of the
system has to be verified along with the changes made to the system.
Challenges Faced During Testing in IOT
The following are the challenges faced during the testing in IOT by the tester:
168
Mesh of Hardware and Software
Architecture of an IOT, where hardware and software components are closely

coupled. Software applications are not only the ones who makes the system it also
depends on the hardware and play a vital role.
In certifying the system not only functionality testing helps but also there are
dependency on each other in terms of the environment or even a data transfer. It
becomes a tedious task as compared to generic system of testing.
Interaction Module for Device
The architecture of both hardware and software are very different so it becomes
mandatory to talk between them before the real time or even close to that time.
Then both are integrated and then testing on security, upgrade issues are the main
obstacle from now on.
Testing on Real Time Data
We know that pilot testing/regulatory testing is must for this kind of system and it
also becomes very tough obtain such data.
As a part of test team getting the data from checkpoints and other from the
deployed system is very difficult.
UI
Based on platform IOT is spread on every one of them and now testing on every
platform is pretty impossible whereas testing can be done on a specific machine
alone for various case.
It is not possible to omit from a device to access UI which do not have the
possibility of simulation. Which is a very tough challenge to overcome.
Availability of Network
Connection of network plays a very important role in IOT as it’s all about the data
being communicated in a very fast manner at all time so it has to be tested for all
kinds of speed.
In order to test the above we need a virtual network simulators which could help
to vary the network load and the connectivity etc. but in a real world data is a real
time depended so it’s very hard to test and solve.
169
Tools Used for IOT Testing Mechanism
There are many tools that are used to test the IOT device lets discuss few of them.
Based on the Target Lets Classify the Tools as Follows:
1. Software:
a. Wireshark: It’s an open source tool used for stimulating the network
related stuff by creating a model and introducing the various variations.
b. Tcpdump: This application does a similar work of a above application but
doesn’t have a GUI and its completely based on command line interface.
2. Hardware:
a. JTAG Dongle: It’s similar to a PC application performing debug operation
on a given target platform code and the various variable step by step.
b. Digital Storage Oscilloscope: This could be used to check various events
based on the time stamp and the variation in power supply and signals.
c. Software Defined Radio: Used to transmit and receive the large amount
of signal of wireless gateways.
In the world that is drastically developing, IoT is a growing market and provides a
lot of opportunities. The time is not far when Internet of Things becomes an essential
requirement for the testers to mark their presence in the development world (Aspect
Oriented Software Engineering).
The IoT-enabled smart gadgets, devise application, and communication module
play an important role in studying and evaluating the performance and behavior of
various IoT services.
Poor design of IoT-enabled devices and services can deteriorate the quality of
functioning of the application and in turn negatively affect the end-user experience.
CASE STUDY
Case Study 1
Title: Blood Bank Application
Introduction and Description

Online Blood Banking is especially purpose for hospital that needs blood to the
patient regularly or for blood bank so to collect blood to be given for different
hospital. Here we maintain the information about the donor in the database. It helps
the doctors to find the donor at right time and help the patient in need.
170
They have tried to maintain few of that information of donor which can be easily
understandable to the doctors which makes them easy to find the donor whenever
essential or for the patient directly. Using this schema donor to be attracted to donate
the blood. Our system encourages the blood communication between the donor and
doctor product is delivered within time and delays are minimized. Proper tools are
used which require low cost for the development of our system.
We have used PHP, HTML to build the system which is generally feasible for the
development of the system as well as suitable for coding which makes our project
more attractive.
Some of the objectives of the project are:
1. Making Time Consumption less for searching for blood in many blood bank
centres’ as user can login and see the various blood bank centre’s on these
application.
2. Customer won’t have to wait for long queue to get blood as they have many
other options for collecting blood at various Blood Bank Centres.
3. To make everything fast, that is from searching for blood bank centres to polio
vaccine reminder and Emergency Notifications.
4. To properly manage detail of each and every customers.
5. Maintain a high level of surveillance.
6. Promote, support and raise the effective establishment of national blood donor
programs and the also eliminating the dependency on the paid blood donation.
7. To encourage every people to donate their blood voluntary on unpaid basis.
8. To ensure that the citizen can access the availability of blood units by using
our applications. And to make every child diseases free and make the future
of our nation bright and wonderful.
Architecture
• Donor: Donor first register himself into the application where he shares his
information with the application and later the data is stored and only required
data is shown on the website where the information like phone number blood
type and his name is shown and can be called when its required by the patient.
• Patient/Receiver: Patient should also register and then search for the required
blood type based on their location where we have used pin code as their
location then search for the donor and contact them so that it saves time and
as well as life in case of emergency and should make sure that the donor is
given the address of the nearby police station.
• Application Manager: It manages donor and patient data by providing and
maintaining the database in the backend where we maintain the centralized
171
database to store the data which provides integrity to the data and can provide
security for the data that is obtained by the users so it provide authorization,
authentication, privacy and other aspects providing the data quality
improvement.
• Some Non-Functional Factors: It provides some of the nonfunctional
requirements which are provide as the additional service to the given system
such as about page which helps the user to understand about the system and
the agreement page to provide safety for the users such that if anything goes
wrong the user himself is responsible for any invalidated data entry or so.
Obstacles in Development of the Project
• Proper GPS location is hard to find as GPS keeps on fluctuating.

• Motivate people to donate blood.
• Authenticity of the application.
• Connectivity of the network in all the area.
Advantages and Disadvantages

So it is a simple and easy to use application that people groups in need at whatever
point basic likewise it gives wellbeing, security, protection and we keep up approval
and confirmation to the application.
In the up and coming form we are running consolidate with Google maps so that
the correct area of the giver can be discovered and can pick him if essential or give
Figure 4. Architecture of simple Blood bank application
172
a taxi to him driven the clinic so this gives an unmistakable preferred standpoint in
sparing existence of the general population.
Cons of this are individuals may abuse the application and can cut down the
fundamental use of the venture so with a specific end goal to overcome in this
application we have an assertion so we can enhance it by the following adaptation.
Case Study 2
Title: Forest Fire Detection
Introduction
Forests are the biggest providers of natural resources on earth and they help maintain
a balance in the ecosystem. But, now a day, they are getting destroyed because of
various reasons. One among the most important reason for depletion of forests is
forest fires. Forest fires are usually detected only after it has spread over a wide
area and has almost consumed most of the place. Sometimes, it is impossible to
control these fires. The result of this action is devastating and it causes irreversible
damage to the environment. Not just that, forest fires contribute to the production
of 30% of carbon dioxide (CO2) in the atmosphere (Bahrudin, & Kassim, 2013), in
addition to irreparable damage to the ecology. Impacts on weather patterns, global
warming, and extinction of rare species of the flora and fauna are just a few of the
many terrible long-term consequences of forest fire. These results are no less than
a disaster.
As forests are filled with trees, dry wood, leaves, etc., they act as a fuel source.
Also, the forests are usually found on the outskirts of the city where human vigilance
is very less. Hence, it is very difficult to manage these forests. These elements form
into highly inflammable material and become the perfect context for initial-fire
ignition. The fire ignition may be caused due to various reasons. To name some, we
can consider: human actions like smoking or barbeque parties; natural reasons such
as very high temperatures on hot summer days; a piece of broken glass working as a
lens focusing the sun light on a small spot for a very long time, leading to ignition of
forest fire. Once the ignition takes place, combustible material act as fuel that feed
the fire’s central spot helping it become massive. The starting stage of ignition is
referred to as “surface fire” stage. This may then lead the fire to spread to adjoining
trees and the flame becomes higher and higher, thus becoming “crown fire.” Mostly,
at this stage, the fire becomes uncontrollable and damage to the locality/ lands in
that area may become excessive and could last for a very long time depending on
existing weather conditions and the terrain.
173
Millions of hectares of forest are destroyed by fire every year. The lands and
locality affected and destroyed by these fires are massive and produce more carbon
monoxide (CO) than the overall automobile traffic. Analyzing and monitoring
of the possible risk areas and early detection of fire can significantly lessen the
reaction time, potential damage and the cost of firefighting. A few known rules
can be applied here: 1 minute ->1 cup of water, 2 minutes -> 100 liters of water,
10 minutes ->1,000 liters of water. The main objective is to detect the forest fire
as fast as possible and its exact localization and early notification to the fire units
is vital. This is the kind of situation, the present invention attempts to remedy, by
detecting of a forest fire at a very early stage, and to enhance the probability of
putting it out before it has grown beyond control or causes any significant damage.
There are a number of detection and monitoring systems used by authorities. These
include observers in the form of patrols or monitoring towers, aerial and satellite
monitoring and increasingly promoted detection and monitoring systems based on
optical camera sensors, and different types of detection sensors or their combination.
Technical Requirement
• ARDUINO UNO
• LCD 16*2 Display
• MQ-7 Semiconductor Sensor for Carbon Monoxide
• Fire Sensor
Figure 5. Architecture of Forest Fire Detection system
174
• Light Sensor and

• Temperature Sensor
Issues and Recommendation

The points in the previous part, interest a few of the central review sources that
have motivated the advancement of the exploration work in this proposition. It
demonstrates that the best accessible answer for forest fire location is utilizing
wireless sensors, just because it can give all required data that impacts nature at
any minute precisely. Remote sensor arrange innovation regularly that conveys vast
number of little, minimal effort sensors that can watch and impact the physical
world around them by collecting the physical data, changing it into electrical signs,
sending it to a remote area to do some examination, and sending the outcomes in
various applications. By along these lines there is no compelling reason to develop
towers or set entangled correspondence connections, for example, microwave and
satellite. It can be sent anyplace even in difficult to reach places.
This innovation can give constant checking, where it can give data at the start
case or at little postponements, contingent upon the hub utilized as a part of wake-
up/rest plan. This innovation takes a shot at short correspondence joins form. Thus,
more precise data with less deferral can be accommodated the fire warriors.
Utilizing this innovation for timberland fire application requires countless
conveyed hubs to give a dependable system.
REFERENCES
Cervantes, H., & Hall, R. S. (2004). Technical Concepts of Service Orientation.

In Service-oriented software system engineering: Challenges and practices. Idea
Group Inc. (IGI).
Kiczales, G., Lamping, J., Mendhekar, A., Maeda, C., Lopes, C., Loingtier, J. M.,
& Irwin, J. (1997). Aspect-oriented programming. ECOOP’97. Proceedings of the
11th European Conference on Object-Oriented Programming, 220–242. 10.1007/
BFb0053381
Mattern, F., & Floerkemeier, C. (2014). From the Internet of Computers to the
Internet of Things. ETH Zurich.
Md, S. B. B., & Abu Kassim, R. (2013). Development of Fire Alarm System using
Raspberry Pi and Arduino Uno. In Electrical, Electronics and System Engineering
(ICEESE), 2013 International Conference on. IEEE.
175
Niekamp. (2011). Software Component Architecture. Gestión de Congresos - CIMNE/

Institute for Scientific Computing, TU Braunschweig.
Schmidt, D.C. (2006). Model-Driven Engineering. IEEE Computer, 39(2).
Sommerville, I. (2011). Software Engineering (9th ed.). Academic Press.
176
177
Chapter 9
IOT and Data Analytics
Solution for Reducing
Pollution, Accidents, and Its
Impact on Environment
Chetan Shetty
Sowmya B. J.
Anemish S.
Seema S.
ABSTRACT
The goal of this chapter is to inspect and consider the answer for accidents and
reactions to the accidents in the urban zones. Modules have been made to manage
the colossal datasets and to bring interesting bits of knowledge into the outcomes.
This is done by utilizing decision tree analysis.
DOI: 10.4018/978-1-5225-6070-8.ch009
IOT and Data Analytics Solution for Reducing Pollution, Accidents, and Its Impact
INTRODUCTION
The overview by NSW demonstrates that from 2005 to 2013, around 30 individuals
are hospitalized by crashes on streets every day. Analysts utilize normal information
mining and prediction models on the measure of accident related data. The goal
of this paper is to inspect and consider the answer for accidents and reactions to
the accidents in the urban zones. Modules have been made to manage the colossal
datasets and to bring out interesting bits of knowledge into the outcomes. The
module including the assurance of the severity of the accidents, based on specific
factors, for example, the city, driving under influence, speed zone, the state of the
light and so forth. The above prediction is done by utilizing Decision Tree analysis
which produces a decision tree due to prediction.
Consistently, many lives are lost because of accidents, chiefly on the grounds
that emergency care fails to reach the accident spot on time. Currently, we totally
depend on bystanders to call for help. At times, accidents may happen in remote areas
where relying upon human’s assistance isn’t conceivable. Likewise, the victim may
not be in a position to call for emergency vehicle. Considering such circumstances,
it would be useful if the way toward distinguishing accidents is computerized.
To robotize the path toward recognizing and reporting accidents, we composed
a model, containing System-On-Chip and distinctive sensors in each vehicle.
Exactly when an accident happens, the data is sent to the Cloud. This data is used
to immediately, advising the concerned specialists, for instance, nearest Ambulance
and on the spot medical care.
Air contamination is the presence of particulates, organic atoms, or other unsafe
materials in the atmosphere, causing to sicknesses to people or harm to other
living life forms i.e. animals and crops, or the natural and manmade structures. Air
contamination can be caused because of different human exercises i.e. industries,
vehicles and due to the consumption of non-renewable energy sources like wood,
coal and so forth or normally. In spite of the fact that there is increment in the
advancement of technology and human race, we have neglected the surroundings
in which we live in. Along these lines we pollute the nature and thereby lessening
the quality of the air in the place we live into. One such significant giver is Motor
Vehicles. Roughly 25% of the unsafe gases that are discharged into air are because
of internal combustion engine. With the expansion in the quantity of vehicles
because of urbanization, air contamination has expanded quickly in the previous
couple of years. The essential toxins transmitted from these vehicles are carbon
monoxide, oxides of nitrogen and unburned hydrocarbons. CO is thought to be
the most harmful among all these. Due to these harmful gases causes different
sicknesses i.e. cancer, asthma, Cardiovascular Disease, diabetes, bronchitis and
furthermore putting the elderly and the children at a higher hazard. Thus different
178
measures are taken to lessen the vehicular contamination. The variables that add
to vehicular contamination are poor fuel quality, old vehicles, insufficient upkeep,
old car technology and traffic. Subsequently vehicles that are more fuel efficient
and those that create less discharges are a portion of the methods by which we can
diminish transport related air contamination. Discharge from vehicles can’t be totally
reduces, it unquestionably can be controlled.
Rapid industrialization and urbanization have achieved an exponential development
of IC engine vehicles everywhere throughout the world and India is no special case.
Such developing urbanization, joined with rising number of vehicles, which has
led to a new field for exploration regarding traffic analysis, prediction and traffic
control. The expansive measure of information accessible today makes information
mining strategies to a great degree helpful in rush hour gridlock related research.
The work aims for proposing the plan of a model which would address the above
issue. Bangalore is the sixth most excruciating city on the planet for traffic jams.
The normal speed of private vehicles is 25 km/hr and the speed of public vehicles
is 15 km/hr. A 20 km drive in the city can take as long as 3 hours in the peak hours.
This Exponential development in the huge number of vehicles and absence of city
wide development plan has led to traffic analysis, prediction and traffic control.
This undertaking tries to see the traffic video information so as to give an outline of
clog level at various intersections over the city. This is done adequately by utilizing
data mining strategies
A car, as observed from a layman’s view is only a means for transportation, to
go from point A to point B. In any case, from the vehicle’s point of view, it is a
source that encounters a ton of ongoing variables. The vehicle stalls out in rush
hour gridlock, traverse long distance through fluctuating climate conditions and
air quality, accidents. In the cutting edge age of the car, it is outfitted with choice
sensors-a rain sensor, a light sensor, a temperature sensor, a humidity sensor, an
collision sensor, sensors for different functions, alongside is an embedded system
that logs this information. An issue of real concern is the breakdown of public
transport or other huge vehicles amidst the street. Quite a while is taken to take care
of these vehicles, in this way prompting serious congested roads and burden to a
ton of workers. This issue can be successfully tended to by: 1) caution the vehicle
proprietors well ahead of time of the vehicles with deteriorating health, and; 2)
alarming the closest service center so they can dispatch help if there should arise
an occurrence of a breakdown.
The meteorological offices anticipate climate, particularly rains, utilizing the
information from cloud cover (acquired by means of satellite pictures) and furthermore
in view of their past predictions. Be that as it may, these prediction relies on real time
data and the precision is missing the goal by a lot. Thinking about the ubiquity of
vehicles in a locale, and the way that the greater part of them are as of now outfitted
179
with humidity sensors, we can give all the more constant rain information to these
organizations. This information can supplement the satellite information to make
more exact forecasts.
Data mining is a computational procedure that aides in finding patterns in
substantial informational collections. Data sets can be mined at various levels with
analysis, representation, gauge and streamlining systems so as to uncover its inborn
properties and enhance better decision making. Data mining includes strategies at
the crossing point of simulated learning, machine learning, insights and database
frameworks.
A definitive objective of data mining process is to separate information and
patterns from a substantial informational index and change it into an understandable
form for future use. A specific field in which data mining can be connected is street
activity information. This field creates gigantic measures of data consistently and
with the developing number of automobiles the information produced should be
mined so as to extricate just the applicable data which is effortlessly justifiable and
valuable. This mined information can be additionally grouped and can be utilized for
different research zones, for example, traffic control, Accidents analysis, seriousness
of mischances and so on. One such region can be in traffic control and prediction of
traffic alongside mischance investigation for clear course for emergency vehicles.
This will be useful in controlling the expanding issue of traffic and will likewise
be useful in anticipating pointless loss of time for emergency vehicles as in that
circumstance every moment is valuable.
LITERATURE SURVEY
(Sangeetha, Archana, Ramya & Ramya, 2014) Emphasizes on the prerequisites on the
emergency vehicle side i.e., for arriving at the accident site in time. This is done by
controlling the signal lights at whatever point a rescue vehicle is recognized adjacent
movement intersections in order to give an unmistakable way to the emergency
vehicle by giving green light to the path in which the emergency vehicle is right
now moving. The creators have developed 2 units to be specific: emergency vehicle
unit, to impart the position of the rescue vehicle and movement intersection unit, to
control the activity lights as and when the rescue vehicle draws close to its region.
In the rescue vehicle unit, a GPRS 3G modem is introduced in order to give position
of the emergency vehicle and in addition to get the directions of the accidents site.
The position is displayed as latitude and longitude, which is shown on the LCD.
In the rush hour gridlock intersection unit, the signal received from the emergency
vehicle will make the hardware change the signals to green. Subsequently the
essential belief system in this work is the programmed control of traffic signals. In
180
this work, other imperative transportation vehicles which need to reach on time to
the conveyance site will endure because of the disturbance in the traffic lights and
furthermore it will require investment to disentangle the genuine position in view
of latitude and longitude to pin point the real area and to give the location to the
emergency vehicle driver.
In (Gotadki, Mohan, Attarwala & Gajare, 2014), the creators have executed
smart rescue vehicle framework by giving instruments to quantify imperative patient
data, for example, heart rate and body temperature. The rescue vehicle is likewise
furnished with the ability of changing traffic signals to give a smoother ride of
emergency vehicle. A visual framework is outlined and actualized in the doctor’s
facilities concerned with the goal that the fundamental arrangements can be made
as appropriate for the casualty and as fast as could be possible.
The plan of the work is as per the following. There are 3 principle parts to
be specific namely Ambulance unit, Hospital unit and Traffic signal unit. In the
emergency vehicle unit sensors to check patient’s condition for example, heart beat/
pulse rate and body temperature is installed on to the micro controller. It likewise has
a simple to computerized converter and a Zigbee module. Zigbee module keeps up
energy effectiveness, subsequently it functions admirably on low power batteries. The
simple to computerized converter changes over the simple signal recognized from
the inserted sensors so necessary operation can be performed. The microcontroller
peruses the parameters and shows on the LCD. It is then sent to the Zigbee module
to send the data to the hospital.
The traffic signal unit comprises of RFID reader, microcontroller, relay driver
and light flag. RFID reader identifies the signals emitted by the emergency vehicle
and every rescue vehicle will have unique RFID. After receiving the signals, the
signals are passed on to the microcontroller which is then passed on to the hand-off
to change the traffic signals. The healing center unit comprises of Zigbee which is
associated with a visual framework to show the patient parameters so the doctor’s
facility staff can influence the necessary arrangements to even before the arrival
of the patient.
The work proposed by the creators in (Jadhav, Satam & Salvi, 2016) comprises
of an android application which acts like a caution framework amid medical need.
At whatever point a man encounters a medical need, the application sends an alert
message to his family and his medical care professional so the prescribed medicines
and the measures to be taken reduce the casualties can be sent. An interior database
is kept up with the goal that the gathered can be sent to trusted health care analyst so
the required examination can be done. It utilizes 3G network as a hotspot for speedy
data transmission and GPS and remote highlights to give the location of the patient.
The design of the application is as follows. Personal Information Gathering, this
module gathers the required personal information from the user. Account details, to
181
manage the details and also to edit any information necessary. Emergency system,
this module sends the emergency message to the patient’s family and his doctor along
with the nature of the emergency and also the location of the patient. Prescriptions,
the information in this module is sent by the doctor to the patient prescribing the
required medicines for those particular emergencies.
In (Moje, Kumbhar, Shinde & Korke, 2016) the creators have proposed keeping
sensors on vehicles in order to recognize impact. At whatever point a crash is
identified the data in regards to the position of the casualty is sent to the main server,
where the primary server chooses the closest emergency vehicle, closest health care
center and the shortest course to the site. Subsequent to ascertaining the required
parameters, the location is sent to the emergency vehicle unit which is embedded in
the rescue vehicle to get the patient. An activity control unit is additionally actualized
to influence the movement to make the signals green at whatever point a rescue
vehicle is recognized close to the unit with the goal that the emergency vehicle can
reach the clinic as fast as possible.
The fundamental outline of the work is as per the following. The vehicle unit,
at whatever point an impact is recognized the location of the casualty through GPS
module is sent to the main server. The main server, at whatever point the data is
gotten from the vehicle unit, the server checks the database to discover the list of
ambulances, finds the closest emergency vehicle, closest medical facility and the
briefest course from the rescue vehicle to the casualty. The rescue vehicle unit
likewise comprises of GPS module to pass on its location to the main server and
furthermore a map with the shortest route to the casualty. The traffic signal unit after
accepting the signals emitted from the emergency vehicle unit judges the distance
and turns the signal green quickly so the rescue vehicle can pass the activity with
no hiccups. The work proposed by the creators in (Manjunath, Nikschal, Mohta &
Sindhuja, 2016) tries to give an answer for the to the traffic congestion which are the
principle factor for the smooth going of crisis vehicles like ambulances. This work
utilizes RFID module to limit the impediments caused by the changing traffic light
for the crisis vehicle yet in addition in the smooth entry of the crisis vehicles. The
proposed framework comprises of RFID, GSM, UART and RS-232 microcontrollers
for the communication and data transmission.
The plan of the proposed work is as per the following. The RFID is for the most
part used to find the closest crisis vehicle and furthermore to identify the priority
of the crisis vehicle. For whatever length of time that the said RFID of the specific
vehicle is close to the region of the getting unit implanted close to the movement
intersections, the traffic signal on that path will be green and every single other path
will be red in order to give a smooth movement stream to the rescue vehicle or some
other crisis vehicle. When the crisis vehicle leaves the region of the intersection the
traffic unit resume normal activity. The proposed work sparkles in the way containing
182
multi-paths. At whatever point the said crisis vehicle go through way containing
different paths, just the activity signals for that path is controlled in order to not
turn out to be excessively nuisance general society.
(Samani & Zhi, 2016) Emphasizes the requirement for getting an outer robot
instance of extreme fatalities, for example, serious heart attacks. This paper focuses
on just the fatalities caused by heart attack. They have actualized anautomated
external defibrillator, with the goal that at whatever point the casualty experiences
a severe heart attack, quick medical aid can be given. The robot works utilizing
GSM network, a mobile application, a sensor connected to the casualty’s body and
a WAN to control and monitor the robot.
(Ali & Eid, 2015) Emphasizes the need to maintain a strategic distance from
traffic at whatever point a accident happens and such mischance must be advised to
the closest medical facility with the goal that the ambulances can be sent in a brisk
and convenient way. At whatever point a mischance is recognized the specialists will
be advised of the accident to take activities against the developing traffic congestion
because of the mishap, with the goal that the casualty can be transported to the
healing facility as quickly as time permits. The work uses fluffy logic to identify
accidents and in addition to reroute the traffic without making any problem the
overall population.
(Megalingam, Nair, & Prakhya, n.d.) Emphasizes on mishap identification and
answering to the closest crisis specialist organization so the specialist organization
can advise the same to the doctor’s facility and furthermore to the police. The sensor
used is accelerometer. The yield from the server is inputted to the PIC16F877A
microcontroller. The microcontroller transmits the information through RF
transmitter module to the Emergency Service supplier control room. The control
room is introduced with the RF recipient and at whatever point it gets data important
advances are taken to tell the police and the hospital.
(Amin, Jalil, & Reaz, n.d.) Emphasizes on identifying accidents and revealing
them utilizing GPS, GPRS and GSM tech. This work screens the speed of the
vehicle through the GPS framework. The GPS unit will continue contrasting the
speed of the vehicle and the speed of the vehicle one moment some time recently. At
whatever point the speed comes to beneath certain edge it will assume an accident
has happened and sends the warnings to the concerned specialists through GSM
and GPRS modules.
(Prabha, Sunitha, & Anitha, 2014) Emphasizes on automatic accident location
detection utilizing GSM and GPS module. This work has used piezoelectric sensor
to distinguish accidents. At whatever point a accident is identified, the data will be
passed on to the ARM7 microcontroller and the required messages will be sent to
the concerned experts which are stored in the microcontroller.
183
(Almadani, Bin-Yahya, & Shakshuki, 2015) Emphasizes on giving different

treatment to the patient inside the rescue vehicle. This is done if the patient condition
is critical and some kind of emergency treatment must be given to stabilize the
state of the patient. This work uses different biosensors to screen the state of the
casualty. These sensors form a network so the data can be gathered in a brief and
systematic way. This data is then sent to the control room introduce in the healing
center so medical care professionals can judge the state of the patient and instruct
the emergency vehicle faculty so that treatment can be given to the casualty.
As the fixation is on aversion and expectation of air contamination from vehicles,
we utilize two gas sensors i.e. MQ-7 and MQ-135. MQ-7 is an essentially as CO gas
sensor as CO is one of the primary substance in the vehicular fumes. It is a continuous
work where a demo application has been made in which Arduino processor is utilized
and a controller board is made where every one of these gadgets get incorporated
and work in like manner. The vehicle is controlled by this circuit. At the point when
a vehicle achieves certain edge contamination level then a SMS is created and sent
to the pre-characterized number put away in the memory through the GSM module
about the day and age which he has been apportioned for the adjusting of the vehicle.
The GPS module is utilized to find the vehicle position where it is stopped. This
paper shows a successful use of innovation by which we spare our condition by
controlling the contamination of vehicles.
The point of (Kulkarni & Teja, 2014) is that the system depends on a smart sensor
microcontroller with a network capable application processor that downloads the
toxins level to a PC for additionally processing. The system screens and transmits
the parameters of climatic condition to a command center (administrator’s server).
From this paper, we got the idea to transmit the information gathered from the
sensors to the administrator’s server, for further analysis and this would likewise
keep track about the proprietor’s vehicular discharges. This information gathered
from the sensors will be useful in information portrayals and rankings for the specific
districts/territories.
(Al-Ali, Zualkernan, & Aloul, 2010) gives us bits of knowledge about the power
administration for the system. Any sensor framework must be exceedingly efficient
as far as power utilization and management. The lesser the framework utilizes power;
the more extended time it can be used. Likewise, from this paper, we thought of the
possibility of having an android application which will give you routing between
regions not founded on traffic but rather in light of contamination levels. This thought
of having an android application gives an edge to this system as well as will help
the proprietor to have his very own records to vehicle’s emanation. This android
application will likewise be the methods for spreading the mindfulness about the
control of air contamination by our “Tip of the day” highlight of the system.
184
Derhorng et al. (2004) have proposed a Computer based reenactment and

representation instrument that discovered new algorithm for incident location and
systems for incident management. They examined circumstances, for example traffic
incident investigation utilizing different information mining techniques. From the
investigation, it created the impression that information mining gave the chance of
better understanding the effects of an occurrence, and was as an imperative instrument
for expanding effectiveness of their analysis.
Ying et al. (2006) have proposed a system on the Gary-Chicago-Milwaukee
(GCM) Corridor Transportation System, it was done to collect the interstate activity.
They had put intelligent sensors on roadways to gather real traffic streams data to
accomplish the idea of savvy streets and utilized the idea of distributed traffic stream
mining system. The objective was to outline the distributed activity stream mining
system to screen the present roadway.
Nicholas et al. (2011). In their exploration article they researched the chain of
events that prompt a crash in light of investigating tiny information. The advancement
of techniques independent of the event of the impact for street wellbeing had likewise
been explored by them. The examination was done for the most part utilizing decision
trees and k-means algorithm and in light of the outcomes designs were identified.
The clustering analysis yield prove that not all conflicts ought to be utilized as
surrogates for all crashes and indicated how gatherings of comparable clashes and
crashes could be recognized.
Wei et al. (2012) have proposed an Artificial Neural Network(ANN) that
was utilized to recognize the activity conditions in the rush hour gridlock data is
determined using the information mining technology. The approach they utilized was
Self-Organized Feature Map (SOFM) to bunch the traffic information through an
unsupervised learning and gave the names to these data. The order result demonstrated
that the concealed examples could be distinguished effectively by the SOFM and
consequently a dependable ANN model could be built with those marked groups.
Jayasudha et al. (2012) have proposed to decrease the death rate and increment the
rate of loss of lives by utilizing a few instruments, techniques or various algorithm
of information mining utilizing the activity information. They did the investigation
of different papers that have been done in accident databases by analysing them
with data mining techniques. They made use of sensors to assemble the points of
interest from the GPS and put away it in the database after which it was grouped,
identified and sent to the driver as an alert.
As per look into by Krishnaveni et al. (2008), in the year 2011, they utilized
characterization models to foresee the level of damage because of road accidents.
They looked at Naive Bayes Bayesian classifier, AdaBoostM1 Meta classifier, PART
Rule, classifier, J48 Decision Tree classifier and Random Forest Tree classifier for
arranging the level of damage of different car crashes. The primary goal of their
185
exploration was to discover the pertinence of information mining procedures to help

street car crash examination in avoiding and controlling accidents .Random Forest
calculation was observed to be the most productive.
Tibebe et al. (2007) they concentrated on developing adaptive regression trees to
build a decision tree to deal with street activity analysis and the level of seriousness
of the damage because of accidents. They did the investigation with the assistance
of the traffic information of Addis Ababa, Ethopia. They utilized coordinated data
mining for prediction and undirected data mining to distinguish the patterns.
Miao et al. (Miao Chong, 2004). The significant concentration was to decide the
seriousness of damage because of an accident by utilizing a practical model. This
research was for broadening earlier research about and effectively predicting the
seriousness and the most essential variables influencing the same amid an accident.
Of all these the decision tree approach was the best regarding precision and it elevated
the decision tree approach in vehicle information mining.
This paper concentrated on the significance of classification algorithm in prediction
of the vehicular impact patterns in the idiosyncrasy of grouping of crashes. Different
classification algorithm, for example, C4.5, C-RT, CS-MC4, Decision List, ID3,
Naïve Bayes and RndTree were connected in anticipating vehicle impact patterns.
Of all the arrangement systems utilized the arbitrary tree approach was the best and
another outcome acquired was that the classifier precision was enhanced when the
component choice strategy was connected with it. The Internet of Things, viewed
as “the following huge mechanical transformation”, is assessed to create about US
$746 billion income opportunity worldwide in the assembling business (counting car)
(Worldwide Internet of Things Spending by Vertical Market 2014-2018 Forecast,”
IDC, June 2014). The IoT is a intelligent system which encourages trade of data and
correspondence through the data detecting gadgets by interfacing everything to the
Internet. It accomplishes the objective of clever distinguishing, finding, following,
observing, and overseeing things (Stankovic, 2014) It is an augmentation and
development of Internet-based system and grows the correspondence from human
and human to human and things or things and things. In the IoT worldview, many
items encompassing us (for our situation, vehicles) will be associated into systems
in some frame. Customary practices and procedures in the vehicle business are
being changed always with the appearance of IoT. It has empowered car industry
to have coordinate contact with the clients by grasping the most recent innovation
and making vehicle overhauling a smooth assignment. In Predictive Maintenance,
the imperative vehicle data and cautioning alarms furnished through telematics
abilities are joined with other information focuses to empower propelled continuous
examination to distinguish patterns, foresee machine wellbeing, and more (Intel
Galileo2 versus Raspberry Pi2).
186
Vong et al. (2011) have connected IoT in vehicles to tell auto proprietors and
legislative experts of the emission data of the automobiles so they can get their cars
fixed before the compulsory vehicle emission test is done. Alam et al. (International
Conference on System Science and Engineering, 2011) have depicted the Internet of
Vehicles (IoV) as one of the key individuals from IoT. Utilizing this tech, to improve
safety, proficiency, infotainment and comfort related data with other vehicle and
infrastructures utilizing vehicular ad-hoc network (VANETs).
DESIGN AND IMPLEMENTATION OF

DIFFERENT CASE STUDIES
Case 1: E-Medical Emergency Monitoring System
A. Graphical User Interface
The following pictures depict the graphical user interface of the E-Medical Emergency
Monitoring App. Figure 2 shows the GUI of Ambulance Service Provider (ASP) side,
which mainly consists of the potential victims that might have met with accident.
Figure 3 shows the GUI of Hospital side of the app which shows the victim to be
admitted and Figure 4 shows the GUI of Ambulance side of the app
The android app consists of 3 modules namely: ASP part, Ambulance part and
the hospital part of the app. The algorithm for the work is depicted below.
Algorithm for the Collision Detection Module
• Step 1: Setup the sensors and the GPS/GPRS module on the Arduino Uno
board.
• Step 2: Monitor the sensor data continuously.
• Step 3: When the data collected by the sensors is beyond the collision
threshold activate the GPS/GPRS module.
• Step 4: Determine the location of the accident through the GPS module and
store it.
• Step 5: Through GPRS send the location of the victim as well as the details
of the victim to the real time database.
Algorithm for the Ambulance Service Provider (ASP) Part of the App
• Step 1: Log in to the app with the required credentials.

• Step 2: Give a unique username to any new ambulances if at all present.
187
Figure 1. Represents the workflow the system
Figure 2. GUI of ASP side of the app
188
Figure 3. GUI of Hospital side of the app
Figure 4. Navigation displayed on the Ambulance side of the app
189
• Step 3: Connect to the real-time database to detect if any accidents have

occurred.
• Step 4: When an accident is detected provide an option to call the victim for
conformation.
• Step 5: If the supposed victim confirms that he does not require emergency
service, abort the job.
• Step 6: If the victim confirms the emergency service requirement, or if there
is no reply from the supposed victim, assign an off duty ambulance to the
victim and notify the nearest hospital of the required number of beds.
Algorithm for the Ambulance Part of the App
• Step 1: Login to the app using the unique username provided by the ASP.
• Step 2: Locate the accident occurred place and navigate to the victim using
Maps.
• Step 3: After reaching the location determine the actual number of victims
and notify the hospitals of the same.
• Step 4: Drop the patient at the nearest hospital and drop the subsequent
victims to the next nearest hospitals, if any.
• Step 5: After dropping all the victims, abort the duty thus alerting the ASP as
well as the concerned hospitals about the completed job.
Algorithm for the Hospital Part of the App
• Step 1: Login to the app using the required credentials.

• Step 2: Whenever an alert is received from the ASP or other hospital, book
the required number of beds.
• Step 3: If less than required number of beds are available, then perform
partial booking.
• Step 4: If partial booking is selected, notify the next nearest hospital of
remaining required number of beds.
• Step 5: Steps from 2-4 carried out in the subsequent hospitals if at all partial
booking is selected by the subsequent hospitals.
Case 2: Prediction and Controlling of Air Pollution Caused

Due to Automobiles Using IoT and Data Analytics Techniques
The objective lies in reducing the air pollution and finding a technical solution for
the same. The project aims at enabling facilities such as:
190
1. Message intimation by the system to the owner of the vehicle for the service
that is required by the vehicle on excess of the carbon monoxide and/or nitrogen
oxides.
2. Warning message to be sent to the owner as a reminder to the inactivity towards
servicing of the vehicle.
3. A device to be developed that blocks the supply of fuel to the engine thereby
deactivating the vehicle if the owner fails to get the vehicle serviced after series
of intimations.
4. Ranking the various areas of the city based on the air pollution levels. Graphical
representations for various parts of the city based on various parameters should
be provided.
5. Predictions should be provided based on the current levels of the pollution and
certain precautionary measures to be suggested.
6. Routing of vehicles based on Air pollution at different locations.
The project overall has been done in three different modules. The three modules
are:
1. IOT module,
2. Data Analytics module, and;
3. Android app module.
The flow of the project in shown in Figure 5. The overall functioning of the
project is as follows:
The sensors i.e. MQ-7 and MQ-135 sense the outflow levels from the vehicular
fumes. These data are then sent to the Arduino for handling. In the event that the
emanation levels are underneath the predefined safe esteem or hasn’t crossed limit,
the system will again gather the following arrangement of values. In the event
that the emission levels gathered from sensors have crossed the limit, at that point
notification is sent to the proprietor of the vehicle about the surpassing discharge
level and is designated a time for getting his vehicle adjusted. Sensors are as yet
gathering the information from vehicle. In the event that the vehicle emission still
keeps on surpassing the limit, suggestion is sent to the proprietor as an update for
overhauling. Indeed, even after the second notification if emission level surpasses
the limit then his fuel supply is cutoff by the utilization of solenoid valve activating
it through a relay circuit. The information gathered by the sensors are sent to a server
for effective analysis.
Presently, data analytics module manages by giving the ranking of the territories
in light of the contamination levels in regions. This information is gathered from
the Karnataka State Pollution Control Board and Sunitha Narain, head of center for
191
Figure 5. Workflow of the system
science and Environment. Indian focal information gathered from Center for Science
and Environment. The information is analyzed and afterwards a ranking is created
for the zones. This ranking is utilized by the App module for navigation in view
of the contamination levels in regions. The administrator keeps up the most recent
year’s information and also the dynamic information gathered from the sensors and
has a dashboard which is available just to the head which has the diverse graphs
and tables. This dashboard has a great UI and is utilized for effective data analysis.
At long last, the Android App module is an application which is produced to
proposes a route for navigation between two locations. The route depends on the
192
level of contamination in the course and propose the most ideal briefest, slightest
polluted course. Application additionally has client validation and gives basic and
simple UI.
The Figure 6 shows the Arduino IDE where the CO and NO extraction code
are run and the output is seen through serial monitor which can be accessed in the
backend.
After the connection with the server, the information gathered from sensors are
sent to the web server for observing and storage of gathered information. The Figure
7 demonstrates the website page which will enable the admin to monitor and control
the system. The web server gives the data about the CO and NO on various days in
various areas. Data gathered from the sensors can be seen in visualizations. All the
required information is put away on the Cloud. We got Dataset from Karnataka state
contamination control board for the year 2015 of 6 unique regions in Bengaluru. The
stored information can be analysed at anytime and place. The Figure 5 demonstrates
the information gathered from these sensors and are region wise. And furthermore
it presents the date, NO, CO and PM emission values. This data is stored in the
AWS and is accessed by utilizing the administrator webpage. This is made open
just to the administrator.
Figure 6. Snapshot of Arduino IDE
193
Figure 7. Snapshot of web page
The graph in the Figure 8 gives the analysis of CO discharge level for the year
2015 month to month premise. This graph dipicts to the change in CO levels for
consistently. What’s more, in which month the CO discharges are low/high. The
diagram appeared in the Figure 8 gives the analysis of CO emission level for the
year 2015 month to month basis.
194
Figure 8. Visualizations on dashboard
An android application is additionally created for client’s. In which they login as

a client and can get the route between two locations in view of minimum pollution
levels. This depends on the information gathered. It additionally furnishes the time
span alongside the direction. Figure 9 demonstrates the guide which gives the route
in view of least pollution levels Notifications on approaches to diminish pollution
are sent every day to make awareness among individuals. And furthermore different
news, online journals are shared utilizing this application to lessen air pollution.
Case 3: Traffic Analysis, Prediction and

Congestion Control Using Data Mining
To analyse the traffic data we created a model and trained it using various machine
learning algorithms. The dataset is divided into two groups, the first group is utilized
to train the model and the second group is used to test numerous scenarios. Since,
the input form for WEKA is arff (attribute relation file format) or csv(comma
separated values), so the matrices are converted into csv format and then using the
arff loader option in WEKA the appropriate arff format is generated. The training
set has a string type column which indicate the congestion level by the string- hc
(high congestion), nc(no congestion) and mc(moderate congestion). To begin with
ZeroR, a machine learning algorithm which is the simplest classification method
relying on the target value was used. The below logic is used for the ZeroR algorithm:
Construct a frequency table for the target.
ZeroR:
195
Figure 9. Android App
1. Construct a frequency table for the target.

2. Select most frequent value from the table.
Post completing the baseline value using the ZeroR algorithm, we used the J48
algorithm, which is an open source Java Implementation. The logic behind this
approach has been detailed below:
196
J48:
1. Check for base cases.

2. For each attribute a. Find the normalized information gain from splitting on
a.
3. Let a best be the attribute with the highest normalized information gain.
4. Create a decision node that splits on a best.
5. Recourse on the sub lists obtained by splitting on a best, and add those nodes
as children of node.
The last algorithm used to train the model was Naïve Bayes, which for each
class, the mean and variance of every attribute of all instances belonging to that
class is computed.
Naïve Bayes:
1. for i 0 to no of classes do
2. Calculate prior probabilities.
3. end for
4. for i 0 to no of attributes do
5. Calculate mean and variance for all the tuples.
6. end for
7. Input test set.
8. for each tuple in the test set do
9. Compute the posterior probabilities for all the classes.
10. Classify the tuple into the class with the maximum posterior probability.
11. end for
The test set matrices have a string type column with a question mark (“?”) in all
rows. This is then fed as an input file to the prediction model generated earlier. The
model then predicts the value of the column with the “?” and fills in the appropriate
string- hc (high congestion), nc (no congestion) or mc(medium congestion) which
is in line with the logic fed into the training model. The predicted output of Weka
is converted back to normal .csv format. Dijkstra’s algorithm is used to find the
shortest path to the destination from the source in terms of distance among the
available routes.
197
Step 1: Assign to every node a tentative distance value: set it to zero for our initial
node and to infinity for all other nodes.
Step 2: Set the initial node as current. Mark all other nodes unvisited. Create a set of
all the unvisited nodes called the unvisited set. For the current node, consider
all of its unvisited neighbours and calculate their tentative distances.
Step 3: Compare the newly calculated tentative distance to the current assigned
value and assign the smaller one.
Step 4: When we are done considering all of the neighbours of the current node,
mark the current node as visited and remove it from the unvisited set. A visited
node will never be checked again. If the destination node has been marked
visited or if the smallest tentative distance among the nodes in the unvisited
set is infinity then stop. The algorithm has finished.
Step 5: Otherwise, select the unvisited node that is marked with the smallest tentative
distance, set it as the new “current node”, and go back to step 3.
After the evaluation results obtained from the initial analysis on Paramics the
next step involves capturing the map of New Bel road from open street map and
importing it on SUMO (Simulation of Urban Mobility), and analysing the traffic
conditions for this situation. The prediction should be accurate as that output will
be given as input to our self-written Route Selection algorithm.
SUMO allows us to simulate traffic on a real map and we can capture the traffic
simulations and analyse them. So the input for this will be screen grabs ofSUMO.
Stationary vehicles can also be generated in SUMO and those have also been taken
into consideration as they provide a distinct advantage for our project over google
maps where stationary vehicles are not considered while finding out the congestion
as infrared rays generated by cars are used to find the congestion. Also, priority
vehicles such as the emergency vehicles, authority vehicles are also shown with
the help of SUMO as to how they will reach the destination in case of congestion.
The two outputs of Weka and Dijkstra’s is used in our ROUTE SELECTION
algorithm which is used to find the shortest path to the destination based on the
distance and the congestion level so as to provide a less congested shorter route for
the vehicles. The logic behind the algorithm is detailed below:
1. Copy the output of WEKA in an array.

2. Divide the predicted value into no of routes.
3. for each route do
4. for i←1 to no of predicted values in each route do
198
5. Find which is maximum class label.

6. end for
7. Assign to the route, the congestion equal to label corresponding to maximum
obtained.
8. end for
9. if shortest route has high congestion compared to others
10. if distance difference with the consecutive shortest distances is less than double
11. take second shortest.
12. else take shortest
13. else if shortest routes has medium congestion and other routes no congestion
and distance difference less than double than
14. take other route.
15. else take shortest route
16. else take shortest route
The above logic is implemented in the output feed from the Dijkstra algorithm and
it displays the route to be selected by the simulated vehicle in case of congestion. In
this section, we have presented the experimental results of the proposed system for
traffic flow prediction. The implementation setup of the proposed system comprises
of, spatial data extraction from simulated images using MATLAB and prediction
done with ZeroR, Naïve Bayes and J48 classifiers using WEKA. The input to the
system is the simulated traffic generated in the PARAMICS and SUMO simulator
which is converted to matrix form and used for mining and predicting the congestion
level. The expected output is a matrix with the string type column filled by predicted
congestion level as indicated by- hc, nc or mc. If the prediction accuracy is more
than 90% the model passes the test stage.
Initially zeroR was used for training the classifier; it gives an accuracy of 90%
but it takesthe maximum value during prediction so cannot be used for prediction.
The maximum accuracy was achieved using J48 which gives above 93% during
prediction. Based on the results obtained for the simulation dataset the next real
data testing was approved.
The confusion matrix for prediction:
a b c ← classified as
413 1 11 | a = nc
5 65 5 |b = hc
14 2 56 |c = mc
199
In the confusion matrix, column a represents the count of ‘nc’, b represents count
of ‘hc’ and c represents count of ‘mc’.T he first row shows that 413 instances out
of 425 were correctly classified as nc, 1 as hc and 11 as mc. The second row shows
that 65 instances out of 75 were correctly classified as hc, 5 as mc and 5 as nc. The
third row shows that 56 instances out of 72 were correctly classified asmc, 2 as hc
and 14 as nc.
The ROC curve for the three classes hc, nc and mc is shown in Figures 10-12.
Once the simulated data was correctly classified, the output obtained is used in
our ROUTE SELECTION algorithm which is used to find the shortest path to the
destination based on the distance and the congestion level so as to provide a less
congested shorter route for the vehicles. The below layout of an actual street ‘New
BEL Road, Bangalore, India’ was imported into SUMO using OpenStreetMap.
Using SUMO, vehicles and pedestrian traffic was simulated into the imported
structure and multiple scenarios were designed. To test our Route Selection algorithm,
we implemented our algorithm for a priority vehicle to reach from Point A to Point
B and take the fastest route with less congestion.
Figure 10. Threshold curve for the class value nc
200
Figure 11. Threshold curve for the class value mc
Figure 12. Threshold curve for the class value hc
201
Figure 13. Vehicles and pedestrian traffic
Case 4: Data Analytics on Accident Data

for Smarter Cities and Safer Lives
A. Analysis of Severity of Accidents Based

on Consumption of Alcohol
In this module, we analyzed the seriousness of the accident because of DUI. The
value on the X-pivot relate to the severity of accidents being highly, moderate and
less severe. The Y-pivot demonstrates the number of accidents which have happened
because of the DUI and non-consumption of liquor. The conclusion which is derived
from this is that greater part of the situations when the driver of the vehicle has
consumed alcohol, the severity was more, which would have caused accidents. The
number of severe accidents which have happened because of DUI is almost double
the number of serious accidents happened due to the non-utilization of alcohol.
Figure 15 demonstrates the graph for this analysis.
B. Analysis of Severity of Accidents Based on the Condition of Light
In this way, accidents were higher in number and were more extreme, when it was
pitch-dark and road lights were missing (more noteworthy than 40,000 in number).
The X-axis demonstrates the different light conditions, for example, total darkness
202
Figure 14. Workflow of the system
Figure 15. Analysis of Severity of Accidents based on consumption of alcohol
with no road lights, day break, sunset and evening. The number of accidents caused
amid the nightfall time is more than the mishaps amid the sunrise and evening. The
number of direct accidents happened amid the first light and nightfall are especially
considerable as well as they are substantial in number.
203
Figure 16. Analysis of Accidents based on the light conditions
C. Analysis of Severity of the Accidents Based on the Speed Zones
The value 1, 2 and 3 on the X-axis relate to the severity of the accidents being high,
modestly and less severe. The number of serious accidents happens in the speed zones
where admissible driving speed is 60 km/hr. The reason could be due to excessive
number of vehicles moving in those zones. The number of accidents occurred in the
zones where admissible speed is about 50 and 80 km/hr. is likewise considerable.
The greater part of moderately severe accidents happens in the zones where as far
as possible is 70 km/hr, trailed by 40 km/hr. One imperative derivation from this is
that most accidents don’t happen in the fast zones however they are in the normal
speed zones. The number of less severe accidents is relatively very lesser in number.
D. Decision Tree (Manjunath, Nikschal, Mohta & Sindhuja,

2016) to Find the Severity of the Accident Using Consumption of
Alcohol and Light Condition as the Various Independent Factors
In this module, the decision tree algorithm was executed to observe the reason for
mishaps in the metropolitan regions. The alcohol time was used as the root node, if
the liquor is consumed and there is any person on foot on the road, the severity of
accidents turns out to be high, if not, it is low. The right side compares to the non-
204
Figure 17. Analysis of Severity based on the speed zones
utilization of alcohol. On the off chance that the state of light is 0, which implies
that it is totally dark and the severity is high if more than one individual has been
injured; in the other case it has low severity. In the event that the light condition
is 2, 3 or 4, which implies that amid nightfall, day break and evening, the mishaps
are moderately serious.
Figure 18. Decision Tree to find the severity of the accident using consumption of
alcohol and light condition
205
The decision tree algorithm makes use of entropy which is computed utilizing
a formula, indicated prior. Utilizing the value of the entropy, we find the largest
information gain value. The information gain depicts to only the degree to which the
prediction of the events happens appropriately utilizing the technique or algorithm
utilized for training the machine. The comparison of the values of the deviations for
the module 4.4 is depicted in the table below. The value of the deviations diminished
from 0.4337385 to 0.4247297 after performing pruning and optimizing the tree.
CONCLUSION
The analysis of the accident data is performed quantitatively and qualitatively. The
sequences of conditions which lead to severe accidents are examined using the
decision trees. Also, regions of high density accidents, the severity of the accident
and the attributing factors were examined. The project in its early iteration can glean
insight into locations which are accident prone and the causes for it. This predictive
model can be implemented in several mobile applications which can minimize risk
of accidents. It also attempts to research the possible causes of accident and the
possible solutions to them. And provides scope for improved traffic scenario and
gives users such as daily commuters and ambulances, an option to take alternate
routes and save time and life. The initial video generated using Paramics was used
as a base case for checking the efficiency of the classifier. The simulated data
resulted in an accuracy of above 90% so the next step was to work on a real map
and implement a method for finding alternate routes. The map of New BEL Road
was imported from OpenStreetMap to SUMO. Dijkstra algorithm, which finds
out the shortest routes from a starting point to destination was used along with
our ROUTE SELECTIONAL GORITHM, to find out all the alternate routes and
select the one with less congestion. This method can be applied at a larger scale
to real time traffic data in order to analyse and predict congestion using data from
the Traffic Management Centre. The entire process can be made automatic along
with deploying sensors at the junctions in order to collect the real time vast data
and do the analysis on it. Also, the final output achieved can be either given back
to the Traffic Management Centre which they can display on bill boards in order
to provide information to the drivers or a mobile app can be designed which allows
users to enter their destination and directs them to faster, congestion free routes.
206
REFERENCES
Al-Ali, Zualkernan, & Aloul. (2010). Mobile GPRS-Sensors Array for Air Pollution
Monitoring. IEEE Sensors Journal, 10(10).
Alam. (2015). Towards Social Internet of Vehicles: Concept, Architecture and
Applications. IEEE.
Ali & Eid. (2015). An Automated System for Accident Detection. Instrumentation
and Measurement Technology Conference, 2015 IEEE Conference.
Almadani, B.-Y., & Shakshuki. (2015). E-Ambulance: Real-Time Integration Platform
for Heterogeneous Medical Telemetry System. 5th International Conference on
Current and Future Trends of Information and Communication Technologies in
Healthcare.
Amin, S. (n.d.). [Accident Detection and Reporting System using GPS, GPRS and
GSM technology. Informatics, Electronics & Vision.]. Jalil, & Reaz.
Barai. (2003). Data Mining Applications In Transport Engineering. Retrieved from
http://tf.nist.gov/seminars/WSTS/PDFs/10_Cisco_FBonomi_ConnectedVehicles.
pdf
Chong, Abraham, & Paprzycki. (2004). Traffic Accident Analysis Using Decision
Trees and Neural Networks. Academic Press.
Der-Horng, Shin-Ting, & Chandrasekar. (2004). Applying data mining techniques for
traffic incident analysis. Journal of The Institution of Engineers, Singapore, 44(2).
Gotadki, Mohan, Attarwala, & Gajare. (2014, April). Intelligent Ambulance.
International Journal of Engineering and Technical Research.
He, Lu, & Wang. (2012). A New Method For Traffic Forecasting Based On Data
Mining Technology With Artificial Intelligent Algorithms. Research Journal of
Applied Sciences, Engineering and Technology.
Jadhav, Satam, & Salvi. (2016, February). An Android-Based Emergency Alarm
Message and Healthcare Management System. International Journal of Advanced
Research in Computer and Communication Engineering.
Krishnaveni & Hemalatha. (2011). A perspective analysis of traffic accident using
data mining techniques. Academic Press.
207
Kulkarni & Teja. (2014). Automated System for Air Pollution Detection and Control
in Vehicles. International Journal of Advanced Research in Electrical, Electronics
and Instrumentation Engineering, 3(9).
Liu, Choudhary, Zhou, & Khokhar. (2006). Distributed Stream Mining System for
Highway traffic. Data (København).
Manjunath, N., & Nikschal, V. M. (2016, May). Design of an Automated Traffic
Control System for Emergency Vehicle Clearance. International Journal of Emerging
Research in Management and Technology.
Megalingam, Nair, & Prakhya. (n.d.). Wireless Vehicular Accident Detection and
Reporting System. Mechanical and Electrical Technology.
Mell, P., & Grance, T. (2009). The NIST Definition of Cloud Computing. Retrieved
from http://www.nist.gov/itl/cloud/upload/cloud-def-v15.pdf
Moje, Kumbhar, Shinde, & Korke. (2016, April). Automatic Ambulance Rescue
System. International Journal of Innovative Research in Electrical, Electronics,
Instrumentation and Control Engineering.
Oh, Lee, & Kote. (2003). Real Time Video Data Mining For Surveillance Video
Streams. Academic Press.
Prabha, Sunitha, & Anitha. (2014). Automatic Vehicle Accident Detection and
Messaging System using GSM and GPS modem. IJAREEIE, 3(7).
Srinivasa Prasad & Ramakrishna. (n.d.). An Efficient Traffic Forecasting System
Based on Spatial Data and Decision Trees. The International Arab Journal of
Information Technology, 11(2).
Roseline, Devapriya, & Sumathi. (2013). Pollution Monitoring using Sensors and
Wireless Sensor Networks: A Survey. International Journal of Application or
Innovation in Engineering & Management, 2(7).
Samani & Zhi. (2016, January 4). Robotic Automated External Defibrillator
Ambulance for Emergency Medical Service in Smart Cities. IEEE Access : Practical
Innovations, Open Solutions.
Sangeetha, Archana, Ramya, & Ramya. (2014, February). Automatic Ambulance
Rescue with Intelligent Traffic Light System. IOSR Journal of Engineering.
208
Saunier, N. (2011). Investigating Collision Factors by Mining Microscopic Data on

Vehicle Conflict and Collisions. Academic Press.
Shanthi & Ramani. (2011). Classification of Vehicle Collision Patterns in Road
Accidents Using Data Mining Algorithms. Academic Press.
Stankovic, J. A. (2014, February). Research directions for the Internet of Things.
IEEE Internet Things J., 1(1), 3–9. doi:10.1109/JIOT.2014.2312291
Tesma. (2014, March). Rule Mining And Classification Of Road Traffic Accidents
Using Adaptive Regression Trees. Academic Press.
Vong. (2011). Framework of Vehicle Emission Inspection and Control through RFID
and Traffic Lights. Proceedings of.
Yu, M., Zhang, D., Cheng, Y., & Wang, M. (2011). An RFID Electronic Tag based
Automatic Vehicle Identification System for Traffic IOT Applications. Chinese Control
and Decision Conference (CCDC), 4192-4197. 10.1109/CCDC.2011.5968962
209
210
Chapter 10
IOT and Data Analytics
Solution for Smart Agriculture
Sowmya B. J.
Chetan Shetty
Netravati V. Cholappagol
Seema S.
ABSTRACT
This chapter gives the real-time solutions to the farmers by providing smart solutions
for irrigation, disease monitoring, and decision supporting systems (which involves
giving suggestions and solutions to the farmers by monitoring soil conditions, rain,
weather, and overall quality of crop growth and the effect on the growth of the crop
due to infertile soil or bad climatic conditions). These solutions are provided using
the IOT and data analytics technology.
INTRODUCTION
In the farming, field crop necessities management from claiming pesticides, fertilizers.
What’s more watering system for finer growth, precision agriculture (PA) will be.
The idea utilized for the same design. The data for example,. Temperature, humidity,
fertilizer, and soil moisture might a chance to be furnished. To provide decision
DOI: 10.4018/978-1-5225-6070-8.ch010
IOT and Data Analytics Solution for Smart Agriculture
support system for maximizing the crop growth development with the optimized
utilization of accessible assets and without. Influencing an nature’s domain.
Those sensor hubs set in the ranch for data gathering ought to dependably think
about Vitality utilization. Information sensed by sensors is send with incorporated
substance to examination and upgrade and comes about of the Investigation need
aid communicated back on to Ranch staff.
We proposed an automated prediction version which analyzes the massive
information units of ancient records the use of Big data analytics. Large information
analytics is the procedure of inspecting massive amount of facts comes from variety
of resources like sensors facts, weather forecasting, and social media information
with sort of codecs to find the hidden patterns, unknown correlations and also useful
precious information. Economic models generally estimate changes in marketplace
developments and situations underneath climate variant. The venture is aiming to
discover solutions to troubles, consisting of climate change-precipitated global meals
lack of confidence, to predicting and mitigating the impact of severe weather events
on worldwide finance. On the grounds that we’ve sensors everywhere in area which
are used to display and degree climate, plants, cloud cowl, ice cover, precipitation,
sea floor temperature, and many greater geophysical parameters. those wide-ranging
information collections give us an increasing number of deeper and broader coverage
of weather trade, both temporally and geospatially.
As smart machines and sensors manifest on farms and cultivate information develop
in amount and degree, cultivating procedures will turn out to be progressively data
driven and information empowered. Fast improvements in the Internet of Things
and Cloud Computing are driving the marvel of what is called smart Farming While
Precision Agriculture is simply considering in-field fluctuation, Smart Farming goes
past that by constructing administration undertakings with respect to area as well
as on information, improved by setting and circumstance mindfulness, activated
by constant occasions. Ongoing helping reconfiguration highlights are required to
do lithe activities, particularly in instances of all of a sudden changed operational
conditions or different conditions (e.g. climate or malady caution). These highlights
ordinarily incorporate insightful help in execution, support and utilization of the
innovation. Figure 1 condenses the idea of Smart Farming along the administration
cycle as a Cyber physical framework, which implies that shrewd gadgets - associated
to the Internet - are controlling the ranch framework. Keen gadgets expand regular
apparatuses (e.g. rain measure, tractor, scratch pad) by including independent
setting mindfulness by all sort of sensors, worked in knowledge, skilled to execute
independent activities or doing this remotely. In this photo it is as of now proposed
that robots can play an essential part in charge, however it can be normal that the
part of people in investigation what’s more, arranging is progressively helped by
211
Figure 1. The cyber-physical management cycle of Smart Farming
machines so that the cyberphysical cycle turns out to be relatively self-ruling.

People will dependably be engaged with the entire procedure yet progressively at a
considerably higher knowledge level, leaving most operational exercises to machines.
RELATED WORK
In GSM Based Automated Irrigation Control using Rain gun Irrigation System
(Suresh, Gopinath, Govindaraju, Devika, & Vanitha, 2014) utilizing programmed
microcontroller based rain weapon water system framework in which the water
system will happen just when there will be extraordinary prerequisite of water that
spare an expansive amount of water. These framework secured bring down scope of
farming area and not monetarily reasonable. The System Supports abundance Amount
of water in the land and uses GSM to send message and an android application is
been utilized they have utilized a technique to overcome under water system, over
water system that causes draining and loss of supplement substance of soil they
have additionally guaranteed that Microcontroller utilized can expand System Life
and lower the power Consumption. There framework is simply restricted to the
robotization of water system framework and needs in additional common highlights.
In GSM based Automatic Irrigation Control System for Efficient Use of Resources
and Crop Planning by Using an Android Mobile (Pavithra & Srinath, 2014) describes,
• The framework bolsters water administration choice, utilized for observing

the entire framework with GSM(RS-232) module.
212
• The framework ceaselessly screens the water level (Water level Sensor) in
the tank and gives exact measure of water required to the plant or tree (trim).
• The framework checks the temperature, and stickiness of soil to hold the
supplement organization of the dirt oversaw for legitimate development of
plant.
• Low cost and viable with less power utilization utilizing sensors for remote
observing and controlling gadgets which are controlled by means of SMS
utilizing a GSM utilizing android portable.
Farming in an India is subject to summer precipitation climate. The impact of

precipitation in summer on generation of agribusiness crops is examined in (Kumar,
2004). In this paper, the past harvests information is broken down to analyze trim
atmosphere relationship. From the outcomes said in this paper, it can expressed that
three elements can specifically make the impact on development of horticulture items
in India, for example, storm precipitation, temperature of pacific and Indian seas,
and weight of ocean level. Besides, the outcomes demonstrate generation rate and
measure of storm precipitation all finished India is consistent, barring a few cases.
For forecast utilizing KF, sensor hubs need to shape set of bunches. The bunch
head will get information about various parameters of the condition from leaf hub
who is mindful to gather and transmit it to head. In PKF, it is accepted that bunch
head can anticipate current information sent by leaf hub with adequate mistake
utilizing information got by leaf hub already. The PKF disguise discussion between
a leaf hub and its group head as for time (Huang, Yu, Osewold & Garcia-Ortiz,
2016). Commonly information gathered by leaf hub is influenced by clamor, so it
is required to expel commotion from this information. To expel the clamor from
the detected information each leaf hub utilize KF channel and create best comes
about. To decrease vitality utilization required for correspondence, bunch head
infers Kalman ideal esteems for leaf hub utilizing basic predicator PKF. The leaf
hub executes predicator at customary time interim by following perception caught
by group head and contrasts those qualities and its ideal incentive to ensure precise
predication investigation. In the event that the forecast mistake is more noteworthy
than guaranteed edge, the current insignificant esteem is going to the bunch head. The
match of predicators with ideal esteem is utilized by PKF. Note that, this indicator
is proportional to a k-advance ahead Kalman indicator.
An example space is recursively divided utilizing classifier known as Decision
tree. The Decision tree is coordinated tree established tree in which there is one
fundamental root hub which has no approaching edges. Every other hub have precisely
one approaching edge. To assemble the choice tree, a discrete capacity with the
contribution of a quality esteem is utilized where occasion space is isolated into at
least two sub-spaces by the inner hub (Rokach & Maimon,n.d.).
213
In Irrigation Control System Using Android and GSM for Efficient Use of Water
and Power (Pasha & Yogesha, 2014). C Automated irrigation system uses valves to
turn motor ON and OFF. These valves may be easily automated by using controllers.
Automating farm or nursery irrigation allows farmers to apply the right amount of
water at the right time, regardless of the availability of labor to turn valves on and
off. In addition, farmers using automation equipment are able to reduce runoff from
over watering saturated soils, avoid irrigating at the wrong time of day, which will
improve crop performance by ensuring adequate water and nutrients when needed.
Those valves may be easily automated by using controllers. Automating farm or
nursery irrigation allows farmers to apply the right amount of water at the right time,
regardless of the availability of labor to turn valves on and off.
They lack in a featured mobile application developed for users with appropriate
user interface. It only allows the user to monitor and maintain the moisture level
remotely irrespective of time. From the point of view of working at remote place the
developed microcontroller based irrigation system can work constantly for indefinite
time period, even in certain abnormal circumstances. The factual gauging models
can be created to appraise the future climate conditions. The model is basic because
of the way that it employments simple mathematical condition utilizing Multiple
Linear Relapse (MLR) to anticipate most extreme temperature, least temperature and
relative mugginess (Paras & Mathur, 2012). Machine learning strategies accomplished
preferable execution over conventional measurable strategies. So present Support
Vector Machines (SVM) for climate expectation (Radhika & Shashi, 2009) and
improved bolster vector relapse demonstrate with more un-interpretable piece works
in the space of estimating the climate conditions (Usha Rani & Rao, 2015). IoT based
farming creation framework observing framework to examine edit condition and
furthermore gives strategy to enhance the effectiveness of choice making by dissecting
harvest measurements, conjecture farming generation utilizing IoT sensors (Lee,
Hwang, & Yoe, 2013). Expectation demonstrate in light of large scale atmosphere
(Septiawan, Komaruddin, & Budi, 2012) depends on recorded information and an
investigative calculation with field observing and traceability display. Neighborhood
time arrangement investigation advance factual model furthermore, NWP presents
dynamical model of climate forms, another approach joins these two strategies and
presenting cross breeds demonstrate. NWP based precise limited here and now
climate forecast framework (Corne Dissanayake & Peacock, 2014) foreseeing the
estimations of meteorological variable. Another Prediction demonstrate in view of
enormous information examination executes cross breed strategy which improves
a FCM bunching calculation for complex qualities of the enterprises (Yang & Kim,
2014).
To choose the yield and to anticipate creation rate of the product counterfeit neural
system utilize data gathered by sensors from the homestead. This data incorporates
214
parameters, for example, soil, temperature, weight, precipitation, and moistness.

The impact of these parameters on trim development is examined, and comes about
are assessed in paper (Kumar, Kumar, Ashrit, Deshpande, & Hansen, 2004). It is
watched that the climatic parameter, soil write, and soil organization can affect
on creation rate of harvest. The strategy examined in this paper moreover predicts
reasonable harvest generation rate ahead of time. Artificial neural networks are
utilized as viable apparatus for displaying and forecast of product creation rate and
enhance the exactness of yield expectation. For anticipating generation of rice trim, an
astute apparatus is produced with the assistance of machine learning approach. This
apparatus is utilized as a part of arrangement and grouping (Huang, Yu, Osewold,
& Garcia-Ortitz, 2016). Rice ranch information are arranged utilizing Bolster vector
machine learning method. Part based bunching calculation is utilized for discovering
bunch in atmosphere information. To control perplexing, high dimensional and
non-straightly distinguishable information Kernel based techniques are relevant.
The effects of different impacting parameters on the rice yield are assessed utilizing
connection investigation what’s more, utilizing relapse investigation forecast about
product yield rate is finished. Support vector machine is utilized for uproarious
information. Because of every one of these highlights strategy from this paper
utilized as a wise framework for foreseeing rice yield.
Irrigation by help of freshwater resources in agricultural areas has a crucial
importance. Traditional instrumentation based on discrete and wired solutions,
presents many difficulties on measuring and control systems especially over the large
geographical areas. Cost effective solar power can be the answer for all our energy
needs. Conserves electricity by reducing the usage of grid power and conserves water
by reducing water losses. Discourage weeds, saves water and time, statistical data can
be used to control diseases and fungal growth, simplest model. This system is just
limited to the automation of irrigation system and lacks in extra ordinary features.
IMPLEMENTATION AND RESULTS
Case 1: Weather Forecast Analysis
Different Sensors are used here:
1. DHT11 Temperature and Humidity Sensor
The DTH11 is a primary, ultra-low-cost digital temperature and humidity sensor. it

makes use of a capacitive humidity sensor and a thermistor to degree the surrounding
air, and spits out a virtual sign at the information pin (no analog input pins wished).
215
It’s pretty easy to apply, but requires careful timing to seize information. the most
effective real drawback of this sensor is you can best get new information from it
once each 2 seconds, so whilst using our library, sensor readings can be up to 2
seconds old.
2. BMP180 Barometric Pressure/Temperature/Altitude Sensor- 5V Ready
This precision sensor from Bosch is the best low-cost sensing solution for
measuring barometric pressure and temperature. Because pressure changes with
altitude you can also use it as an altimeter! The sensor is soldered onto a PCB with
a 3.3V regulator, I2C level shifter and pull-up resistors on the I2C pins.
3. HC-05 - Bluetooth to Serial Port Module
Serial port Bluetooth, Drop-in replacement for wired serial networks, translucent
usage. You can use it simply for a serial port replacement to establish connection
between MCU and GPS, PC to your surrounded project and etc. And now, we provide
HC-05 and HC-06. HC-05 could be setting to Master or Slave by user. HC-06 has
be designed Master or Slave when the factory, user couldn’t change the role.
4. ESP8266 Wi-Fi Module
This component is very low cost, but very simple. It doesn’t provision SSL or
communication over SPI - just UART! It doesn’t have 5V to 3V logic level unstable.
so you’ll probably want to pick up a logic level shifter, it also doesn’t have a regulator
on board and it can use big spikes of 300mA or more current at 3.3V so if using
with an Arduino, an external 3V regulator is essential
5. Arduino Ethernet Shield
The Arduino Ethernet Shield permits an Arduino board to attach to the internet.
It is based on the Wiznet W5100 ethernet chip (datasheet). The Wiznet W5100
delivers a network (IP) stack accomplished with both TCP and UDP. It supports up
to four simultaneous socket connections. Use the Ethernet library to write drawings
which connect to the internet by means of the shield. The Ethernet shield joins to an
Arduino board using extended wire-wrap headers which extend through the shield.
This keeps the pin layout intact and allows another shield to be stacked on top.
This includes c programming with Arduino Uno sensor which is compatible with
many packages to implement the project .This employs native interfacing technique
to connect components.
216
1. Thing Speak API
The user can log into the account on thingspeak. He can revise his channel
immediately if he makes his ravine inn for this the use requires any invention such
as a pc, laptop etc with an internet association. The use can see the charting of the
temperature, moisture and pressure as well as the detail computed by the thingspeak
app
2. Android App
A first-time customer of the mobile request should see a page with two buttons
one enables him to list the Bluetooth device in the area of which he can pair with
the weather station by selecting it and then he can connect with it using the connect
button .when he/she connects the application starts to receive and displays it in the
textbox. If the user is satisfied with the reading he/she should be able to disconnect
from the device through the disconnect button which now occupies the position
previously occupied by connect button. The following is a sample code for the user
Interface
voidstartEthernet()
{
client. stop();
Serial.println (“Connecting Arduino to network...”);
Serial.println ();
delay (1000);
// connect to network and obtain an IP address using DHCP
if (Ethernet.begin(mac) == 0)
{
Serial.println (“DHCP Failed, reset Arduino to try
Again”);
Serial.println ();
}
else
{
Serial.println (“Arduino connected to network using
DHCP”);
Serial.println ();
}
delay(1000);
217
The following Figures 2 and 3 shows the graph which shows the humidity and
temperature variation in a particular date using the user interface Thingspeak.
It was found that the latest line of production of the former is Hardware incompatible
with the Arduino Uno. This is because the module runs at Baud rate much higher
than that compatible with Software Serial Communication. The alternative Ethernet
shield is used to communicate over the internet. The Thingspeak API seems to be
highly efficient with successful visualization. The communication with the android
app also was successful in rendering the sensor data in text format. The following
Figure 4 shows the Aurdino interface for the farmers.
Case 2: Smart Agriculture System
In this Information is collected from storage system and field through sensors which
are interfaced with raspberry pi, the information acquired is sent to cloud. The data
or information is stored in cloud then which is mailed to the farmer.
Figure 2. Graphical Visualization on Thingspeak
218
Figure 3. Graphical Visualization on Thingspeak
Different Modules
Ubidots
Ubidots offers a REST API through which the Data Sources, Variables and Values,
can be generated, read, modified or erased.It is open source application where, which
provide us the feature to store data on the cloud and also giving us the feature to
send data on the mobile or email to the particular user.
Email
Email is used to receive data from system which is being sent from the cloud and
the user could see the message and take appropriate steps.
Putty
PuTTY is a free and open-source terminal emulator, serial console and network
file transfer application. It supports many network protocols, including SCP, SSH,
219
Figure 4. Various parameter information
Telnet, login and raw socket connection. It can also provide connection to a serial
port. The name “PuTTY” has no definitive meaning. PuTTY was originally written
for Microsoft Windows, but it has been ported to various other operating systems.
Authorized ports are available for some Unix-likeplatforms, with work-in-progress
ports to Classic Mac OS and Mac OS X and unauthorized ports have been contributed
to platforms such as Symbian,Windows Mobile and Windows Phone.
Implementation Using the Sensors
The motion detector sensor detects the presence of infrared in its surrounding areas
thus detect the presence of the living organism in the field. The soil moisture sensor
is being used to detect the presence of the humidity in the soil. The information
about the temperature and humidity sensor is given by DTH 11. The data is being
analysed and we have used the UBIDOTS where we are creating variables and
putting up the condition based on which email is being sent to the user and so that
the user is being notified about the environment and its surrounding.
220
Figure 5. Collection of Information
Sample Algorithm for the Implementation
Read temperature_sensor()
Read humidity_sensor()
Read motion_sensor()
Read soil_moisture_sensor()
Upload Data to Cloud
If temperature_sensed>20:
Notify the user
If motion_sensed ==1:
Send email to user about the
intruder
If soil_moisture_sensed ==0
condition of the land
If humidity_sensed>80
humidity of storage system
221
In this algorithm we read the data from the sensor and that data is send to the
raspberry pi. In raspberry pi we will decide whether or when to send data into the
cloud. The data is uploaded to the cloud (ubidots). Ubidot offer us to create some
event on the basis of data uploaded on the cloud. On the basis of the condition we
will notify the farmer by sending alert message to the cloud.
Information About the Implementation of Modules
In this project we have used four modules in which we are checking the condition
on the data received on the cloud. The four modules are:
1. Sensing Data from the Sensor: In this module we are checking the pins
voltage and based on the data we send data.
2. Sending the Data to the Cloud: Checking the condition for the data and
sending data to the cloud.
3. Creating Event: On the cloud we are creating some event.
4. Notify the User: On the basis of event we notify the user.
Figure 6 describes the readings of temperature from the sensors

Figure 7 shows the data acquired from motion sensors.
Figure 8 shows the humidity readings from the sorrounding.
Figure 9 shows the moisture content of the soil.
Figure 6. Reading data from DTH11
222
Figure 7. Reading presence of animals
Figure 8. Reading data from DTH11
223
Figure 9. Reading data from Soil
Case 3: IOT Based Smart Irrigation System

using Sunflower Tracking Model
The Sunflower Model
Solar tracking will be done with the help of Stepper motor. Microcontroller will
be programmed with 1.8 degree step angle to control the rotation of stepper motor.
Output of microcontroller sends to buffer for temporary storage, stepper motor
driver to drive the stepper motor. Solar panel will be placed upon the stepper motor.
Number of Modules
• Temperature Sensors
• Moisture Sensors
• Conductivity Sensors
• Water Sensors
• DTMF or the Mobile unit
• Solar panel
224
Modules Description
• Temperature Sensors: A Temperature Sensors is a type of controller whose

resistance fluctuates significantly with temperature, additionally we require
standard resistors. It adapts the crops to appropriate temperature required for
its proper growth.
• Moisture Sensors: Soil moistness sensors measure the water content in soil.
A soil moisture probe is made up of multiple soil moisture sensors. If the
soil moisture content is less, then the soil moisture sensor will command
the microcontroller to moisturize the soil such that the crops recollects its
fertility and stay healthy.
• Conductivity Sensors: The Conductivity sensors are located in each and
every corner of the field such that it checks whether the land is irrigated
entirelyor not and if not, then it will send a message to the concerned user
via the microcontroller. In short they are used to check the water conductivity
overall the field to be irrigated.
• Water Sensors: The Water sensors are used to detect the level of water
necessary for the crop. If the level of water is less than required then it
commands the microcontroller to provide the necessary water. If the level of
water reaches maximum, then it commands the microcontroller to flip a sheet
horizontally such that the excess water runs on the sheet and gets collected
in a dig.
• DTMF or The Mobile Unit: DTMF stands for “Dual Tone Multiple
Frequency” signal. The DTMF signal is an arithmetical sum of two different
frequencies; those are higher row frequency group and Lower column
frequency group. The DTMF technology will be having Encoder and Decoder.
The Encoder will be used to convert Digital signal to electrical signal. The
Decoder will be used convert the electrical signal to digital signal.
• Solar Panel: Solar panel refers to either a photovoltaic module, a solar thermal
energy panel, or to a set of solar photovoltaic (PV) modules electrically
joined and mounted on a supporting structure. A PV module is a packaged,
connected assembly of solar cells.
Irrigation method uses valves to turn irrigation ON and OFF. These valves may
be definitely computerized by using checkers and solenoids. Automating farm or
nursery irrigation permits farmers to apply the correct amount of water at the accurate
time, irrespective of the availability of labor to turn valves on and off. In addition,
farmers using automation equipment are able to reduce runoff from over watering
saturated soils, avoid irrigating at the wrong time of day, which will improve crop
performance by ensuring sufficient water and nutrients when needed. Automatic Drip
225
Irrigation is a valuable tool for accurate soil moisture regulator in highly specialized
greenhouse vegetable production and it is a simple, precise method for irrigation.
It also supports in time saving, removal of human error in modifying available
soil moisture altitudes and to get the best out of their net profits. Irrigation is the
artificial application of water to the soil usually for assisting in growing crops. In
crop production it is mainly used in dry areas and in periods of rainfall shortfalls,
but also to protect plants against frost. Types of Irrigation are
• Surface Irrigation
• Localized Irrigation
• Drip Irrigation
• Sprinkler Irrigation
• Power Supply Unit: This section necessities two voltages viz., +12 V &
+5 V, as working voltages. Hence particularly designed power supply is
constructed to get controlled power supplies.
• Temperature Sensor (Thermistor): A thermistor is a type of regulator
whose resistance fluctuates significantly with temperature, more so than in
standard resistors. The word is a portmanteau of thermal and resistor.
• Moisture Sensor: Soil moisture sensors measure the water content in soil. A
soil moisture probe is made up of multiple soil moisture sensors.
• Solar Panel: Solar panel refers either to a photovoltaic module, a solar
thermal energy panel, or to a set of solar photovoltaic (PV) modules
electrically connected and mounted on a supporting structure. A PV module
is a packaged, connected assembly of solar cells.
• DTMF Decoder:The DTMF decoder used is CM8870. It is used to decode
the mobile’s audio signal, i.e., the keypad tone. When the user presses a
button in the keypad of the mobile, it generates two tones at the same time.
These tones are taken from a table comprising of 2 frequency one is row
frequency and other is column frequency.
• Relays: It is a electromagnetic device which is used to drive the load connected
through the relay and the output of relay can be connected to controller or
load for further processing.
• Indicator: This stage offers visual suggestion of which relay is activated and
deactivated, by glowing respective LED or Buzzer.
• DTMF or The Mobile Unit: DTMF stands for “Dual Tone Multiple
Frequency” signal. The DTMF signal is an algebraic summation of two
different frequencies; those are higher row frequency group and Lower
column frequency group. The DTMF technology will be having Encoder and
Decoder. The Encoder will be used to convert Digital signal to electrical
signal. The Decoder will be used convert the electrical signal to digital signal.
226
This is used to monitor the field to achieve appropriate irrigation by the help of
wireless sensors. Temperature sensor, moisture sensor and conductivity sensor is
used monitor the field with respect to that specific condition. If any factor of the
field varied that will be noticed by individual sensor and sensed signal will be send
to buffer for temporary storage, driver to drive the relay and relay for switching via
monostable multivibrator. Relay starts the DC engine to irrigate the field properly.
Arduino Uno collects the indication which was monitored from sensor, analyses and
activates GSM module to send information to concern person as a communication
and to update about observed factor to concern person Gmail server to attain IOT.
Water sensor will be used to display the water level and if the water content exceeds
the requirement then controller starts the flip to achieve rain harvesting. This module
will be having mobile component to receive the call from concern person after two
rings because that unit will be kept in automatic receive mode. After receiving the
call the person has to activate the key to turn ON the motor. That activated indication
will be decoded by DTMF decoder and sends to buffer for temporary storage, driver
to drive the relay and relay switches the motor to ON situation. Even the concerned
person can also turn OFF the motor by activating another key. The Smart Irrigation
System Setup is as shown in Figure 10 with the sunflower model
Figure 11 shows the flap system rose to cover the plants at times of heavy rain
Figure 12 shows LCD displays the different conditions sensed by the sensor
Figure 10. Smart irrigation system
227
Figure 11. Flap over system
Figure 12. Moisture Low message in the Display
228
Case 4: Environment Change Prediction to Adapt

Climate Smart Agriculture Using Big Data Analytics
The proposed system shown in Figure 13 demonstrate to discover answers for issues,
for example, environmental change-initiated worldwide nourishment instability, to
foreseeing and moderating the effect of extraordinary climate occasions on worldwide
back. Since we have sensors wherever in space which are utilized to screen and
measure climate, vegetation, overcast cover, ice cover, precipitation, ocean surface
temperature, and some more geophysical parameters. These boundless information
accumulations give us progressively more profound and more extensive scope of
atmosphere change, both transiently and geospatially. This great sensor conveys
an immense rate of unstructured information (petabytes of information from day
by day). These gigantic unstructured information require the run of the mill huge
information apparatuses for information stockpiling, handling, dissecting, picturing,
also, anticipating. The design of the proposed framework in which MapReduce in
the utilization of Hadoop to screen and breaking down the enormous information
gathered from different sources like climate guaging, sensor information, advertise
patterns furthermore, online networking information, at that point result must be
displayed to Data researcher for anticipate future climate with high-determination
recreations, to help agriculturists in adjusting to environmental change, what’s more,
to ensure the world’s horticulture business.
Implementation
Implementation includes following modules:
Figure 13. Shows the idea of smart agriculture
229
• Data Acquisition Module
This module gathers information from different sources like sensor information,
climate determining, online networking information and advertises patterns. These
Meteorological information can be issued physically, information can be gained
with access of meteorological information procurement hardware, the little gotten
information are first put away in a prophet database, at the point when little information
are accumulated to a specific number, the little information will be moved into the
capacity module, exchanged information will be consequently erased.
• Data Storage Module
It is in charge of capacity of metadata and information sets with repeated

duplicate, which give reinforcement office. HDFS is a capacity holder and isn’t
constrained by an information. Little information in the information procurement
module collected to a specific sum will be put in the capacity module all the time.
Figure 14. Integrated working of different modules
230
This module is handling stage incorporates two sections of information perusing/

breaking down and foundation of conjecture result. The information perusing is done
principally by Hive. Hive is a structure for information warehousing over Hadoop. It
was made to make it feasible for experts with solid SQL aptitudes to keep running
on the immense volumes of information put away in HDFS. Hive keeps running
on workstations and change over SQL questions into arrangement of MapReduce
occupations for execution on a Hadoop bunch. MapReduce is an execution motor
reasonable for expansive information handling and can altogether enhance the
reaction speed for returning question comes about. In the second part we actualize
forecast work for build up gauge information through k-implies bunch calculation.
Here we make utilization of apache mahout. It is an open-source versatile machine
learning library. Mahout gives a proficient method for actualizing unsupervised
machine learning calculation. The information of recent years are utilized to make
forecast about future.
• Query Analysis Module
Query module will be shown in this module in a imagined mode. Speaking to

complex information with diagrams what’s more, diagrams is a fundamental piece
of the information examination process, and we influence utilization of flo tend
to instrument to make delightful and one of a kind information visualizations. In
the initial step climate datasets will be gathered from different sources and these
information things are further pre-processed to make a powerful contribution to
forecast calculation. After information cleaning, stack into HDFS at that point
apply inquiry utilizing Hive for investigation. We can likewise run Pig content for
breaking down this information what’s more, result is giving as a contribution to
flotend to draw the diagram for examination. We make utilization of strategic relapse
calculation for forecast and it can be executed utilizing Apache Mahout Machine
learning library. At that point check the exactness of the forecast.
Workflow of Proposed System
Shown in Figure 15.
Prediction Algorithm: Logistic Algorithm
Strategic Regression executes in two noteworthy stages:
231
Figure 15. Proposed model
1. Train the Model: This progression is tied in with making a show utilizing some
prepare information, which can additionally be utilized for the arrangement of
any info information rather I would state test information.
2. Test the Model: This progression tests the created display in stage 1 by
assessing the after effects of grouping of test information, and estimating the
exactness, scores and disarray framework.
Supervised Learning algorithm is used for prediction because it provides most

accuracy rather than traditional prediction technique.
Model of supervised learning algorithm (Figure 16).
232
Figure 16. Supervised Learning Methodology
Directed Learning calculation is utilized for expectation since it gives most

exactness as opposed to customary forecast strategy Logistic relapse fits an
extraordinary s-formed bend by taking the direct relapse, which could create any
y-esteem between short unendingness and in addition to boundlessness, and changing
it with the capacity:
p = Exp(y) / (1 + Exp(y))
which produces p-values among 0 (as y approaches minus infinity) and 1 (as y
approaches plus infinity). This now becomes a special kind of non-linear regression,
which is what this page performs. Online-linear-regression (D, number of iterations)
Initialize weights w= (w0, w1, W2 …. Wd)

For i=1:1: number of iterations
do select a data point from Di=(xi, yi) from D
seta=1/i
updateweight vector
w<-w + a(yi-f(xi, W))xi
end for returnweights w
233
The upsides of utilizing on the web straight relapse calculation is anything but
difficult to execute, consistent information streams, adapts to changes in the model
after some time.
This embraces examination of precipitation and temperature informational index
of Tumkur area district over the past 100 years (1901-2000) to lead the determining
test and foresee the information of the coming one month from now or year ahead
of time, and uses the above climate expectation framework which empowers the
strategic relapse calculation to be acknowledged to make predictions. Figure below
demonstrates the diagram portrayal of investigation. Here we utilize Pig content to
examination what’s more, the yield is providing for as a contribution to flo tend code
to produce a chart. Here we dissect recent years climate information of Tumkur area
and result will be created. Here we examine temperature informational indexes and
figure year wise furthermore, month to month normal figuring for Tumkur region.
The investigation of various atmosphere factors like greatest temperature, normal
temperature, least temperature, precipitation, overcast cover. Here examination
depends on authentic information just for Tumkur district. The distinctive parameters
utilized as an input variable to run a calculation. Pass is the quantity of times to
disregard enters information. Lambda is the measure of coefficient rot to utilize.
Figure 17. Analysis of annual and monthly temperature data of Tumkur region
234
Figure 18. Weather data analysis for Tumkur region
Rate demonstrates Learning rate. Precision will rely upon these parameters. Table
underneath demonstrates exactness increments while changing the comparing
variable esteem.
CONCLUSION
This system discussed here gives solution for the Smart Agriculture and reduces
the water use because it provides irrigation as per the requirement of the crop.
This system is automated irrigation system so it reduces the human involvement.
This irrigation system was found to be feasible and cost effective for optimizing
water resources for agricultural production. The irrigation system can be adjusted
to a variety of specific crop needs and requires minimum maintenance. Using this
system we can monitor the status of all the sensors (Soil-moisture, Temperature,
Water level) and also the ON/OFF status of the motor.
235
REFERENCES
Corne, D. (2014). Accurate Localized Short Term Weather Prediction for Renewables
Planning. IEEE.
Dahikar, S. S., & Rode, S. V. (2014). Agricultural crop yield prediction using
artificial neural network approach. Int J Innov Res Electr Electron Instrum Control
Eng, 2(1), 683–686.
Huang, Y., Yu, W., Osewold, C., & Garcia-Ortiz, A. (2016). Analysis of PKF: A
communication cost reduction scheme for wireless sensor networks. IEEE Transactions
on Wireless Communications, 15(2), 843–856. doi:10.1109/TWC.2015.2479234
Kumar, K. K., Kumar, K. R., Ashrit, R. G., Deshpande, N. R., & Hansen, J. W.
(2004). Climate impacts on Indian agriculture. International Journal of Climatology,
24(11), 1375–1393. doi:10.1002/joc.1081
Lee, M., Hwang, J., & Yoe, H. (2013). Agricultural Production System based on IoT.
IEEE 16th International Conference on Computational Science and Engineering.
10.1109/CSE.2013.126
Li, L., Ma, Z., & Li, L. (2013). Hadoopbased ARIMA Algorithm and its Application
in Weather Forecast. International Journal of Database Theory and Application,
6(5), 119-132. 10.14257/ijdta.2013.6.5.11
Paras & Mathur. (2012, September). A Simple Weather Forecasting Model Using
Mathematical Regression. Indian Research Journal of Extension Education.
Pavithra & Srinath. (2014). GSM based Automatic Irrigation Control System for
Efficient Use of Resources and Crop Planning by Using an Android Mobile. IOSR
Journal of Mechanical and Civil Engineering, 11(1), 49-55.
Radhika & Shashi. (2009). Atmospheric Temperature Prediction using Support
Vector Machines. International Journal of Computer Theory and Engineering, 1(1).
Rokach & Maimon. (n.d.). Decision trees. In Data Mining and Knowledge Discovery
Handbook. London: Springer.
Sap, M. N., & Awan, A. M. (2006). Development of an intelligent prediction tool
for rice yield based on machine learning techniques. Jurnal Teknologi Maklumat,
18(2), 73–93.
Septiawan, Komaruddin, & Budi. (2012). Prediction Model for Chilli Productivity
Based on Climate and Productivity Data. UKSim-AMSS 6th European Modelling
Symposium.
236
Shabadi, Patil, Nikita, Shruti, Smitha, & Swati. (2014). Irrigation Control System
Using Android and GSM for Efficient Use of Water and Power. International Journal
of Advanced Research in Computer Science and Software Engineering, 4(7).
Shiraz Pasha & Yogesha. (2014). Microcontroller Based Automated Irrigation
System. The International Journal Of Engineering And Science, 3(7).
Suresh, Gopinath, Govindaraju, Devika, & Vanitha. (2014). GSM based Automated
Irrigation Control using Raingun Irrigation System. International Journal of Advanced
Research in Computer and Communication Engineering, 3(2).
Usha, Rama, & Rao. (2015). An Enhanced Support Vector Regression Model.
International Journal of Advanced Research in Computer Engineering & Technology,
4(5).
Waga, D. (2014). Environmental Conditions’ Big Data Management and Cloud
Computing Analytics for Sustainable Agriculture. World Journal of Computer
Application and Technology, 2(3), 73–81.
Yang & Kim. (2014). A Prediction Model based on Big Data Analysis Using Hybrid
FCM Clustering. The 9th International Conference for Internet Technology and
Secured Transactions.
237
238
Compilation of References
Aazam, M., & Huh, E. N. (2014). Fog computing and smart gateway based
communication for cloud of things. In Future Internet of Things and Cloud (FiCloud).
IEEE.
Aazam, M., & Huh, E. N. (2016). Fog computing: The cloud-iot/ioe middleware
Aazam, M., St-Hilaire, M., & Lung, C. H. (2016). Mefore: Qoe based resource
estimation at fog to enhance qos in iot. 23rd International Conference on
Telecommunications (ICT). 10.1109/ICT.2016.7500362
Adibi, S. (2014). Biomedical Sensing Analyzer (BSA) for Mobile-Health (m-Health)
– LTE. IEEE Journal of Biomedical and Health Informatics, 18(1), 345–351.
doi:10.1109/JBHI.2013.2262076 PMID:24403433
Ahmed, A., & Ahmed, E. (2016). A Survey on Mobile Edge Computing. IEEE
International Conference on Itelligent System and Control (ISCO 2016). 10.1109/
ISCO.2016.7727082
Ahmed, E., Gani, A., Sookhak, M., Hamid, S. H., & Xia, F. (2015). Application
Optimization in Mobile Cloud Computing: Motivation, Taxonomies, and Open
Challenges. Journal of Network and Computer Applications, 2, 52–68. doi:10.1016/j.
jnca.2015.02.003
Ahmed, E., & Rehmani, M. H. (2017). Mobile Edge Computing: Opportunities,
Solutions and Challenges. Future Generation Computer Systems, 70, 59–63.
doi:10.1016/j.future.2016.09.015
Akella & Xiong. (2014). Quality of service (qos)-guaranteed network resource
allocation via software defined networking (sdn). In Dependable, Autonomic and
Secure Computing (DASC), 2014 IEEE 12th International Conference on (pp.
7–13). IEEE.
Al-Ali, Zualkernan, & Aloul. (2010). Mobile GPRS-Sensors Array for Air Pollution
Monitoring. IEEE Sensors Journal, 10(10).
Alam. (2015). Towards Social Internet of Vehicles: Concept, Architecture and
Applications. IEEE.
Ali & Eid. (2015). An Automated System for Accident Detection. Instrumentation
and Measurement Technology Conference, 2015 IEEE Conference.
Alliance, C. S. (2016). The Treacherous 12 Cloud Computing Top Threats in 2016.
Retrieved from https://downloads.cloudsecurityalliance.org/assets/ research/top-
threats/Treacherous-12_Cloud-Computing_Top-Threats. pdf
Almadani, B.-Y., & Shakshuki. (2015). E-Ambulance: Real-Time Integration Platform
for Heterogeneous Medical Telemetry System. 5th International Conference on
Current and Future Trends of Information and Communication Technologies in
Healthcare.
Amin, S. (n.d.). [Accident Detection and Reporting System using GPS, GPRS and
GSM technology. Informatics, Electronics & Vision.]. Jalil, & Reaz.
Atzori, L., Iera, A., & Morabito, G. (2010). The internet of things: A survey. Computer
Networks, 54(15), 2787–2805. doi:10.1016/j.comnet.2010.05.010
Bahl, P., Han, Y. R., Li, E. L., & Satyanarayanan, M. (2012). Advancing the State
of Mobile Cloud Computing. Proceedings of the 3rd ACM workshop on Mobile
cloud computing and services (MCS 2012), 21 – 28. 10.1145/2307849.2307856
Baktir, A. C., Ozgovde, A., & Ersoy, C. (2017). How Can Edge Computing Benefit
from Software-Defined Networking: A Survey, Use Cases & Future Directions.
IEEE Communications Surveys and Tutorials, 19(4), 2359–2391. doi:10.1109/
COMST.2017.2717482
Barai. (2003). Data Mining Applications In Transport Engineering. Retrieved from
http://tf.nist.gov/seminars/WSTS/PDFs/10_Cisco_FBonomi_ConnectedVehicles.
pdf
Beck, M., Werner, M., Feld, S., & Schimper, S. (2014). Mobile edge computing: A
taxonomy. Sixth International Conference on Advances in Future Internet.
Bonomi, Milito, Zhu, & Addepalli. (2012). Fog computing andits role in the internet
of things. In Proceedings of The First Edition of The Mcc Workshop on Mobile
Cloud Computing. ACM.
239
Bonomi, Milito, Zhu, & Addepalli. (n.d.). Fog Computing and Its Role in the Internet
of Things. Cisco Systems Inc.
Bonomi, F. (2011). Connected vehicles, the internet of things, and fog computing.
The Eighth ACM International Workshop on Vehicular Inter-Networking (VANET).
Bonomi, F. (2014). Fog Computing: A Platform for Internet of Things and Analytics.
In N. Bessis & C. Dobre (Eds.), Big Data and Internet of Things: A Roadmap for
Smart Environments (pp. 169–186). Springer. doi:10.1007/978-3-319-05029-4_7
for Internet of Things and Analytics. Springer International Publishing.
Bonomi, F., Milito, R., Zhu, J., & Addepalli, S. (2012). Fog Computing and its Role
in the Internet of Things. Proceedings of the First Edition of the ACM SIGCOMM
Workshop on Mobile Cloud Computing (MCC 2012). 10.1145/2342509.2342513
Brech, B., Jamison, J., Shao, L., & Wightwick, G. (2013). The Interconnecting of
Everything. IBM Corp.
Brown, G. (2013). Converging Telecom and IT in the LTE RAN. White paper at
Heavy Reading on behalf on Samsung Cloud Computing Principles and Paradigms.
Cao, Y. (2015). FAST: A Fog Computing Assisted Distributed Analytics System
to Monitor Fall for Stroke Mitigation. Proc. 10th IEEE Int’l Conf. Networking,
Architecture and Storage (NAS 15), 2–11. 10.1109/NAS.2015.7255196
Carla, Diala, & Sami, Roch, Glitho, Morrow, & Polakos. (2017). A Comprehensive
Survey on Fog Computing: State-of-the-art and Research Challenges. IEEE
Communications Surveys and Tutorials.
Cervantes, H., & Hall, R. S. (2004). Technical Concepts of Service Orientation.
In Service-oriented software system engineering: Challenges and practices. Idea
Group Inc. (IGI).
Chang & Srirama. (n.d.). Indie Fog: An Efficient Fog-Computing Infrastructure for
the Internet of Things. University of Melbourne and Manjrasoft Pty Ltd.
Chavali, L. N. (2014). Cloud Computing in Agriculture. In Agricultural Bioinformatics.
Springer. doi:10.1007/978-81-322-1880-7_12
240
Chen, N., Chen, Y., You, Y., Ling, H., Liang, P., & Zimmermann, R. (2016). Dynamic
urban surveillance video stream processing using fog computing. In Multimedia Big
Data (BigMM), 2016 IEEE Second International Conference on. IEEE. 10.1109/
BigMM.2016.53
Chen, R. Y. (2017). An intelligent value stream-based approach to collaboration
of food traceability cyber physical system by fog computing. Food Control, 71,
124–136. doi:10.1016/j.foodcont.2016.06.042
Chiang, M. (2015). Fog Networking: An Overview on Research Opportunities.
White Paper.
Chiang, M. (2015, December). White Paper on Fog Networking: An Overview on
Research Opportunities. Academic Press.
Chiang, M., & Zhang, T. (2016). FogandIoT:Anoverviewofresearchopportunities.
IEEE Internet Things J., 3(6), 1–11.
Chong, Abraham, & Paprzycki. (2004). Traffic Accident Analysis Using Decision
Trees and Neural Networks. Academic Press.
Cirani, S., Ferrari, G., Iotti, N., & Picone, M. (2015). The IoT hub: a fog node
for seamless management of heterogeneous connected smart objects. Sensing,
Communication, and Networking-Workshops (SECONWorkshops). 12th Annual
IEEE International Conference on IEEE. 10.1109/SECONW.2015.7328145
Cisco IOx and Fog Applications. (2017). Retrieved from https://www.cisco.com/c/
en_in/ solutions/internet-of-things/iot-fog-applications.html
Cisco, Tech. Rep. (2014). Cisco Delivers Vision Of Fog Computing To Accelerate
Value From Billions Of Connected Devices. Cisco.
CISCO. (2015). Fog computing & the internet of things: extend the cloud to where
the things are. Retrieved from https://www.cisco.com/c/dam/en_us/solutions/trends/
iot/docs /computing-overview.pdf
Cisco. (2017). Fog Computing and Internet of Things: Extend the Cloud to Where
the Things Are. Available: http://www.cisco.com/c/dam/en_us/solutions/trends/iot/
docs/computing-overview.pdf
Cisco. (n.d.). New cisco internet of things (iot) system provides a foundation for the
transformation of industries. Retrieved from https://newsroom.cisco.com
241
Corne, D. (2014). Accurate Localized Short Term Weather Prediction for Renewables
Planning. IEEE.
Corser & Aledhari. (2017). Internet Of Things(IOT)Security Best Practices. IEEE
Internet Technology Policy Community White Paper.
Cristea, V., Dobre, C., & Pop, F. (2013). Context-aware Environments for the Internet
of Things. Internet of Things and Inter-cooperative Computational Technologies for
Collective Intelligence Studies in Computational Intelligence, 460, 25–49.
Dahikar, S. S., & Rode, S. V. (2014). Agricultural crop yield prediction using
artificial neural network approach. Int J Innov Res Electr Electron Instrum Control
Eng, 2(1), 683–686.
Dastjerdi, Gupta, Calheiros, Ghosh, & Buyya. (2016). Fog Computing: Principles,
Architectures, and Applications. Internet of Things Principles and Paradigms, 61–75.
Der-Horng, Shin-Ting, & Chandrasekar. (2004). Applying data mining techniques for
traffic incident analysis. Journal of The Institution of Engineers, Singapore, 44(2).
Dialogic Corporation. (2010). Introduction to Cloud Computing. Dialogic Corporation
White Paper.
Díaz, M., Martín, C., & Rubio, B. (2016, May). State-of-the-art,challenges,andopen
issues in the integration of Internet of Things and cloud computing. Journal of
Network and Computer Applications, 67, 99–117. doi:10.1016/j.jnca.2016.01.010
Dihn, T. H., Lee, C., Niyato, D., & Wang, P. (2011). A Survey of Mobile Cloud
Computing: Architecture, Applications, and Approaches. Wireless Communications
and Mobile Computing, Wiley, 13(18), 1587–1611.
Do, C. T., Tran, N. H., Pham, C., Alam, M. G. R., Son, J. H., & Hong, C. S. (2015).
A proximal algorithm for joint resource allocation and minimizing carbon footprint
in geo-distributed fog computing. In 2015 International Conference on Information
Networking (ICOIN). IEEE. 10.1109/ICOIN.2015.7057905
Dsouza, A., & Taguinod, M. (2014). Policy-driven security management for fog
computing: Preliminary framework and a case study. Information Reuse and
Integration (IRI), 2014 IEEE 15th International Conference, 16-23.
Dsouza, C., Ahn, G. J., & Taguinod, M. (2014, August). Policy-driven security
management for fog computing: Preliminary framework and a case study. In
Information Reuse and Integration (IRI), 2014 IEEE 15th International Conference
on (pp. 16-23). IEEE.
242
Dubey, H., Yang, J., Constant, N., Amiri, A. M., Yang, Q., & Makodiya, K. (2015,
October). Fog data: Enhancing telehealth big data through fog computing. In
Proceedings of the ASE BigData & SocialInformatics 2015 (p. 14). ACM.
Ericsson. (2016). Ericsson Mobility Report on the Pulse of the Networked Society.
Stockholm, Sweden: Ericsson.
European Telecommunications Standards Institute Industry Specifications Group,
Mobile-Edge Computing – Service Scenarios. (2017). Retrieved from http://
www.etsi.org/deliver/etsi_gs/MEC-IEG/001_099/004/01.01.01_60/gs_MEC-
IEG004v010101p.pdf
Evans, D. (2011). The internet of things: How the next evolution of the internet is
changing everything. CISCO. Int. J. Internet, 3(2), 123–132.
Fakeeh. (2016). Privacy and Security Problems in Fog Computing. Communications
on Applied Electronics. Foundation of Computer Science FCS.
Farahani, R. Z., & Hekmatfar, M. (2009). Facility Location, Concepts, Models,
Algorithms and Case Studies. Contributions to management science. Heidelberg,
Germany: Physica-Verlag.
Fernández-Lozano, J. J., Martín-Guzmán, M., Martín-Ávila, J., & García-Cerezo,
A. (2015). A wireless sensor network for urban traffic characterization and trend
monitoring. Sensors (Basel), 15(10), 26143–26169. doi:10.3390151026143
PMID:26501278
Fowler, S. (2013). Survey on Mobile Cloud Computing Challenges Ahead. IEEE
CommSoft E-Letters, 2(1), 13–17.
Gao, L., Luan, T. H., Liu, B. W. Z., & Yu, S. (2017). Fog Computing and Its
Applications in 5G. In 5G Mobile Communications (pp. 571-593). Springer
International Publishing Switzerland.
Gennaro, R., Gentry, C., & Parno, B. (2010). Non-interactive verifiable computing:
Outsourcing computation to untrusted workers. In Advances in Cryptology--CRYPTO
2010. Springer.
H. (2015). Fog Computing in Healthcare Internet of Things: A Case Study on ECG
Feature Extraction. In Computer and Information Technology; Ubiquitous Computing
and Communications; Dependable, Autonomic and Secure Computing; Pervasive
Intelligence and Computing. CIT/IUCC/DASC/PICOM.
243
Giang, N. K., Blackstock, M., Lea, R., & Leung, V. C. M. (2015). Developing iot
applications in the fog: A distributed dataflow approach. 5th International Conference
on the Internet of Things (IOT). 10.1109/IOT.2015.7356560
H. (2015). Fog Computing in Healthcare Internet-of-Things: A Case Study on
ECG Feature Extraction. 15th IEEE International Conference on Computer and
Information Technology, 356-363. 10.1109/CIT/IUCC/DASC/PICOM.2015.51
Goel, Bhagwan, Chaturvedi, Rai, & Pandey. (2015). Application of cloud computing
in agricultural sector. 4th International Conference on Agriculture & Horticulture.
Agrotechnol
Gotadki, Mohan, Attarwala, & Gajare. (2014, April). Intelligent Ambulance.
International Journal of Engineering and Technical Research.
Goyal, S. (2014). Public vs Private vs Hybrid vs Community – Cloud Computing:
A Critial Review. International Journal of Computer Network and Information
Security, 6(3), 20–29. doi:10.5815/ijcnis.2014.03.03
Gu, L., Zeng, D., Guo, S., Barnawi, A., Xiang, Y. (2015). Cost-efficient resource
management in fog computing supported medical cps. IEEE Transactions on
Emerging Topics in Computing.
Gupta, H., Chakraborty, S., Ghosh, K. S., & Buyya, R. (2017). Fog Computing in
5G Networks: An Application Perspective. In Cloud and Fog Computing in 5G
Mobile Networks: Emerging advances and applications (pp. 23-54). The Institute
of Engineering and Technology (IET), IET Telecommunications Series 70.
Haruna, Manis, Oladipo, & Ariwa. (2017). User Mobility and Resource Scheduling
and Management in Fog Computing to Support IoT Devices. IEEE.
He, Lu, & Wang. (2012). A New Method For Traffic Forecasting Based On Data
Mining Technology With Artificial Intelligent Algorithms. Research Journal of
Applied Sciences, Engineering and Technology.
Heuser, S., Negro, M., Pendyala, P. K., & Sadeghi, A. R. (2016). Droidauditor:
“Forensic analysis of application-layer privilege escalation attacks on android.”
Technical report. TU Darmstadt.
Hoang, D. B., & Chen, L. (2010). Mobile cloud for assistive healthcare (mocash).
Services Computing Conference (APSCC).
244
Hong, K., Lillethun, D., Ramachandran, U., Ottenwälder, B., & Koldehofe, B. (2013).
Mobile fog: A programming model for large-scale applications on the internet of
things. In Proceedings of the second ACM SIGCOMM workshop on Mobile cloud
computing. ACM. 10.1145/2491266.2491270
Hossain, M. S., & Atiquzzaman, M. (2013). Cost analysis of mobility protocols.
Huang, Y., Yu, W., Osewold, C., & Garcia-Ortiz, A. (2016). Analysis of PKF: A
communication cost reduction scheme for wireless sensor networks. IEEE Transactions
on Wireless Communications, 15(2), 843–856. doi:10.1109/TWC.2015.2479234
Intharawijitr, K., Iida, K., & Koga, H. (2016). Analysis of fog model considering
computing and communication latency in 5G cellular networks. IEEE. doi:10.1109/
PERCOMW.2016.7457059
ITU-T. (2013). Cloud Computing Framework and High Level Requirements. ITU-T
ITU-T. (2013). Cloud Computing Infrastructure Requirements. ITU-T
Jadhav, Satam, & Salvi. (2016, February). An Android-Based Emergency Alarm
Message and Healthcare Management System. International Journal of Advanced
Research in Computer and Communication Engineering.
Jovović, I., & Forenbacher, I. M. P. (2015). Massive Machine-Type Communications:
An Overview and Perspectives Towards 5G. The 3rd International Virtual Research
Conference In Technical Disciplines.
Kang, K., Wang, C., & Luo, T. (2016). Fog computing for vehicular ad-hoc networks:
Paradigms, scenarios, and issues. Journal of China Universities of Posts and
Telecommunications, 23(2), 56–96. doi:10.1016/S1005-8885(16)60021-3
Khan, Yongrui, & Qin. (2017). Fog computing security: a review of current
applications and security solutions. Journal of Cloud Computing.
245
Khan, A. R., Othman, M., Madani, S. A., & Khan, S. U. (2014). Survey of Mobile
Cloud Computing Application Models. IEEE.
Kiczales, G., Lamping, J., Mendhekar, A., Maeda, C., Lopes, C., Loingtier, J. M.,
& Irwin, J. (1997). Aspect-oriented programming. ECOOP’97. Proceedings of the
11th European Conference on Object-Oriented Programming, 220–242. 10.1007/
BFb0053381
Kirat, D., Vigna, G., & Kruegel, C. (2014), Barecloud: Bare-metal analysis-based
evasive malware detection. USENIX Security Symposium, 287–301.
Kitanov, S., & Janevski, T. (2014). State of the Art: Mobile Cloud Computing.
Proceedings of the Sixth IEEE International Conference on Computational
Intelligence, Communication Systems and Networks 2014 (CICSYN 2014), 153-158.
10.1109/CICSyN.2014.41
Kitanov, S., Monteiro, E., & Janevski, T. (2016). 5G and the Fog – Survey of Related
Technologies and Research Directions. Proceedings of the 18th Mediterranean IEEE
Electrotechnical Conference MELECON 2016. 10.1109/MELCON.2016.7495388
Krishnaveni & Hemalatha. (2011). A perspective analysis of traffic accident using
data mining techniques. Academic Press.
Kulkarni & Teja. (2014). Automated System for Air Pollution Detection and Control
in Vehicles. International Journal of Advanced Research in Electrical, Electronics
and Instrumentation Engineering, 3(9).
Kumar, K. K., Kumar, K. R., Ashrit, R. G., Deshpande, N. R., & Hansen, J. W.
(2004). Climate impacts on Indian agriculture. International Journal of Climatology,
24(11), 1375–1393. doi:10.1002/joc.1081
Lebied, M. (2017). 6 Cloud Computing Challenges Businesses are facing in these days.
Retrieved from: https://www.datapine.com/blog/top-6-cloud-computing-challenges
Lee, M., Hwang, J., & Yoe, H. (2013). Agricultural Production System based on IoT.
IEEE 16th International Conference on Computational Science and Engineering.
10.1109/CSE.2013.126
Li, L., Ma, Z., & Li, L. (2013). Hadoopbased ARIMA Algorithm and its Application
in Weather Forecast. International Journal of Database Theory and Application,
6(5), 119-132. 10.14257/ijdta.2013.6.5.11
246
Li, J. (2015) EHOPES: Data-Centered Fog Platform for Smart Living. Int’l
Telecommunication Networks and Applications Conf. (ITNAC 15), 308–313. 10.1109/
ATNAC.2015.7366831
Li, M., Yu, S., Ren, K., & Lou, W. (2010). Securing personal health records in cloud
computing: Patient-centric and fine-grained data access control in multi-owner
settings. In International Conference on Security and Privacy in Communication
Systems. Springer. 10.1007/978-3-642-16161-2_6
Liu, Choudhary, Zhou, & Khokhar. (2006). Distributed Stream Mining System for
Highway traffic. Data (København).
Liu, G., Hou, X., Jin, J., Wang, F., Wang, Q., Hao, Y., ... Deng, A. (2017). 3-D-MIMO
With Massive Antennas Paves the Way to 5G Enhanced Mobile Broadband: From
System Design to Field Trials. IEEE Journal on Selected Areas in Communications,
35(6), 1222–1233. doi:10.1109/JSAC.2017.2687998
Liu, Y. J. E., Fieldsend, E. J., & Min, G. (2017). A Framework of Fog Computing:
Architecture, Challenges and Optimization. IEEE Access: Practical Innovations,
Open Solutions, 5, 25445–25454. doi:10.1109/ACCESS.2017.2766923
Madakam, S., & Date, H. (2016). Security Mechanisms for Connectivity of Smart
Devices in the Internet of Things. Springer. doi:10.1007/978-3-319-33124-9_2
Maher. (2015). IoT, from Cloud to Fog Computing. Retrieved from https://blogs.
cisco.com/ perspectives/iot-from-cloud-to-fog-computing
Mahmud, R., Kotagiri, R., & Buyya, R. (2018). Fog Computing: A Taxonomy,
Survey and Future Directions. In B. Di Martino, K. C. Li, L. Yang, & A. Esposito
(Eds.), Internet of Everything. Internet of Things (Technology, Communications and
Computing). Singapore: Springer. doi:10.1007/978-981-10-5861-5_5
Makwana. (2016). Introduction to Fog Computing. AI-eHive.
Manjunath, N., & Nikschal, V. M. (2016, May). Design of an Automated Traffic
Control System for Emergency Vehicle Clearance. International Journal of Emerging
Research in Management and Technology.
Mattern, F., & Floerkemeier, C. (2014). From the Internet of Computers to the
Internet of Things. ETH Zurich.
247
Md, S. B. B., & Abu Kassim, R. (2013). Development of Fire Alarm System using
Raspberry Pi and Arduino Uno. In Electrical, Electronics and System Engineering
(ICEESE), 2013 International Conference on. IEEE.
Megalingam, Nair, & Prakhya. (n.d.). Wireless Vehicular Accident Detection and
Reporting System. Mechanical and Electrical Technology.
Mell, P., & Grance, T. (2009). The NIST Definition of Cloud Computing. Retrieved
from http://www.nist.gov/itl/cloud/upload/cloud-def-v15.pdf
Meola, A. (2016). How the Internet of Things will affect the security & privacy?
Retrieved from http://www.businessinsider.com/internet-of-things-security-privacy-
2016-8?IR=T
Mitchell, S., Villa, N., Stewart-Weeks, M., & Lange, A. (2013). The Internet of
Everything for Cities Connecting People, Process, Data, and Things to Improve
the ‘Livability’ of Cities and Communities. CISCO, white paper.
Moje, Kumbhar, Shinde, & Korke. (2016, April). Automatic Ambulance Rescue
System. International Journal of Innovative Research in Electrical, Electronics,
Instrumentation and Control Engineering.
Monteiro, A., Dubey, H., Mahler, L., Yang, Q., & Mankodiya, K. (2016). Fit a fog
computing device for speech tele-treatments. arXiv preprint arXiv:1605.06236
Mouradian C., Naboulsi D., Yangui S., Glitho H. R., Morrow J. M., & Polakos A. P.
(2017). A Comprehensive Survey on Fog Computing: State-of-the-art and Research
Challenges. IEEE Communications Surveys and Tutorials, 1 – 51.
Munir, A., Kansakar, P., & Khan, S. U. (2017). IFCIoT: Integrated Fog Cloud IoT:
A novel architectural paradigm for the future Internet of Things. IEEE Consumer
Electronics Magazine, 6(3), 74–82. doi:10.1109/MCE.2017.2684981
Nam, T., & Pardo, T. A. (2011, September). Smart city as urban innovation: Focusing
on management, policy, and context. In Proceedings of the 5th international
conference on theory and practice of electronic governance (pp. 185-194). ACM.
10.1145/2072069.2072100
NebbioloTechnologies. (2015). Fog vs Edge Computing v1.1. Milpitas, CA: Nebbiolo
Technologies Inc.
248
Ni, Zhang, Lin, & Shen. (n.d.). Securing Fog Computing for Internet of Things
Applications: Challenges and Solutions. IEEE Communications Surveys & Tutorials.
DOI 10.1109/COMST.2017.2762345
Niekamp. (2011). Software Component Architecture. Gestión de Congresos - CIMNE/
Institute for Scientific Computing, TU Braunschweig.
NIST. (2013). NIST Cloud Computing Standards Roadmap. NIST Special Publication
500-291, Version 2.
Oh, Lee, & Kote. (2003). Real Time Video Data Mining For Surveillance Video
Streams. Academic Press.
Okay, F. Y., & Ozdemir, S. (2016), A Fog Computing based Smart Grid Model,
Networks, Computers and Communications (ISNCC). 2016 International Symposium
IEEE, 1-6.
OpenFog Consortium. (2017). OpenFog Reference Architecture for Fog Computing.
OpenFog Consortium Architecture Working Group.
OpenFog Reference Architecture for Fog Computing, Produced by the OpenFog
Consortium Architecture Working Group. (2017, February). Retrieved from http://
www.OpenFogConsortium.org
Oueis, J., Strinati, E. C., Sardellitti, S., & Barbarossa, S. (2015). Small cell clustering
for efficient distributed fog computing: A multi-user case. In Vehicular Technology
Conference (VTC Fall). IEEE. doi:10.1109/VTCFall.2015.7391144
Paras & Mathur. (2012, September). A Simple Weather Forecasting Model Using
Mathematical Regression. Indian Research Journal of Extension Education.
Pavithra & Srinath. (2014). GSM based Automatic Irrigation Control System for
Efficient Use of Resources and Crop Planning by Using an Android Mobile. IOSR
Journal of Mechanical and Civil Engineering, 11(1), 49-55.
Peng, M., Li, Y., Jiang, J., Li, J., & Wang, C. (2014). Heterogeneous cloud radio
access networks: A new perspective for enhancing spectral and energy efficiencies.
IEEE Wireless Communications, 21(6), 126–135. doi:10.1109/MWC.2014.7000980
Peng, M., Li, Y., Zhao, Z., & Wang, C. (2015). System architecture and key
technologies for 5G heterogeneous cloud radio access networks. IEEE Network,
29(2), 6–14. doi:10.1109/MNET.2015.7064897 PMID:26504265
249
Peng, M., Yan, S., Zhang, K., & Wang, C. (2016). Fog Computing based Radio
Access Networks: Issues and Challenges. IEEE Network, 30(4), 46–53. doi:10.1109/
MNET.2016.7513863
Pfaff, B. J., Pettit, J., & Koponen, T. (2015). The design and implementation of open
vSwitch. Networked Systems Design and Implementation.
Prabha, Sunitha, & Anitha. (2014). Automatic Vehicle Accident Detection and
Messaging System using GSM and GPS modem. IJAREEIE, 3(7).
Prieto González, L., Jaedicke, C., Schubert, J., & Stantchev, V. (2016). “Fog computing
architectures for healthcare”: Wireless performance and semantic opportunities. J
Inf Commun Ethics Soc, 14(4), 334–349. doi:10.1108/JICES-05-2016-0014
Qureshi, S. S., Ahmad, T., & Shuja-ul-islam, K. R. (2011). Mobile Cloud Computing
as Future for Mobile Applications – Implementation Methods and Challenging
Issues. Proceedings of IEEE Conference on Cloud Computing and Intelligence
Systems (CCIS).
Quwaider, M., & Jararweh, Y. (2013, December). Cloudlet-based for big data
collection in body area networks. In Internet Technology and Secured Transactions
(ICITST), 2013 8th International Conference for (pp. 137-141). IEEE.
Radhika & Shashi. (2009). Atmospheric Temperature Prediction using Support
Vector Machines. International Journal of Computer Theory and Engineering, 1(1).
Rauniyar, A., Engelstad, P., Feng, B., & Thanh, D. V. (2016). Crowdsourcing
based Disaster Management using Fog Computing in Internet of Things Paradigm.
Collaboration and Internet Computing (CIC), 2016 IEEE 2nd International
Conference, 490-494. 10.1109/CIC.2016.074
Raval, K. (2016). The Big Data of IoT – Understanding Privacy and Security
Threats. Retrieved from: https://www.letsnurture.com/blog/the-big-data-of-iot-
understanding-privacy-and-security-threats.html
Ren, K., Lou, W., & Zhang, Y. (2008). Leds: Providing location-aware end-to-end
data security in wireless sensor networks. IEEE Transactions on Mobile Computing,
7(5), 585–598. doi:10.1109/TMC.2007.70753
Rohila & Singla. (2016). An effective review of fog computing using virtualization.
International Journal of Innovation Research of Computer and Communication
Engineering, 4(4).
250
Rokach & Maimon. (n.d.). Decision trees. In Data Mining and Knowledge Discovery
Handbook. London: Springer.
Roman, R., Lopez, J., & Manbo, M. (2018). Mobile Edge Computing, Fog et al.: A
Survey and Analysis of Security, Threats and Challenges (Vol. 78). Futur. Gener.
Comp. Syst.
Roseline, Devapriya, & Sumathi. (2013). Pollution Monitoring using Sensors and
Wireless Sensor Networks: A Survey. International Journal of Application or
Innovation in Engineering & Management, 2(7).
Sabarish, K., & Shaji, R. (2014). A Scalable Cloud Enabled Mobile Governance
Framework. Global Humanitarian Technology Conference - South Asia Satellite
(GHTC-SAS). 10.1109/GHTC-SAS.2014.6967554
Samani & Zhi. (2016, January 4). Robotic Automated External Defibrillator
Ambulance for Emergency Medical Service in Smart Cities. IEEE Access : Practical
Innovations, Open Solutions.
Sangeetha, Archana, Ramya, & Ramya. (2014, February). Automatic Ambulance
Rescue with Intelligent Traffic Light System. IOSR Journal of Engineering.
Sap, M. N., & Awan, A. M. (2006). Development of an intelligent prediction tool
for rice yield based on machine learning techniques. Jurnal Teknologi Maklumat,
18(2), 73–93.
Saqib, A., Anwar, R. W., Hussain, O. K., Ahmad, M., Ngadi, M. A., Mohamad, M.
M., & ... . (2015). Cyber security for cyber physcial systems: A trust-based approach.
J Theor Appl Inf Technol, 71(2), 144–152.
Sarkar, S., Chatterjee, S., & Misra, S. (Forthcoming). Assessment of the suitability
of fog computing in the context of Internet of things. IEEE Trans. Cloud Comput.
Satyanarayanan, M., Bahl, P., Caceres, R., & Davies, N. (2009). The Case for VM-
based Cloudlets in Mobile Computing. IEEE Pervasive Computing, 8(4), 14–23.
doi:10.1109/MPRV.2009.82
Satyanarayanan, M., Schuster, R., Ebling, M., Fettweis, G., Flinck, H., Joshi, K.,
& Sabnani, K. (2015). An open ecosystem for mobile-cloud convergence. IEEE
Communications Magazine, 53(3), 63–70. doi:10.1109/MCOM.2015.7060484
251
Saunier, N. (2011). Investigating Collision Factors by Mining Microscopic Data on

Vehicle Conflict and Collisions. Academic Press.
Schmidt, D.C. (2006). Model-Driven Engineering. IEEE Computer, 39(2).
Sekma, N. C., Dridi, N., & Elleuch, A. (2015). Analyses toward a prediction system
for a large-scale volunteer computing system. 2015 World Congress on Information
Technology and Computer Applications (WCITCA). 10.1109/WCITCA.2015.7367017
Septiawan, Komaruddin, & Budi. (2012). Prediction Model for Chilli Productivity
Based on Climate and Productivity Data. UKSim-AMSS 6th European Modelling
Symposium.
Shabadi, Patil, Nikita, Shruti, Smitha, & Swati. (2014). Irrigation Control System
Using Android and GSM for Efficient Use of Water and Power. International Journal
of Advanced Research in Computer Science and Software Engineering, 4(7).
Shanthi & Ramani. (2011). Classification of Vehicle Collision Patterns in Road
Accidents Using Data Mining Algorithms. Academic Press.
Shi, Y., Ding, G., Wang, H., Roman, H. E., & Lu, S. (2015). The fog computing
service for healthcare. In Future Information and Communication Technologies for
Ubiquitous HealthCare (Ubi-HealthTech), 2015 2nd International Symposium on.
IEEE. 10.1109/Ubi-HealthTech.2015.7203325
Shiraz Pasha & Yogesha. (2014). Microcontroller Based Automated Irrigation
System. The International Journal Of Engineering And Science, 3(7).
Shi, W., & Dustdar, S. (2016). The promise of edge computing. Computer, 49(5),
78–81. doi:10.1109/MC.2016.145
Singh, S., Chiu, Y., & Yang, Y. T. (2016). Mobile Edge Fog Computing in 5G Era:
Architecture and Implementation. 2016 International Computer Symposium.
Sommerville, I. (2011). Software Engineering (9th ed.). Academic Press.
Sqalli, H. M., Al-saeedi, M., Binbeshr, F., & Siddiqui, M. (2012). UCloud: Simulated
Hybrid Cloud for a University Environment. IEEE 1st International Conf. on Cloud
Networking (CLOUDNET).
Srinivasa Prasad & Ramakrishna. (n.d.). An Efficient Traffic Forecasting System
Based on Spatial Data and Decision Trees. The International Arab Journal of
Information Technology, 11(2).
252
Stankovic, J. A. (2014, February). Research directions for the Internet of Things.

IEEE Internet Things J., 1(1), 3–9. doi:10.1109/JIOT.2014.2312291
Stantchev, V. (2015). Smart Items, Fog and Cloud Computing as Enablers of
Servitization in Healthcare. J. Sensors & Transducers, 185(2), 121–128.
Stantchev, V., Barnawi, A., Ghulam, S., Schubert, J., & Tamm, G. (2015). Smart
items, fog and cloud computing as enablers of servitization in healthcare. Sensors
Transducers, 185(2), 121.
Stojmenovic, I., & Wen, S. (2014). The Fog Computing Paradigm: Scenarios and
Security Issues. Federated Conference on Computer Science and Information
Systems, 2(5).
Stojmenovic, I., & Wen, S. (2014). The fog computing paradigm: Scenarios and
security issues. In Computer Science and Information Systems (FedCSIS), 2014
Federated Conference On. IEEE. 10.15439/2014F503
Stojmenovic, I. (2014, November). Fog computing: A cloud to the ground support
for smart things and machine-to-machine networks. In Telecommunication Networks
and Applications Conference (ATNAC), 2014 Australasian (pp. 117-122). IEEE.
10.1109/ATNAC.2014.7020884
Stojmenovic, I., Wen, S., Huang, X., & Luan, H. (2015). An overview of fog
computing and its security issues. Concurrency and Computation.
Stolfo, S. J., Salem, M. B., & Keromytis, A. D. (2012). Fog Computing: Mitigating
Insider Data Theft Attacks in the Cloud. Proceeding of IEEE Symposium on Security
and Privacy Workshops (SPW). 10.1109/SPW.2012.19
Sun, G., & Shen, J. (2014). Facilitating Social Collaboration in Mobile Cloud-based
learning: A Teamwork as a Service (TaaS) Approach. IEEE Transactions on Learning
Technologies, 7(3), 207–220. doi:10.1109/TLT.2014.2340402
Suresh, Gopinath, Govindaraju, Devika, & Vanitha. (2014). GSM based Automated
Irrigation Control using Raingun Irrigation System. International Journal of Advanced
Research in Computer and Communication Engineering, 3(2).
Suthar, S. (2016). How is Security important to IoT devices? Retrieved from: http://
www.cxotoday.com/story/how-is-security-important-to-iot-devices/
253
Tang, B., Chen, Z., Hefferman, G., Wei, T., He, H., & Yang, Q. (2015, October).
A hierarchical distributed fog computing architecture for big data analysis in smart
cities. In Proceedings of the ASE BigData & SocialInformatics 2015 (p. 28). ACM.
Tesma. (2014, March). Rule Mining And Classification Of Road Traffic Accidents
Using Adaptive Regression Trees. Academic Press.
Usha, Rama, & Rao. (2015). An Enhanced Support Vector Regression Model.
International Journal of Advanced Research in Computer Engineering & Technology,
4(5).
Vaquero, L. M., & Rodero-Merino, L. (2014). Finding your Way in the Fog: Towards
a Comprehensive Definition of Fog Computing. ACM SIGCOMM Computer
Communication Review Newsletter, 44(5), 27–32. doi:10.1145/2677046.2677052
Varalakshmi, L., Sudha, G. F., & Jaikishan, G. (2014). A selective encryption and
energy efficient clustering scheme for video streaming in wireless sensor networks.
Varshney, P., & Simmhan, Y. (2017). Demystifying Fog Computing: Characterizing
Architectures, Applications and Abstractions. arXiv:1702.06331
Vong. (2011). Framework of Vehicle Emission Inspection and Control through RFID
and Traffic Lights. Proceedings of.
Waga, D. (2014). Environmental Conditions’ Big Data Management and Cloud
Computing Analytics for Sustainable Agriculture. World Journal of Computer
Application and Technology, 2(3), 73–81.
Wei, X., Gomez, L., Neamtiu, I., & Faloutsos, M. (2012). Malicious android
applications in the enterprise: What do they do and how do we fix it? In Data
Engineering Workshops (ICDEW), 2012 IEEE 28th International Conference on.
IEEE. 10.1109/ICDEW.2012.81
Wu, J., Yuen, C., Cheung, N.-M., Chen, J., & Chen, C. (2015). Enabling Adaptive
High-frame-rate Video Streaming in Mobile Cloud Gaming Applications. IEEE
Transactions on Circuits and Systems for Video Technology, 1-1.
Yang & Kim. (2014). A Prediction Model based on Big Data Analysis Using Hybrid
FCM Clustering. The 9th International Conference for Internet Technology and
Secured Transactions.
254
Yaseen, Q., AlBalas, F., Jararweh, Y., & Al-Ayyoub, M. (2016, September). A fog
computing based system for selective forwarding detection in mobile wireless sensor
networks. In Foundations and Applications of Self* Systems, IEEE International
Workshops on (pp. 256-262). IEEE. doi:10.1109/FAS-W.2016.60
Yilmaz, O. N., Wang, Y.-P. E., & Johansson, N. A. (2015). Analysis of ultra-
reliable and low-latency 5G communication for a factory automation use case.
IEEE International Conference on Communication Workshop (ICCW). 10.1109/
ICCW.2015.7247339
Yi, S., Hao, Z., Qin, Z., & Li, Q. (2015). Fog Computing- Platform and Applications.
IEEE Workshop on Hot Topics in Web Systems and Technologies.
Yi, S., Qin, Z., & Li, Q. (2015). Security and privacy issues of fog computing:
A survey. In International Conference on Wireless Algorithms, Systems, and
Yi, S., Zi, J. H., Qin, Z., & Li, Q. (2015). Fog Computing: Platform and Applications.
Workshop on Hot Topics in Web Systems and Technologies. IEEE.
Yu, Ren, & Wenjing. (2010). Achieving secure,scalable,and fine grained data access
control in Cloud Computing, Infocom. 2010 Proceedings IEEE, 1-9.
Yu, M., Zhang, D., Cheng, Y., & Wang, M. (2011). An RFID Electronic Tag based
Automatic Vehicle Identification System for Traffic IOT Applications. Chinese Control
and Decision Conference (CCDC), 4192-4197. 10.1109/CCDC.2011.5968962
Yu, Y. (2016). Mobile Edge Computing Towards 5G: Vision, Recent Progress,
and Open Challenges. China Communications, 13(Supplement No. 2), 89–99.
doi:10.1109/CC.2016.7833463
Zao, J. (2014) Augmented Brain Computer Interaction Based on Fog Computing
and Linked Data. Proc. 10th IEEE Int’l Conf. Intelligent Environments (IE 14),
374–377. 10.1109/IE.2014.54
doi:10.1109/TC.2016.2536019
Zhang, Z., Zhang, J., & Ying, L. (n.d.). Multimedia streaming in cooperative mobile
social networks. Academic Press.
255
Zhang, S., Zhang, S., Chen, X., & Huo, X. (2010). Cloud Computing, Research
and Development Trend. Proceedings of the 2nd IEEE International Conference
on Future Networks.
Zhang, W., Lin, B., Yin, Q., & Zhao, T. (2016). Infrastructure deployment and
optimization of fog network based on microdc and lrpon integration. Peer-to-Peer
Networking and Applications Springer.
Zhang, Y., Niyato, D., Wang, P., & Dong, I. K. (2016). Optimal energy management
policy of mobile energy gateway. IEEE Transactions on Vehicular Technology, 65(5),
3685–3699. doi:10.1109/TVT.2015.2445833
Zhu, J. (2013). Improving web sites performance using edge servers in fog computing
architecture. In Service Oriented System Engineering (SOSE), 2013 IEEE 7th
International Symposium on. IEEE.
Zhu, J., Chan, D. S., Prabhu, M. S., Natarajan, P., Hu, H., & Bonomi, F. (2013).
Improving Web Sites Performance Using Edge Servers in Fog Computing
Architecture. Service Oriented System Engineering (SOSE), IEEE 7th International
Symposium, 320-323.
256
257
Related References
To continue our tradition of advancing information science and technology research,

we have compiled a list of recommended IGI Global readings. These references will
provide additional information and guidance to further enrich your knowledge and
assist you with your own research and future publications.
Aasi, P., Rusu, L., & Vieru, D. (2017). The Role of Culture in IT Governance Five
Focus Areas: A Literature Review. International Journal of IT/Business Alignment
and Governance, 8(2), 42-61. doi:10.4018/IJITBAG.2017070103
Abdrabo, A. A. (2018). Egypt’s Knowledge-Based Development: Opportunities,
Challenges, and Future Possibilities. In A. Alraouf (Ed.), Knowledge-Based
Urban Development in the Middle East (pp. 80–101). Hershey, PA: IGI Global.
doi:10.4018/978-1-5225-3734-2.ch005
Abu Doush, I., & Alhami, I. (2018). Evaluating the Accessibility of Computer
Laboratories, Libraries, and Websites in Jordanian Universities and Colleges.
International Journal of Information Systems and Social Change, 9(2), 44–60.
doi:10.4018/IJISSC.2018040104
Adeboye, A. (2016). Perceived Use and Acceptance of Cloud Enterprise Resource
Planning (ERP) Implementation in the Manufacturing Industries. International
Journal of Strategic Information Technology and Applications, 7(3), 24–40.
doi:10.4018/IJSITA.2016070102
Related References
Adegbore, A. M., Quadri, M. O., & Oyewo, O. R. (2018). A Theoretical Approach

to the Adoption of Electronic Resource Management Systems (ERMS) in Nigerian
University Libraries. In A. Tella & T. Kwanya (Eds.), Handbook of Research on
Managing Intellectual Property in Digital Libraries (pp. 292–311). Hershey, PA:
IGI Global. doi:10.4018/978-1-5225-3093-0.ch015
Adhikari, M., & Roy, D. (2016). Green Computing. In G. Deka, G. Siddesh, K.
Srinivasa, & L. Patnaik (Eds.), Emerging Research Surrounding Power Consumption
and Performance Issues in Utility Computing (pp. 84–108). Hershey, PA: IGI Global.
doi:10.4018/978-1-4666-8853-7.ch005
Afolabi, O. A. (2018). Myths and Challenges of Building an Effective Digital Library
in Developing Nations: An African Perspective. In A. Tella & T. Kwanya (Eds.),
Handbook of Research on Managing Intellectual Property in Digital Libraries (pp.
51–79). Hershey, PA: IGI Global. doi:10.4018/978-1-5225-3093-0.ch004
Agarwal, R., Singh, A., & Sen, S. (2016). Role of Molecular Docking in Computer-
Aided Drug Design and Development. In S. Dastmalchi, M. Hamzeh-Mivehroud, &
B. Sokouti (Eds.), Applied Case Studies and Solutions in Molecular Docking-Based
Drug Design (pp. 1–28). Hershey, PA: IGI Global. doi:10.4018/978-1-5225-0362-
0.ch001
Ali, O., & Soar, J. (2016). Technology Innovation Adoption Theories. In L. Al-
Hakim, X. Wu, A. Koronios, & Y. Shou (Eds.), Handbook of Research on Driving
Competitive Advantage through Sustainable, Lean, and Disruptive Innovation (pp.
Alsharo, M. (2017). Attitudes Towards Cloud Computing Adoption in Emerging
Economies. International Journal of Cloud Applications and Computing, 7(3),
44–58. doi:10.4018/IJCAC.2017070102
Amer, T. S., & Johnson, T. L. (2016). Information Technology Progress Indicators:
Temporal Expectancy, User Preference, and the Perception of Process Duration.
International Journal of Technology and Human Interaction, 12(4), 1–14. doi:10.4018/
IJTHI.2016100101
Amer, T. S., & Johnson, T. L. (2017). Information Technology Progress Indicators:
Research Employing Psychological Frameworks. In A. Mesquita (Ed.), Research
Paradigms and Contemporary Perspectives on Human-Technology Interaction (pp.
258
Related References
Anchugam, C. V., & Thangadurai, K. (2016). Introduction to Network Security. In D.

G., M. Singh, & M. Jayanthi (Eds.), Network Security Attacks and Countermeasures
(pp. 1-48). Hershey, PA: IGI Global. doi:10.4018/978-1-4666-8761-5.ch001
Anchugam, C. V., & Thangadurai, K. (2016). Classification of Network Attacks and
Countermeasures of Different Attacks. In D. G., M. Singh, & M. Jayanthi (Eds.),
Network Security Attacks and Countermeasures (pp. 115-156). Hershey, PA: IGI
Global. doi:10.4018/978-1-4666-8761-5.ch004
Anohah, E. (2016). Pedagogy and Design of Online Learning Environment in
Computer Science Education for High Schools. International Journal of Online
Pedagogy and Course Design, 6(3), 39–51. doi:10.4018/IJOPCD.2016070104
Anohah, E. (2017). Paradigm and Architecture of Computing Augmented Learning
Management System for Computer Science Education. International Journal of
Online Pedagogy and Course Design, 7(2), 60–70. doi:10.4018/IJOPCD.2017040105
Anohah, E., & Suhonen, J. (2017). Trends of Mobile Learning in Computing
Education from 2006 to 2014: A Systematic Review of Research Publications.
International Journal of Mobile and Blended Learning, 9(1), 16–33. doi:10.4018/
IJMBL.2017010102
Assis-Hassid, S., Heart, T., Reychav, I., & Pliskin, J. S. (2016). Modelling
Factors Affecting Patient-Doctor-Computer Communication in Primary Care.
International Journal of Reliable and Quality E-Healthcare, 5(1), 1–17. doi:10.4018/
IJRQEH.2016010101
Bailey, E. K. (2017). Applying Learning Theories to Computer Technology Supported
Instruction. In M. Grassetti & S. Brookby (Eds.), Advancing Next-Generation Teacher
Education through Digital Tools and Applications (pp. 61–81). Hershey, PA: IGI
Global. doi:10.4018/978-1-5225-0965-3.ch004
Balasubramanian, K. (2016). Attacks on Online Banking and Commerce. In K.
Balasubramanian, K. Mala, & M. Rajakani (Eds.), Cryptographic Solutions for
Secure Online Banking and Commerce (pp. 1–19). Hershey, PA: IGI Global.
doi:10.4018/978-1-5225-0273-9.ch001
Baldwin, S., Opoku-Agyemang, K., & Roy, D. (2016). Games People Play: A
Trilateral Collaboration Researching Computer Gaming across Cultures. In K.
Valentine & L. Jensen (Eds.), Examining the Evolution of Gaming and Its Impact
on Social, Cultural, and Political Perspectives (pp. 364–376). Hershey, PA: IGI
Global. doi:10.4018/978-1-5225-0261-6.ch017
259
Related References
Banerjee, S., Sing, T. Y., Chowdhury, A. R., & Anwar, H. (2018). Let’s Go Green:
Towards a Taxonomy of Green Computing Enablers for Business Sustainability. In
M. Khosrow-Pour (Ed.), Green Computing Strategies for Competitive Advantage and
Business Sustainability (pp. 89–109). Hershey, PA: IGI Global. doi:10.4018/978-
1-5225-5017-4.ch005
Basham, R. (2018). Information Science and Technology in Crisis Response and
Management. In M. Khosrow-Pour, D.B.A. (Ed.), Encyclopedia of Information
Science and Technology, Fourth Edition (pp. 1407-1418). Hershey, PA: IGI Global.
doi:10.4018/978-1-5225-2255-3.ch121
Batyashe, T., & Iyamu, T. (2018). Architectural Framework for the Implementation
of Information Technology Governance in Organisations. In M. Khosrow-Pour,
D.B.A. (Ed.), Encyclopedia of Information Science and Technology, Fourth Edition
Bekleyen, N., & Çelik, S. (2017). Attitudes of Adult EFL Learners towards Preparing
for a Language Test via CALL. In D. Tafazoli & M. Romero (Eds.), Multiculturalism
and Technology-Enhanced Language Learning (pp. 214–229). Hershey, PA: IGI
Global. doi:10.4018/978-1-5225-1882-2.ch013
Bennett, A., Eglash, R., Lachney, M., & Babbitt, W. (2016). Design Agency:
Diversifying Computer Science at the Intersections of Creativity and Culture. In M.
Raisinghani (Ed.), Revolutionizing Education through Web-Based Instruction (pp.
Bergeron, F., Croteau, A., Uwizeyemungu, S., & Raymond, L. (2017). A Framework
for Research on Information Technology Governance in SMEs. In S. De Haes & W.
Van Grembergen (Eds.), Strategic IT Governance and Alignment in Business Settings
(pp. 53–81). Hershey, PA: IGI Global. doi:10.4018/978-1-5225-0861-8.ch003
Bhatt, G. D., Wang, Z., & Rodger, J. A. (2017). Information Systems Capabilities and
Their Effects on Competitive Advantages: A Study of Chinese Companies. Information
Resources Management Journal, 30(3), 41–57. doi:10.4018/IRMJ.2017070103
Bogdanoski, M., Stoilkovski, M., & Risteski, A. (2016). Novel First Responder
Digital Forensics Tool as a Support to Law Enforcement. In M. Hadji-Janev & M.
Bogdanoski (Eds.), Handbook of Research on Civil Society and National Security in
the Era of Cyber Warfare (pp. 352–376). Hershey, PA: IGI Global. doi:10.4018/978-
1-4666-8793-6.ch016
260
Related References
Boontarig, W., Papasratorn, B., & Chutimaskul, W. (2016). The Unified Model for
Acceptance and Use of Health Information on Online Social Networks: Evidence
from Thailand. International Journal of E-Health and Medical Communications,
7(1), 31–47. doi:10.4018/IJEHMC.2016010102
Brown, S., & Yuan, X. (2016). Techniques for Retaining Computer Science Students
at Historical Black Colleges and Universities. In C. Prince & R. Ford (Eds.), Setting
a New Agenda for Student Engagement and Retention in Historically Black Colleges
and Universities (pp. 251–268). Hershey, PA: IGI Global. doi:10.4018/978-1-5225-
0308-8.ch014
Burcoff, A., & Shamir, L. (2017). Computer Analysis of Pablo Picasso’s Artistic
Style. International Journal of Art, Culture and Design Technologies, 6(1), 1–18.
doi:10.4018/IJACDT.2017010101
Byker, E. J. (2017). I Play I Learn: Introducing Technological Play Theory. In
C. Martin & D. Polly (Eds.), Handbook of Research on Teacher Education and
Professional Development (pp. 297–306). Hershey, PA: IGI Global. doi:10.4018/978-
1-5225-1067-3.ch016
Calongne, C. M., Stricker, A. G., Truman, B., & Arenas, F. J. (2017). Cognitive
Apprenticeship and Computer Science Education in Cyberspace: Reimagining the
Past. In A. Stricker, C. Calongne, B. Truman, & F. Arenas (Eds.), Integrating an
Awareness of Selfhood and Society into Virtual Learning (pp. 180–197). Hershey,
PA: IGI Global. doi:10.4018/978-1-5225-2182-2.ch013
Carlton, E. L., Holsinger, J. W. Jr, & Anunobi, N. (2016). Physician Engagement
with Health Information Technology: Implications for Practice and Professionalism.
International Journal of Computers in Clinical Practice, 1(2), 51–73. doi:10.4018/
IJCCP.2016070103
Carneiro, A. D. (2017). Defending Information Networks in Cyberspace: Some
Notes on Security Needs. In M. Dawson, D. Kisku, P. Gupta, J. Sing, & W. Li
(Eds.), Developing Next-Generation Countermeasures for Homeland Security
Threat Prevention (pp. 354-375). Hershey, PA: IGI Global. doi:10.4018/978-1-
5225-0703-1.ch016
Cavalcanti, J. C. (2016). The New “ABC” of ICTs (Analytics + Big Data + Cloud
Computing): A Complex Trade-Off between IT and CT Costs. In J. Martins & A.
Molnar (Eds.), Handbook of Research on Innovations in Information Retrieval,
Analysis, and Management (pp. 152–186). Hershey, PA: IGI Global. doi:10.4018/978-
1-4666-8833-9.ch006
261
Related References
Chase, J. P., & Yan, Z. (2017). Affect in Statistics Cognition. In Assessing and
Measuring Statistics Cognition in Higher Education Online Environments:
Emerging Research and Opportunities (pp. 144–187). Hershey, PA: IGI Global.
doi:10.4018/978-1-5225-2420-5.ch005
Chen, C. (2016). Effective Learning Strategies for the 21st Century: Implications
for the E-Learning. In M. Anderson & C. Gavan (Eds.), Developing Effective
Educational Experiences through Learning Analytics (pp. 143–169). Hershey, PA:
IGI Global. doi:10.4018/978-1-4666-9983-0.ch006
Chen, E. T. (2016). Examining the Influence of Information Technology on Modern
Health Care. In P. Manolitzas, E. Grigoroudis, N. Matsatsinis, & D. Yannacopoulos
(Eds.), Effective Methods for Modern Healthcare Service Quality and Evaluation
Cimermanova, I. (2017). Computer-Assisted Learning in Slovakia. In D. Tafazoli &
M. Romero (Eds.), Multiculturalism and Technology-Enhanced Language Learning
Cipolla-Ficarra, F. V., & Cipolla-Ficarra, M. (2018). Computer Animation for
Ingenious Revival. In F. Cipolla-Ficarra, M. Ficarra, M. Cipolla-Ficarra, A. Quiroga,
J. Alma, & J. Carré (Eds.), Technology-Enhanced Human Interaction in Modern
Society (pp. 159–181). Hershey, PA: IGI Global. doi:10.4018/978-1-5225-3437-2.
ch008
Cockrell, S., Damron, T. S., Melton, A. M., & Smith, A. D. (2018). Offshoring IT. In M.
Khosrow-Pour, D.B.A. (Ed.), Encyclopedia of Information Science and Technology,
Fourth Edition (pp. 5476-5489). Hershey, PA: IGI Global. doi:10.4018/978-1-5225-
2255-3.ch476
Coffey, J. W. (2018). Logic and Proof in Computer Science: Categories and Limits of
Proof Techniques. In J. Horne (Ed.), Philosophical Perceptions on Logic and Order
Dale, M. (2017). Re-Thinking the Challenges of Enterprise Architecture
Implementation. In M. Tavana (Ed.), Enterprise Information Systems and the
Digitalization of Business Functions (pp. 205–221). Hershey, PA: IGI Global.
doi:10.4018/978-1-5225-2382-6.ch009
262
Related References
Das, A., Dasgupta, R., & Bagchi, A. (2016). Overview of Cellular Computing-
Basic Principles and Applications. In J. Mandal, S. Mukhopadhyay, & T. Pal
(Eds.), Handbook of Research on Natural Computing for Optimization Problems
De Maere, K., De Haes, S., & von Kutzschenbach, M. (2017). CIO Perspectives
on Organizational Learning within the Context of IT Governance. International
Journal of IT/Business Alignment and Governance, 8(1), 32-47. doi:10.4018/
IJITBAG.2017010103
Demir, K., Çaka, C., Yaman, N. D., İslamoğlu, H., & Kuzu, A. (2018). Examining
the Current Definitions of Computational Thinking. In H. Ozcinar, G. Wong, & H.
Ozturk (Eds.), Teaching Computational Thinking in Primary Education (pp. 36–64).
Hershey, PA: IGI Global. doi:10.4018/978-1-5225-3200-2.ch003
Deng, X., Hung, Y., & Lin, C. D. (2017). Design and Analysis of Computer
Experiments. In S. Saha, A. Mandal, A. Narasimhamurthy, S. V, & S. Sangam (Eds.),
Handbook of Research on Applied Cybernetics and Systems Science (pp. 264-279).
Denner, J., Martinez, J., & Thiry, H. (2017). Strategies for Engaging Hispanic/
Latino Youth in the US in Computer Science. In Y. Rankin & J. Thomas (Eds.),
Moving Students of Color from Consumers to Producers of Technology (pp. 24–48).
Devi, A. (2017). Cyber Crime and Cyber Security: A Quick Glance. In R. Kumar, P.
Pattnaik, & P. Pandey (Eds.), Detecting and Mitigating Robotic Cyber Security Risks
Dores, A. R., Barbosa, F., Guerreiro, S., Almeida, I., & Carvalho, I. P. (2016).
Computer-Based Neuropsychological Rehabilitation: Virtual Reality and Serious
Games. In M. Cruz-Cunha, I. Miranda, R. Martinho, & R. Rijo (Eds.), Encyclopedia of
E-Health and Telemedicine (pp. 473–485). Hershey, PA: IGI Global. doi:10.4018/978-
1-4666-9978-6.ch037
Doshi, N., & Schaefer, G. (2016). Computer-Aided Analysis of Nailfold Capillaroscopy
Images. In D. Fotiadis (Ed.), Handbook of Research on Trends in the Diagnosis
and Treatment of Chronic Conditions (pp. 146–158). Hershey, PA: IGI Global.
doi:10.4018/978-1-4666-8828-5.ch007
263
Related References
Doyle, D. J., & Fahy, P. J. (2018). Interactivity in Distance Education and Computer-
Aided Learning, With Medical Education Examples. In M. Khosrow-Pour, D.B.A.
(Ed.), Encyclopedia of Information Science and Technology, Fourth Edition (pp.
5829-5840). Hershey, PA: IGI Global. doi:10.4018/978-1-5225-2255-3.ch507
Elias, N. I., & Walker, T. W. (2017). Factors that Contribute to Continued Use of
E-Training among Healthcare Professionals. In F. Topor (Ed.), Handbook of Research
on Individualism and Identity in the Globalized Digital Age (pp. 403–429). Hershey,
Eloy, S., Dias, M. S., Lopes, P. F., & Vilar, E. (2016). Digital Technologies in
Architecture and Engineering: Exploring an Engaged Interaction within Curricula.
In D. Fonseca & E. Redondo (Eds.), Handbook of Research on Applied E-Learning
in Engineering and Architecture Education (pp. 368–402). Hershey, PA: IGI Global.
doi:10.4018/978-1-4666-8803-2.ch017
Estrela, V. V., Magalhães, H. A., & Saotome, O. (2016). Total Variation Applications
in Computer Vision. In N. Kamila (Ed.), Handbook of Research on Emerging
Perspectives in Intelligent Pattern Recognition, Analysis, and Image Processing (pp.
Filipovic, N., Radovic, M., Nikolic, D. D., Saveljic, I., Milosevic, Z., Exarchos,
T. P., ... Parodi, O. (2016). Computer Predictive Model for Plaque Formation and
Progression in the Artery. In D. Fotiadis (Ed.), Handbook of Research on Trends in
the Diagnosis and Treatment of Chronic Conditions (pp. 279–300). Hershey, PA:
IGI Global. doi:10.4018/978-1-4666-8828-5.ch013
Fisher, R. L. (2018). Computer-Assisted Indian Matrimonial Services. In M.
2255-3.ch358
Fleenor, H. G., & Hodhod, R. (2016). Assessment of Learning and Technology:
Computer Science Education. In V. Wang (Ed.), Handbook of Research on Learning
Outcomes and Opportunities in the Digital Age (pp. 51–78). Hershey, PA: IGI
Global. doi:10.4018/978-1-4666-9577-1.ch003
García-Valcárcel, A., & Mena, J. (2016). Information Technology as a Way To Support
Collaborative Learning: What In-Service Teachers Think, Know and Do. Journal
of Information Technology Research, 9(1), 1–17. doi:10.4018/JITR.2016010101
264
Related References
Gardner-McCune, C., & Jimenez, Y. (2017). Historical App Developers: Integrating

CS into K-12 through Cross-Disciplinary Projects. In Y. Rankin & J. Thomas (Eds.),
Moving Students of Color from Consumers to Producers of Technology (pp. 85–112).
Garvey, G. P. (2016). Exploring Perception, Cognition, and Neural Pathways of
Stereo Vision and the Split–Brain Human Computer Interface. In A. Ursyn (Ed.),
Knowledge Visualization and Visual Literacy in Science Education (pp. 28–76).
Ghafele, R., & Gibert, B. (2018). Open Growth: The Economic Impact of Open
Source Software in the USA. In M. Khosrow-Pour (Ed.), Optimizing Contemporary
Application and Processes in Open Source Software (pp. 164–197). Hershey, PA:
IGI Global. doi:10.4018/978-1-5225-5314-4.ch007
Ghobakhloo, M., & Azar, A. (2018). Information Technology Resources, the
Organizational Capability of Lean-Agile Manufacturing, and Business Performance.
Information Resources Management Journal, 31(2), 47–74. doi:10.4018/
IRMJ.2018040103
Gianni, M., & Gotzamani, K. (2016). Integrated Management Systems and Information
Management Systems: Common Threads. In P. Papajorgji, F. Pinet, A. Guimarães,
& J. Papathanasiou (Eds.), Automated Enterprise Systems for Maximizing Business
Performance (pp. 195–214). Hershey, PA: IGI Global. doi:10.4018/978-1-4666-
8841-4.ch011
Gikandi, J. W. (2017). Computer-Supported Collaborative Learning and Assessment:
A Strategy for Developing Online Learning Communities in Continuing Education.
In J. Keengwe & G. Onchwari (Eds.), Handbook of Research on Learner-Centered
Pedagogy in Teacher Education and Professional Development (pp. 309–333).
Gokhale, A. A., & Machina, K. F. (2017). Development of a Scale to Measure
Attitudes toward Information Technology. In L. Tomei (Ed.), Exploring the New
Era of Technology-Infused Education (pp. 49–64). Hershey, PA: IGI Global.
doi:10.4018/978-1-5225-1709-2.ch004
Grace, A., O’Donoghue, J., Mahony, C., Heffernan, T., Molony, D., & Carroll, T.
(2016). Computerized Decision Support Systems for Multimorbidity Care: An Urgent
Call for Research and Development. In M. Cruz-Cunha, I. Miranda, R. Martinho, &
R. Rijo (Eds.), Encyclopedia of E-Health and Telemedicine (pp. 486–494). Hershey,
265
Related References
Gupta, A., & Singh, O. (2016). Computer Aided Modeling and Finite Element Analysis
of Human Elbow. International Journal of Biomedical and Clinical Engineering,
5(1), 31–38. doi:10.4018/IJBCE.2016010104
H., S. K. (2016). Classification of Cybercrimes and Punishments under the Information
Technology Act, 2000. In S. Geetha, & A. Phamila (Eds.), Combating Security
Breaches and Criminal Activity in the Digital Sphere (pp. 57-66). Hershey, PA: IGI
Global. doi:10.4018/978-1-5225-0193-0.ch004
Hafeez-Baig, A., Gururajan, R., & Wickramasinghe, N. (2017). Readiness as a
Novel Construct of Readiness Acceptance Model (RAM) for the Wireless Handheld
Technology. In N. Wickramasinghe (Ed.), Handbook of Research on Healthcare
Administration and Management (pp. 578–595). Hershey, PA: IGI Global.
doi:10.4018/978-1-5225-0920-2.ch035
Hanafizadeh, P., Ghandchi, S., & Asgarimehr, M. (2017). Impact of Information
Technology on Lifestyle: A Literature Review and Classification. International
Journal of Virtual Communities and Social Networking, 9(2), 1–23. doi:10.4018/
IJVCSN.2017040101
Harlow, D. B., Dwyer, H., Hansen, A. K., Hill, C., Iveland, A., Leak, A. E., &
Franklin, D. M. (2016). Computer Programming in Elementary and Middle School:
Connections across Content. In M. Urban & D. Falvo (Eds.), Improving K-12 STEM
Education Outcomes through Technological Integration (pp. 337–361). Hershey,
Haseski, H. İ., Ilic, U., & Tuğtekin, U. (2018). Computational Thinking in Educational
Digital Games: An Assessment Tool Proposal. In H. Ozcinar, G. Wong, & H. Ozturk
(Eds.), Teaching Computational Thinking in Primary Education (pp. 256–287).
Hee, W. J., Jalleh, G., Lai, H., & Lin, C. (2017). E-Commerce and IT Projects:
Evaluation and Management Issues in Australian and Taiwanese Hospitals.
International Journal of Public Health Management and Ethics, 2(1), 69–90.
doi:10.4018/IJPHME.2017010104
Hernandez, A. A. (2017). Green Information Technology Usage: Awareness and
Practices of Philippine IT Professionals. International Journal of Enterprise
Information Systems, 13(4), 90–103. doi:10.4018/IJEIS.2017100106
266
Related References
Hernandez, A. A., & Ona, S. E. (2016). Green IT Adoption: Lessons from the
Philippines Business Process Outsourcing Industry. International Journal of Social
Ecology and Sustainable Development, 7(1), 1–34. doi:10.4018/IJSESD.2016010101
Hernandez, M. A., Marin, E. C., Garcia-Rodriguez, J., Azorin-Lopez, J., & Cazorla,
M. (2017). Automatic Learning Improves Human-Robot Interaction in Productive
Environments: A Review. International Journal of Computer Vision and Image
Processing, 7(3), 65–75. doi:10.4018/IJCVIP.2017070106
Horne-Popp, L. M., Tessone, E. B., & Welker, J. (2018). If You Build It, They Will
Come: Creating a Library Statistics Dashboard for Decision-Making. In L. Costello
& M. Powers (Eds.), Developing In-House Digital Tools in Library Spaces (pp.
Hossan, C. G., & Ryan, J. C. (2016). Factors Affecting e-Government Technology
Adoption Behaviour in a Voluntary Environment. International Journal of Electronic
Government Research, 12(1), 24–49. doi:10.4018/IJEGR.2016010102
Hu, H., Hu, P. J., & Al-Gahtani, S. S. (2017). User Acceptance of Computer
Technology at Work in Arabian Culture: A Model Comparison Approach. In M.
Khosrow-Pour (Ed.), Handbook of Research on Technology Adoption, Social Policy,
and Global Integration (pp. 205–228). Hershey, PA: IGI Global. doi:10.4018/978-
1-5225-2668-1.ch011
Huie, C. P. (2016). Perceptions of Business Intelligence Professionals about Factors
Related to Business Intelligence input in Decision Making. International Journal
of Business Analytics, 3(3), 1–24. doi:10.4018/IJBAN.2016070101
Hung, S., Huang, W., Yen, D. C., Chang, S., & Lu, C. (2016). Effect of Information
Service Competence and Contextual Factors on the Effectiveness of Strategic
Information Systems Planning in Hospitals. Journal of Global Information
Management, 24(1), 14–36. doi:10.4018/JGIM.2016010102
Ifinedo, P. (2017). Using an Extended Theory of Planned Behavior to Study Nurses’
Adoption of Healthcare Information Systems in Nova Scotia. International Journal
of Technology Diffusion, 8(1), 1–17. doi:10.4018/IJTD.2017010101
Ilie, V., & Sneha, S. (2018). A Three Country Study for Understanding Physicians’
Engagement With Electronic Information Resources Pre and Post System
Implementation. Journal of Global Information Management, 26(2), 48–73.
doi:10.4018/JGIM.2018040103
267
Related References
Inoue-Smith, Y. (2017). Perceived Ease in Using Technology Predicts Teacher

Candidates’ Preferences for Online Resources. International Journal of Online
Pedagogy and Course Design, 7(3), 17–28. doi:10.4018/IJOPCD.2017070102
Islam, A. A. (2016). Development and Validation of the Technology Adoption and
Gratification (TAG) Model in Higher Education: A Cross-Cultural Study Between
Malaysia and China. International Journal of Technology and Human Interaction,
12(3), 78–105. doi:10.4018/IJTHI.2016070106
Islam, A. Y. (2017). Technology Satisfaction in an Academic Context: Moderating
Effect of Gender. In A. Mesquita (Ed.), Research Paradigms and Contemporary
Perspectives on Human-Technology Interaction (pp. 187–211). Hershey, PA: IGI
Global. doi:10.4018/978-1-5225-1868-6.ch009
Jamil, G. L., & Jamil, C. C. (2017). Information and Knowledge Management
Perspective Contributions for Fashion Studies: Observing Logistics and Supply
Chain Management Processes. In G. Jamil, A. Soares, & C. Pessoa (Eds.), Handbook
of Research on Information Management for Effective Logistics and Supply Chains
Jamil, G. L., Jamil, L. C., Vieira, A. A., & Xavier, A. J. (2016). Challenges in Modelling
Healthcare Services: A Study Case of Information Architecture Perspectives. In G.
Jamil, J. Poças Rascão, F. Ribeiro, & A. Malheiro da Silva (Eds.), Handbook of
Research on Information Architecture and Management in Modern Organizations
Janakova, M. (2018). Big Data and Simulations for the Solution of Controversies in
Small Businesses. In M. Khosrow-Pour, D.B.A. (Ed.), Encyclopedia of Information
doi:10.4018/978-1-5225-2255-3.ch598
Jha, D. G. (2016). Preparing for Information Technology Driven Changes. In S.
Tiwari & L. Nafees (Eds.), Innovative Management Education Pedagogies for
Preparing Next-Generation Leaders (pp. 258–274). Hershey, PA: IGI Global.
doi:10.4018/978-1-4666-9691-4.ch015
Jhawar, A., & Garg, S. K. (2018). Logistics Improvement by Investment in
Information Technology Using System Dynamics. In A. Azar & S. Vaidyanathan
(Eds.), Advances in System Dynamics and Control (pp. 528–567). Hershey, PA: IGI
Global. doi:10.4018/978-1-5225-4077-9.ch017
268
Related References
Kalelioğlu, F., Gülbahar, Y., & Doğan, D. (2018). Teaching How to Think Like
a Programmer: Emerging Insights. In H. Ozcinar, G. Wong, & H. Ozturk (Eds.),
Teaching Computational Thinking in Primary Education (pp. 18–35). Hershey, PA:
IGI Global. doi:10.4018/978-1-5225-3200-2.ch002
Kamberi, S. (2017). A Girls-Only Online Virtual World Environment and its
Implications for Game-Based Learning. In A. Stricker, C. Calongne, B. Truman,
& F. Arenas (Eds.), Integrating an Awareness of Selfhood and Society into Virtual
Learning (pp. 74–95). Hershey, PA: IGI Global. doi:10.4018/978-1-5225-2182-2.
ch006
Kamel, S., & Rizk, N. (2017). ICT Strategy Development: From Design to
Implementation – Case of Egypt. In C. Howard & K. Hargiss (Eds.), Strategic
Information Systems and Technologies in Modern Organizations (pp. 239–257).
Kamel, S. H. (2018). The Potential Role of the Software Industry in Supporting
Economic Development. In M. Khosrow-Pour, D.B.A. (Ed.), Encyclopedia of
Information Science and Technology, Fourth Edition (pp. 7259-7269). Hershey,
Karon, R. (2016). Utilisation of Health Information Systems for Service Delivery in
the Namibian Environment. In T. Iyamu & A. Tatnall (Eds.), Maximizing Healthcare
Delivery and Management through Technology Integration (pp. 169–183). Hershey,
Kawata, S. (2018). Computer-Assisted Parallel Program Generation. In M. Khosrow-
Pour, D.B.A. (Ed.), Encyclopedia of Information Science and Technology, Fourth
Edition (pp. 4583-4593). Hershey, PA: IGI Global. doi:10.4018/978-1-5225-2255-3.
ch398
Khanam, S., Siddiqui, J., & Talib, F. (2016). A DEMATEL Approach for Prioritizing
the TQM Enablers and IT Resources in the Indian ICT Industry. International
Journal of Applied Management Sciences and Engineering, 3(1), 11–29. doi:10.4018/
IJAMSE.2016010102
Khari, M., Shrivastava, G., Gupta, S., & Gupta, R. (2017). Role of Cyber Security
in Today’s Scenario. In R. Kumar, P. Pattnaik, & P. Pandey (Eds.), Detecting and
Mitigating Robotic Cyber Security Risks (pp. 177–191). Hershey, PA: IGI Global.
doi:10.4018/978-1-5225-2154-9.ch013
269
Related References
Khouja, M., Rodriguez, I. B., Ben Halima, Y., & Moalla, S. (2018). IT Governance
in Higher Education Institutions: A Systematic Literature Review. International
Journal of Human Capital and Information Technology Professionals, 9(2), 52–67.
doi:10.4018/IJHCITP.2018040104
Kim, S., Chang, M., Choi, N., Park, J., & Kim, H. (2016). The Direct and Indirect
Effects of Computer Uses on Student Success in Math. International Journal of Cyber
Behavior, Psychology and Learning, 6(3), 48–64. doi:10.4018/IJCBPL.2016070104
Kiourt, C., Pavlidis, G., Koutsoudis, A., & Kalles, D. (2017). Realistic Simulation
of Cultural Heritage. International Journal of Computational Methods in Heritage
Science, 1(1), 10–40. doi:10.4018/IJCMHS.2017010102
Korikov, A., & Krivtsov, O. (2016). System of People-Computer: On the Way
of Creation of Human-Oriented Interface. In V. Mkrttchian, A. Bershadsky, A.
Bozhday, M. Kataev, & S. Kataev (Eds.), Handbook of Research on Estimation and
Control Techniques in E-Learning Systems (pp. 458–470). Hershey, PA: IGI Global.
doi:10.4018/978-1-4666-9489-7.ch032
Köse, U. (2017). An Augmented-Reality-Based Intelligent Mobile Application
for Open Computer Education. In G. Kurubacak & H. Altinpulluk (Eds.), Mobile
Technologies and Augmented Reality in Open Education (pp. 154–174). Hershey,
Lahmiri, S. (2018). Information Technology Outsourcing Risk Factors and Provider
Selection. In M. Gupta, R. Sharman, J. Walp, & P. Mulgund (Eds.), Information
Technology Risk Management and Compliance in Modern Organizations (pp.
Landriscina, F. (2017). Computer-Supported Imagination: The Interplay Between
Computer and Mental Simulation in Understanding Scientific Concepts. In I. Levin &
D. Tsybulsky (Eds.), Digital Tools and Solutions for Inquiry-Based STEM Learning
Lau, S. K., Winley, G. K., Leung, N. K., Tsang, N., & Lau, S. Y. (2016). An Exploratory
Study of Expectation in IT Skills in a Developing Nation: Vietnam. Journal of Global
Information Management, 24(1), 1–13. doi:10.4018/JGIM.2016010101
Lavranos, C., Kostagiolas, P., & Papadatos, J. (2016). Information Retrieval
Technologies and the “Realities” of Music Information Seeking. In I. Deliyannis, P.
Kostagiolas, & C. Banou (Eds.), Experimental Multimedia Systems for Interactivity
and Strategic Innovation (pp. 102–121). Hershey, PA: IGI Global. doi:10.4018/978-
1-4666-8659-5.ch005
270
Related References
Lee, W. W. (2018). Ethical Computing Continues From Problem to Solution. In M.

2255-3.ch423
Lehto, M. (2016). Cyber Security Education and Research in the Finland’s Universities
and Universities of Applied Sciences. International Journal of Cyber Warfare &
Terrorism, 6(2), 15–31. doi:10.4018/IJCWT.2016040102
Lin, C., Jalleh, G., & Huang, Y. (2016). Evaluating and Managing Electronic
Commerce and Outsourcing Projects in Hospitals. In A. Dwivedi (Ed.), Reshaping
Medical Practice and Care with Health Information Systems (pp. 132–172). Hershey,
Lin, S., Chen, S., & Chuang, S. (2017). Perceived Innovation and Quick Response
Codes in an Online-to-Offline E-Commerce Service Model. International Journal
of E-Adoption, 9(2), 1–16. doi:10.4018/IJEA.2017070101
Liu, M., Wang, Y., Xu, W., & Liu, L. (2017). Automated Scoring of Chinese
Engineering Students’ English Essays. International Journal of Distance Education
Technologies, 15(1), 52–68. doi:10.4018/IJDET.2017010104
Luciano, E. M., Wiedenhöft, G. C., Macadar, M. A., & Pinheiro dos Santos, F. (2016).
Information Technology Governance Adoption: Understanding its Expectations
Through the Lens of Organizational Citizenship. International Journal of IT/Business
Alignment and Governance, 7(2), 22-32. doi:10.4018/IJITBAG.2016070102
Mabe, L. K., & Oladele, O. I. (2017). Application of Information Communication
Technologies for Agricultural Development through Extension Services: A Review. In
T. Tossy (Ed.), Information Technology Integration for Socio-Economic Development
Manogaran, G., Thota, C., & Lopez, D. (2018). Human-Computer Interaction
With Big Data Analytics. In D. Lopez & M. Durai (Eds.), HCI Challenges and
Privacy Preservation in Big Data Security (pp. 1–22). Hershey, PA: IGI Global.
doi:10.4018/978-1-5225-2863-0.ch001
Margolis, J., Goode, J., & Flapan, J. (2017). A Critical Crossroads for Computer
Science for All: “Identifying Talent” or “Building Talent,” and What Difference
Does It Make? In Y. Rankin & J. Thomas (Eds.), Moving Students of Color from
Consumers to Producers of Technology (pp. 1–23). Hershey, PA: IGI Global.
doi:10.4018/978-1-5225-2005-4.ch001
271
Related References
Mbale, J. (2018). Computer Centres Resource Cloud Elasticity-Scalability (CRECES):

Copperbelt University Case Study. In S. Aljawarneh & M. Malhotra (Eds.), Critical
Research on Scalability and Security Issues in Virtual Cloud Environments (pp.
McKee, J. (2018). The Right Information: The Key to Effective Business Planning.
In Business Architectures for Risk Assessment and Strategic Planning: Emerging
Research and Opportunities (pp. 38–52). Hershey, PA: IGI Global. doi:10.4018/978-
1-5225-3392-4.ch003
Mensah, I. K., & Mi, J. (2018). Determinants of Intention to Use Local E-Government
Services in Ghana: The Perspective of Local Government Workers. International
Journal of Technology Diffusion, 9(2), 41–60. doi:10.4018/IJTD.2018040103
Mohamed, J. H. (2018). Scientograph-Based Visualization of Computer Forensics
Research Literature. In J. Jeyasekar & P. Saravanan (Eds.), Innovations in Measuring
and Evaluating Scientific Information (pp. 148–162). Hershey, PA: IGI Global.
doi:10.4018/978-1-5225-3457-0.ch010
Moore, R. L., & Johnson, N. (2017). Earning a Seat at the Table: How IT Departments
Can Partner in Organizational Change and Innovation. International Journal of
Knowledge-Based Organizations, 7(2), 1–12. doi:10.4018/IJKBO.2017040101
Mtebe, J. S., & Kissaka, M. M. (2016). Enhancing the Quality of Computer Science
Education with MOOCs in Sub-Saharan Africa. In J. Keengwe & G. Onchwari (Eds.),
Handbook of Research on Active Learning and the Flipped Classroom Model in
the Digital Age (pp. 366–377). Hershey, PA: IGI Global. doi:10.4018/978-1-4666-
9680-8.ch019
Mukul, M. K., & Bhattaharyya, S. (2017). Brain-Machine Interface: Human-Computer
Interaction. In E. Noughabi, B. Raahemi, A. Albadvi, & B. Far (Eds.), Handbook
of Research on Data Science for Effective Healthcare Practice and Administration
Na, L. (2017). Library and Information Science Education and Graduate Programs
in Academic Libraries. In L. Ruan, Q. Zhu, & Y. Ye (Eds.), Academic Library
Development and Administration in China (pp. 218–229). Hershey, PA: IGI Global.
doi:10.4018/978-1-5225-0550-1.ch013
Nabavi, A., Taghavi-Fard, M. T., Hanafizadeh, P., & Taghva, M. R. (2016). Information
Technology Continuance Intention: A Systematic Literature Review. International
Journal of E-Business Research, 12(1), 58–95. doi:10.4018/IJEBR.2016010104
272
Related References
Nath, R., & Murthy, V. N. (2018). What Accounts for the Differences in Internet
Diffusion Rates Around the World? In M. Khosrow-Pour, D.B.A. (Ed.), Encyclopedia
of Information Science and Technology, Fourth Edition (pp. 8095-8104). Hershey,
Nedelko, Z., & Potocan, V. (2018). The Role of Emerging Information Technologies
for Supporting Supply Chain Management. In M. Khosrow-Pour, D.B.A. (Ed.),
Encyclopedia of Information Science and Technology, Fourth Edition (pp. 5559-
5569). Hershey, PA: IGI Global. doi:10.4018/978-1-5225-2255-3.ch483
Ngafeeson, M. N. (2018). User Resistance to Health Information Technology. In M.
2255-3.ch331
Nozari, H., Najafi, S. E., Jafari-Eskandari, M., & Aliahmadi, A. (2016). Providing
a Model for Virtual Project Management with an Emphasis on IT Projects. In C.
Graham (Ed.), Strategic Management and Leadership for Systems Development in
Virtual Spaces (pp. 43–63). Hershey, PA: IGI Global. doi:10.4018/978-1-4666-
9688-4.ch003
Nurdin, N., Stockdale, R., & Scheepers, H. (2016). Influence of Organizational Factors
in the Sustainability of E-Government: A Case Study of Local E-Government in
Indonesia. In I. Sodhi (Ed.), Trends, Prospects, and Challenges in Asian E-Governance
Odagiri, K. (2017). Introduction of Individual Technology to Constitute the
Current Internet. In Strategic Policy-Based Network Management in Contemporary
Organizations (pp. 20–96). Hershey, PA: IGI Global. doi:10.4018/978-1-68318-
003-6.ch003
Okike, E. U. (2018). Computer Science and Prison Education. In I. Biao (Ed.),
Strategic Learning Ideologies in Prison Education Programs (pp. 246–264). Hershey,
Olelewe, C. J., & Nwafor, I. P. (2017). Level of Computer Appreciation Skills
Acquired for Sustainable Development by Secondary School Students in Nsukka
LGA of Enugu State, Nigeria. In C. Ayo & V. Mbarika (Eds.), Sustainable ICT
Adoption and Integration for Socio-Economic Development (pp. 214–233). Hershey,
273
Related References
Oliveira, M., Maçada, A. C., Curado, C., & Nodari, F. (2017). Infrastructure Profiles
and Knowledge Sharing. International Journal of Technology and Human Interaction,
13(3), 1–12. doi:10.4018/IJTHI.2017070101
Otarkhani, A., Shokouhyar, S., & Pour, S. S. (2017). Analyzing the Impact of
Governance of Enterprise IT on Hospital Performance: Tehran’s (Iran) Hospitals
– A Case Study. International Journal of Healthcare Information Systems and
Informatics, 12(3), 1–20. doi:10.4018/IJHISI.2017070101
Otunla, A. O., & Amuda, C. O. (2018). Nigerian Undergraduate Students’ Computer
Competencies and Use of Information Technology Tools and Resources for Study
Skills and Habits’ Enhancement. In M. Khosrow-Pour, D.B.A. (Ed.), Encyclopedia
Özçınar, H. (2018). A Brief Discussion on Incentives and Barriers to Computational
Thinking Education. In H. Ozcinar, G. Wong, & H. Ozturk (Eds.), Teaching
Computational Thinking in Primary Education (pp. 1–17). Hershey, PA: IGI Global.
doi:10.4018/978-1-5225-3200-2.ch001
Pandey, J. M., Garg, S., Mishra, P., & Mishra, B. P. (2017). Computer Based
Psychological Interventions: Subject to the Efficacy of Psychological Services.
International Journal of Computers in Clinical Practice, 2(1), 25–33. doi:10.4018/
IJCCP.2017010102
Parry, V. K., & Lind, M. L. (2016). Alignment of Business Strategy and Information
Technology Considering Information Technology Governance, Project Portfolio
Control, and Risk Management. International Journal of Information Technology
Project Management, 7(4), 21–37. doi:10.4018/IJITPM.2016100102
Patro, C. (2017). Impulsion of Information Technology on Human Resource Practices.
In P. Ordóñez de Pablos (Ed.), Managerial Strategies and Solutions for Business
Success in Asia (pp. 231–254). Hershey, PA: IGI Global. doi:10.4018/978-1-5225-
1886-0.ch013
Patro, C. S., & Raghunath, K. M. (2017). Information Technology Paraphernalia for
Supply Chain Management Decisions. In M. Tavana (Ed.), Enterprise Information
Systems and the Digitalization of Business Functions (pp. 294–320). Hershey, PA:
IGI Global. doi:10.4018/978-1-5225-2382-6.ch014
274
Related References
Paul, P. K. (2016). Cloud Computing: An Agent of Promoting Interdisciplinary

Sciences, Especially Information Science and I-Schools – Emerging Techno-
Educational Scenario. In L. Chao (Ed.), Handbook of Research on Cloud-Based
STEM Education for Improved Learning Outcomes (pp. 247–258). Hershey, PA:
IGI Global. doi:10.4018/978-1-4666-9924-3.ch016
Paul, P. K. (2018). The Context of IST for Solid Information Retrieval and
Infrastructure Building: Study of Developing Country. International Journal of
Information Retrieval Research, 8(1), 86–100. doi:10.4018/IJIRR.2018010106
Paul, P. K., & Chatterjee, D. (2018). iSchools Promoting “Information Science
and Technology” (IST) Domain Towards Community, Business, and Society
With Contemporary Worldwide Trend and Emerging Potentialities in India. In M.
2255-3.ch410
Pessoa, C. R., & Marques, M. E. (2017). Information Technology and Communication
Management in Supply Chain Management. In G. Jamil, A. Soares, & C. Pessoa
(Eds.), Handbook of Research on Information Management for Effective Logistics
and Supply Chains (pp. 23–33). Hershey, PA: IGI Global. doi:10.4018/978-1-5225-
0973-8.ch002
Pineda, R. G. (2016). Where the Interaction Is Not: Reflections on the Philosophy
of Human-Computer Interaction. International Journal of Art, Culture and Design
Technologies, 5(1), 1–12. doi:10.4018/IJACDT.2016010101
Pineda, R. G. (2018). Remediating Interaction: Towards a Philosophy of Human-
Computer Relationship. In M. Khosrow-Pour (Ed.), Enhancing Art, Culture, and
Design With Technological Integration (pp. 75–98). Hershey, PA: IGI Global.
doi:10.4018/978-1-5225-5023-5.ch004
Poikela, P., & Vuojärvi, H. (2016). Learning ICT-Mediated Communication through
Computer-Based Simulations. In M. Cruz-Cunha, I. Miranda, R. Martinho, & R.
Rijo (Eds.), Encyclopedia of E-Health and Telemedicine (pp. 674–687). Hershey,
Qian, Y. (2017). Computer Simulation in Higher Education: Affordances,
Opportunities, and Outcomes. In P. Vu, S. Fredrickson, & C. Moore (Eds.), Handbook
of Research on Innovative Pedagogies and Technologies for Online Learning in
Higher Education (pp. 236–262). Hershey, PA: IGI Global. doi:10.4018/978-1-
5225-1851-8.ch011
275
Related References
Radant, O., Colomo-Palacios, R., & Stantchev, V. (2016). Factors for the Management
of Scarce Human Resources and Highly Skilled Employees in IT-Departments:
A Systematic Review. Journal of Information Technology Research, 9(1), 65–82.
doi:10.4018/JITR.2016010105
Rahman, N. (2016). Toward Achieving Environmental Sustainability in the Computer
Industry. International Journal of Green Computing, 7(1), 37–54. doi:10.4018/
IJGC.2016010103
Rahman, N. (2017). Lessons from a Successful Data Warehousing Project
Management. International Journal of Information Technology Project Management,
8(4), 30–45. doi:10.4018/IJITPM.2017100103
Rahman, N. (2018). Environmental Sustainability in the Computer Industry for
Competitive Advantage. In M. Khosrow-Pour (Ed.), Green Computing Strategies
for Competitive Advantage and Business Sustainability (pp. 110–130). Hershey, PA:
IGI Global. doi:10.4018/978-1-5225-5017-4.ch006
Rajh, A., & Pavetic, T. (2017). Computer Generated Description as the Required
Digital Competence in Archival Profession. International Journal of Digital Literacy
and Digital Competence, 8(1), 36–49. doi:10.4018/IJDLDC.2017010103
Raman, A., & Goyal, D. P. (2017). Extending IMPLEMENT Framework for Enterprise
Information Systems Implementation to Information System Innovation. In M. Tavana
(Ed.), Enterprise Information Systems and the Digitalization of Business Functions
Rao, Y. S., Rauta, A. K., Saini, H., & Panda, T. C. (2017). Mathematical Model
for Cyber Attack in Computer Network. International Journal of Business Data
Communications and Networking, 13(1), 58–65. doi:10.4018/IJBDCN.2017010105
Rapaport, W. J. (2018). Syntactic Semantics and the Proper Treatment of
Computationalism. In M. Danesi (Ed.), Empirical Research on Semiotics and Visual
Rhetoric (pp. 128–176). Hershey, PA: IGI Global. doi:10.4018/978-1-5225-5622-0.
ch007
Raut, R., Priyadarshinee, P., & Jha, M. (2017). Understanding the Mediation Effect
of Cloud Computing Adoption in Indian Organization: Integrating TAM-TOE- Risk
Model. International Journal of Service Science, Management, Engineering, and
Technology, 8(3), 40–59. doi:10.4018/IJSSMET.2017070103
276
Related References
Regan, E. A., & Wang, J. (2016). Realizing the Value of EHR Systems Critical
Success Factors. International Journal of Healthcare Information Systems and
Informatics, 11(3), 1–18. doi:10.4018/IJHISI.2016070101
Rezaie, S., Mirabedini, S. J., & Abtahi, A. (2018). Designing a Model for
Implementation of Business Intelligence in the Banking Industry. International Journal
of Enterprise Information Systems, 14(1), 77–103. doi:10.4018/IJEIS.2018010105
Rezende, D. A. (2016). Digital City Projects: Information and Public Services Offered
by Chicago (USA) and Curitiba (Brazil). International Journal of Knowledge Society
Research, 7(3), 16–30. doi:10.4018/IJKSR.2016070102
Rezende, D. A. (2018). Strategic Digital City Projects: Innovative Information and
Public Services Offered by Chicago (USA) and Curitiba (Brazil). In M. Lytras, L.
Daniela, & A. Visvizi (Eds.), Enhancing Knowledge Discovery and Innovation in
the Digital Era (pp. 204–223). Hershey, PA: IGI Global. doi:10.4018/978-1-5225-
4191-2.ch012
Riabov, V. V. (2016). Teaching Online Computer-Science Courses in LMS and
Cloud Environment. International Journal of Quality Assurance in Engineering
and Technology Education, 5(4), 12–41. doi:10.4018/IJQAETE.2016100102
Ricordel, V., Wang, J., Da Silva, M. P., & Le Callet, P. (2016). 2D and 3D Visual
Attention for Computer Vision: Concepts, Measurement, and Modeling. In R. Pal
(Ed.), Innovative Research in Attention Modeling and Computer Vision Applications
Rodriguez, A., Rico-Diaz, A. J., Rabuñal, J. R., & Gestal, M. (2017). Fish Tracking
with Computer Vision Techniques: An Application to Vertical Slot Fishways. In
M. S., & V. V. (Eds.), Multi-Core Computer Vision and Image Processing for
Intelligent Applications (pp. 74-104). Hershey, PA: IGI Global. doi:10.4018/978-
1-5225-0889-2.ch003
Romero, J. A. (2018). Sustainable Advantages of Business Value of Information
Technology. In M. Khosrow-Pour, D.B.A. (Ed.), Encyclopedia of Information
doi:10.4018/978-1-5225-2255-3.ch079
Romero, J. A. (2018). The Always-On Business Model and Competitive Advantage.
In N. Bajgoric (Ed.), Always-On Enterprise Information Systems for Modern
Organizations (pp. 23–40). Hershey, PA: IGI Global. doi:10.4018/978-1-5225-
3704-5.ch002
277
Related References
Rosen, Y. (2018). Computer Agent Technologies in Collaborative Learning and

Assessment. In M. Khosrow-Pour, D.B.A. (Ed.), Encyclopedia of Information
doi:10.4018/978-1-5225-2255-3.ch209
Rosen, Y., & Mosharraf, M. (2016). Computer Agent Technologies in Collaborative
Assessments. In Y. Rosen, S. Ferrara, & M. Mosharraf (Eds.), Handbook of Research
on Technology Tools for Real-World Skill Development (pp. 319–343). Hershey, PA:
IGI Global. doi:10.4018/978-1-4666-9441-5.ch012
Roy, D. (2018). Success Factors of Adoption of Mobile Applications in Rural India:
Effect of Service Characteristics on Conceptual Model. In M. Khosrow-Pour (Ed.),
Green Computing Strategies for Competitive Advantage and Business Sustainability
Ruffin, T. R. (2016). Health Information Technology and Change. In V. Wang (Ed.),
Handbook of Research on Advancing Health Education through Technology (pp.
Ruffin, T. R. (2016). Health Information Technology and Quality Management.
International Journal of Information Communication Technologies and Human
Development, 8(4), 56–72. doi:10.4018/IJICTHD.2016100105
Ruffin, T. R., & Hawkins, D. P. (2018). Trends in Health Care Information Technology
and Informatics. In M. Khosrow-Pour, D.B.A. (Ed.), Encyclopedia of Information
doi:10.4018/978-1-5225-2255-3.ch330
Safari, M. R., & Jiang, Q. (2018). The Theory and Practice of IT Governance
Maturity and Strategies Alignment: Evidence From Banking Industry. Journal of
Global Information Management, 26(2), 127–146. doi:10.4018/JGIM.2018040106
Sahin, H. B., & Anagun, S. S. (2018). Educational Computer Games in Math
Teaching: A Learning Culture. In E. Toprak & E. Kumtepe (Eds.), Supporting
Multiculturalism in Open and Distance Learning Spaces (pp. 249–280). Hershey,
Sanna, A., & Valpreda, F. (2017). An Assessment of the Impact of a Collaborative
Didactic Approach and Students’ Background in Teaching Computer Animation.
International Journal of Information and Communication Technology Education,
13(4), 1–16. doi:10.4018/IJICTE.2017100101
278
Related References
Savita, K., Dominic, P., & Ramayah, T. (2016). The Drivers, Practices and Outcomes
of Green Supply Chain Management: Insights from ISO14001 Manufacturing
Firms in Malaysia. International Journal of Information Systems and Supply Chain
Management, 9(2), 35–60. doi:10.4018/IJISSCM.2016040103
Scott, A., Martin, A., & McAlear, F. (2017). Enhancing Participation in Computer
Science among Girls of Color: An Examination of a Preparatory AP Computer
Science Intervention. In Y. Rankin & J. Thomas (Eds.), Moving Students of Color
from Consumers to Producers of Technology (pp. 62–84). Hershey, PA: IGI Global.
doi:10.4018/978-1-5225-2005-4.ch004
Shahsavandi, E., Mayah, G., & Rahbari, H. (2016). Impact of E-Government
on Transparency and Corruption in Iran. In I. Sodhi (Ed.), Trends, Prospects,
and Challenges in Asian E-Governance (pp. 75–94). Hershey, PA: IGI Global.
doi:10.4018/978-1-4666-9536-8.ch004
Siddoo, V., & Wongsai, N. (2017). Factors Influencing the Adoption of ISO/IEC 29110
in Thai Government Projects: A Case Study. International Journal of Information
Technologies and Systems Approach, 10(1), 22–44. doi:10.4018/IJITSA.2017010102
Sidorkina, I., & Rybakov, A. (2016). Computer-Aided Design as Carrier of Set
Development Changes System in E-Course Engineering. In V. Mkrttchian, A.
Bershadsky, A. Bozhday, M. Kataev, & S. Kataev (Eds.), Handbook of Research on
Estimation and Control Techniques in E-Learning Systems (pp. 500–515). Hershey,
Sidorkina, I., & Rybakov, A. (2016). Creating Model of E-Course: As an Object of
Computer-Aided Design. In V. Mkrttchian, A. Bershadsky, A. Bozhday, M. Kataev,
& S. Kataev (Eds.), Handbook of Research on Estimation and Control Techniques
in E-Learning Systems (pp. 286–297). Hershey, PA: IGI Global. doi:10.4018/978-
1-4666-9489-7.ch019
Simões, A. (2017). Using Game Frameworks to Teach Computer Programming. In
R. Alexandre Peixoto de Queirós & M. Pinto (Eds.), Gamification-Based E-Learning
Strategies for Computer Programming Education (pp. 221–236). Hershey, PA: IGI
Global. doi:10.4018/978-1-5225-1034-5.ch010
Sllame, A. M. (2017). Integrating LAB Work With Classes in Computer Network
Courses. In H. Alphin Jr, R. Chan, & J. Lavine (Eds.), The Future of Accessibility
in International Higher Education (pp. 253–275). Hershey, PA: IGI Global.
doi:10.4018/978-1-5225-2560-8.ch015
279
Related References
Smirnov, A., Ponomarev, A., Shilov, N., Kashevnik, A., & Teslya, N. (2018).
Ontology-Based Human-Computer Cloud for Decision Support: Architecture
and Applications in Tourism. International Journal of Embedded and Real-Time
Communication Systems, 9(1), 1–19. doi:10.4018/IJERTCS.2018010101
Smith-Ditizio, A. A., & Smith, A. D. (2018). Computer Fraud Challenges and Its
Legal Implications. In M. Khosrow-Pour, D.B.A. (Ed.), Encyclopedia of Information
doi:10.4018/978-1-5225-2255-3.ch419
Sohani, S. S. (2016). Job Shadowing in Information Technology Projects: A Source
of Competitive Advantage. International Journal of Information Technology Project
Management, 7(1), 47–57. doi:10.4018/IJITPM.2016010104
Sosnin, P. (2018). Figuratively Semantic Support of Human-Computer Interactions.
In Experience-Based Human-Computer Interactions: Emerging Research and
Opportunities (pp. 244–272). Hershey, PA: IGI Global. doi:10.4018/978-1-5225-
2987-3.ch008
Spinelli, R., & Benevolo, C. (2016). From Healthcare Services to E-Health
Applications: A Delivery System-Based Taxonomy. In A. Dwivedi (Ed.), Reshaping
Medical Practice and Care with Health Information Systems (pp. 205–245). Hershey,
Srinivasan, S. (2016). Overview of Clinical Trial and Pharmacovigilance Process
and Areas of Application of Computer System. In P. Chakraborty & A. Nagal (Eds.),
Software Innovations in Clinical Drug Development and Safety (pp. 1–13). Hershey,
Srisawasdi, N. (2016). Motivating Inquiry-Based Learning Through a Combination
of Physical and Virtual Computer-Based Laboratory Experiments in High School
Science. In M. Urban & D. Falvo (Eds.), Improving K-12 STEM Education Outcomes
through Technological Integration (pp. 108–134). Hershey, PA: IGI Global.
doi:10.4018/978-1-4666-9616-7.ch006
Stavridi, S. V., & Hamada, D. R. (2016). Children and Youth Librarians: Competencies
Required in Technology-Based Environment. In J. Yap, M. Perez, M. Ayson, & G.
Entico (Eds.), Special Library Administration, Standardization and Technological
Integration (pp. 25–50). Hershey, PA: IGI Global. doi:10.4018/978-1-4666-9542-9.
ch002
280
Related References
Sung, W., Ahn, J., Kai, S. M., Choi, A., & Black, J. B. (2016). Incorporating Touch-
Based Tablets into Classroom Activities: Fostering Children’s Computational
Thinking through iPad Integrated Instruction. In D. Mentor (Ed.), Handbook of
Research on Mobile Learning in Contemporary Classrooms (pp. 378–406). Hershey,
Syväjärvi, A., Leinonen, J., Kivivirta, V., & Kesti, M. (2017). The Latitude of
Information Management in Local Government: Views of Local Government
Managers. International Journal of Electronic Government Research, 13(1), 69–85.
doi:10.4018/IJEGR.2017010105
Tanque, M., & Foxwell, H. J. (2018). Big Data and Cloud Computing: A Review
of Supply Chain Capabilities and Challenges. In A. Prasad (Ed.), Exploring the
Convergence of Big Data and the Internet of Things (pp. 1–28). Hershey, PA: IGI
Global. doi:10.4018/978-1-5225-2947-7.ch001
Teixeira, A., Gomes, A., & Orvalho, J. G. (2017). Auditory Feedback in a Computer
Game for Blind People. In T. Issa, P. Kommers, T. Issa, P. Isaías, & T. Issa (Eds.),
Smart Technology Applications in Business Environments (pp. 134–158). Hershey,
Thompson, N., McGill, T., & Murray, D. (2018). Affect-Sensitive Computer
Systems. In M. Khosrow-Pour, D.B.A. (Ed.), Encyclopedia of Information Science
and Technology, Fourth Edition (pp. 4124-4135). Hershey, PA: IGI Global.
doi:10.4018/978-1-5225-2255-3.ch357
Trad, A., & Kalpić, D. (2016). The E-Business Transformation Framework for
E-Commerce Control and Monitoring Pattern. In I. Lee (Ed.), Encyclopedia of
E-Commerce Development, Implementation, and Management (pp. 754–777).
Triberti, S., Brivio, E., & Galimberti, C. (2018). On Social Presence: Theories,
Methodologies, and Guidelines for the Innovative Contexts of Computer-Mediated
Learning. In M. Marmon (Ed.), Enhancing Social Presence in Online Learning
Environments (pp. 20–41). Hershey, PA: IGI Global. doi:10.4018/978-1-5225-
3229-3.ch002
Tripathy, B. K. T. R., S., & Mohanty, R. K. (2018). Memetic Algorithms and Their
Applications in Computer Science. In S. Dash, B. Tripathy, & A. Rahman (Eds.),
Handbook of Research on Modeling, Analysis, and Application of Nature-Inspired
Metaheuristic Algorithms (pp. 73-93). Hershey, PA: IGI Global. doi:10.4018/978-
1-5225-2857-9.ch004
281
Related References
Turulja, L., & Bajgoric, N. (2017). Human Resource Management IT and Global
Economy Perspective: Global Human Resource Information Systems. In M.
Khosrow-Pour (Ed.), Handbook of Research on Technology Adoption, Social Policy,
and Global Integration (pp. 377–394). Hershey, PA: IGI Global. doi:10.4018/978-
1-5225-2668-1.ch018
Unwin, D. W., Sanzogni, L., & Sandhu, K. (2017). Developing and Measuring the
Business Case for Health Information Technology. In K. Moahi, K. Bwalya, & P.
Sebina (Eds.), Health Information Systems and the Advancement of Medical Practice
in Developing Countries (pp. 262–290). Hershey, PA: IGI Global. doi:10.4018/978-
1-5225-2262-1.ch015
Vadhanam, B. R. S., M., Sugumaran, V., V., V., & Ramalingam, V. V. (2017).
Computer Vision Based Classification on Commercial Videos. In M. S., & V. V.
(Eds.), Multi-Core Computer Vision and Image Processing for Intelligent Applications
Valverde, R., Torres, B., & Motaghi, H. (2018). A Quantum NeuroIS Data Analytics
Architecture for the Usability Evaluation of Learning Management Systems. In S.
Bhattacharyya (Ed.), Quantum-Inspired Intelligent Systems for Multimedia Data
Analysis (pp. 277–299). Hershey, PA: IGI Global. doi:10.4018/978-1-5225-5219-
2.ch009
Vassilis, E. (2018). Learning and Teaching Methodology: “1:1 Educational
Computing. In K. Koutsopoulos, K. Doukas, & Y. Kotsanis (Eds.), Handbook of
Research on Educational Design and Cloud Computing in Modern Classroom Settings
Wadhwani, A. K., Wadhwani, S., & Singh, T. (2016). Computer Aided Diagnosis
System for Breast Cancer Detection. In Y. Morsi, A. Shukla, & C. Rathore (Eds.),
Optimizing Assistive Technologies for Aging Populations (pp. 378–395). Hershey,
Wang, L., Wu, Y., & Hu, C. (2016). English Teachers’ Practice and Perspectives
on Using Educational Computer Games in EIL Context. International Journal of
Technology and Human Interaction, 12(3), 33–46. doi:10.4018/IJTHI.2016070103
Watfa, M. K., Majeed, H., & Salahuddin, T. (2016). Computer Based E-Healthcare
Clinical Systems: A Comprehensive Survey. International Journal of Privacy and
Health Information Management, 4(1), 50–69. doi:10.4018/IJPHIM.2016010104
282
Related References
Weeger, A., & Haase, U. (2016). Taking up Three Challenges to Business-IT

Alignment Research by the Use of Activity Theory. International Journal of IT/
Business Alignment and Governance, 7(2), 1-21. doi:10.4018/IJITBAG.2016070101
Wexler, B. E. (2017). Computer-Presented and Physical Brain-Training Exercises for
School Children: Improving Executive Functions and Learning. In B. Dubbels (Ed.),
Transforming Gaming and Computer Simulation Technologies across Industries
Williams, D. M., Gani, M. O., Addo, I. D., Majumder, A. J., Tamma, C. P., Wang,
M., ... Chu, C. (2016). Challenges in Developing Applications for Aging Populations.
In Y. Morsi, A. Shukla, & C. Rathore (Eds.), Optimizing Assistive Technologies for
Aging Populations (pp. 1–21). Hershey, PA: IGI Global. doi:10.4018/978-1-4666-
9530-6.ch001
Wimble, M., Singh, H., & Phillips, B. (2018). Understanding Cross-Level Interactions
of Firm-Level Information Technology and Industry Environment: A Multilevel
Model of Business Value. Information Resources Management Journal, 31(1),
1–20. doi:10.4018/IRMJ.2018010101
Wimmer, H., Powell, L., Kilgus, L., & Force, C. (2017). Improving Course Assessment
via Web-based Homework. International Journal of Online Pedagogy and Course
Design, 7(2), 1–19. doi:10.4018/IJOPCD.2017040101
Wong, Y. L., & Siu, K. W. (2018). Assessing Computer-Aided Design Skills. In M.
2255-3.ch642
Wongsurawat, W., & Shrestha, V. (2018). Information Technology, Globalization,
and Local Conditions: Implications for Entrepreneurs in Southeast Asia. In P.
Ordóñez de Pablos (Ed.), Management Strategies and Technology Fluidity in the
Asian Business Sector (pp. 163–176). Hershey, PA: IGI Global. doi:10.4018/978-
1-5225-4056-4.ch010
Yang, Y., Zhu, X., Jin, C., & Li, J. J. (2018). Reforming Classroom Education Through
a QQ Group: A Pilot Experiment at a Primary School in Shanghai. In H. Spires
(Ed.), Digital Transformation and Innovation in Chinese Education (pp. 211–231).
283
Related References
Yilmaz, R., Sezgin, A., Kurnaz, S., & Arslan, Y. Z. (2018). Object-Oriented
Programming in Computer Science. In M. Khosrow-Pour, D.B.A. (Ed.), Encyclopedia
Yu, L. (2018). From Teaching Software Engineering Locally and Globally to Devising
an Internationalized Computer Science Curriculum. In S. Dikli, B. Etheridge, &
R. Rawls (Eds.), Curriculum Internationalization and the Future of Education (pp.
Yuhua, F. (2018). Computer Information Library Clusters. In M. Khosrow-Pour,
D.B.A. (Ed.), Encyclopedia of Information Science and Technology, Fourth Edition
Zare, M. A., Taghavi Fard, M. T., & Hanafizadeh, P. (2016). The Assessment of
Outsourcing IT Services using DEA Technique: A Study of Application Outsourcing
in Research Centers. International Journal of Operations Research and Information
Systems, 7(1), 45–57. doi:10.4018/IJORIS.2016010104
Zhao, J., Wang, Q., Guo, J., Gao, L., & Yang, F. (2016). An Overview on Passive
Image Forensics Technology for Automatic Computer Forgery. International Journal
of Digital Crime and Forensics, 8(4), 14–25. doi:10.4018/IJDCF.2016100102
Zimeras, S. (2016). Computer Virus Models and Analysis in M-Health IT Systems:
Computer Virus Models. In A. Moumtzoglou (Ed.), M-Health Innovations for
Patient-Centered Care (pp. 284–297). Hershey, PA: IGI Global. doi:10.4018/978-
1-4666-9861-1.ch014
Zlatanovska, K. (2016). Hacking and Hacktivism as an Information Communication
System Threat. In M. Hadji-Janev & M. Bogdanoski (Eds.), Handbook of Research
on Civil Society and National Security in the Era of Cyber Warfare (pp. 68–101).
284
285
Index
5G 14, 19, 30, 53 D

A dashboard 192, 195
ALGORITHMIC CHALLENGES 45 E
android app 187, 192, 196, 218
Arduino Uno 216, 218, 227 ECG Extraction 156
authentication 5, 64, 97, 100-102, 120, edge analytics 86, 131
134-135, 137 edge computing 1-2, 13-16, 21-22, 29-30,
54, 108-113, 123, 126, 131, 147
B Edge network 69
Emergency Monitoring System 187
big data 2, 10, 16, 25, 27, 30, 37-38, 86-87,
131, 160-161, 211, 229 F
Blood bank 170, 172
fog 1-2, 16-17, 19-30, 36-40, 42-50, 53-55,
C 57-60, 62-65, 68-72, 74-76, 79-82,
84-91, 94-104, 108-109, 112-114,
CASE STUDIES 75, 129, 141, 154, 187 120, 129-147, 149-154, 156
cloud 1-25, 27-28, 30, 36-37, 39, 44, 47-50, fog computing 1-2, 16-17, 19-26, 28-30,
53-57, 64-65, 68-69, 71, 76, 79-80, 36-37, 39-40, 42, 45, 50, 53-55, 57-60,
84-88, 90-91, 93, 98-99, 102, 104, 62-65, 68-72, 76, 79-80, 82, 84-91,
108-111, 113-117, 120, 123, 126, 129- 94-96, 98-101, 103-104, 108-109,
131, 135-136, 138, 140-147, 150-152, 112-113, 129-135, 138-139, 141-144,
156, 178-179, 193, 211, 218-219, 222 146-147, 149-154, 156
cloud computing 1-17, 19-20, 22-23, 25, fog nodes 23-25, 39-40, 44-46, 48, 64-65,
27, 30, 36-37, 50, 53, 55, 68-69, 79, 69, 72, 75-76, 80, 85-86, 88, 91, 96,
84-85, 87-88, 90-91, 104, 108-111, 98, 100, 102-104, 113, 131, 134-135,
126, 129-131, 135, 138, 140, 142-144, 138-139, 143-145, 147, 150-151, 153
147, 150, 152, 156, 211 future research directions 36-37, 45, 50
Collaborative form 73
Index
G N
gateway 80, 145-146, 165 network traffic 53, 113-114, 126
GPRS 5, 180, 183
GSM module 184, 227 P
H privacy 3, 9, 27, 37, 53-55, 76, 82, 84, 86,
90-91, 94-99, 103-104, 111, 113-114,
healthcare 68, 76, 84, 135-136, 144-145 117, 120-121, 126, 129-131, 136-140,
142-143, 151, 154, 156
I
R
Indie fog 72
Integrated form 73 repository 120-121
internet of everything (IoE) 1, 37 resource continuity 108, 110-111, 113-117,
Internet of Things (IoT) 2, 13, 23-24, 36- 119-122, 124, 126
37, 53, 68-69, 71, 85, 87-88, 109, 112,
129-131, 133, 136, 144, 157, 159-161, S
166, 170, 186, 211
intrusion detection 84, 86, 102-103, 135 security 2-3, 5, 8, 14-15, 20-22, 25, 27-28,
Intrusion Prevention 99 30, 36-37, 53, 64-65, 68-69, 75-76, 84,
86, 90-91, 93, 95-99, 101-104, 111,
L 113-114, 117, 120-121, 126, 129-138,
141, 151, 154, 156, 165, 167, 169, 172
latency 1-2, 10-14, 19-22, 24-25, 30, 37, sensor network 9
39, 46, 49-50, 53, 57, 64-65, 68-69, service level objectives 36, 39, 50
80, 82, 85, 100-101, 104, 112-113, smart meter 79, 133, 142-143, 156
122, 131, 143, 152 Software Engineering 157-159, 161-163,
170
M
T
mobile cloud computing 2-5, 8-9, 11, 13
mobile edge computing 1-2, 13-16, 21- terminal nodes 85, 96, 98-99
22, 30
monitoring 8, 28, 55, 64-65, 79-81, W
99, 109, 121-122, 130, 136,
141, 144-145, 174, 187, 210 Web Optimization 141, 151-152, 156
wireless sensors 144, 175, 227
286

The Rise of Fog Computing in The Digital Era PDF

Uploaded by

Copyright:

Available Formats

The Rise of Fog Computing in The Digital Era PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Rise of Fog Computing in The Digital Era PDF

Uploaded by

Copyright:

Available Formats

The Rise of Fog

Computing in the Digital

A volume in the Advances

Library of Congress Cataloging-in-Publication Data

Names: Srinivasa, K. G., editor. | Lathar, Pankaj, 1974- editor. | Siddesh,

British Cataloguing in Publication Data

All work contributed to this book is new, previously-unpublished material.

For electronic access to this publication, please contact: [email protected].

Editor-in-Chief: Srikanta Patnaik, SOA University, India

Electronic Nose Technologies and Advances in Machine Ofaction

Quantum-Inspired Intelligent Systems for Multimedia Data Analysis

Advancements in Computer Vision and Image Processing

Handbook of Research on Power and Energy System Optimization

Modeling and Simulations for Metamaterials Emerging Research and Opportunities

For an entire list of titles in this series, please visit:

701 East Chocolate Avenue, Hershey, PA 17033, USA

Compilation of References............................................................................... 238

Related References............................................................................................ 257

Compilation of References............................................................................... 238

Related References............................................................................................ 257

With the immense growth of information, we are embracing the prevalence of

Srinivasa would like to thank Prof. K C Tiwari, Principal of CBP Government

The idea of cloud computing is based on a very fundamental principal of reusability

Mobile Cloud Computing

MCC is an infrastructure used by mobile applications where both the data

• Infrastructure Based: Where the hardware infrastructure is static and

Cloud Computing Model and Architecture

(Mobile) cloud computing is made up of complex network and relationships of and

Figure 1. (Mobile) Cloud Computing Model

The Infrastructure Providers provide hardware and software infrastructure, or

Figure 2. Mobile Cloud Computing Architecture

Cloud Computing Service Oriented Architecture

Figure 3. Cloud Computing Service Oriented Architecture

Cloud Computing Deployment Models

Depending on the requirements different deployment models for (mobile) cloud

capital expenditure requirements normally associated with other deployment options.

Cloud Computing Benefits

Figure 4. Hybrid Cloud Computing Model

Cloud Computing Services

CLOUD COMPUTING AND TELECOM OPERATORS

Why Telecom Operators Should Use the Cloud

The Cloud in Telecom Operators

Cloud Radio Access Network (CRAN)

Figure 5. Cloud Integration with Telecommunication Network

Figure 6. Mobile Network with CRAN and HCRAN

MOBILE EDGE COMPUTING

Therefore, a better proposed solution is the Mobile Edge Computing (MEC)

Figure 7. Mobile Edge Computing Architecture

Reasons for the Rise of Fog Computing

1. Real Time Processing and Cyber-Physical System Control: Edge data

3. Increased Efficiency by Pooling of Idle and Unused Local Resources: The

Fog Computing Applications

Fog computing was proposed to support applications whose requirements don’t

• Applications having stringent latency requirements, such as mobile gaming,

Fog Computing Architecture

An overview of Fog Computing architecture is given in Figure 8 (Kitanov, Monteiro,

Figure 8. Cloud Integration with Telecommunication Network

A Comparison Between Cloud and Fog Computing

A comparison between Fog Computing and Cloud Computing is given in (Gao,