Resource Management Techniques A R Publications Pdf Download
Figures - uploaded by Muhammad Majid Butt
Author content
All figure content in this area was uploaded by Muhammad Majid Butt
Content may be subject to copyright.
Discover the world's research
- 20+ million members
- 135+ million publications
- 700k+ research projects
Join for free
Networks of the Future
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
CHAPMAN & HALL/CRC
COMPUTER and INFORMATION SCIENCE SERIES
Series Editor: Sartaj Sahni
ADVERSARIAL REASONING: COMPUTATIONAL APPROACHES TO READING THE OPPONENT'S MIND
Alexander Kott and William M. McEneaney
COMPUTER-AIDED GRAPHING AND SIMULATION TOOLS FOR AUTOCAD USERS
P. A. Simionescu
COMPUTER SIMULATION: A FOUNDATIONAL APPROACH USING PYTHON
Yahya Esmail Osais
DELAUNAY MESH GENERATION
Siu-Wing Cheng, Tamal Krishna Dey, and Jonathan Richard Shewchuk
DISTRIBUTED SENSOR NETWORKS, SECOND EDITION
S. Sitharama Iyengar and Richard R. Brooks
DISTRIBUTED SYSTEMS: AN ALGORITHMIC APPROACH, SECOND EDITION
Sukumar Ghosh
ENERGY-AWARE MEMORY MANAGEMENT FOR EMBEDDED MULTIMEDIA SYSTEMS:
A COMPUTER-AIDED DESIGN APPROACH
Florin Balasa and Dhiraj K. Pradhan
ENERGY EFFICIENT HARDWARE-SOFTWARE CO-SYNTHESIS USING RECONFIGURABLE HARDWARE
Jingzhao Ou and Viktor K. Prasanna
EVOLUTIONARY MULTI-OBJECTIVE SYSTEM DESIGN: THEORY AND APPLICATIONS
Nadia Nedjah, Luiza De Macedo Mourelle, and Heitor Silverio Lopes
FROM ACTION SYSTEMS TO DISTRIBUTED SYSTEMS: THE REFINEMENT APPROACH
Luigia Petre and Emil Sekerinski
FROM INTERNET OF THINGS TO SMART CITIES: ENABLING TECHNOLOGIES
Hongjian Sun, Chao Wang, and Bashar I. Ahmad
FUNDAMENTALS OF NATURAL COMPUTING: BASIC CONCEPTS, ALGORITHMS, AND APPLICATIONS
Leandro Nunes de Castro
HANDBOOK OF ALGORITHMS FOR WIRELESS NETWORKING AND MOBILE COMPUTING
Azzedine Boukerche
PUBLISHED TITLES
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
HANDBOOK OF APPROXIMATION ALGORITHMS AND METAHEURISTICS
Teolo F. Gonzalez
HANDBOOK OF BIOINSPIRED ALGORITHMS AND APPLICATIONS
Stephan Olariu and Albert Y. Zomaya
HANDBOOK OF COMPUTATIONAL MOLECULAR BIOLOGY
Srinivas Aluru
HANDBOOK OF DATA STRUCTURES AND APPLICATIONS
Dinesh P. Mehta and Sartaj Sahni
HANDBOOK OF DYNAMIC SYSTEM MODELING
Paul A. Fishwick
HANDBOOK OF ENERGY-AWARE AND GREEN COMPUTING
Ishfaq Ahmad and Sanjay Ranka
HANDBOOK OF GRAPH THEORY, COMBINATORIAL OPTIMIZATION, AND ALGORITHMS
Krishnaiyan "KT" Thulasiraman, Subramanian Arumugam, Andreas Brandstädt, and Takao Nishizeki
HANDBOOK OF PARALLEL COMPUTING: MODELS, ALGORITHMS AND APPLICATIONS
Sanguthevar Rajasekaran and John Reif
HANDBOOK OF REAL-TIME AND EMBEDDED SYSTEMS
Insup Lee, Joseph Y-T. Leung, and Sang H. Son
HANDBOOK OF SCHEDULING: ALGORITHMS, MODELS, AND PERFORMANCE ANALYSIS
Joseph Y.-T. Leung
HIGH PERFORMANCE COMPUTING IN REMOTE SENSING
Antonio J. Plaza and Chein-I Chang
HUMAN ACTIVITY RECOGNITION: USING WEARABLE SENSORS AND SMARTPHONES
Miguel A. Labrador and Oscar D. Lara Yejas
IMPROVING THE PERFORMANCE OF WIRELESS LANs: A PRACTICAL GUIDE
Nurul Sarkar
INTEGRATION OF SERVICES INTO WORKFLOW APPLICATIONS
Paweł Czarnul
INTRODUCTION TO NETWORK SECURITY
Douglas Jacobson
LOCATION-BASED INFORMATION SYSTEMS: DEVELOPING REAL-TIME TRACKING APPLICATIONS
Miguel A. Labrador, Alfredo J. Pérez, and Pedro M. Wightman
PUBLISHED TITLES CONTINUED
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
METHODS IN ALGORITHMIC ANALYSIS
Vladimir A. Dobrushkin
MULTICORE COMPUTING: ALGORITHMS, ARCHITECTURES, AND APPLICATIONS
Sanguthevar Rajasekaran, Lance Fiondella, Mohamed Ahmed, and Reda A. Ammar
NETWORKS OF THE FUTURE: ARCHITECTURES, TECHNOLOGIES, AND IMPLEMENTATIONS
Mahmoud Elkhodr, Qusay F. Hassan, and Seyed Shahrestani
PERFORMANCE ANALYSIS OF QUEUING AND COMPUTER NETWORKS
G. R. Dattatreya
THE PRACTICAL HANDBOOK OF INTERNET COMPUTING
Munindar P. Singh
SCALABLE AND SECURE INTERNET SERVICES AND ARCHITECTURE
Cheng-Zhong Xu
SOFTWARE APPLICATION DEVELOPMENT: A VISUAL C++® , MFC, AND STL TUTORIAL
Bud Fox, Zhang Wenzu, and Tan May Ling
SPECULATIVE EXECUTION IN HIGH PERFORMANCE COMPUTER ARCHITECTURES
David Kaeli and Pen-Chung Yew
TRUSTWORTHY CYBER-PHYSICAL SYSTEMS ENGINEERING
Alexander Romanovsky and Fuyuki Ishikawa
VEHICULAR NETWORKS: FROM THEORY TO PRACTICE
Stephan Olariu and Michele C. Weigle
X-MACHINES FOR AGENT-BASED MODELING: FLAME PERSPECTIVES
Mariam Kiran
PUBLISHED TITLES CONTINUED
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
Networks of the Future:
Architectures, Technologies, and Implementations
Edited by
Mahmoud Elkhodr
Western Sydney University, Australia
Qusay F. Hassan
Mansoura University, Egypt
Seyed Shahrestani
Western Sydney University, Australia
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
CRC Press
Taylor & Francis Group
6000 Broken Sound Park way NW, Suite 30 0
Boca Raton, FL 33487-2742
© 2018 by Taylor & Francis Group, LLC
CRC Press is an imprint of Taylor & Francis Group, an Informa business
No claim to original U.S. Government works
Printed on acid-free paper
International Standard Book Number-13: 978-1-4987-8397-2 (Hardback)
is book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been made to
publish reliable data a nd information, but t he author and publisher cannot assu me responsibility for the va lidity of al l materials
or the consequences of their use. e authors and publishers have attempted to trace the copyright holders of all material
reproduced i n this publication and apologi ze to copyright holder s if permission to publish in t his form has not been obtained . If
any copyright material has not been acknowledged please write and let us know so we may rectify in any future reprint.
Except as p ermitted und er U.S. Copyrig ht Law, no part of thi s book may be reprint ed, reproduced, tra nsmitted , or utilize d in any
form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming,
and recording, or in any information storage or retrieval system, without written permission from the publishers.
For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://www.
copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-
8400. CCC is a not-for-profit organization that provides licenses and registration for a variety of users. For organizations that
have been granted a photocopy license by the CCC, a separate system of payment has been arranged.
Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for
identification and explanation without intent to infringe.
Visit the Taylor & Francis Web site at
http://www.taylorandfrancis.com
and the CRC Press Web site at
http://www.crcpress.com
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
vii
Contents
Preface...............................................................................................................................................xi
Acknowledgments .......................................................................................................................... xiii
Reviewers .........................................................................................................................................xv
Editors ........................................................................................................................................... xvii
Contributors ....................................................................................................................................xix
PART I Cognitive Radio Networks
Chapter 1 Cognit ive Radio with Spectrum Sensing for Future Networks .................................... 3
Nabil Giweli, Seyed Shahrestani, and Hon Cheung
Chapter 2 Cognitive Radio and Spectrum Sensing ..................................................................... 25
Daniel Malafaia, José Vieira, and Ana Tomé
Chapter 3 Ma chine Learning Techniques for Wideband Spectrum Sensing in Cognitive
Radio Networks .......................................................................................................... 43
Su Tabassum Gul, Asad Ullah Omer, and Abdul Majid
Chapter 4 Reso urce Management Techniques in Licensed Shared Access Networks ............... 69
M. Majid Butt, Jasmina McMenamy, Arman Farhang, Irene Macaluso,
Carlo Galiotto, and Nicola Marchetti
PART II 5G Technologies and Software-Dened Networks
Chapter 5 Software-Den ed Network Security: Breaks and Obstacles .....................................89
Ahmed Dawoud, Seyed Shahristani, and Chun Raun
Chapter 6 Fog Computing Mechan isms in 5G Mobile Networks............................................. 101
Stojan Kitanov and Toni Janevski
Chapter 7 Lightweight Cr yptograph y in 5G Machine-Type Communication .......................... 127
Hüsnü Yıldız, Adnan Kılıç, and Ertan Onur
Chapter 8 Index Modulation: A Promising Technique for 5G and Beyond Wireless
Networks .................................................................................................................. 145
Ertuğ r u l B a s¸ a r
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
viii Contents
Chapter 9 Seamless and Sec ure Communication for 5G Subscribers in 5G-WLAN
Heterogeneous Networks .......................................................................................... 167
Amit Kumar and Hari Om
Chapter 10 Simulators, Testbeds, and Prototypes of 5G Mobile Networking Architectures ..... 185
Shahram Mollahasan, Alperen Eroğlu, Ömer Yamaç, and Ertan Onur
PART III Efcient Solutions for Future Heterogenous Networks
Chapter 11 A Fuzzy Logic–Based QoS Evaluation Method for Heterogeneous Networks .......203
Farnaz Farid, Seyed Shahrestani, and Chun Ruan
Chapter 12 Network Virtualization for Next-Generation Computing and Communication
Infrastructures : Scalable Mapping Algorithm and Self-Healing Solution .............. 243
Qiang Yang
Chapter 13 Maximizing the Lifetime of Wireless Sensor Networks by Optimal Network
Design ....................................................................................................................... 277
Keqin Li
Chapter 14 Bandwidth Allocation Scheme with QoS Provisioning for Heterogeneous
Optical and Wireless Networks ................................................................................ 301
Siti H. Mohammad, Nadiatulhuda Zulkii, Sevia Mahdaliza Idrus,
and Arnidza Ramli
Chapter 15 Energy Conservation Techniques for Passive Optical Networks ............................. 319
Rizwan Aslam Butt, Sevia Mahdaliza Idrus, and Nadiatulhuda Zulkii
Chapter 16 Energy Efciency in Wireless Body Sensor Networks ............................................ 339
Ali Hassan Sodhro, Giancarlo Fortino, Sandeep Pirbhulal,
Mir Muhammad Lodro, and Madad Ali Shah
Chapter 17 Efcient Modulation Schemes for Visible Light Communication Systems ............. 355
Navera Karim Memon and Fahim A. Umrani
PART IV Big Data and the Internet of Things
Chapter 18 A Data Aware Scheme for Scheduling Big Data Applications with SAVANNA
Hadoop ..................................................................................................................... 377
K. Hemant Kumar Reddy, Himansu Das, and Diptendu Sinha Roy
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
ix Contents
Chapter 19 Big Data Computing Using Cloud-Based Technologies: Challenges and Future
Perspectives .............................................................................................................. 393
Samiya Khan, Kashish A. Shakil, and Mansaf Alam
Chapter 20 A Multidimensional Sensitivity-Based Anonymization Method of Big Data .......... 415
Mohammed Al-Zobbi, Seyed Shahrestani, and Chun Ruan
Chapter 21 A Quick Perspective on the Current State of IoT Security: A Survey ..................... 431
Musa G. Samaila, João B. F. Sequeiros, Acácio F. P. P. Correia,
Mário M. Freire, and Pedro R. M. Inácio
Chapter 22 A Semidistributed Metaheuristic Algorithm for Collaborative Beamforming in
the Internet of Things ............................................................................................... 465
Suhanya Jayaprakasam, Sharul Kamal Abdul Rahim, and Chee Yen Leow
Index .............................................................................................................................................. 481
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
xi
Preface
The Internet constitutes the largest heterogeneous network and infrastructure in existence. It is
estimated that more than 3.4 billion people accessed the Internet in 2016. The number of mobile
subscriptions has already exceeded the world population. The estimated 2.5 exabytes (2.5 × 1018
bytes) of global data exchanged per month in 2014 is forecast to grow at a staggering compound
annual growth rate of 57% to reach 24.3 exabytes per month in 2019. This rapid explosion of data
can be attributed to several factors, including the advances in wireless technologies and cellular
systems, and the widespread adoption of smart devices, fueling the development of the Internet of
Thing s (IoT).
With the ubiquitous diffusion of the IoT, cloud computing, 5G, and other evolved wireless tech-
nologies into our daily lives, the world will see the Internet of the future expanding and growing
even more rapidly. Recent gures estimate that the number of connected devices to the Internet will
rise to 50 billion by 2020. The IoT is a fast-growing heterogeneous network of connected sensors and
actuators attached to a wide variety of everyday objects. Mobile and wireless technologies including
traditional wireless local access networks (WLANs); low-and ultra-low-power technologies; and
short-and long-range technologies will continue to drive the progress of communications and con-
nectivity. The rapid growth of smart devices that connect to each other and to the Internet through
cellular and wireless communication technologies forms the future of networking. Pervasive con-
nectivity will use technologies such as 5G systems, cognitive radio (CR), software-dened net-
works, and cloud computing amongst many others. These technologies facilitate communication
among the growing number of connected devices, leading to the generation of huge volumes of data.
Processing and analysis of such "Big Data" bring about many opportunities, which, as usual, come
with many challenges, such as those relating to efcient power consumption, security, privacy, man-
agement, and quality of service. This book is about the technologies, opportunities, and challenges
that can drive and shape the networks of the future. We try to provide answers to fundamental and
pressing research challenges including architectural shifts, concepts, mitigation solutions and tech-
niques, and key technologies in the areas of networking.
The book consists of 22 chapters written by some established international researchers and
experts in their eld from various countries. These chapters went through multiple review cycles
and were handpicked based on their quality, clarity, and the subjects we believe are of interest to
the reader. It is divided into four parts. Part I consists of ve chapters. It starts with a discussion
on CR technologies as promising solutions for improving spectrum utilization to manage the ever-
increasing trafc of wireless networks. This is followed by exploring the advances in CR spectrum
sensing techniques and resource management methods.
Part II presents the latest developments and research in the areas of 5G technologies and soft-
ware-dened networks (SDN). After highlighting some of the challenges that SDN faces, various
opportunities and solutions that address them are discussed. It also discusses SDN security solu-
tions for policy enforcement and verication, and explores the application of SDNs in the network
intrusion detection context. This part of the book then moves to discuss and present solutions to the
most pressing challenges facing the adoption of 5G technologies. In this direction, the new para-
digm known as fog computing is examined in the context of 5G networks. A new re-authentication
schema for multiradio access handover in 5G networks is also presented. This part then concludes
with a chapter that compares and investigates the existing and developing 5G simulators, 5G test
beds, projects, and other 5G-based federated platforms.
Part III is focused on efcient solutions for future heterogeneous networks. It consists of a collec-
tion of six chapters that discuss self-healing solutions, dealing with network virtualization, QoS in
heterogeneous networks, and energy-efcient techniques for passive optical networks and wireless
sensor networks.
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
xii Preface
The nal part of this book covers the areas of IoT and Big Data. It consists of ve chapters that
discuss the latest developments and future perspectives of Big Data and the IoT paradigms. The rst
three chapters discuss topics such as data anonymization, which is presented as one of the pioneer
solutions that can minimize privacy risks associated with Big Data. This part also includes a chap-
ter that advocates employing a data location-aware application scheme to improve the performance
of data transfer among clusters. Part IV of this book then ends with two chapters on IoT. The rst
chapter presents a survey on the current state of IoT security. The second chapter discusses the latest
research on beamforming technologies in the IoT.
This book is intended for a broad audience. It is a collection of works that researchers and
graduate students may nd useful in exploring the latest trends in networking and communications.
It can also be used as a resource for self-study by advanced students. The book can also be of
value to cross-domain practicing researchers, professionals, and business stakeholders who may be
interested in knowing about the future networking landscape.
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
xiii
Acknowledgments
We would like to thank everyone who participated in this project and made this book a reality. In
particular, we would like to acknowledge the hard work of authors and their cooperation during the
revisions of their chapters.
We would also like to acknowledge the valuable comments of the reviewers which have enabled
us to select these chapters out of the so many chapters we received and also improve their quality.
Lastly, we would like to specially thank the editorial team at CRC Press/Taylor & Francis Group,
particularly, Randi Cohen for her support throughout the entire process and Todd Perry who greatly
managed the production of the book. We also thank Balasubramanian Shanmugam, project man-
ager from DiacriTech, and his team for taking care of the copyediting process of this book.
Mahmoud Elkhodr
Qusay F. Hassan
Seyed Shahrestani
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
xv
Reviewers
1. Steven Gordon, Senior Lecturer, Central Queensland University, Australia
2. Yufeng Lin, Associate Proferssor, Central Queensland University, Australia
3. Ertuğ rul Ba şar , Assistant Professor, Istanbul Technical University, Turkey
4. Elias Yaacoub, Associate Professor, Arab Open University, Lebanon
5. Chintan M. Bhatt, Assistant Professor Chandubhai S. Patel Institute of Technology, India
6. Mehregan Mahdevi, Director and Lecturer, Victoria University, Australia
7. Jahan Hassan , Senior Lecturer, Royal Melbourne Institute of Technology, Australia
8. Nabil Giweli, Lecturer, Western Sydney University, Australia
9. Farnaz Farid , Lecturer, Western Sydney University, Australia
10. Mohamed Al Zoobi , Lecturer, Western Sydney University, Australia
11. Belal Alsinglawi, Western Sydney University, Australia
12. Rachid Hamadi , Lecturer, Royal Melbourne Institute of Technology, Australia
13. Ahmed Dawood, Lecturer, Western Sydney University, Australia
14. Omar Mubin, Senior Lecturer, Western Sydney University, Australia
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
xvii
Editors
Dr. Mahmoud Elkhodr completed his PhD degree in information and communication technolo-
gies at Western Sydney University (Western), Australia. Mahmoud was awarded the International
Postgraduate Research Scholarship (IPRS) and Australian Postgraduate Award (APA) in 2012–
2015. He was awarded the High Achieving Graduate Award twice, in 2010 and 2012. Mahmoud has
authored several journal articles and book chapters and presented at prestigious conference venues.
He is currently editing two books on the future of networking and 5G technologies to be published
in 2017. His research interests include the Internet of Things, e-health, human–computer interac-
tions, security, and privacy.
Dr. Qusay F. Hassan is an independent researcher and a technology evangelist with 15 years of
professional experience in ICT. He is currently a systems analyst at the United States Agency for
International Development in Cairo, Egypt, where he deals with large-scale and complex systems.
Dr. Hassan received his BS, MS, and PhD from Mansoura University, Egypt, in computer sci-
ence and information systems in 2003, 2008, and 2015, respectively. His research interests are
varied, including IoT, SOA, high-performance computing, cloud computing, and grid computing.
Dr. Hassan has authored and coauthored a number of journal and conference papers as well as
book chapters. He recently published a book, Applications in Next-Generation High Performance
Computing (published in 2016 by IGI Global), and he is currently editing two new books on the
Internet of Things to be published in 2017. Dr. Hassan is an IEEE senior member and a member of
the editorial board of a number of associations.
Dr. Seyed Shahrestani completed his PhD degree in electrical and information engineering at
the University of Sydney, Australia. He joined Western Sydney University (Western), Australia,
in 1999, where he is currently a senior lecturer. He is also the head of the Networking, Security,
and Cloud Research (NSCR) group at Western. His main teaching and research interests include
networking, management and security of networked systems, articial intelligence applications,
health ICT, IoT, and smart environments. He is also highly active in higher degree research training
supervision, with successful results.
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
xix
Contributors
Sharul Kamal Abdul Rahim
Wireless Communication Centre
Universiti Teknologi Malaysia
Johor, Malaysia
Mansaf Alam
Department of Computer Science
Jamia Millia Islamia
New Delhi, India
Mohammed Al-Zobbi
School of Computing, Engineering and
Mathematics
Western Sydney University
New South Wales, Australia
Ertuğ r u l B a s¸ a r
Faculty of Electrical and Electronics
Engineering
Istanbul Technical University
Maslak, Istanbul
M. Majid Butt
CONNECT Centre
Trinity College Dublin
Dublin, Ireland
Rizwan Aslam Butt
Department of Electrical Engineering
University Technology Malaysia
Johor, Malaysia
Hon Cheung
School of Computing, Engineering and
Mathematics
Western Sydney University
Sydney, Australia
Acácio F. P. P. Correia
Department of Computer Science
University of Beira Interior
Covilhã, Portugal
Himansu Das
Department of Computer Science &
Engineering
Kalinga Institute of Industrial Technology
Bhubaneswar, India
Ahmed Dawoud
School of Computing, Engineering and
Mathematics
Western Sydney University
Sydney, Australia
Alperen Eroğ lu
Department of Computer Engineering
Middle East Technical University
Ankara, Tu r key
Arman Farhang
CONNECT Centre
Trinity College Dublin
Dublin, Ireland
Farnaz Farid
School of Computing,
Engineering and Mathematics
Western Sydney University
Sydney, Australia
Giancarlo Fortino
Department of Informatics, Modeling,
Electronics and Systems
University of Calabria
Rende, Italy
Mário M. Freire
Department of Computer Science
University of Beira Interior
Covilhã, Portugal
Carlo Galiotto
CONNECT Centre
Trinity College Dublin
Dublin, Ireland
Nabil Giweli
School of Computing, Engineering and
Mathematics
Western Sydney University
Sydney, Australia
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
xx Contributors
Su Tabassum Gul
Department of Electrical Engineering
Pakistan Institute of Engineering and Applied
Sciences (PIEAS)
Islamabad, Pakistan
K. Hemant Kumar Reddy
Department of Computer Science &
Engineering
National Institute of Science and Technology
Berhampur, India
Sevia Mahdaliza Idrus
Department of Electrical Engineering
University Technology Malaysia
Johor, Malaysia
Pedro R. M. Inácio
Department of Computer Science
University of Beira Interior
Covilhã, Portugal
Toni Janevski
Faculty of Electrical Engineering and
Information Technologies
Saints Cyril and Methodius University
Skopje, Republic of Macedonia
Suhanya Jayaprakasam
Wireless Systems Laboratory, Engineering
Building
Hanyang University
Seoul, South Korea
Samiya Khan
Department of Computer Science
Jamia Millia Islamia
New Delhi, India
Adnan Kılıç
Department of Computer Engineering
Middle East Technical University
Ankara, Tu r key
Stojan Kitanov
Faculty of Informatics
Mother Teresa University
Skopje, Republic of Macedonia
Amit Kumar
epartment of Computer Science and
Engineering
Indian Institute of Technology (Indian School
of Mines)
Dhanbad, Jharkhand
Chee Yen Leow
Wireless Communication Centre
Universiti Teknologi Malaysia
Johor, Malaysia
Keqin Li
Department of Computer Science
State University ofNew York
New Paltz, New York
Mir Muhamm ad Lodro
Electrical Engineering Department
Sukkur IBA University
Sukkur, Sindh, Pakistan
Irene Macaluso
CONNECT Centre
Trinity College Dublin
Dublin, Ireland
Abdul Maji d
Department of Computer & Information
Sciences (DCIS)
Pakistan Institute of Engineering and Applied
Sciences (PIEAS)
Islamabad, Pakistan
Daniel Malafaia
Department of Electronics,
Telecommunications and Informatics
University of Aveiro
Aveiro, Portugal
Nicola Marchetti
CONNECT Centre
Trinity College Dublin
Dublin, Ireland
Jasmina McMenamy
CONNECT Centre
Trinity College Dublin
Dublin, Ireland
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
xxi Contributors
Navera Karim Memon
Department of Telecommunication Engineering
Mehran University of Engineering & Technology
Jamshoro, Pakistan
Siti H. Mohammad
Faculty of Electrical Engineering
Universiti Teknologi Malaysia
Skudai, Malaysia
Shahram Mollahasani
Department of Computer Engineering
Middle East Technical University
Ankara, Tu r key
Hari Om
Department of Computer Science and
Engineering
Indian Institute of Technology (Indian School
of Mines)
Dhanbad, Jharkhand
Asad Ullah Omer
Department of Electrical Engineering
Pakistan Institute of Engineering and Applied
Sciences (PIEAS)
Islamabad, Pakistan
Ertan Onur
Department of Computer Engineering
Middle East Technical University
Ankara, Tu r key
Sandeep Pirbhulal
Shenzhen Institutes of Advanced Technology
Chinese Academy of Sciences
Shenzhen, China
Arnidza Ramli
Faculty of Electrical Engineering
Universiti Teknologi Malaysia
Skudai, Malaysia
Diptendu Sinha Roy
Department of Computer Science &
Engineering
National Institute of Science and Technology
Berhampur, India
Chun Ruan
School of Computing, Engineering and
Mathematics
Western Sydney University
New South Wales, Australia
Musa G. Samaila
Department of Computer Science
University of Beira Interior
Covilhã, Portugal
João B. F. Sequeiros
Department of Computer Science
University of Beira Interior
Covilhã, Portugal
Madad Ali S hah
Electrical Engineering Department
Sukkur IBA University
Sukkur, Sindh, Pakistan
Kashish A. Shakil
Department of Computer Science
Jamia Millia Islamia
New Delhi, India
Ali Hassan Sodhro
Electrical Engineering Department
Sukkur IBA University
Sukkur, Sindh, Pakistan
Ana Tomé
Department of Electronics,
Telecommunications and Informatics
University of Aveiro
Aveiro, Portugal
Fahim A. Umrani
Department of Telecommunication Engineering
Mehran University of Engineering &
Tech nology
Jamshoro, Pakistan
José Vieira
Department of Electronics,
Telecommunications and Informatics
University of Aveiro
Aveiro, Portugal
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
xxii Contributors
Ömer Yam aç
Department of Computer Engineering
Middle East Technical University
Ankara, Tu r key
Qiang Yang
College of Electrical Engineering
Zhejiang University
Hangzhou, China
Hüsnü Yıldız
Department of Computer Engineering
Middle East Technical University
Ankara, Tu r key
Nadiatulhuda Zulkii
Department of Electrical Engineering
University Technology Malaysia
Johor, Malaysia
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
69
Resource Management
Techniques in Licensed
Shared Access Networks
M. Majid Butt, Jasmina McMenamy, Arman Farhang,
Irene Macaluso, Carlo Galiotto, and Nicola Marchetti
The advent of mobile Internet has led to a phenomenal growth in mobile data trafc over the past few
years. This trend is expected to continue considering the envisioned services of the fth-generation
(5G) mobile communication systems that will be required to provide ubiquitous connectivity, support
of various verticals, and ten-fold improvements in data rates and latency compared to 4G. Spectrum
is therefore at the heart of 5G, and its exible use and better utilization are two of the key components
when addressing its scarcity and fragmented availability. For this, spectrum-sharing paradigms such
as licensed shared access (LSA)—a licensing approach designed to enable sharing of spectrum bands
with low incumbent activity—become increasingly important. LSA builds on the concept of vertical
sharing in which a licensed entity, called an LSA licensee, utilizes spectrum resources unused by the
incumbent network(s). LSA rules ensure the protection of the incumbent from harmful interference by
the transmissions from the LSA licensees. Moreover, with LSA, the aim is also to provide consistency
in quality of service (QoS) for the LSA licensees, typically mobile network operators (MNOs), by
enabling exclusive access to spectrum resources not otherwise used by the incumbent.
Starting from the fundamental aspects of LSA, this chapter extends the discussion to the proposed
advances in LSA spectrum management framework within the European Advanced Dynamic Spectrum
5G mobile networks Employing Licensed shared access (ADEL) project [1]. This chapter also provides
an overview of the literature on spectrum management aspects in LSA and presents distinct resource
management algorithms, two of which consider fairness, and the third evaluates an auction-based spec-
trum allocation. The chapter is organized in the following way: Section4.1 describes the architecture
and central aspects of LSA; Section4.2 reviews the existing literature on LSA and spectrum man-
agement; Section4.3 addresses two spectrum allocation algorithms based on fairness—one provides
4
CONTENTS
4.1 LSA Fundamentals and Architecture ..................................................................................... 70
4.2 Brief Literature Overview of LSA Spectrum Sharing............................................................ 72
4.3 Fair Spectrum Allocation Schemes ........................................................................................ 73
4.3.1 Strictly Fair Scheme ................................................................................................... 73
4.3.2 Long-Term Fair Scheme ............................................................................................. 75
4.4 Auction-Based Spectrum Allocation ...................................................................................... 78
4.4.1 Enhanced Auction-Assisted LSA Architecture .......................................................... 78
4.4.2 Auction Procedure ...................................................................................................... 79
4.4.3 Numerical Analysis .................................................................................................... 81
4.4.4 Comparison with Fixed Sharing ................................................................................. 83
4.5 Conclusions ............................................................................................................................. 84
Acknowledgments ............................................................................................................................84
References ........................................................................................................................................ 84
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
70 Networks of the Future
strictly fair spectrum allocation among licensee networks, while the second provides fairness on a long-
term basis only; Section4.4 presents an auction-based LSA spectrum allocation, which in addition to
spectrum sharing, also considers infrastructure sharing; and Section5 concludes the chapter.
4.1 LSA FUNDAMENTALS AND ARCHITECTURE
It has been widely accepted that spectrum sharing is an essential requirement to support data trafc
growth and ubiquitous and high-bandwidth connectivity. In that respect, the past decade and a half
has seen numerous regulatory and standardization initiatives and technological advances that would
enable dynamic access to spectrum and its more efcient utilization. Progress in technologies such
as cognitive radio (CR)* and software-dened radio (SDR) † [2] is perceived as key for this paradigm
shift. Nevertheless, reaching the vision of a fully developed dynamic access to spectrum requires
advances on many fronts, including a well-dened regulatory environment and commercial viabil-
ity as well as a broad adoption of CR, SDR, and other enabling technologies [3].
In the regulatory domain during the past decade, three main spectrum-sharing models have
emerged, namely TV white spaces (WSs) [4–6], a three-tier spectrum access system (SAS)–based
sharing model in the 3.5 GHz band in the US [7], and LSA in Europe. This chapter focuses on LSA.
Our starting point is the latest status of the work of regulatory and standardization bodies in the
eld, upon which we propose enhancements in the direction of dynamic LSA.
Since its inception, LSA has been a topic of interest for regulatory and standardization bodies, cel-
lular operators, and the academic community. The concept of LSA stems from the industry initiative,
authorized shared access (ASA) [8], to acquire access to additional spectrum for mobile broadband
service that would be provided on a shared basis under an exclusive licensing regime. Initially, fre-
quency bands of 2.3 GHz and 3.8 GHz were sought. LSA, as dened by the European Conference of
Postal and Telecommunications Administrations (CEPT)‡ in [9], is a spectrum management tool that
enables the sharing of selected frequency bands between the incumbents and licensed users—LSA
licensees. The incumbents are the current holders of the right to use the spectrum. The currently
designated band in Europe for LSA use is 2.3–2.4 GHz, whose harmonization was completed in
2014 [10]. Typical incumbents include program making and special events (PMSE) applications,
telemetry, and other governmental use [11]. The aim with the LSA framework is to protect the incum-
bents from harmful interference while providing predictable QoS to the LSA licensees through an
exclusive use of the LSA-designated spectrum. The use of the band was not restricted to MNOs,
although in the rst instance, it was envisaged that they would implement the rst use cases. It is
also foreseen that the LSA licensee and the incumbent will provide different types of services [9].
The access to spectrum by the LSA licensee is determined based on an agreement that species the
terms of the use of the band, including the requirements on vacating the band upon the incumbent's
request. While involvement of a national regulatory authority (NRA) in setting up the LSA agree-
ment between the incumbent and the LSA licensee will vary from country to country, the NRA is
responsible for granting the LSA licensee the individual right to use the LSA spectrum.
In relation to the standardization of LSA, the reference design, architecture, and interfaces are speci-
ed within the European Telecommunications Standards Institute (ETSI) (www.etsi.org/). Examples of
high-level architecture, functional requirements, and deployment scenarios by MNOs, including pro-
posed operational parameters such as transmit power, channel bandwidth, spectrum emission masks,
receiver sensitivity, etc., are provided in [12]. In [13], system requirements for the operation of mobile
broadband service in the LSA band are presented, including functional and other requirements such
as protection of the incumbent, security aspects, and performance. It also species exclusion zones
* CR is dened as a radio aware of its environment, with the ability to learn from it, identify the best spectrum opportunity
for efcient and reliable communication, and adjust its operating parameters accordingly.
† With SDR, some or all of the radio functionality is realized in software.
‡ CEPT is a European association that coordinates the activities related to radio spectrum, telecommunication, and postal
regulations. In relation to LSA, it ensures technical harmonization between different national administrations.
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
71 Resource Management Techniques in Licensed Shared Access Networks
and restriction zones in relation to the LSA licensees (i.e., MNOs) operations. Exclusion zones are
geographical areas where MNOs cannot transmit on the LSA spectrum, whereas in restriction zones,
MNOs can transmit albeit under controlled conditions on such parameters as power levels or antennas.
The document also requires that, in addition to a scheduled mode of operation, the LSA system must
support on-demand operation. That means that in the event of an emergency, an LSA licensee also
needs to be able to release spectrum, according to the specied conditions. ETSI system requirements
were also reviewed and evaluated from the implementation perspective in [14]. In [15], ETSI reference
architecture for LSA is presented. As in [9], it envisages the introduction of two new architectural
building blocks: the LSA repository and the LSA controller. The LSA repository contains information
on the incumbents' use of spectrum and the requirements on their protection. Its task is to provide the
spectrum availability information to the LSA controllers, but it can also receive and store acknowledg-
ment information from an LSA controller. The LSA controller, on the other hand, retrieves information
from the LSA repository about the spectrum the incumbent uses and manages the access of the LSA
licensee to the available spectrum. The LSA controller may interface one or more LSA repositories
and LSA licensees. While [15] does not stipulate in which domain the LSA repository may be located
(i.e., whether it is managed by the NRA, the incumbent, or delegated to a third party), it does specify
that the LSA controller is within an LSA licensee domain. In this way, it enables the LSA controller to
interact with the Operations and Maintenance (O&M) centre system of the LSA licensee to support the
reconguration of the appropriate transmitters, according to the information from the LSA repository.
The LSA regulation and standardization activities currently focus on long-term sharing arrange-
ments based on xed-channel plans. With the aim to progress these activities toward a more
dynamic approach, the European ADEL project [1] proposes an architecture that supports dynamic
LSA congurations, targeting better overall spectrum utilization through the use of advanced radio
resource management (RRM) techniques and sensing reasoning. To this end, the basic two-node
LSA architecture is complemented with additional modules, as depicted in Figure4.1, enabling
detection of the changes in the radio environment as well as adaptation to these changes that could
be caused either by the incumbents or by the LSA licensees. The architecture also allows coordina-
tion of access of multiple LSA licensees to the LSA band.
In addition to the LSA repository, which contains information only about the incumbent's spec-
trum, the ADEL project proposes the use of one or more collaborative spectrum–sensing networks
to provide periodic updates about the radio environment. These information sources will be updat-
ing a radio environment map (REM), whose role is to reect the radio environment as accurately as
LSA licensee
no. 1
NRA
LSA
authentication
server
Request
manager LSA
RRM
Radio
environment
map
Sensing
reasoning
LSA
repository
Incumbent
no. 1
Incumbent
no. 2
Sensing
network no. 1
Sensing
network no. 2
Incumbent
no. L
Sensing
LSA sharing
agreement
LSA controller
LSA controller
LSA controller
LSA licensee
no. 2
LSA licensee
no. L LSA
billing
LSA band manager
Spectrum usage information
Fine for policy violation
FIGURE4.1 The LSA system architecture proposed by ADEL [16].
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
72 Networks of the Future
possible. When there is an LSA licensee request for spectrum, the information contained in the REM
will be used by the LSA band manager to allocate adequate resources (frequency and power) to that
particular LSA licensee. The proposed architecture also includes modules dealing with authentica-
tion, storage of the LSA sharing agreement rules, and spectrum usage accounting and billing. These
functional modules may be implemented by the same, or by different, physical modules. A detailed
description of the functional modules of the proposed LSA system can be found in [17].
The LSA functional architecture discussed here is the one presented in [17] and addresses mul-
tiple LSA licensees and multiple incumbent dynamic congurations. It contains a building block
responsible for coordinating the access of multiple LSA licensees to the LSA band, thus avoiding
the need to have a xed-band plan, as prescribed by the ETSI standard [15] and by the CORE+
single-licensee trials (http://core.willab.). This architecture is also ETSI compliant since each LSA
licensee has an LSA controller responsible for translating the spectrum availability information,
provided by the LSA band manager, into networking reconguration commands.
The LSA band manager contains two sub-blocks: the request manager, which performs priority
management according to the LSA spectrum usage rules, and the LSA RRM block, which performs
the computation of available resources for assignment to the LSA licensees, based on spectrum
usage rules and the information stored in the repository.
4.2 BRIEF LITERATURE OVERVIEW OF LSA SPECTRUM SHARING
Several works have appeared in recent years focusing on different aspects of LSA systems. While
some of them are focused on trials, such as [18] and [19], others, which are discussed briey here,
represent research investigations into cellular system performance and advances to LSA. A scheme
for ofoading macro-cell trafc to a small-cell network using LSA as a basis is proposed in [20].
Based on a game-theoretic approach and taking into account individual utilities of the macro- and
small-cell networks, the scheme determines the number of small cells that will be used for ofoading
using LSA spectrum while increasing energy efciency. The authors in [21] consider how two param-
eters of cellular networks, power and antenna tilt, can be optimized to meet the conditions to operate
in the LSA band, considering different incumbent services (wireless cameras, video links) and their
requirements. The authors use measurements by the MNO's user terminals and additional test points
to estimate interference levels caused by MNO's transmissions to determine the feasibility of using the
LSA spectrum. Their results show that the best use of LSA spectrum takes place when the locations of
the incumbent's users are close to the MNO's users. Outage probability of an LTE system is evaluated
in different deployment scenarios, such as in macro and heterogeneous networks with various node
densities as in [22], taking into account cumulative interference power in the incumbent region. The
authors argue a signicant reduction in the size of geographical borders between an LSA licensee and
the incumbent when the LSA licensee deploys small cells instead of macro base stations. In [23], the
authors provide an interference management scheme, based on a REM, to combat interference caused
to the incumbent on the uplink in an LSA system. A distributed antenna system (DAS) architecture
in a network virtualization context using fractional frequency reuse is considered in [24]. The paper
compares the capacity of cell-edge users between two cases. One is the case when LSA spectrum can
be used in combination with single-user multiple-input multiple-output (MIMO) with joint transmis-
sion by two remote antennas. In the other scenario, the LSA spectrum is not available, but all users
can avail of multi-user MIMO transmission with coordinated beamforming. The paper derives the
ratio between the required LSA bandwidth and the cell-edge bandwidth to support a decision on when
it is more efcient to use LSA to meet the capacity requirements. In [25], a multicarrier waveform–
based, exible inter-operator spectrum-sharing concept is proposed for 5G communication systems.
There, multiple operators obtain access to the shared band, which can be an LSA spectrum band. By
adapting waveforms with respect to the out-of-fragment radiation masks, the authors show that the
inter-operator interference can be avoided. A one-cell 3GPP LTE system using LSA is studied in [26],
in which the authors propose a methodology to model the unreliable operation of an LSA frequency
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
73 Resource Management Techniques in Licensed Shared Access Networks
band by employing a multiline queuing system with unreliable servers. Opportunistic beamforming in
the LSA band is proposed in [27], in which an LSA licensee coexists with an incumbent in a single-
cell scenario. The LSA licensee has instantaneous information on the performance of the incumbent
system to protect the incumbents' user's QoS. In [28], cloud RAN and massive MIMO in the context
of LSA were analyzed, where the authors evaluate the trade-offs between spectrum and antennas.
The authors of [29] propose a two-tier evolutionary game for dynamic allocation of spectrum
resources, enabling the coexistence of incumbents and LSA licensees. The authors present a mecha-
nism for fair decision-making regarding spectrum allocation to LSA licensees, taking into account
spectrum demand. An auction-based approach to spectrum sharing in LSA is presented in [30].
There, the authors propose a mechanism to allocate the incumbent's unused spectrum to the access
points belonging to a number of LSA licensees. The mechanism, LSA auction (LSAA), combines
independent set selection by bidding and a group bid. The goal is a policy aiming for revenue and
market regularity. An auction-based approach for spectrum and infrastructure sharing is also pro-
posed in [31]. There, the authors design a hierarchical, combinatorial auction mechanism, based
on a Vickrey-Clarke-Groves (VCG) auction, and consider the infrastructure providers and cellular
virtual network operators (VNOs). The authors evaluate the allocation with three degrees of free-
dom (i.e., frequency, power, and antennas) and propose a computationally tractable solution. In [32],
an auction mechanism with a mixed graph, which can further quantify and tackle the interference
between the LSA licensees, is proposed. Furthermore, to improve the revenue, the merging of bid
comparisons is done when grouping nodes in the interference graph.
In the following sections, we consider a more dynamic nature of spectrum access in LSA. As
mentioned earlier in Section4.1, the currently envisaged sharing arrangements between the incum-
bent and the LSA licensee are foreseen to be maintained in the long term. Here, our aim is to encour-
age faster allocation (and release) of the spectrum and consider spectrum sharing based on a more
immediate MNO's spectrum demands. In that, we discuss and evaluate two distinct approaches to
spectrum allocation—one based on fairness and the other based on an auction mechanism.
4.3 FAIR SPECTRUM ALLOCATION SCHEMES
In this section, we assume that there is no formal bidding process involved at the time of spectrum
allocation and that the MNOs have agreed a priori on a fair use of shared resources such that every
MNO pays the same price and agrees on receiving a fair proportion of the available LSA spectrum.
As every MNO is offering the same price for spectrum access, the utility function for the LSA system
is to distribute the available spectrum fairly in the "long and short term." We propose spectrum-
sharing mechanisms which aim at satisfying spectrum requirements of the allocated MNOs (as much
as possible) at a particular spectrum allocation instant and allocating spectrum in a fair manner.
4.3.1 StRiCtly faiR SCheme
This scheme aims to provide a fair share of the spectrum to each competing MNO at each spectrum
allocation instant; that is, available spectrum is distributed among all MNOs (with demand) based on
previous allocation history. Each MNO with spectrum demand gets an offer of a nonzero spectrum.
Denoting by
{1 ,, } ∈ the MNO index out of N MNOs, we dene the priority index (PI ),
PIn , for MNO n as
limBWawardedto MNO
SumBWallocatedbytheincumbent
lim
()
()
lim
()
1
1
1
11
PI n
Bj
Bj
Bj
nW
W
j
W
n
a
j
WW
j
W
n
a
j
W
n
N
n
a
∑
∑
=
==
→∞
→∞
=
→∞
=
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
74 Networks of the Future
where
n
and
denote the bandwidth awarded to MNO n and the total offered bandwidth by
the incumbent at the j th spectrum allocation instant, respectively. Note that we assume that all the
offered spectrum is allocated to the MNOs.
Denoting as
, for simplicity, the available spectrum at a single allocation instant,
, the proposed
spectrum allocation algorithm operates in the following steps [17]:
1. Initialize the assigned spectrum to every MNO
a with zero in round
.
2. In round i, divide the bandwidth, B, in proportion to the PI for each MNO with
demand >
Bn
d, that is, the MNO n is allocated spectrum in inverse proportion to its PI
such that
∑
=⋅ −
−
=
BB PI
PI
ni
an
n
N
n
1
1
,
1
3. If the spectrum demand for any MNO n is less than ,
B
a, the bandwidth −
,
ni
an
d becomes
the residual bandwidth Bn
r, which is zero otherwise. All the MNOs with >
,
ni
an
ddo not
take further part in the allocation.
4. After completing the allocation procedure in each round i, update the assigned and
requested spectrum by
min ,
,
BB
n
an
ani
an
d
=+ and =− min(
,
BB
n
dn
dni
an
d,
.
5. Set
==
n
r
n
N
1for the next round, and go back to Step 2.
6. The process terminates when
or
0,
n
d
This algorithm allocates spectrum in a fair fashion to each MNO in every allocation round,
since PI depends on spectrum allocation history for every MNO. The drawback of this strictly
enforced fairness is that the allocated spectrum to a single MNO may not be sufcient to meet
its spectrum demands if the number of MNOs is large, thereby making allocated spectrum less
useful.
We use Monte Carlo simulations to evaluate the performance of this algorithm and demon-
strate its short- and long-term fairness characteristics. The window size, W, for computing the PI
is set to 20 allocation instants to ensure more short-term fairness. The shorter the window size,
the more short-term fairness the algorithm will achieve. As the PI computation for each MNO
requires bandwidth allocation in the last W instants, the simulations are initialized by having
W – 1 time slots with zero spectrum allocation and the W th time slot with allocation depending on
a random PI (chosen between 0 and 1) for every MNO. Without loss of generality, in the simula-
tions, N = 4, and the incumbent spectrum B is normalized to 100 units. At each spectrum alloca-
tion instant, MNOs 1, 2, and 3 choose the demand randomly out of a vector of values [50,100]
with uniform probability, while MNO 4 always requires 100 units (resulting in a signicantly
larger average demand). Ten thousand spectrum allocation instants are simulated to compute the
mean spectrum allocation for each MNO.
Figure4.2 shows the performance of the proposed spectrum allocation algorithm, plotting the
spectrum allocation instants 21–220, where the rst 20 instants were initialized with zero spectrum
allocation and random PI. As all of the MNOs behave symmetrically (including MNO 4), the spec-
trum allocation statistics are plotted for one MNO only.
The algorithm provides a strictly equal share of available bandwidth from the incumbent to each
MNO (25% for N = 4) and provides long-term fairness in spite of excessive demand from MNO 4.
The short-term allocation for each operator for the rst 200 allocation instants is evaluated as well.
To study the short-term behavior of the proposed algorithm, let us dene the moving average of the
allocated spectrum to an MNO in a time slot, t , by
() 1
1
Bt W
jtW
=
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
75 Resource Management Techniques in Licensed Shared Access Networks
It is evident that the algorithm allocates spectrum to each MNO in such a way that its mov-
ing average (evaluated over W allocation instants) converges to its mean very quickly. After the
initialization phase, the algorithm starts dividing the instantly available spectrum equally among
the competing MNOs as the PIs for all the MNOs converge to the same values. The instantaneous
allocation remains constant at 0.25 B for N = 4 (strictly fair) if the minimum demand for every MNO
is greater than 0.25 B (it is 0.5 B in this example). However, if the minimum possible demand is
less than 0.25B, the instantaneous allocation cannot be 25% all the time, and the moving average
slightly diverges from the mean, recovering very quickly in future allocation instances.
4.3.2 long -teRm f aiR SCheme
The proposed spectrum allocation algorithm operates in a proportionally fair manner and assigns
spectrum to the operators based on their allocation history in the past, as before. In contrast to the
short-term fair algorithm, this algorithm does not aim to provide fair spectrum at every spectrum
allocation instant (by providing a nonzero spectrum). However, this algorithm is fair in the long
run and aims to meet the spectrum requirements of the MNOs as much as possible at a specic
spectrum allocation instant.
Based on PI for each operator, apply the following algorithm [33]:
1. Sort the MNOs with respect to PI in increasing order and queue them.
2. Offer as much spectrum as possible to the operator at the head of the queue (HOQ) (and
with the smallest PI ) asks for, and remove it from the queue.
3. If the allocated spectrum is less than the MNO's demand, the MNO can refuse to accept
the offer.
4. If the MNO accepts the offer, the MNO uses the offered spectrum.
5. If the incumbent spectrum is still available after assignment to the selected HOQ MNO,
go back to step 2.
6. Terminate the algorithm either when there are no MNOs with any spectrum demand or
when the incumbent-offered spectrum is fully distributed among the MNOs.
Instantaneous allocation
25
20
15
10
5
020 40 60 80 100 120
Iterations
Spectrum allocation %
140 160 180 200
Moving average
Long-term average
FIGURE4.2 Performance evaluation for the strictly fair spectrum allocation algorithm.
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
76 Networks of the Future
The owchart for the L1 algorithm is shown in Figure4.3.
We use Monte Carlo simulations to evaluate the performance of the proposed algorithm. The
simulation parameters are the same as for the strictly fair algorithm evaluation. Without loss of
generality, we assume that an MNO accepts whatever spectrum is offered by the LSA band manager
after running the spectrum allocation algorithm.
Allocation completed
Compute PI for all OPs
Sort OPs in increasing order
of PI
Make best possible spectrum
offer to HOQ OP and remove
from queue
Offer accepted?
Allocate channels to the
selected OP
Still channels
available and queue
is non-empty?
Yes
No
No
FIGURE4.3 Flowchart for long-term fair L1 algorithm.
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
77 Resource Management Techniques in Licensed Shared Access Networks
Figure 4.4 shows the mean spectrum allocation to four MNOs. It is clear that the algorithm
divides the spectrum among the MNOs uniformly and is fair in the long term (as the strictly fair
algorithm was).
Figure4.5 shows the instantaneous spectrum allocation statistics for the proposed algorithm. As
all of the MNOs have symmetrical demand and allocation statistics, we plot statistics for MNO 1
only. The instantaneous allocation for the operator varies between zero and its demand. As appar-
ent from Figure4.5, when the MNO is allocated full spectrum, it has little chance of accessing the
spectrum in the next few allocation instants. Similarly, a long sequence of zero allocation is usu-
ally followed by full allocation. This justies the algorithm's aim to achieve fairness in spectrum
25%
25% 25%
25%
MNO 1
MNO 2
MNO 3
MNO 4
FIGURE4.4 Spectrum allocation for long-term fair L1 spectrum allocation algorithm.
0
10
20
30
40
50
60
70
80
90
Instantaneous allocation
Moving average
Long-term average
20 40 60 80 1001 20 1401 60 18
FIGURE4.5 Performance evaluation for short-term spectrum allocation for MNO 1. The spectrum alloca-
tion instants 21–200 are plotted.
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
78 Networks of the Future
allocation for the MNOs in addition to meeting spectrum demands of the MNOs temporarily. It is
evident that the moving average of the allocated spectrum for MNO 1 converges to its mean after
very few iterations and diverges marginally afterward, which conrms that the algorithm provides
reasonably good fairness in a short time span.
It is worth noting that both of the algorithms do not provide any excessive spectrum to
MNO4, which has greater than average demand. On the other hand, if spectrum allocation is
provided without taking spectrum allocation history into account, MNO 4 may get additional
spectrum during its turn (e.g., on a round robin basis), which will result in an unfair mean
spectrum allocation.
4.4 AUCTION-BASED SPECTRUM ALLOCATION
In this section, we introduce an auction-based spectrum management as presented in [34]. We
explore the aspects of sharing not only spectrum but also infrastructure by virtual network opera-
tors (VNOs), which will be constructing networks using resources from a shared pool, such as base
stations, spectrum, core network components, cloud resources, etc. We use the existing LSA spec-
trum-sharing framework as a basis and propose an auction-based mechanism to allocate resources
to the VNOs. As we have seen in Section4.1, in LSA, the spectrum resources are orthogonally
assigned to maintain service quality. Regarding the infrastructure, we consider a cloud-based, mas-
sive-MIMO system, in which multiple VNOs can share all the antennas. Antennas are connected
to the centralized processing units that reside in the cloud and create a baseband pool. The cloud
and fronthaul physical resources are logically separated and shared between VNOs, creating vir-
tual base stations [35]. In this way, each VNO has a virtual slice comprised of infrastructure and
radio resources, enabling them to provide distinct services to their users. This section presents the
case in which all VNOs offer the same service to their users, evaluated through average user rate.
The infrastructure provider may be third party or may be a public network provider. This approach
to spectrum and infrastructure management is in line with the radio access network (RAN) shar-
ing scenario proposed by 3GPP in [36], where the participating operators share RAN by utilizing
orthogonal portions of the licensed spectrum.
4.4.1 enhanCed auCtion-aSSiSted lSa aRChiteCtuRe
The proposed enhanced LSA architecture that supports spectrum and infrastructure sharing is
depicted in Figure4.6. It consists of three main building blocks: the LSA architecture with an LSA
controller and an LSA repository, an auctioneer, and an infrastructure provider for a given area.
The enhanced LSA architecture, therefore, extends the conventional one and also incorporates
cloud RAN, virtualization, and software-dened radio/network concepts [37]. Cloud RAN envis-
ages cloud-based baseband processing, where baseband resources are pooled and shared among
different remote radios. Virtualization can be considered a next stage in the evolution of cloud
RAN [35], allowing multiple operators to share common infrastructure (baseband, transport, and
access) resources as well as spectrum resources. Software radio/networking enhances virtualiza-
tion, enabling direct programmability of the network. In the context of this chapter, virtualization
envisages providing distinct wireless network resources, such as antennas, baseband, fronthaul, and
spectrum, to different VNOs. The resources are logically separated, enabling each VNO to man-
age their resource allocation policy. Here, the spectrum is a public resource, whereas infrastructure
may be provided by a third party (e.g. an infrastructure provider) or may be a part of a public
(cellular) network. If present, the infrastructure provider is responsible for dening the terms of
infrastructure sharing. Should the infrastructure be a part of a public network, the NRA role would
need to be extended to set the terms of infrastructure sharing. Furthermore, the new LSA licens-
ees are virtual cellular operators, which now do not own infrastructure. We also envisage that the
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
79 Resource Management Techniques in Licensed Shared Access Networks
sharing arrangements involve the auction mechanism, where a third party (i.e., the auctioneer) is
introduced on behalf of the NRA and infrastructure provider to manage both the spectrum and the
infrastructure sharing. The auction mechanism follows the LSA spectrum-sharing rules, where the
temporal allocation of spectrum follows the statistics of the incumbent(s) in the band. Concerning
the infrastructure sharing, we consider cloud-based, massive-MIMO antennas as a resource that
multiple VNOs can share at the same time. Based on the input from the auctioneer and the LSA
repository, the wireless resource controller assigns spectrum and infrastructure resources to each
VNO (i.e., the appropriate channels, the number of antennas, and the required cloud and fronthaul
resources). The wireless resource controller instructs the resource manager to manage the assigned
resources. Considering that resources may belong to different entities, the wireless resource control-
ler and resource manager may consist of separate logical units that each control/manage spectrum
or infrastructure. It should also be noted that, in general, independent providers may provide differ-
ent resources (i.e., antennas, cloud, and fronthaul). In this chapter, a single infrastructure provider
is responsible for all the resources.
4.4.2 auCtion pRoCeduRe
The auction here is similar to the one in [38], in which a clock auction is performed by a third-party
auctioneer for the combined acquisition of spectrum and antennas. In our auction, the bidders (i.e.,
VNOs) also bid for spectrum and infrastructure resources. Each VNO serves the same number
of (its own) users. Furthermore, to comply with the LSA framework and according to the LTE
standard, the available spectrum is channelized into blocks of 5 MHz. Thanks to massive MIMO,
users of the same VNO can reuse the same spectrum. However, the LSA framework stipulates the
orthogonal use of spectrum by VNOs. It should be noted that allocation of resources is valid for the
period determined by the type of incumbent and their usage of spectrum. In the case of appearance
of an incumbent in a given band, there are a few options as to how the resources can be reassigned,
namely (i) the residual spectrum from the current auction can be reassigned to the VNOs that are
affected by the appearance of an incumbent, (ii) the auction can be repeated over the updated
available spectrum, or (iii) the affected VNOs will be left without LSA spectrum, waiting for the
incumbent to evacuate the band.
NRA
Incumbent
LSA
repository
Wireless
resource
controller
Resource
manager
LSA element
New element/entity
LSA element with
extended functionality
Auctioneer
LSA2
LSA1 VNO 1 , VNO2 ... VNO N
LSA3
provider Cloud resources
FIGURE4.6 Proposed enhanced LSA architecture, supported by the auction mechanism. Dashed lines rep-
resent the existing LSA interfaces, whereas full lines represent new, required interfaces.
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
80 Networks of the Future
We consider N VNOs,
users to be served by each VNO, and
distributed antennas. In
our model, VNOs lease antennas at a xed price and acquire access to spectrum via an auction
mechanism. It should be noted that when referring to the antenna price, here, we also refer to
the required cloud and fronthaul resources. The xed price associated with the usage of each
antenna affects spectrum utilization. Since spectrum and antennas are partially interchange-
able resources[38], the demand for spectrum will vary with the cost of antennas. As a case in
point, if the cost of antennas is too high, the remaining budget might not be sufcient to acquire
the necessary spectral resources for delivering a given rate. In this chapter, we have adopted a
clock auction for the assignment of resources to the VNOs. The clock auction operates in two
phases, namely, the price discovery (clock) phase and the nal assignment phase. The price of
spectrum monotonically increases in each round, and VNOs indicate the packages of spectrum
and antennas they are willing to buy at a given price. In particular, if the auctioneer detects
excess demand for the spectrum after a round of bidding has closed, it increases the posted
spectrum price and opens another round of bidding. In our model, in each round, each VNO can
XOR two package bids. Each VNO computes the rst package bid as the number of antennas
and 5 MHz blocks that minimize its cost within its budget constraint while providing its users
a minimum rate. The cost is a linear combination of the number of antennas and bandwidth
at the prices indicated by the auctioneer. Except for spectrum channelization, this is the same
model discussed in [40]. However, since the price of spectrum increases at each round, we also
consider a second bidding strategy, which models a less aggressive competitive bidding for the
spectrum resources. To calculate the second package bid, each VNO starts from the rst set
of a number of antennas and spectrum requirements and attempts to minimize the bandwidth
requirement by incrementing the number of antennas, provided that the minimum rate require -
ment is satised. Then, each VNO checks if the cost is less than or equal to its available budget
and submits a package bid to the auctioneer. This procedure is shown in Figure 4.7, where
ij
and
ij
are the number of antennas and spectrum blocks
of width 5 MHz that VNO
submits to the auctioneer at bid package
, Ma
is the maximum
FIGURE4.7 Algorithm for the auction-based resource allocation.
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
81 Resource Management Techniques in Licensed Shared Access Networks
number of available antennas,
Min, is the minimum required rate for the
th VNO, and
i
is the budget for VNO
.
The clock phase of the auction ends when all excess demand is removed from the market. In
the ideal situation, both supply and demand completely match. However, this is unlikely to happen
in complex multi-item-unit, multi-item-type auctions. Consequently, the approach that is used in
this part of the auction may result in the oversupply of spectrum. Namely, this situation will occur
if the bidder's private valuation of the minimum required rate is lower than the corresponding
cost to acquire spectrum and antennas at the requested price. If this situation arises, the bids are
assigned using a revenue-maximizing approach (i.e., using a winner determination algorithm). This
algorithm determines which combination of the bids that stood at the last clock price that caused
excess demand will maximize the auctioneer's revenue. The winner determination problem can be
formulated as follows:
∑∑
+
==
yca cb y
ij i
N
j
ai jb ij ij
Maximize
11
2
Subjectt
11
2
yb B
i
N
j
ij ij
∑∑
≤
==
1, 1, 2,
1
2
yi N
j
ij
∑
≤∀
=
0,1, 1, 2, ,,
,2
yi Nj
ij
∈∀ ∈… ∈
where
ij if package
of bidder
is accepted, otherwise
ij .
and
are the costs per
antenna and spectrum block, respectively. Finally,
is the total available bandwidth.
4.4.3 numeRiCal analySiS
The simulated scenario is based on the auctioning strategy explained previously. The scenario
includes 15 VNOs competing in a bid to acquire spectrum and infrastructure to meet their requested
minimum rate.
The minimum requested rate is the same for all the operators. We consider 10 users per VNO
that are randomly distributed in a given area. A total of 64 antennas is available for sharing
between the VNOs. The total available spectrum is 50 MHz, where each VNO can acquire spec-
trum in blocks of 5 MHz, according to the LSA rules. The budget of each VNO is proportional to
the rate that it wants to offer to its users—the higher the rate, the greater its budget. The results
of the simulated scenario are depicted in Figures 4.8 and 4.9. Figure4.8 depicts the number of
required antennas as a function of the minimum rate and antenna price. Figure4.9 illustrates two
directly related aspects—the required bandwidth and the number of VNOs that can be served. To
better understand the trends, the gures should be considered together. Looking along the x -axis
in both gures, we can see that for the lowest considered user rate, the same number of antennas
and spectrum are required, regardless of the antenna price. In this case, the spectrum is abundant,
as VNOs cannot lease less than 5MHz of spectrum or less than 10 antennas.* This spectrum is,
therefore, sufcient to provide the required rate, with the minimum number of antennas. In total,
* In C-RAN systems, the number of antennas need to be more than or equal to the number of users to be served by
eachVNO.
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
82 Networks of the Future
10 VNOs are served. As the minimum considered user rate increases (looking along the y -a xis),
the number of antennas increases up to the highest possible number. The VNOs can still serve
their users with 5 MHz of spectrum but with an increasing number of antennas. This is the case
until the maximum number of antennas is reached and as long as the antenna price is less than a
495
445
395
345
295
245
195
145 70 210 350
Antenna price as a percentage of budget per kbps (%)
490 630 770 910 1050 5
3 VNOs
10 VNOs
5 VNOs
FIGURE4.8 Number of antennas assigned to each allocated V NO in correspondence to a particular rate require-
ment (y -axis) and antenna cost (x-axis). Antenna cost incorporates the required cloud and fronthaul resources.
495
445
395
345
295
245
195
145 70 210 350
Antenna price as a percentage of budget per kbps (%)
490 630 770 910 1050 10
15
20
25
30
35
40
45
50
55
60
FIGURE4.9 Number of VNOs in correspondence to a particular rate requirement (y-axis) and antenna cost
(x -axis). Again, antenna cost incorporates the required cloud and fronthaul resources.
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
83 Resource Management Techniques in Licensed Shared Access Networks
certain value (approximately250% of the budget per kbps). For higher antenna prices, it is more
cost-effective for VNOs to buy more spectrum than to further increase the number of anten-
nas. The spectrum requirement, therefore, jumps to the region of 10 MHz when 5 VNOs can be
served. This trend repeats itself with 10 and 15 MHz of spectrum, serving 5 and 3 VNOs, respec-
tively. It should be noted that this periodicity with the number of antennas is observed only in the
case in which discrete spectrum blocks are considered. It is one of the main differences between
this and the case in which the continuous spectrum is considered [38].
4.4.4 CompaRiSon With fixed ShaRing
In this subsection, we compare the results of auction-based sharing with the xed-based allocation
of resources. In that, we consider two approaches—one with the orthogonal and equal allocation
of both spectrum and antennas and the other with the orthogonal and equal allocation of spectrum,
where all VNOs can utilize all the antennas. Figure4.10 depicts the number of served VNOs versus
the rate requirement for the considered approaches. It should be noted that two different antenna
price values are evaluated for the auction-based sharing. The case of xed sharing with the equal
and orthogonal allocation of the spectrum where all the antennas are shared can be considered as a
benchmark in terms of system efciency (i.e., the number of VNOs that can be served with a given
minimum rate requirement, but excluding the cost of infrastructure). Namely, as in our study, all
VNOs have the same rate requirements, under the assumption of orthogonal spectrum allocation,
using all antennas and equally dividing the spectrum among the VNOs is the optimal solution in
terms of system efciency.
The xed-sharing case with the orthogonal usage of antennas serves the lowest number of VNOs,
regardless of the rate. This degradation in the number of VNOs that can be served occurs because
virtualization is not exploited (i.e., each VNO uses a smaller number of antennas). As shown in
Figure4.10, the auction-based approach for two different antenna prices under consideration outper-
forms the xed-sharing case with the orthogonal utilization of antennas. Furthermore, with a cost
of antennas that is less than approximately 250% of the budget, we can always achieve the optimal
performance through the auction-based approach.
10
9
8
7
6
5
4
3
2
1
0450400350300
Rate (Mbps)
Number of VNOs
250200150100
FS
FS (no antenna sharing)
Auction (antenna price <250%)
Auction (antenna price = 800%)
FIGURE4.10 Fixed sharing versus auction-based sharing.
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
84 Networks of the Future
4.5 CONCLUSIONS
In this chapter, we have described the LSA architecture, followed by the overview of LSA literature
and a description of two distinct approaches aimed to promote a more dynamic LSA spectrum allo-
cation. The rst approach aims at providing spectrum to multiple MNOs based on fairness. Here,
two algorithms are evaluated numerically, and the results show quantitatively that we can guarantee
fairness in spectrum allocation regardless of the demand from the MNOs.
We also propose an enhanced, auction-assisted LSA framework, which encompasses not only
spectrum but also infrastructure (i.e., cloud-based, massive-MIMO antennas). There, we identify
the key architectural aspects required to enhance the LSA framework to avail of this technology. In
our numerical evaluation, we observe periodic patterns in the antenna allocations to VNOs when
considering a range of minimum rate requirements and antenna prices. Furthermore, we show that
the auction-based approach outperforms xed static sharing with the orthogonal use of spectrum
and antennas. Finally, we show that for cases in which the cost of antennas is less than a certain
percentage of the budget (per kbps), we can achieve optimal performance in terms of the number of
VNOs being served.
It is evident that LSA and other LSA systems will play a key role in dealing with the spec-
trum scarcity problem in 5G networks and beyond. While LSA has been designed to offer high
predictability and certainty for both MNOs and the incumbents, there is progress to be made
on the use of licensed shared bands and coexistence aspects. Although the existing MNOs can
avail of the existing infrastructure and customer base, they are still cognizant of using shared
(licensed) bands. Namely, operational and implemental aspects yet need to be proven for enter -
ing and vacating spectrum, management of exclusion and protection zones [39], security [3],
and scalability and network-wide deployments that will be supported by the automated opera-
tions [40]. Furthermore, the initial costs related to LSA implementation need to be proven to be
lower than with the exclusive, dedicated spectrum models. Hence, more advances are needed by
industry and academia in relation to business model design and implemental aspects to make
LSA a true success.
ACKNOWLEDGMENTS
The project ADEL acknowledges the nancial support of the Seventh Framework Programme for
Research of the European Commission under grant number 619647. We also acknowledge support
from the Science Foundation Ireland under grants No. 13/RC/2077 and No. 10/CE/i853.
REFERENCES
1. European Commission–funded FP-7 Framework Project, Advanced Dynamic Spectrum 5G mobile net-
works Employing Licensed shared access (ADEL). Available at: www.fp7-adel.eu/.
2. J. Mitola III, Cognitive Radio. An Integrated Agent Architecture for Software Dened Radio, Stockholm:
Ph.D. dissertation, Royal Institute of Technology (KTH), 2000.
3. S. Bhattarai, J. M. J. Park, B. Gao, K. Bian, and W. Lehr. An overview of dynamic spectrum sharing:
Ongoing initiatives, challenges, and a roadmap for future research, IEEE Transactions on Cognitive
Communications and Networking, 2(2), 11, 2016.
4. FCC, Unlicensed Use of TV Band and 600 MHz Band Spectrum [ET Docket No. 14–165; FCC
15–99],2015.
5. CEPT Report 159, Technical and Operational Requirements for the Possible Operation of Cognitive
Radio Systems in the White Spaces of the Frequency Band 470–790 MHz, ECC, 2011.
6. CEPT Report 186, Technical and Operational Requirements for the Operation of White Space Devices
under Geo-location Approach, ECC, 2013.
7. FCC, Further Notice of Proposed Rulemaking (FNPRM) in the Matter of the amendment of the
Commission's Rules with Regard to Commercial Operations in the 3550–3650 MHz Band [GN Docket
No. 12–354], April 2015.
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
85 Resource Management Techniques in Licensed Shared Access Networks
8. Nokia Siemens Networks; Qualcomm, Authorized Shared Access: An evolutionary spectrum authoriza-
tion scheme for sustainable economic growth and consumer benet [Input Document FM(11)116 to the
72nd Meeting of the WG FM], May 2011.
9. CEPT, Licensed Shared Access (LSA) ECC Report 205, February 2014.
10. CEPT, ECC Decision (14)02: Harmonised Technical and Regulatory Conditions for the Use of the Band
2300–2400 MHz for Mobile/Fixed Communications Networks MFCN, ECC, June 2014.
11. CEPT Report 56, Technological and Regulatory Options Facilitating Sharing between Wireless
Broadband Applications (WBB) and the Relevant Incumbent Services/Applications in the 2.3 GHz Band,
ECC, March 2015.
12. ETSI, SR 103 13 Electromagnetic Compatibility and Radio Spectrum Matters (ERM) System Reference
Document (SRDoc): Mobile Broadband Services in 2300–2400 MHz Frequency Band under the Licensed
Shared Access Regime, Special Report, V1.1.1, July 2013.
13. ETSI, TS 103 154 Recongurable Radio Systems (RRS): System Requirements for Operation of Mobile
Broadband Systems in the 2 300 MHz–2 400 MHz Band under Licensed Shared Access (LSA), V1.1.1,
October 2014.
14. M. Mustonen, M. Matinmikko, M. Palola, T. Rautio, and S. Yrjölä. Analysis of requirements from standardiza-
tion for licensed shared access (LSA) system implementation, in IEEE International Symposium on Dynamic
Spectrum Access Networks (DySPAN), Stockholm: IEEE Conference Publications, September 2015.
15. ETSI, TS 103 235 Recongurable Radio Systems (RRS): System Architecture and High Level Procedures
for Operation of Licensed Shared Access in the 2300 MHz–2400 MHz Band, December 2015.
16. V. Frascolla, A. J. Morgado, A. Gomes, M. M. Butt, N. Marchetti, K. Voulgaris, and C. B. Papadias.
Dynamic licensed shared access: A new architecture and spectrum allocation techniques, in IEEE
Vehicular Technology Conference (VTC-fall), Montreal, CA: IEEE Conference Publications, 2016.
17. A. Morgado, A. Gomes, V. Frascolla, K. Ntougias, C. Papadias, et al. Dynamic LSA for 5G networks,
in Proceeding of European Conference Networks and Communications, Paris: IEEE Conference
Publications, June 2015.
18. M. Palola, M. Matinmikko, J. Prokkola, M. Mustonen, M. Heikkilä, et al. Live eld trial of licensed
shared access (LSA) concept using LTE network in 2.3 GHz band, in IEEE International Symposium
on Dynamic Spectrum Access Networks (DYSPAN), McLean, VA: IEEE Conference Publications, 2014.
19. P. Masek, E. Mokrov, A. Pyattaev, K. Zeman, A. Ponomarenko-Timofeev, et al. Experimental evaluation
of dynamic licensed shared access operation in live 3GPP LTE system, in Globecom , Washington, DC:
IEEE Conference Publications, December 2016.
20. M. Hafeez and J. M. H. Elmirghani. Green licensed-shared access, IEEE Journal on Selected Areas in
Communications, 33(12), 2579–2595, September 2015.
21. E. Pérez, K. J. Friederichs, I. Viering, and J. Diego Naranjo. Optimization of authorised/licensed shared
access resources, in 9th International Conference on Cognitive Radio Oriented Wireless Networks and
Communications (CROWNCOM), Oulu, FL: IEEE Conference Publications, June 2014.
22. T. Wirth, B. Holfeld, D. Wieruch, R. Halfmann, and K.-J. Friederichs. System level performance of cel-
lular networks utilizing ASA/LSA mechanisms, in 1st International Workshop on Cognitive Cellular
Systems (CCS), Rhine River, DE: IEEE Conference Publications, September 2014.
23. R. C. Dwarakanath, J. D. Naranjo, and A. Ravanshid. Modeling of interference maps for licensed shared
access in LTE-advanced networks supporting carrier aggregation, in IFIP Wireless Days (WD), Valencia,
SP: IEEE Conference Publications, November 2013.
24. Y. He, E. Dutkiewicz, G. Fang, and M. D. Mueck. Licensed shared access in distributed antenna systems
enabling network virtualization, in 1st International Conference on 5G for Ubiquitous Connectivity ,
Akaslompolo, FL: IEEE Conference Publications, November 2014.
25. J. Luo, J. Eichinger, Z. Zhao, and E. Schulz. Multi-carrier waveform based exible inter-operator spec-
trum sharing for 5G systems, in IEEE International Symposium on Dynamic Access Networks, McLean,
VA: IEEE Conference Publications, April 2014.
26. V. Y. Borodakiy, K. E. Samouylov, I. A. Gudkova, D. Y. Ostrikova, A. A. Ponomarenko-Timofeev, et al.
Modeling unreliable LSA operation in 3GPP LTE cellular networks, in 6th International Congress on
Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT), St. Petersburg, RU:
IEEE Conference Publications, October 2014.
27. N. Taramas, G. C. Alexandropoulos, and C. B. Papadias. Opportunistic beamforming for secondary users
in licensed shared access networks, in 6th International Symposium on Communications, Control and
Signal Processing (ISCCSP), Athens, GR: IEEE Conference Publications, May 2014.
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
86 Networks of the Future
28. I. Gomez-Miguelez, E. Avdic, N. Marchetti, I. Macaluso, and L. Doyle. Cloud-RAN platform for LSA
in 5G networks-tradeoff within the infrastructure, in 6th International Symposium on Communications,
Control and Signal Processing (ISCCSP), Athens, GR: IEEE Conference Publications, 2014.
29. A. Saadat, G. Fang, and W. Ni. A two-tier evolutionary game theoretic approach to dynamic spectrum
sharing through licensed shared access, in IEEE CIT/IUCC/DASC/PICOM, Liverpool, UK: IEEE
Conference Publications, October 2015.
30. H. Wang, E. Dutkiewicz, G. Fang, and M. Mueck. Spectrum sharing based on truthful auction in licensed
shared access systems, in 82nd Vehicular Technology Conference (VTC Fall), Boston, MA: IEEE
Conference Publications, September 2015.
31. K. Zhu and E. Hossain. Virtualization of 5G cellular networks as a hierarchical combinatorial auction,
IEEE Transactions on Mobile Computing, 2640–2654, October 2016.
32. H. Wang, E. Dutkiewicz, G. Fang, and M. D. Mueck. Framework of joint auction and mixed graph
for licensed shared access systems, in IEEE International Symposium on Dynamic Spectrum Access
Networks (DySPAN), Stockholm, SE: IEEE Conference Publications, September 2015.
33. M. M. Butt, C. Galiotto, and N. Marchetti. Fair dynamic spectrum access in licensed shared access
systems, in EEE International Symposium on Personal, Indoor and Mobile Radio Communications
(PIMRC), Valencia, ES: IEEE Conference Publications, September 2016.
34. J. McMenamy, A. Fahrang, N. Marchetti, and I. Macaluso. Enhanced auction-assisted LSA, in
International Symposium on Wireless Communication Systems (ISWCS), Poznan, PL: IEEE Conference
Publications, September 2016.
35. A. Checko, H. L. Christiansen, Y. Yan, L. Scolari, G. Kardaras, M. S. Berger, and L. Dittmann. Cloud
RAN for mobile networks: A technology overview, IEEE Communications Surveys Tutorials, 7(1), 405–
426, 2015.
36. 3GPP, Study on radio access network (RAN) sharing enhancements, 3GPP TR 22.852 v13.1.0 ,
September2014.
37. C. Liang and F. R. Yu. Wireless network virtualization: A survey, some research issues and challenges,
IEEE Communications Surveys Tutorials, 17(1), 358–380, First quarter 2015.
38. H. Ahmadi, I. Macaluso, I. Gomez-Miguele, L. Doyle, and L. A. DaSilva, Substitutability of spectrum
and cloud-based antennas in virtualized wireless networks, IEEE Wireless Communications Magazine,
99, 2–8, 2016.
39. M. D. Mueck, V. Frascolla, and B. Badic. Licensed shared access—State-of-the-art and current chal-
lenges, in 1st International Workshop on Cognitive Cellular Systems (CCS), Rhine River, DE: IEEE
Conference Publications, September 2014.
40. S. Yrjölä, P. Ahokangas, and M. Matinmikko. Evaluation of recent spectrum sharing concepts from busi-
ness model scalability point of view, in IEEE International Symposium on Dynamic Spectrum Access
Networks (DYSPAN), Stockholm, SE: IEEE Conference Publications, September 2015.
Review Copy – Not for Redistribution
Muhammad Majid Butt - Trinity College Dublin - 9/29/17
ResearchGate has not been able to resolve any citations for this publication.
Some of the new trends emerging in future wireless networks enable a vastly increased fluidity in accessing a wide range of resources, thus supporting flexible network composition and dynamic allocation of resources to virtual network operators (VNOs). In this work we study a new resource allocation opportunity that is enabled by the cloud radio access network architecture. In particular, we investigate the relationship between the cloud-based antennas and spectrum as two important resources in virtualized wireless networks. We analyze the interplay between spectrum and antennas in the context of an auction-based allocation mechanism through which VNOs can bid for a combination of the two types of resources. Our analysis shows that the complementarity and partial substitutability of the two resources significantly impact the results of the allocation of those resources and uncovers the possibility of divergent interests between the spectrum and the infrastructure providers.
- Huiyang Wang
- Eryk Dutkiewicz
- Gengfa Fang
-
![Markus Mueck]()
Licensed Shared Access offers an opportunity to further increase data rates in 5G networks. Considering that different commercial operators have no knowledge of each other, their base stations should be coordinated by a management entity to enable them to access the licensed shared spectrum without interference. An auction mechanism is often used as an efficient tool for resource allocation where rivalry is high. In this paper, we propose an on-line auction framework using a mixed graph due to the spatial reusability of spectrum. This proposed scheme allows each base station to make a concession by asking for a second shrinking interference-free area if its initial area overlaps some other base stations. We use a mixed graph to model the interference between them, because a mixed graph can quantify the interference much closer to the practical cases than an undirected graph does. We also propose to take the bid comparison into account when grouping the independent nodes in the interference graph. These two feathers together make the spectrum spatial efficiency improved, which leads to a higher revenue and a better buyer satisfaction.
Licensed Shared Access (LSA) is a novel flexible regulatory framework, which introduces a shared licensed use of a spectrum band to complement the existing exclusively licensed and license-exempt use. LSA as a general concept was first introduced by the European Commission already in 2011 and it has gained growing interest in standardization and regulatory forums in Europe since. To highlight its potential, European regulators have recently recognized LSA as a promising approach to provide mobile network operators (MNOs) access to the 2.3–2.4 GHz band. In this case, the protection of incumbent users introduces new requirements for information exchange between the incumbents and the mobile network. European Telecommunications Standards Institute's (ETSI) Reconfigurable Radio Systems (RRS) group has defined the requirements, which the LSA system needs to fulfil in order to enable mobile access to the 2.3–2.4 GHz band. This paper places LSA in the regulation and standardization landscape and presents a comprehensive overview of the activities in all relevant forums including their interrelations, to demonstrate the development of the concept. It specifically focuses on the standardization requirements on the LSA system, analyzes and maps the requirements from the ETSI RRS into the different functional blocks of the LSA architecture, and envisions how these can be taken into account in the system implementation. While the incumbent protection places a large number of new requirements on the mobile system design, the LSA system implementation is seen to be feasible by utilizing the existing LTE and LTE-Advanced features, and by developing the required new functionalities according to the standardization requirements.
We are in the midst of a major paradigm shift in how we manage radio spectrum. This paradigm shift is necessitated by the growth of wireless services of all types and the demand pressure imposed on limited spectrum resources under legacy management regimes. The shift is feasible because of advances in radio and networking technologies that make it possible to share spectrum dynamically in all possible dimensions-i.e., across frequencies, time, location, users, uses, and networks. Realizing the full potential of this shift to Dynamic Spectrum Sharing will require the co-evolution of wireless technologies, markets, and regulatory policies; a process which is occurring on a global scale. This paper provides a current overview of major technological and regulatory reforms that are leading the way toward a global paradigm shift to more flexible, dynamic, market-based ways to manage and share radio spectrum resources. We focus on current efforts to implement database-driven approaches for managing the shared co-existence of users with heterogeneous access and interference protection rights, and discuss open research challenges.
Posted by: fivewidget03.blogspot.com
Source: https://www.researchgate.net/publication/320183231_Resource_Management_Techniques_in_Licensed_Shared_Access_Networks