107,99 €
Mehr erfahren.
- Herausgeber: John Wiley & Sons
- Kategorie: Wissenschaft und neue Technologien
- Serie: IEEE Press
- Sprache: Englisch
An up-to-date and comprehensive guide to mobile edge computing and communications
Mobile Edge Computing and Communications offers a practical guide to mobile edge computing and communications (MEC). With contributions from noted experts on the topic, the book covers the design, deployment, and operational aspects of this rapidly growing domain. The text provides the information needed to understand the mainstream system architectures and integration methods that have been proposed in MEC. In addition, the book clearly illustrates critical lifecycle functions and stages of MEC, and shows how to deploy MEC in 5G and beyond mobile networks.
Comprehensive in scope, the book contains discussions on the challenges and opportunities of mobile edge computing and communications concepts combined with the most relevant emerging applications and services. The authors provide insights for all relative stakeholders of mobile networks such as mobile network operators. This important book:
- Provides a comprehensive walkthrough of mobile edge computing and communications
- Includes detailed analysis of current edge applications and technology foundation
- Presents information on driving forces and future directions of MEC
- Provides an authentic source of information from industry experts to drive the future of computing
Written for mobile network operators, ICT service developers, academic researchers, undergraduate and graduate students, Mobile Edge Computing and Communications offers a guide to the current and future of MEC that will enable a completely new paradigm for future computing and communications.
Sie lesen das E-Book in den Legimi-Apps auf:
Seitenzahl: 531
Veröffentlichungsjahr: 2024
Ähnliche
Table of Contents
Cover
Table of Contents
Title Page
Copyright
About the Authors
Preface
To Our Readers
Acknowledgments
Acronyms
Part I: Introduction and Edge Applications
1 Introduction
1.1 Focus of the Book
1.2 The Vision of Edge
1.3 Stakeholders and Related Paradigms
2 Case for Multi-access Edge Computing*
2.1 Case for Multi-access Edge Computing
2.2 MEC Reference Architecture
2.3 MEC Standardization
2.4 MEC Use Cases and Applications
2.5 Related Computing Paradigms to MEC
2.6 Comparison with Related Computing Paradigms
2.7 Summary
Note
3 Edge-Based Video Analytics and Mobile Augmented Reality*
3.1 Video Analytics Overview
3.2 Edge Video Analytics
3.3 Edge-Enabled Augmented Reality (AR)
3.4 Architecture Options for Mobile Augmented Reality
3.5 Applications of MEC-Based MAR
3.6 Technical Aspects of MEC-Based MAR
3.7 Summary and Outlook
Note
4 for eHealthcare: Empowering Next-Generation Healthcare Systems*
4.1 Introduction
4.2 MEC Enabled Use Cases
4.3 Realizing Use Cases with MEC
4.4 Deployment Challenges and Solutions
4.5 Conclusion
Note
Part II: Technology Foundation
5 Edge AI*
5.1 Overview of Edge AI
5.2 Training and Inference in Edge AI
5.3 Edge AI in Practice
5.4 Summary and Outlook
Note
6 Edge Offloading
6.1 Motivation and Fundamental Concept
6.2 Edge Offloading
6.3 Edge Offloading in Practice
6.4 Summary and Outlook
7 Edge Storage and Caching
7.1 Edge Storage
7.2 Edge Caching
7.3 Edge Caching Towards Information-Centric Networking
7.4 Conclusion
8 Edge Communication, Hardware and Enabling Frameworks*
8.1 Advancement of Communication Technologies
8.2 Evolution of Communication Networks
8.3 Edge Computing Landscape
8.4 Edge Computing System Solutions
8.5 Edge Hardware and Devices
8.6 Edge Accelerators
8.7 Industry Frameworks
8.8 Concluding Remarks
Note
Part III: Integration and Driving Forces
9 Driving Forces for Edge Integration
9.1 Internet of Things Toward Internet of Everything
9.2 The Growth of Small Data
9.3 AI/ML Based Communication Technologies
9.4 Novel Technologies and Applications
9.5 Conclusion
10 Edge Security*
10.1 Introduction
10.2 Security Threats in MEC
10.3 Threat Vectors Related to the Access Network
10.4 Threat Vectors Related to the Mobile Edge Network (MEN)
10.5 Threat Vectors Related to the Core Network
10.6 Architectural Threat Vectors
10.7 Security Requirements of MEC
10.8 Possible Security Solutions for MEC
Note
11 Edge Privacy*
11.1 Introduction
11.2 Privacy in 5G and Beyond
11.3 MEC Privacy
11.4 Objectives for Privacy Preservation in MEC
11.5 Privacy Protection Mechanisms for MEC
11.6 Future Directions
Notes
12 Telecom and Cloud Integration*
12.1 Introduction
12.2 Network Softwarization in 5G Networks
12.3 MEC Integration in 5G Backhaul
12.4 Key Technologies for MEC 5G Integration
12.5 Conclusion
Note
Part IV: Outlook
13 Conclusion and Outlook for
References
Index
End User License Agreement
List of Tables
Chapter 4
Table 4.1 Requirements of MEC Use Cases.
Chapter 6
Table 6.1 Metric for Assessing Data Offloading.
Table 6.2 Metric for Assessing Computation Offloading.
List of Illustrations
Chapter 2
Figure 2.1 What Is MEC.
Figure 2.2 Benefits of MEC Compared to Cloud Computing.
Figure 2.3 Evolution of MEC.
Figure 2.4 MEC Reference Architecture.
Figure 2.5 Standardization Timeline of MEC.
Figure 2.6 Critical Infrastructure Connectivity to MEC Platform.
Figure 2.7 eMBB and Video Streaming Applications with MEC.
Figure 2.8 MTC Integration with MEC.
Figure 2.9 ITS Integration with MEC.
Figure 2.10 Augmented Reality Integration with MEC.
Figure 2.11 UAV-Based MEC Deployments.
Chapter 3
Figure 3.1 Miligram’s Reality–Virtuality Continuum.
Figure 3.2 Components of a Generic AR System.
Figure 3.3 Characteristics of Augmented Reality vs. Mobile Augmented Reality...
Figure 3.4 A Mobile Augmented Reality Service Scenario with MEC processing....
Figure 3.5 Cloud Based System Architecture for a MAR System.
Figure 3.6 MEC Based System Architecture for a MAR System.
Figure 3.7 Localized System Architecture for a MAR System Using a Dedicated ...
Figure 3.8 Localized System Architecture for a MAR System Using only MAR Dev...
Figure 3.9 Hybrid System Architecture for a MAR System.
Figure 3.10 Applicability of Different Architectures Based on the Characteri...
Chapter 4
Figure 4.1 MEC Based Health Services.
Figure 4.2 Realization of Use Cases with MEC.
Chapter 5
Figure 5.1 Overview of Edge AI and Cloud AI Architectures.
Figure 5.2 Computing Continuum.
Figure 5.3 Scope of Seven Levels for Edge AI.
Chapter 6
Figure 6.1 Infrastructure-driven and device-driven offloading.
Figure 6.2 Edge Offloading: Cloud-to-Edge Paradigm.
Chapter 7
Figure 7.1 Edge Storage.
Figure 7.2 Edge Caching.
Figure 7.3 Information-Centric Networking.
Chapter 8
Figure 8.1 Towards THz Communication.
Figure 8.2 Federated Learning.
Figure 8.3 Evolution of Mobile Networks Towards .
Figure 8.4 6G Technologies for .
Figure 8.5 Advantages of Using Edge AI.
Figure 8.6 Popular Edge AI Chip Manufacturers and Devices Categorized Accord...
Figure 8.7 Primary Metrics for Edge HW Design Considerations.
Figure 8.8 Hardware Correlation in Edge Applications.
Figure 8.9 Processing Units and Alternatives for AI Model Deployment at Edge...
Figure 8.10 Edgent Framework.
Chapter 9
Figure 9.1 Driving Trends Toward MEC Integrated Beyond 5G.
Chapter 10
Figure 10.1 Identified Threat Vectors in an MEC Deployment.
Chapter 11
Figure 11.1 Privacy by Design.
Chapter 12
Figure 12.1 Four Phases Toward Softwarization of Mobile Networks.
Figure 12.2 MEC in NFV Architecture.
Figure 12.3 The Use of SDN for MEC.
Figure 12.4 Use of Network Slicing and MEC in Different 5G Applications.
Guide
Cover
Table of Contents
Title Page
Copyright
About the Authors
Preface
To Our Readers
Acknowledgments
Acronyms
Begin Reading
References
Index
End User License Agreement
Pages
iii
iv
xiii
xiv
xv
xvi
xvii
xviii
xix
xxi
1
3
4
5
6
7
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
69
71
72
73
74
75
76
77
78
79
80
81
83
84
85
86
87
88
89
90
91
92
93
94
95
97
98
99
100
101
102
103
104
105
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
139
140
141
142
143
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
207
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
Mobile Edge Computing and Communications
Driving Forces, Technology Foundation, and Application Areas
Aaron Yi Ding
Delft University of Technology
Netherlands
Chamitha De Alwis
University of Bedfordshire
United Kingdom
Madhusanka Liyanage
University College Dublin
Ireland
This edition first published 2025© 2025 John Wiley & Sons Ltd.
All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by law. Advice on how to obtain permission to reuse material from this title is available at http://www.wiley.com/go/permissions.
The right of Aaron Yi Ding, Chamitha De Alwis, and Madhusanka Liyanage to be identified as the authors of this work has been asserted in accordance with law.
Registered OfficesJohn Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030, USAJohn Wiley & Sons Ltd, New Era House, 8 Oldlands Way, Bognor Regis, West Sussex, PO22 9NQ, UK
For details of our global editorial offices, customer services, and more information about Wiley products visit us at www.wiley.com.
Wiley also publishes its books in a variety of electronic formats and by print-on-demand. Some content that appears in standard print versions of this book may not be available in other formats.
Trademarks: Wiley and the Wiley logo are trademarks or registered trademarks of John Wiley & Sons, Inc. and/or its affiliates in the United States and other countries and may not be used without written permission. All other trademarks are the property of their respective owners. John Wiley & Sons, Inc. is not associated with any product or vendor mentioned in this book.
Limit of Liability/Disclaimer of WarrantyIn view of ongoing research, equipment modifications, changes in governmental regulations, and the constant flow of information relating to the use of experimental reagents, equipment, and devices, the reader is urged to review and evaluate the information provided in the package insert or instructions for each chemical, piece of equipment, reagent, or device for, among other things, any changes in the instructions or indication of usage and for added warnings and precautions. While the publisher and authors have used their best efforts in preparing this work, they make no representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives, written sales materials or promotional statements for this work. The fact that an organization, website, or product is referred to in this work as a citation and/or potential source of further information does not mean that the publisher and authors endorse the information or services the organization, website, or product may provide or recommendations it may make. This work is sold with the understanding that the publisher is not engaged in rendering professional services. The advice and strategies contained herein may not be suitable for your situation. You should consult with a specialist where appropriate. Further, readers should be aware that websites listed in this work may have changed or disappeared between when this work was written and when it is read. Neither the publisher nor authors shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages.
Library of Congress Cataloging-in-Publication Data Applied for:
HardBack ISBN: 9781119611523
Cover Design: WileyCover Image: © buffaloboy/Shutterstock
About the Authors
Aaron Yi Ding leads the Cyber Physical Intelligence (CPI) Lab as tenured Associate Professor at TU Delft and Associate Professor (permanent) in Computer Science at the University of Helsinki. He has over 17 years of R&D practices in EU, Switzerland, UK and USA. Prior to TU Delft, he has worked at TU Munich with Jörg Ott, at the University of Cambridge with Jon Crowcroft, at Columbia University with Henning Schulzrinne, and on sabbatical at ETH Zurich with Adrian Perrig. His PhD is with Sasu Tarkoma (Helsinki) and Jon Crowcroft (Cambridge). Funded by the Nokia Foundation, vital part of his PhD programme is completed at the University of Cambridge and Columbia University in New York. He has supervised 100+ students and won EU research grants as Consortium Director and Principal Investigator (PI) in tight collaboration with industrial companies of Nokia, Ericsson, Deutsche Telekom, Broadcom, Telia, NEC, Telefonica and WithSecure. His research focuses on edge computing, edge AI, and data-driven IoT services. For contributing to mobile edge computing with over 90 peer-reviewed publications, he received best paper awards and recognition from ACM SIGCOMM, ACM EdgeSys, ACM SenSys CCIoT, ACM IoT, IEEE INFOCOM, IEEE CVPR MAT, and TU Delft. Being an active member of ACM, IEEE, IETF, he is the founder of ACM EdgeSys, Associate Editor for ACM TIOT, IEEE OJ-ITS and Springer Nature Computing. He has served on the chairing and programme committees for numerous prestigious conferences including ACM SIGCOMM, ACM MobiCom, ACM UbiComp, ACM WWW, ACM CoNEXT, ACM MobiSys, ACM/IEEE SEC, and IEEE INFOCOM. He is a recipient of the esteemed Nokia Foundation Scholarship. More info: https://aaronyiding.github.io.
Chamitha De Alwis, a Senior Member of the IEEE, currently holds the position of Senior Lecturer in the School of Computer Science and Technology at the University of Bedfordshire, United Kingdom. His academic journey includes the completion of a PhD in Electronic Engineering from the University of Surrey, United Kingdom, in 2014, as well as a distinguished BSc with First Class Honors in Electronic and Telecommunication Engineering from the University of Moratuwa, Sri Lanka, in 2009. Chamitha is recognized for his leadership and academic contributions as the Founder Head of the Department of Electrical and Electronic Engineering at the University of Sri Jayewardenepura, Sri Lanka, where he concurrently served as a Senior Lecturer. Beyond academia, he has demonstrated his expertise in roles as a consultant and network engineer in the dynamic field of telecommunications. Maintaining an active research profile, he boasts a commendable track record of scholarly publications and has been the recipient of several competitive grants. Notably, he has played a pivotal role in numerous research projects, including those funded by the EU FP7. His research pursuits are centered around critical areas of inquiry, with a primary focus on network security, the evolution of 5G/6G technologies, and the transformative landscape of blockchain.
Madhusanka Liyanage an Assistant Professor/Ad Astra Fellow and Director of Graduate Research at the School of Computer Science, University College Dublin, Ireland. He is also a Adjunct Professor at the University of Oulu, Finland, University of Ruhuna, Sri Lanka, and University of Sri Jarawardhenepura, Sri Lanka. He received his Doctor of Technology degree in communication engineering from the University of Oulu, Finland, in 2016. He also received the prestigious Marie Skłodowska-Curie Actions Individual Fellowship and the Government of Ireland Postdoctoral Fellowship during 2018–2020. During 2015–2018, he has been a Visiting Research Fellow at the CSIRO, Australia, the Infolabs21, Lancaster University, UK, Computer Science and Engineering, The University of New South Wales, Australia, School of IT, University of Sydney, Australia, LIP6, Sorbonne University, France, and Computer Science and Engineering, The University of Oxford, UK. He is also a senior member of IEEE. In 2020, he received the “2020 IEEE ComSoc Outstanding Young Researcher” award by IEEE ComSoc EMEA. In 2021 and 2022, he was ranked among the World’s Top 2% Scientists (2020 and 2021) in the List prepared by Elsevier BV, Stanford University, USA. Also, he was awarded an Irish Research Council (IRC) Research Ally Prize as part of the IRC Researcher of the Year 2021 awards for his positive impact as a supervisor. In 2022, he received “2022 The Tom Brazil Excellence in Research Award” by SFI CONNECT Center. Moreover, Madhusanka received a Special Commendation for IRC Early Career Researcher of 2022 by Irish Research Council, Ireland. He is also an expert consultant at European Union Agency for Cybersecurity (ENISA). In 2021, Liyanage was elevated as a Funded Investigator of the Science Foundation Ireland CONNECT Research Centre, Ireland. Moreover, he is an expert reviewer at different funding agencies in France, Qatar, UAE, Sri Lanka, and Kazakhstan. His research interest includes 5G/6G, Blockchain, Network security, Artificial Intelligence (AI), Explainable AI, Federated Learning (FL), Network Slicing, Internet of Things (IoT), Multi-access Edge Computing (MEC) and Cyber-Physical Systems (CPS). More info: www.madhusanka.com.
Preface
This book entails a special journey that is accomplished throughout the pandemics of 2020–2023. As the adoption of information and communication technologies has been accelerated by the growing trend of remote working and learning, the book is developed in constant reflections on the opportunities and challenges behind this rapid transition.
Recently, we have witnessed a remarkable proliferation of new applications and services in the context of Internet-of-Things (IoT) and mobile computing. With the mission of being smart, the various concepts under the umbrella of IoT, including smart factory, smart cities, and autonomous driving, have introduced new ways for people to interact with the physical world. At the same time, mobile computing has evolved to new dimensions with more advanced paradigms such as virtual reality/augmented reality (VR/AR). On the one hand, these applications and services are expected to contribute significantly to the economy and human life quality. On the other hand, a lot of new challenges have been imposed or magnified, among which real-time big data processing and analysis become prominent. Despite the fact that mobile devices are becoming more and more powerful nowadays, handling the constantly generated data flows in real time on those devices for providing application intelligence is still undesirable or non-optimal, one of the reasons being the limited battery lifetime of the mobile device.
With the help of cloud computing, Mobile Cloud Computing (MCC) targets the processing capability gap. Unfortunately, the uncertainty of the network connection to the remote cloud renders most of the MCC solutions impractical, especially for the aforementioned applications. Meanwhile, wireless communication has advanced to a large extent in the past decade. Under the 5G regime, various technologies have been developed that can achieve properties of ultra-high bandwidth, ultra-low latency, and high reliability. Benefiting from these new communication technologies, MCC seems to become applicable but the reality is crucial as the Internet—being part of the remote connection, comes with no such nice properties of 5G, which does not seem to change soon. In the end, only the last hop connectivity can be expected to embrace the new opportunities brought by the 5G adoption.
The situation described above naturally leads to the advent of edge computing—a technology concept that aims to provide intelligence from the edge of the networks. In this book, we deliberately use the term in a more general manner for the essential technologies of mobile edge computing and communications. The idea of is to generalize the traditional cloud computing into more distributed environments with close proximity (even in a single hop) to the mobile and IoT devices. Thanks to the 5G advancement, the computing and storage resources at the network edge now can facilitate the real-time data processing for emerging applications with performance guarantees. Note that such computing/storage resources are already partially available on the infrastructure as a result of the recent network campaign called Network Function Virtualization (NFV), where network functions are virtualized and run on general-purpose servers deployed within the edge network. In certain sense, edge computing can be regarded as an extension of NFV from the core network to mobile and IoT applications.
Many enabling technologies, such as Software Defined Networking (SDN), network slicing, Information Centric Networking (ICN), and their synergies have also been extensively explored by both the academia and industry. From the computing perspective, new distributed computing infrastructures that feature locality-awareness have been proposed, adopting new programming/management concepts such as microservice or serverless computing. In addition, multiple communication technologies (e.g., those under 5G, WiFi, VLC, and LoRa) have also been examined for their potentials of supporting different applications in . The integration of both computing and communication parts is complex and becomes unprecedentedly challenging, despite that the aforementioned enabling technologies can provide significant simplification to system management and control. With all the research and development efforts, is expected to tremendously benefit a large variety of emerging mobile applications and services and will enable a completely new paradigm for future computing and communications.
Following the Top-Down approach to start from the “layer” of applications and dive further into the underlying technologies and integration, this book is divided into four parts:
Part I
:
The first part of the book introduces emerging “edge native” applications for readers to comprehend the practical usage of edge computing.
Part II
:
The second part dives deeper into the technology foundation that supports the edge applications, including system frameworks, enabling technologies and edge hardware.
Part III
:
The third part covers the technology driving forces and the integration with cloud and mobile ecosystems. Security and privacy aspects of are discussed.
Part IV
:
The final part illustrates open challenges and future directions for .
Instead of being a competing paradigm, the future success of is bundled with the cloud and mobile ecosystems by benefiting from the mature cloud and mobile access infrastructures. In a nutshell, is an exciting domain that is growing fast and fostering numerous opportunities. We wish you enjoy this book and have fun in reading it.
Delft, 2024
Aaron Yi Ding
Chamitha De Alwis
Madhusanka Liyanage
To Our Readers
This book covers mobile edge computing and communications, an emerging computing-communication paradigm that augments today’s centralized cloud by distributing processing and storage capabilities into mobile/access networks, close to where end users connect. Drawing influence from cloud and cellular domains, this computing-communication paradigm is a fast growing phase. Standardization efforts have been made by the European Telecommunications Standards Institute (ETSI) Industry Specification Group (ISG) in terms of mobile edge computing/multi-access edge computing.
In this book, we deliberately use the term in a more generic manner to address the essential technologies for mobile edge computing and communications. This book provides a walk-through of applications, technology foundation and driving forces for .
We take a “Top-Down” approach in this book. As a reader, you will first learn the emerging “edge native” applications to comprehend the practical usage of (Part I). Then, you can dive deeper into the technology foundation (Part II), the integration and driving forces (Part III). Finally, the book concludes with open challenges and outlook on future computing directions (Part IV).
The content of this book appeals to a wide range of audience in the ICT domain, including:
ICT and IoT Service Developers:
As computing requirements and user demands can outpace the traditional tools available, this book provides practical information on the frameworks and technical enablers of , which can facilitate developers to create innovative ICT applications catering to future demands.
Mobile Network Operators (MNOs):
MNOs are currently looking to adopt new technologies such as edge computing to offer responsive services to their customers. This book offers the guidelines, methods, tools and mechanisms that can benefit their development and deployment in the next generation mobile infrastructure, i.e., 5G and beyond.
Academic Researchers:
Mobile edge computing has already been an area of active research and study for major educational institutions across the world. This book meets the growing demand from academics who can benefit from the latest literature reference and use cases, especially in their daily teaching and research on this critical domain of interest.
Computing Professionals and Industrial Experts:
Edge computing is going to cross the traditional borders and is going to have visible impact to enterprises and organizations who are planning to transform into IoT-driven business. It would be critical for computing professionals and industrial experts such as IT architects to start aligning their technology adoption and architectures to the future needs of computing and standards. This books offers resources to design and build an edge-driven architecture and maintain it.
Advanced Undergraduate and Graduate Students:
For educational institutes offering computing and networking lectures, this book provides an essential source of up-to-date information for fundamental technologies in . The use cases and practical examples shall motivate curious students to explore real-world problems in operational environments. The book is hence appropriate for advanced undergraduate graduate level courses.
Through the book, readers can get familiar with the mainstream system architectures and integration methods that have been proposed in recent years. The book also discusses the open challenges and outlook from design, deployment, and operational aspects. Given the rapid pace of edge technology advancement, this book is not intended to be encyclopedic and likely to be an evolving “piece” for the foreseeable future. Overall, this book shall offer insights for active actors and stakeholders of mobile networking and cloud computing. By serving as a reference material and introductory guide on mobile edge computing and communications, this book shall appeal to a wide intended audience including academic researchers, service developers, computing professionals, curious university students, network operators and industrial experts.
Acknowledgments
To my wife Mia for her trust, encouragement and firm supports during those tough years. I express my sincere gratitude for the understanding and support from Sandra at Wiley. I appreciate the initial discussions with Lin Wang and Dirk Kutscher at the start of the journey for preparing this book.
Aaron Yi Ding
This book would not have been possible without the great help and supports of many. The concept of publishing this book on mobile edge computing and communications came into light during our research work in projects of EU H2020 SPATIAL project (Grant No. 101021808) and EU Marie Curie ITN APROPOS project (Grant No. 956090).
We would also like to acknowledge all the partners of those projects. Furthermore, we would like to thank our universities, Delft University of Technology, University of Bedfordshire and University College Dublin, for all the support extended towards the successful completion of this book. We would also like to thank chapter contributors, including Dewant Katare, Pasika Ranaweera, Yushan Siriwardhane, Anshuman Kalla for their invaluable contribution to complete this book. We also thank all the reviewers for helping us select suitable chapters for this book.
We appreciate the efforts of Professor for writing a nice foreword to the book. We are also grateful to Sandra Grayson, Teresa Netzler, Nandhini Karuppiah, and the whole John Wiley & Sons team for their support towards getting this book published.
Last but not least, we would offer our heartiest gratitude to our families, who gladly allowed us to share our time with them towards the completion of this book.
Aaron Yi Ding
Chamitha De Alwis
Madhusanka Liyanage
Acronyms
mobile edge computing and communications
BHCA
busy hour call attempts
BR
bandwidth reservation
b.u.
bandwidth unit(s)
CDN
content delivery network (CDN)
CDTM
connection dependent threshold model
CS
complete sharing
DiffServ
differentiated services
ETSI
European Telecommunications Standards Institute
FIFO
first in-first out
GoS
Grade of Service
ICT
Information and Communication Technology
IoT
Internet of Things
IP
Internet protocol
ITU-T
International Telecommunication Unit-Standardization sector
MMPP
Markov modulated poisson process
MPLS
multiple protocol labeling switching
PASTA
Poisson arrivals see time averages
probability distribution function
PFS
product form solution
QoS
Quality of Service
RED
random early detection
RLA
reduced load approximation
TCP
transport control protocol
UDP
user datagram protocol
Part IIntroduction and Edge Applications
‘What exactly is an Edge App?’ – this is probably one of the most common questions when you start the journey on edge computing and communication. Before we dive into the fundamental technologies, this part of the book will illustrate the focus of the book 1 and introduce the applications of for readers to comprehend the practical usage, including Multi-Access Edge Computing in Chapter 2, edge video analytics and augmented reality in Chapter 3, and smart healthcare in Chapter 4.
1Introduction
1.1 Focus of the Book
Besides the latest research on computing design and standardization efforts made by the European Telecommunications Standards Institute (ETSI) Industry Specification Group (ISG), the driving forces behind edge computing and its technology foundations deserve a thorough illustration and analysis. Along with the advancement of Internet-of-Things (IoT) and mobile communications, the rapid development on “Edge” is calling for a handbook that covers the service/application perspectives and how to integrate “Edge” in the future computing infrastructures.
Given the “Edge” domain is in a fast growing phase, in this book, we deliberately use the term in a more generic manner to enclose the essential technologies for mobile edge computing and communications.
One of the key reasons for a seemingly old idea to take off after many decades is that other enabling technologies make it possible. This holds especially for . For the rapid advancement of , this book is not intended to be encyclopedic and likely to be an evolving “piece” for the foreseeable future. In this respect, readers’ suggestions, feedback are more than welcome. By serving as a reference material and introductory guide on mobile edge computing and communications, this book shall appeal to a wide intended audience including academic researchers, service developers, computing professionals, curious university students, network operators and industrial experts.
In this book, we take the system and incremental deployment perspective, which is not to be mixed with incremental research. The research can be disruptive or ground-breaking (i.e., non-incremental) but still lend itself to incremental deployment and integration to existing mobile and cloud infrastructures. The research can hence go beyond the constraints of existing deployment context by envisioning what could be possible. Ultimately, it comes down to the deployment of that leads to long-term impact. Overall, this book shall offer practical deployment insights for actors and stakeholders in the future mobile and cloud infrastructures.
1.2 The Vision of Edge
The vision for edge computing is brought out as early as 2005 [1] where the importance of network edge is highlighted as a new source of creative energy for system and applications, especially given the growing demand in mobile access networks. The initial concept is further formulated in recent years, along with a visible trend for data analytics moving toward this “edge.” Gartner predicted that by 2023, over 50% of the primary responsibility of data analytics stakeholders will comprise data that is created, managed and analyzed in edge environments. The advantages may include greater data management flexibility, speed, governance, and resilience. In addition, the edge capabilities can support use cases ranging from real-time event analytics to autonomous driving services [2].
Albeit its great potential, the fundamental concept of edge computing, referred as in this book, is not new. By looking at the history of computing, it is not the first time for the network edge to receive tremendous attention. For instance, we have seen numerous computing design before the hype of , including the peer-to-peer (P2P) networks, content delivery networks (CDN), and mobile cloud computing (MCC). It is hence appropriate to remind our readers that one of the key reasons for a seemingly “old” idea, such as edge computing, to take off after many years of undergoing is that other enabling technologies are making it possible. Those enablers and technology foundation are exactly what we intend to illustrate throughout this book.
In fact, the vision is largely built upon its promises to the rising demands for high bandwidth, low latency, better privacy and reliability. Those promises have been argued recently by experts as to whether it is merely an academic enthusiastic pursuit [3]. At the same time, as the Internet is transformed into a more ethical system, is gradually showing its potential in enhancing data privacy and sovereignty [4].
Regardless of the controversial opinions [3, 5, 6], is clearly far from reaching its “Hay Day.” Many uncertainties and challenges are intertwined with the opportunities along its advancement. From a pragmatic perspective, this book conveys our vision to facilitate future development of towards a coherent edge-to-cloud continuum. This continuum path needs more collaborative efforts with experts from multiple disciplines even outside the technical sphere (e.g., law, ethics and public policies), for achieving the full potential of .
Why Do We Need Edge?
Regarding a frequently asked question “Why do we need edge?,” the general motives and potential benefits include:
Latency and Bandwidth:
brings the latency benefit by placing computation units closer to the data sources and end users. Given the bandwidth bottleneck to access a distant cloud via the Internet, can conserve bandwidth by processing data locally, hence avoiding transferring excessive amount of data to the Internet. This helps alleviate network congestion. The latency and bandwidth benefits are crucial for real-time services that also demand higher bandwidth, such as virtual reality (VR) gaming and autonomous driving.
Proliferation and Personalisation:
is envisioned to capture high quality data in a distributed manner from the massive deployment of IoT, augmented reality (AR)/VR devices, and smart vehicles. can achieve higher quality by filtering out the data “noise,” labeling the data with more context and with better sampling. The high-quality data with locality context is demanded by analytic services for personalisation (e.g., for end users/clients).
Privacy and Sovereignty:
The edge-native computing and analytics (often referred as Edge AI) can keep the data ownership and control closer to the end users. This can when computation is managed and task distribution controlled from the user’s own devices, and suitable security and privacy protection methods are in use. By utilizing local context, can strike a balance between privacy and usability, while allowing ethical data management.
Energy Saving and Sustainability:
avoids transmitting redundant data traffic to consume network resources. The energy required for transmitting data is proportional to the distance that data travels from the source (end user devices) to the Internet or centralized cloud sites. By cutting down the data volume to be transmitted and processed by the network units along the path, can promote the energy saving of the Internet. In addition, the data flowing through the Internet is becoming one the primary drivers for CO
2
emissions. can help reduce the carbon footprint.
Mobility and Collaboration:
can support end user devices continue functioning even when they are disconnected from the cloud. This is particularly valuable for mobile scenarios where network connectivity is intermittent or unreliable. In addition, is complementing the corporate data centers that facilitate computing, storage, networking, and data analytic functions at locations such as collaboratively training large-scale machine learning models.
What and Where is the Edge?
To approach the question “What is the Edge?,” it can often trigger heated discussions. Can it be a lightweight computing server, a layer of networking devices, or a set of IoT oriented services? For some of you who are studying for the first time, this can be rather confusing.
One major reason for such blurring interpretation is the fact that is in rapid development. The concept of “edge” is hence enriched and evolved along the way, sometimes mixed with different flavours. For example, a general expectation for edge is to extend the cloud computing capability to the access networks. For this prospect, can take the form of lightweight computing servers that can offload the tasks from data centers (cloud) for performance considerations. Meanwhile, for mobile network operators that are enhancing their access infrastructure, can also be regarded as a layer of network devices (e.g., routers) in the proximity to mobile devices and data sources. In addition, there is an increasing pressure to offer low-latency and high-bandwidth applications such as augmented reality (AR), virtual reality (VR), Internet-of-Things (IoT), and autonomous driving. From the service offering angle, can be regarded as a “meta” service that enables numerous advanced applications demanded by the growing amount of connected things.
Closely related to the “What” question, “Where is the Edge?” also depends on which stakeholder we are asking. For an Internet service provider (ISP) that owns and operates the network access infrastructure, can be a layer that consists of last hop gateways to connect end users to ISP’s infrastructure. For instance, mobile operators can utilize to construct a computing layer on their base stations, so the enabled base station is regarded as where the edge is “physically” located. For cloud service providers such as Amazon and Google, is the extended computing and storage infrastructure closer to their customers for supporting real-time services, analytics, and content delivery. For instance, Content Delivery Networks (CDNs) can be regarded as where the edge is placed, for reducing the distance and latency of content delivery. In the future, may come in different form factors for different scenarios. The devices/entities with computational capability to perform computing near the data source could be considered as part of edge infrastructure This is in line with the motive of edge computing.
1.3 Stakeholders and Related Paradigms
Beyond the “What” and “Where” questions, it is important to understand the major stakeholders and related paradigms of . The major stakeholders and actors involved in include:
Mobile Network Operators (MNOs):
MNOs are currently looking to adopt edge computing to offer responsive mobile services to their customers. For telecom’s next generation mobile infrastructure, MNOs can harness the advantage of edge to create an integrated computing-communication hub in the “last-mile” network for efficient data processing, analytics, and communications.
Cloud Service Providers:
To meet the demands from a wide-range of latency-sensitive and data-intensive applications that will emerge in the near future, cloud service providers are adopting edge designs to enhance their service performance. This is in line with the new industry trend of computing continuum, which could lead to the convergence of cloud and edge.
Academic Researchers and Industrial Experts:
has created active research and practical solutions for academic researchers and industrial experts, respectively, to tackle challenges in computing sustainability and data sovereignty.
ICT Service/Application Developers:
As computing requirements and user demands can outpace the traditional tools, ICT service developers are constantly seeking new technologies such as that can facilitate them to create innovative ICT applications that cater to future demands from users, government regulations, and international standards.
Since is in its making phase, there are related computing designs that have been explored and inspired the trend of edge. The related computing concepts include P2P, CDN, MCC, Cloudlets, Fog and Mist:
P2P
: The concept of Peer-to-Peer (P2P) is driven by decentralization. It is intended to address the limitations of classic client–server design in terms of performance bottleneck and single point of failure on the server side. As a decentralized networking design, P2P can enable direct communication between distributed devices for better scalability, reliability and resource utilization. P2P has provided inspiration to from decentralization perspective.
CDN
: To optimize the latency in content-oriented ICT/web services, Content Delivery Network (CDN) forms a distributed proxy layer that can cache content and serve such content via CDN servers geographically closest to the end users. The latency benefit of CDN has inspired to achieve low latency, load balancing and service availability in a distributed environment.
MCC
: The design of Mobile Cloud Computing (MCC) integrates cloud computing with mobile computing to support efficient computational offloading. By harnessing the computing and storage capacity in the remote cloud, MCC can enable devices with limited processing power and storage to run resource-intensive applications. MCC has inspired to enhance end user experience in terms of resource offloading and battery saving on the end devices.
Cloudlets, Fog and Mist
: The idea of Cloudlets
[7]
is to deploy small-scale data processing units close to the edge of access network. Cloudlets aims to tackle latency and bandwidth challenges for applications that often need to exchange data with remote cloud servers. In line with the goal of improving latency and bandwidth, Fog computing
[8]
aims to transform the network edge to a distributed computing infrastructure for the rapid growth of IoT. Similar to Fog computing, the term Mist is geared towards embedded computing devices deployed in the access networks. In general, Cloudlets, Fog and Mist intend to extend the Cloud computing paradigm by allowing data to be processed and analyzed across various levels of the network hierarchy. The concepts of Cloudlets, Fog and Mist formed the groundwork for computing continuum that inspired to complement both centralized cloud and distributed end devices by offering an intermediate layer of edge resources to balance the benefits of local processing with the advantages of cloud services.
In the following chapters of Part 1 we will illustrate the applications of to help readers comprehend the practical usage of edge technologies.
2Case for Multi-access Edge Computing*
2.1 Case for Multi-access Edge Computing
Multi-access Edge Computing (MEC) is a pioneer edge computing paradigm introduced by the European Telecommunications Standards Institute (ETSI) to overcome the intricacies of impending mobile communication networks. As the name derives, the focus of the MEC is to cater multi or diversified access to the existing mobile network for expanding its capability to meet the requirements of novel Internet of Things (IoT) based services. IoT, being the pinnacle of the Information Technology (IT) and electronics industries in the recent past, has raised the guaranteed intrinsic specifications of mobile and Internet technologies drastically. To have the “Things” such as physical devices/apparatus/components, sensors, actuators, or embedded software systems communicated through the Internet for IoT services, MEC provides the means to solve their scalability, interoperability, and compatibility challenges.
2.1.1 What Is MEC?
The main goal of MEC is to extend the Cloud Computing (CC) capabilities to the edge of the mobile network to overcome the constraints of prevailing cloud services [9]. Hitherto, CC was the solution that unburdened the service providers from the inconvenience of maintaining their own data centers. MEC creates a server/computing environment in mobile Base Stations (BSs) or 5G based gNodeBs (gNBs) integrated to the mobile network infrastructure [10]. As Figure 2.1 illustrates, MEC is complementing the corporate data centers that facilitate computing, storage, networking, and data analytic functions at locations in proximity to the data source [11]. In fact, these data centers localized in the mobile edge network are supported by the cloud servers, and 5G core network entities located at the mobile core network to cater enhanced service access to the radio access network (RAN).
2.1.2 Why We Need MEC?
Current cloud-based services require connectivity to the centralized cloud platforms to convey control information and authentication credentials for authorization mechanisms. This connection, which is generally linked through the Internet with featured higher bandwidth consumption and embedded cumbersome cryptographic primitives; restricts the real-time service guarantees of forthcoming mobile technologies. Figure 2.2 illustrates the significance of the MEC paradigm concerning such impediments. Thus, the resource-enriched edge servers of MEC systems are promising real-time service guarantees for novel applications. Since the intrinsic processing and storage functions can be provided with the MEC edge platforms, cloud connectivity requirement becomes diminutive. It is obvious that Mobile Network Operators (MNOs) should invest more in forming the MEC edge server environment. The communication capacity of the MEC-enabled BSs should also be improved with the investment into the edge server environment. Thus, despite the investment put forward by the MNOs, the number of BSs can be reduced due to the improved communication range that serves better scalability [12].
Figure 2.1 What Is MEC.
The forthcoming fifth-generation (5G) mobile technology and its requirements serve as one of the rationales for the emergence of MEC. The performance metrics of data rates up to 10 Gb/s, service level latency below 1 ms, ultra-high reliability of 99.99999%, reduced energy consumption of 90%, and support for 300,000 devices within a single cell are guaranteed for 5G [9, 13]. To meet these requirements, service infrastructure (currently operating at clouds) should be migrated to a proximate location. Thus, the MEC paradigm is formed and designed with the above intentions.
As Figure 2.2 illustrates, storage and processing infrastructure facilitated with the networking platform in MEC deployment is ensuring the benefits of ultra-low latency, locational awareness, proximate data outsourcing, and improved capacity in the edge devices. These features enable higher bandwidth and real-time responsiveness to the subscriber applications. These features enable the launching of novel 5 G-based applications or use cases such as Ultra High Definition (UHD) video streaming, Augmented Reality (AR), Virtual Reality (VR), Mixed Reality (MR), Tactile Internet, Machine Type Communication (MTC), Machine-to-Machine (M2M), Industrial IoT (IIoT), Unmanned Aerial Vehicle (UAV), and Vehicle-to-Everything (V2E). These terms will be introduced later in the chapter.
Figure 2.2 Benefits of MEC Compared to Cloud Computing.
Source: Yeamake/Adobe Stock.
2.1.3 Evolution of MEC
The data storage and processing services have evolved from mainframes (BITNET-1981) to dedicated servers, and to cloud computing for improving the convenience of data center maintenance. The growth in demand for Big data and data science over the years leading to IoT emergence has prompted the Internet Service Providers (ISPs) to invest more in their data centers; though faced issues in erecting dedicated vicinities for the purpose while expanding the capabilities of the network technologies to improve the network utilization. With cloud computing, the subscriber is facilitated with Infrastructure as a Service (IaaS), Platforms as a Service (PaaS), and Software as a Service (SaaS) directives by the Cloud Service Provider (CSP). Infact, the subscriber is free of data center maintenance and management concerns. Though, centralized and geographically dispersed placement of cloud infrastructure is raising concerns for the IoT paradigm and 5 G-related novel applications as specified in Section 2.1.2. Thus, MEC or any other edge computing paradigm is envisaged to overcome these issues of cloud computing by shifting the data center infrastructure to a proximate location. Figure 2.3 depicts the discussed evolution of MEC in an illustrative context.
Figure 2.3 Evolution of MEC.
2.2 MEC Reference Architecture
ETSI has defined the reference architecture for MEC, which is illustrated in Figure 2.4[14]. This architecture is formed with the idea of leveraging the deployed MEC infrastructure to launch diverse services in an autonomous manner. The aspects of subscriber access through the mobile network, user access management, virtualization technologies for launching service instances, proper monitoring and isolation of virtual entities for convenient service handling and delivery, core and edge network segregation for better integration with prevailing mobile network layout, and orchestration of the entire MEC serviceable platform are considered in formulating this architecture. Though there is an updated version of the MEC architecture being published by pioneer research groups/institutions on integrating Network Function Virtualization (NFV) and 5G use cases, the version in Figure 2.4 depicts the fundamental formation of a MEC deployment. Conversely, NFV is a concept introduced for decoupling network functions and services from the hardware apparatus for moving them to the softwarized domain; where softwarized instances called Virtualized Network Functions (VNFs) are acting as programmable network functions that can execute independent of the hardware dependencies [13]. MEC reference architecture mainly has two levels: Mobile Edge System Level and Mobile Edge Host (MEH) Level. This segregation is allowing the MEC service providers to link the MEC system-level entities with the 5G core network entities to merge the two networks as a singular one.
Figure 2.4 MEC Reference Architecture.
2.2.1 Mobile Edge System Level
The MEC system level is the governing domain of the MEC system, where several MEC edge levels can be administered through a single system level. However, the governing capability of the MEC system level will depend on its resources and the scope and nature of the MNO domain. The 5G core network entities are tethered to the system level of the MEC (i.e., through mobile edge orchestrator [MEO]) for traffic steering and mobility management purposes [15].
2.2.1.1 User Equipment (UE) and Applications
User equipments (UEs) are any kind of device that provides an interface for MEC subscribers to interact with the MEC service framework. These UEs can be varied from a wearable IoT apparatus to UAVs, to an autonomous vehicle that might operate with human interaction or vice versa. The significance of the UEs are that they should be compatible with the communication access medium of the MEC system, either via 3GPP or non-3GPP means. 3GPP extends from Global System for Mobile Communication (GSM) or 2G to 5G New Radio (NR) Radio Access Technologies (RATs); where non-3GPP technologies are Worldwide Interoperability for Microwave Access (WiMAX), Code Division Multiple Access (CDMA), Wireless Local Area Networks (WLANs), or fixed networks such as Public Switched Telephone Networks (PSTNs). We can visualize that the UE Applications (UE Apps) running on the UE, interacts with the Mobile Edge Applications (ME Apps) launched at the MEC edge level. In fact, UE Apps are mapped with the ME Apps for entities operating at the access network and the mobile edge network, respectively. However, UE Apps are only capable of interacting with the mobile edge system through a user application life-cycle management proxy.
2.2.1.2 Customer Facing Service Portal (CFSP)
Customer Facing Services (CFSs) are a type of service category where the service and its specifications are exposed to the customer for perceiving a commercial view in an e-trading domain [16]. In contrast, Resource Facing Services (RFSs) are invincible to the customers because they act as internal and indirect processes even to the service providers. In the context of MEC and 5G, high-end service providers (i.e., automobile companies: autonomous vehicles, gaming companies: AR-based games, online trading and delivery companies: UAV based delivering) might leverage the resources in the MEC environment for launching their services. In such a scenario, the service is not offered by the MEC service provider, but by a third-party institution. Thus, for maintaining transparency in this third-party trading domain, exposing the service specifications and consensus regarding the service conduct (i.e., Service Level Agreements—SLAs) is a profound necessity from the customers’ point of view. Hence, such services can be categorized as CFSs. In fact, customer facing service portal (CFSP) is an entity or an interface operated by the MEC operator for facilitating or outsourcing the MEC platform for third-party services, so that customers can directly negotiate the service delivery with the third-party service provider. CFSP is capable of recalling SLA information from the provisioned applications.
2.2.1.3 Operation Support System (OSS)
The operation support system (OSS) is the main access registry of the holistic MEC system. The service requests forwarded via the CFSP or UE application life-cycle management proxy are evaluated for approval of access. Generally, OSS entities are responsible for monitoring, controlling, analyzing, and managing network-type services. In the MEC context, OSS decides whether the user requests are granted access or not. The granted requests are directed to the MEC orchestrator for further processing. Though, OSS decides the instigation and termination states of any ME App registered under it. In addition, OSS is tasked with service or application discovery and relocation functions among the MEC system or any external clouds. Thus, this entity has a record of all the MEC applications registered under the MEC operator in addition to their statistics, while UE App to ME App mappings are envisaged to be conducted here. In fact, SLA negotiations can take place during the OSS process phases where a global perception of the SLA specifications, standardization (i.e., in terms of Quality of Service—QoS, or Quality of Experience—QoE), and measurement can be embedded into the OSS constructs due to its access to both UE App and ME App perspectives.
2.2.1.4 User Application Lifecycle Management Proxy (UALCMP)
This proxy is only accessible within the mobile network. The UE App requests are directed to this entity. The handling of multiple UEs for their application requests are pursued by this component. As the OSS is the access authority from the MEC perspective, user application lifecycle management proxy (UALCMP) is the access portal of the access network. In fact, UALCMP decides the instigation and termination states (i.e., life-cycle) of the UE Apps. Since it is a proxy, anonymization and user ID handling are probable functions. Further, authentication and authorization handling of the MEC system is pursued by this entity, which makes this ingress point a prominent location for cyber intruders.
2.2.1.5 Mobile Edge Orchestrator (MEO)
This component carries out the core functionality of the MEC architecture. Orchestration is the cardinal function of any virtualized deployment. It is in fact the function of automated configuration, coordination, and management of the entire softwarized system. Orchestration in cloud computing does not require a higher level of automation due to its centralized administrator involvement from the cloud service providers. On the contrary, automation is an utmost requirement for the edge computing paradigms, which eases the burden of the MEC operator. The emphasis on automation improves the responsiveness of the edge paradigms to meet the requirements of the emerging 5G and IoT applications. In the context of MEC, the orchestrator is responsible for delivering automated service provisioning, service mobility handling, interoperability among converged networks, and service optimization [13]. The granted requests from the OSS are forwarded for processing in the MEO. MEO maintains a holistic view of the MEC system focusing on: deployed MEC hosts at each edge level under its governance, access capacity of the system level, edge-level resource utilization, network utilization, available MEC services and applications, protocols, and topology. Being the main orchestrator, MEO performs the monitoring functions and constructs leading to system updates. MEO is responsible for the below functionalities;
Monitoring and Verifying
: Governing the virtualization infrastructure manager for application handling while checking the integrity, authenticity of the packages and validating application rules and requirements.
Service Mapping
: Selecting the appropriate MEHs for requested services considering service level requirements of latency, resources, and access capabilities.
Service Handling
: Triggering instantiation and termination of the applications.
Application relocation.
2.2.2 Mobile Edge Host Level
The host level or the Edge level of the MEC system is the action domain of the MEC system. This dynamic environment, envisaged to be deployed via virtualization technologies is the actual EDGE part of the MEC concept. As there can be many edge levels governed under a system level, redundancy of the system-level entities is a challenge for the MEC concept. Further, the backhaul channels connecting the edge and system-level entities should be secured beyond the capabilities of resourceful cyber intruders. This level mainly consists of Mobile Edge Hosts (MEH), Mobile Edge Platform Managers (MEPM), and the Virtualization Infrastructure Manager (VIM).
2.2.2.1 Mobile Edge Host (MEH)
This is an entity that contains the corresponding ME Apps and the Mobile Edge Platform (MEP) on top of the underlying Virtualization Infrastructure (VI). In fact, MEH is operating on the data plane of the 5G mobile network, where it is a packet-based transmission platform that features a common encapsulation compatible with both fronthaul and backhaul traffic domains. Thus, access to the data plane guarantees direct connectivity for ME Apps towards the UE domain, in addition to the Internet. MEH mainly provides computing, storage, and networking resources to execute the ME Apps. In recent research advancements, MEHs are launched as Virtual Machines (VMs) that inherit formidable levels of resources and are specified for different service types [17].
Mobile Edge Applications (ME Apps) They are executed as software-only entities in the MEH. Different ME Apps are launched as service instances within the MEH to service UE Apps at the subscriber end. These ME Apps are to be pre-configured and hibernated within the MEH and dispense once the requirement arises with the UE requests. As the MEHs are envisaged to be launched as VMs, ME Apps can be deployed as lightweight virtualized entities such as docker containers [18]. The underlying virtualization infrastructure can be formed by a hybrid approach of integrating hypervisor-based and lightweight virtualization technologies. ME Apps should be optimized for resource utilization.
Mobile Edge Platform (MEP) Dispense functionality to run the ME Apps while facilitating the essential environment to discover, advertise, and consume mobile edge services. The traffic rules obtained from the MEPM are notified to the data plane by the MEP as the traffic rules controller, and the IP address translation is conducted for UE tokens. MEP Handles the Domain Name System (DNS) proxy/server according to the records conveyed from the MEPM. In a scenario that requires the interconnection of two or more MEHs (i.e., for a high-end mobile edge service that requires more than one MEH to implement the complete service function chain), the internal connections between the MEHs are established via the MEPs. In addition, the service registry element within the MEP maps the ME Apps to the mobile edge service IDs and triggers the internal resource allocations at the VI.
Virtualization Infrastructure (VI) Links are drawn towards the VIM and the MEP. This includes the data plane which executes the traffic rules set notified by the MEP and routes the traffic through ME Apps, DNS server/proxy, 3GPP network, local, and external networks. The most important task of the VI is to provision the intrinsic virtual resources to the ME Apps while monitoring their utilization. VI is notifying the VIM of all the states regarding virtual resources while corresponding resource allocations are conducted with the knowledge of the service registry in the MEP.
2.2.2.2 Mobile Edge Platform Manager (MEPM)
MEPM can be considered as the orchestrator at the edge level, where its main task is to notify the MEO regarding the status of the MEC edge platform. The function of the MEPM resembles the orchestration function in NFV (i.e., NFV Orchestrator—NFVO), where ME Apps or MEHs can be considered mapped entities to VNFs depending on their application deployment. The main functions of the MEPM can be listed below;
Govern ME App life-cycle while informing the MEO of monitoring stats of each application.
Element management provided for all the entities in the MEC edge level.
Managing the application rules and requirements while performing the tasks of service authorization, traffic rules, DNS configuration, and resolving conflicts.
Record the fault reports and performance measurements sent from the VIM.
2.2.2.3 Virtualization Infrastructure Manager (VIM)
VIM is responsible for allocating, managing, and releasing visualized resources (compute, storage, and networking) of the holistic virtualization infrastructure at the MEC edge level. In fact, VIM is the hypervisor of the MEC edge platform, that is tasked with handling and managing the virtual resources. VIM is tasked to properly measure the resource consumption in terms of memory, processing, and networking; and to benchmark the general operation of the MEC edge platform for the sake of detecting any anomalous behavior in a compromised scenario. The functions of the VIM can be summarized as:
