Wireless Blockchain Principles Technologies and Applications 1st Edition Bin Cao Lei Zhang Mugen Peng Muhammad Ali Imran Eds

Wireless Blockchain Principles
Technologies and Applications 1st

Edition Bin Cao Lei Zhang Mugen Peng
Muhammad Ali Imran Eds
Visit to download the full and correct content document:
https://ebookmeta.com/product/wireless-blockchain-principles-technologies-and-appli
cations-1st-edition-bin-cao-lei-zhang-mugen-peng-muhammad-ali-imran-eds/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
Wireless Blockchain Principles Technologies and

Applications 1st Edition Bin Cao (Editor)
https://ebookmeta.com/product/wireless-blockchain-principles-
technologies-and-applications-1st-edition-bin-cao-editor/
Autonomous Airborne Wireless Networks (IEEE Press) 1st

Edition Muhammad Ali Imran (Editor)
https://ebookmeta.com/product/autonomous-airborne-wireless-
networks-ieee-press-1st-edition-muhammad-ali-imran-editor/
Eco-Friendly Corrosion Inhibitors: Principles,

Designing and Applications 1st Edition Lei Guo
https://ebookmeta.com/product/eco-friendly-corrosion-inhibitors-
principles-designing-and-applications-1st-edition-lei-guo/
Networked Microgrids 1st Edition Peng Zhang
https://ebookmeta.com/product/networked-microgrids-1st-edition-
peng-zhang/
Blockchain Consensus: An Introduction to Classical,
Blockchain, and Quantum Consensus Protocols 1st Edition
Imran Bashir
https://ebookmeta.com/product/blockchain-consensus-an-
introduction-to-classical-blockchain-and-quantum-consensus-
protocols-1st-edition-imran-bashir/
Wireless Communication with Artificial Intelligence:

Emerging Trends and Applications (Wireless
Communications and Networking Technologies) 1st Edition
Anuj Singal (Editor)
https://ebookmeta.com/product/wireless-communication-with-
artificial-intelligence-emerging-trends-and-applications-
wireless-communications-and-networking-technologies-1st-edition-
anuj-singal-editor/
3-D Computer Vision: Principles, Algorithms and

Applications 1st Edition Yu-Jin Zhang
https://ebookmeta.com/product/3-d-computer-vision-principles-
algorithms-and-applications-1st-edition-yu-jin-zhang/
Molecular Characterization of Polymers: A Fundamental

Guide 1st Edition Muhammad Imran Malik
https://ebookmeta.com/product/molecular-characterization-of-
polymers-a-fundamental-guide-1st-edition-muhammad-imran-malik/
Wireless Communication: Advancements and Challenges

(Wireless Communications and Networking Technologies)
1st Edition Prashant Ranjan
https://ebookmeta.com/product/wireless-communication-
advancements-and-challenges-wireless-communications-and-
networking-technologies-1st-edition-prashant-ranjan/
Wireless Blockchain
Wireless Blockchain
Principles, Technologies and Applications
Edited by
Bin Cao
Beijing University of Posts and Telecommunications
Beijing, China
Lei Zhang
University of Glasgow
Glasgow, UK
Mugen Peng
Beijing University of Posts and Telecommunications
Beijing, China
Muhammad Ali Imran

University of Glasgow
Glasgow, UK
Copyright © 2022 by John Wiley & Sons Ltd. All rights reserved.
Published by John Wiley & Sons Ltd., Chichester, United Kingdom.

Published simultaneously in Canada.
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, scanning, or otherwise, except as permitted
under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written
permission of the Publisher, or authorization through payment of the appropriate per-copy fee to the
Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, (978) 750-8400, fax (978)
750-4470, or on the web at www.copyright.com. Requests to the Publisher for permission should be
addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 07030,
(201) 748-6011, fax (201) 748-6008, or online at http://www.wiley.com/go/permission.
Limit of Liability/Disclaimer of Warranty: 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.
For general information on our other products and services or for technical support, please contact our
Customer Care Department within the United States at (800) 762-2974, outside the United States at (317)
572-3993 or fax (317) 572-4002.
Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not
be available in electronic formats. For more information about Wiley products, visit our web site at
www.wiley.com.
Library of Congress Cataloging-in-Publication Data
Names: Cao, Bin, editor. | Zhang, Lei, editor. | Peng, Mugen, editor. |
Imran, Muhammad Ali, editor.
Title: Wireless blockchain : principles, technologies and applications /
Bin Cao, Beijing University of Posts and Telecommunications, Beijing,
China, Lei Zhang, University of Glasgow, Glasgow, UK, Mugen Peng,
Beijing University of Posts and Telecommunications, Beijing, China,
Muhammad Ali Imran, University of Glasgow, Glasgow, UK.
Description: Chichester, United Kingdom ; Hoboken : Wiley-IEEE Press,
[2022] | Includes bibliographical references and index.
Identifiers: LCCN 2021034990 (print) | LCCN 2021034991 (ebook) | ISBN
9781119790808 (cloth) | ISBN 9781119790815 (adobe pdf) | ISBN
9781119790822 (epub)
Subjects: LCSH: Blockchains (Databases) | Wireless communication
systems–Industrial applications. | Personal communication service
systems.
Classification: LCC QA76.9.B56 W57 2022 (print) | LCC QA76.9.B56 (ebook)
| DDC 005.74–dc23
LC record available at https://lccn.loc.gov/2021034990
LC ebook record available at https://lccn.loc.gov/2021034991
Cover Design: Wiley

Cover Image: © phive/Shutterstock
Set in 9.5/12.5pt STIXTwoText by Straive, Chennai, India
10 9 8 7 6 5 4 3 2 1
v
Contents
List of Contributors xiii

Preface xvii
Abbreviations xxiii
1 What is Blockchain Radio Access Network? 1

Xintong Ling, Yuwei Le, Jiaheng Wang, Zhi Ding, and Xiqi Gao
1.1 Introduction 1
1.2 What is B-RAN 3
1.2.1 B-RAN Framework 3
1.2.2 Consensus Mechanism 6
1.2.3 Implementation 6
1.3 Mining Model 7
1.3.1 Hash-Based Mining 7
1.3.2 Modeling of Hash Trials 7
1.3.3 Threat Model 10
1.4 B-RAN Queuing Model 10
1.5 Latency Analysis of B-RAN 12
1.5.1 Steady-State Analysis 12
1.5.2 Average Service Latency 16
1.6 Security Considerations 18
1.6.1 Alternative History Attack 18
1.6.2 Probability of a Successful Attack 19
1.7 Latency-Security Trade-off 20
1.8 Conclusions and Future Works 22
1.8.1 Network Effect and Congest Effect 22
1.8.2 Chicken and Eggs 22
1.8.3 Decentralization and Centralization 22
1.8.4 Beyond Bitcoin Blockchain 22
References 23
2 Consensus Algorithm Analysis in Blockchain: PoW and Raft 27

Taotao Wang, Dongyan Huang, and Shengli Zhang
2.1 Introduction 27
2.2 Mining Strategy Analysis for the PoW Consensus-Based Blockchain 30
vi Contents
2.2.1 Blockchain Preliminaries 30

2.2.2 Proof of Work and Mining 30
2.2.3 Honest Mining Strategy 31
2.2.4 PoW Blockchain Mining Model 32
2.2.4.1 State 33
2.2.4.2 Action 33
2.2.4.3 Transition and Reward 34
2.2.4.4 Objective Function 39
2.2.4.5 Honest Mining 40
2.2.4.6 Selfish Mining 40
2.2.4.7 Lead Stubborn Mining 40
2.2.4.8 Optimal Mining 41
2.2.5 Mining Through RL 41
2.2.5.1 Preliminaries for Original Reinforcement Learning Algorithm 41
2.2.5.2 New Reinforcement Learning Algorithm for Mining 42
2.2.6 Performance Evaluations 44
2.3 Performance Analysis of the Raft Consensus Algorithm 52
2.3.1 Review of Raft Algorithm 52
2.3.2 System Model 53
2.3.3 Network Model 53
2.3.4 Network Split Probability 55
2.3.5 Average Number of Replies 57
2.3.6 Expected Number of Received Heartbeats for a Follower 57
2.3.7 Time to Transition to Candidate 58
2.3.8 Time to Elect a New Leader 59
2.3.9 Simulation Results 60
2.3.10 Discussion 67
2.3.10.1 Extended Model 67
2.3.10.2 System Availability and Consensus Efficiency 68
2.4 Conclusion 69
Appendix A.2 69
References 70
3 A Low Communication Complexity Double-layer PBFT Consensus 73

Chenglin Feng, Wenyu Li, Bowen Yang, Yao Sun, and Lei Zhang
3.1 Introduction 73
3.1.1 PBFT Applied to Blockchain 74
3.1.2 From CFT to BFT 74
3.1.2.1 State Machine Replication 74
3.1.2.2 Primary Copy 75
3.1.2.3 Quorum Voting 75
3.1.3 Byzantine Generals Problem 76
3.1.4 Byzantine Consensus Protocols 76
3.1.4.1 Two-Phase Commit 76
3.1.4.2 View Stamp 76
Contents vii
3.1.4.3 PBFT Protocol 76

3.1.5 Motivations 78
3.1.6 Chapter Organizations 78
3.2 Double-Layer PBFT-Based Protocol 79
3.2.1 Consensus Flow 79
3.2.1.1 The Client 79
3.2.1.2 First-Layer Protocol 81
3.2.1.3 Second-Layer Protocol 81
3.2.2 Faulty Primary Elimination 82
3.2.2.1 Faulty Primary Detection 82
3.2.2.2 View Change 83
3.2.3 Garbage Cleaning 84
3.3 Communication Reduction 84
3.3.1 Operation Synchronization 85
3.3.2 Safety and Liveness 85
3.4 Communication Complexity of Double-Layer PBFT 85
3.5 Security Threshold Analysis 86
3.5.1 Faulty Probability Determined 87
3.5.2 Faulty Number Determined 89
3.6 Conclusion 90
References 90
4 Blockchain-Driven Internet of Things 93

Bin Cao, Weikang Liu, and Mugen Peng
4.1 Introduction 93
4.1.1 Challenges and Issues in IoT 93
4.1.2 Advantages of Blockchain for IoT 94
4.1.3 Integration of IoT and Blockchain 94
4.2 Consensus Mechanism in Blockchain 96
4.2.1 PoW 96
4.2.2 PoS 97
4.2.3 Limitations of PoW and PoS for IoT 98
4.2.3.1 Resource Consumption 98
4.2.3.2 Transaction Fee 98
4.2.3.3 Throughput Limitation 98
4.2.3.4 Confirmation Delay 98
4.2.4 PBFT 98
4.2.5 DAG 100
4.2.5.1 Tangle 101
4.2.5.2 Hashgraph 102
4.3 Applications of Blockchain in IoT 102
4.3.1 Supply Chain 102
4.3.1.1 Introduction 102
4.3.1.2 Modified Blockchain 103
4.3.1.3 Integrated Architecture 104
viii Contents
4.3.1.4 Security Analysis 105

4.3.2 Smart City 106
4.3.2.1 Introduction 106
4.3.2.2 Smart Contract System 107
4.3.2.3 Main Functions of the Framework 109
4.3.2.4 Discussion 110
4.4 Issues and Challenges of Blockchain in IoT 111
4.4.1 Resource Constraints 111
4.4.2 Security Vulnerability 111
4.4.3 Privacy Leakage 112
4.4.4 Incentive Mechanism 112
4.5 Conclusion 112
References 112
5 Hyperledger Blockchain-Based Distributed Marketplaces for

5G Networks 117
Nima Afraz, Marco Ruffini, and Hamed Ahmadi
5.1 Introduction 117
5.2 Marketplaces in Telecommunications 118
5.2.1 Wireless Spectrum Allocation 119
5.2.2 Network Slicing 119
5.2.3 Passive optical networks (PON) Sharing 120
5.2.4 Enterprise Blockchain: Hyperledger Fabric 121
5.2.4.1 Shared Ledger 122
5.2.4.2 Organizations 122
5.2.4.3 Consensus Protocol 122
5.2.4.4 Network Peers 122
5.2.4.5 Smart Contracts (chaincodes) 123
5.2.4.6 Channels 123
5.3 Distributed Resource Sharing Market 123
5.3.1 Market Mechanism (Auction) 125
5.3.2 Preliminaries 125
5.4 Experimental Design and Results 126
5.4.1 Experimental Blockchain Deployment 127
5.4.1.1 Cloud Infrastructure 127
5.4.1.2 Container Orchestration: Docker Swarm 127
5.4.2 Blockchain Performance Evaluation 127
5.4.3 Benchmark Apparatus 128
5.4.3.1 Hyperledger Caliper 130
5.4.3.2 Data Collection: Prometheus Monitor 130
5.4.4 Experimental Results 131
5.4.4.1 Maximum Transaction Throughput 131
5.4.4.2 Block Size 131
5.4.4.3 Network Size 131
5.5 Conclusions 133
References 133
Contents ix
6 Blockchain for Spectrum Management in 6G Networks 137

Asuquo A. Okon, Olusegun S. Sholiyi, Jaafar M. H. Elmirghani, and Kumudu
Munasighe
6.2 Background 139
6.2.1 Rise of Micro-operators 139
6.2.2 Case for Novel Spectrum Sharing Models 140
6.2.2.1 Blockchain for Spectrum Sharing 141
6.2.2.2 Blockchain in 6G Networks 142
6.3 Architecture of an Integrated SDN and Blockchain Model 143
6.3.1 SDN Platform Design 143
6.3.2 Blockchain Network Layer Design 144
6.3.3 Network Operation and Spectrum Management 146
6.4 Simulation Design 149
6.5 Results and Analysis 152
6.5.1 Radio Access Network and Throughput 152
6.5.2 Blockchain Performance 154
6.5.3 Blockchain Scalability Performance 155
6.6 Conclusion 156
Acknowledgments 156
References 157
7 Integration of MEC and Blockchain 161

Bin Cao, Weikang Liu, and Mugen Peng
7.2 Typical Framework 162
7.2.1 Blockchain-Enabled MEC 162
7.2.1.1 Background 162
7.2.1.2 Framework Description 162
7.2.2 MEC-Based Blockchain 164
7.2.2.2 Framework Description 164
7.3 Use Cases 166
7.3.1 Security Federated Learning via MEC-Enabled Blockchain Network 166
7.3.1.2 Blockchain-Driven Federated Learning 167
7.3.1.3 Experimental Results 168
7.3.2 Blockchain-Assisted Secure Authentication for Cross-Domain Industrial
IoT 170
7.3.2.2 Blockchain-Driven Cross-Domain Authentication 170
7.3.2.3 Experimental Results 172
7.4 Conclusion 174
References 174
x Contents
8 Performance Analysis on Wireless Blockchain IoT System 179

Yao Sun, Lei Zhang, Paulo Klaine, Bin Cao, and Muhammad Ali Imran
8.2 System Model 181
8.2.1 Blockchain-Enabled IoT Network Model 181
8.2.2 Wireless Communication Model 183
8.3 Performance Analysis in Blockchain-Enabled Wireless IoT Networks 184
8.3.1 Probability Density Function of SINR 185
8.3.2 TDP Transmission Successful Rate 187
8.3.3 Overall Communication Throughput 189
8.4 Optimal FN Deployment 189
8.5 Security Performance Analysis 190
8.5.1 Eclipse Attacks 190
8.5.2 Random Link Attacks 192
8.5.3 Random FN Attacks 192
8.6 Numerical Results and Discussion 192
8.6.1 Simulation Settings 193
8.6.2 Performance Evaluation without Attacks 193
8.7 Chapter Summary 197
References 197
9 Utilizing Blockchain as a Citizen-Utility for Future Smart Grids 201

Samuel Karumba, Volkan Dedeoglu, Ali Dorri, Raja Jurdak, and Salil S. Kanhere
9.2 DET Using Citizen-Utilities 204
9.2.1 Prosumer Community Groups 204
9.2.1.1 Microgrids 205
9.2.1.2 Virtual Power Plants (VPP) 206
9.2.1.3 Vehicular Energy Networks (VEN) 206
9.2.2 Demand Side Management 207
9.2.2.1 Energy Efficiency 208
9.2.2.2 Demand Response 209
9.2.2.3 Spinning Reserves 210
9.2.3 Open Research Challenges 211
9.2.3.1 Scalability and IoT Overhead Issues 211
9.2.3.2 Privacy Leakage Issues 212
9.2.3.3 Standardization and Interoperability Issues 212
9.3 Improved Citizen-Utilities 213
9.3.1 Toward Scalable Citizen-Utilities 213
9.3.1.1 Challenges 213
9.3.1.2 HARB Framework-Based Citizen-Utility 214
9.3.2 Toward Privacy-Preserving Citizen-Utilities 216
9.3.2.1 Threat Model 217
9.3.2.2 PDCH System 219
9.4 Conclusions 220
References 221
Contents xi
10 Blockchain-enabled COVID-19 Contact Tracing Solutions 225

Hong Kang, Zaixin Zhang, Junyi Dong, Hao Xu, Paulo Valente Klaine, and Lei Zhang
10.2 Preliminaries of BeepTrace 228
10.2.1 Motivation 228
10.2.1.1 Comprehensive Privacy Protection 229
10.2.1.2 Performance is Uncompromising 229
10.2.1.3 Broad Community Participation 229
10.2.1.4 Inclusiveness and Openness 230
10.2.2 Two Implementations are Based on Different Matching Protocols 230
10.3 Modes of BeepTrace 231
10.3.1 BeepTrace-Active 231
10.3.1.1 Active Mode Workflow 231
10.3.1.2 Privacy Protection of BeepTrace-Active 232
10.3.2 BeepTrace-Passive 233
10.3.2.1 Two-Chain Architecture and Workflow 233
10.3.2.2 Privacy Protection in BeepTrace-Passive 235
10.4 Future Opportunity and Conclusions 237
10.4.1 Preliminary Approach 237
10.4.2 Future Directions 238
10.4.2.1 Network Throughput and Scalability 238
10.4.2.2 Technology for Elders and Minors 239
10.4.2.3 Battery Drainage and Storage Optimization 240
10.4.2.4 Social and Economic Aspects 240
10.4.3 Concluding Remarks 240
References 241
11 Blockchain Medical Data Sharing 245

Qi Xia, Jianbin Gao, and Sandro Amofa
11.1.1 General Overview 248
11.1.2 Defining Challenges 248
11.1.2.1 Data Security 248
11.1.2.2 Data Privacy 248
11.1.2.3 Source Identity 248
11.1.2.4 Data Utility 249
11.1.2.5 Data Interoperability 249
11.1.2.6 Trust 249
11.1.2.7 Data Provenance 249
11.1.2.8 Authenticity 250
11.1.3 Sharing Paradigms 250
11.1.3.1 Institution-to-Institution Data Sharing 251
11.1.3.2 Patient-to-Institution Data Sharing 256
11.1.3.3 Patient-to-Patient Data Sharing 257
11.1.4 Special Use Cases 260
xii Contents
11.1.4.1 Precision Medicine 261

11.1.4.2 Monetization of Medical Data 263
11.1.4.3 Patient Record Regeneration 264
11.1.5 Conclusion 266
Acknowledgments 266
References 266
12 Decentralized Content Vetting in Social Network with

Blockchain 269
Subhasis Thakur and John G. Breslin
12.2 Related Literature 270
12.3 Content Propagation Models in Social Network 271
12.4 Content Vetting with Blockchains 273
12.4.1 Overview of the Solution 273
12.4.2 Unidirectional Offline Channel 273
12.4.3 Content Vetting with Blockchains 275
12.5 Optimized Channel Networks 278
12.6 Simulations of Content Propagation 280
12.7 Evaluation with Simulations of Social Network 286
12.8 Conclusion 293
Acknowledgment 293
References 294
Index 297
xiii
List of Contributors
Nima Afraz Bin Cao

CONNECT Center, Trinity College State Key Laboratory of Networking and
Dublin Switching Technology, Beijing University
Ireland of Posts and Telecommunications
Beijing
and China
School of Computer Science
University College Dublin Volkan Dedeoglu
Dublin Data61, CSIRO
Ireland Brisbane
Australia
Hamed Ahmadi
Department of Electronic Engineering Zhi Ding
University of York Department of Electrical and Computer
York Engineering, University of California
UK Davis, CA
USA
Sandro Amofa
University of Electronic Science and Junyi Dong
Technology of China James Watt School of Engineering
Chengdu University of Glasgow
China Glasgow
UK
John G. Breslin
National University of Ireland Ali Dorri
Galway School of Computer Science, QUT
Ireland Brisbane
Australia
xiv List of Contributors
Jaafar M. H. Elmirghani Raja Jurdak

School of Electronic and Electrical School of Computer Science, QUT
Engineering, University of Leeds Brisbane
Leeds Australia
UK
and
Chenglin Feng Data61, CSIRO
College of Science and Engineering Brisbane
University of Glasgow Australia
Glasgow
UK Hong Kang
James Watt School of Engineering
Jianbin Gao University of Glasgow
University of Electronic Science and Glasgow
Technology of China UK
Chengdu
China Salil S. Kanhere
School of Computer Science and
Xiqi Gao Engineering, UNSW
National Mobile Communications Sydney
Research Laboratory, Southeast University Australia
Nanjing
China Samuel Karumba
and School of Computer Science and
Engineering, UNSW
Purple Mountain Laboratories Sydney
Nanjing, Jiangsu Australia
China
Paulo Valente Klaine
Dongyan Huang James Watt School of Engineering
College of Information and University of Glasgow
Communications, Guilin University of Glasgow
Electronic Technology UK
Guilin
China Yuwei Le
National Mobile Communications
Muhammad Ali Imran Research Laboratory, Southeast University
James Watt School of Engineering Nanjing
University of Glasgow China
Glasgow
UK
List of Contributors xv
Wenyu Li Marco Ruffini

College of Science and Engineering CONNECT Center, Trinity College
University of Glasgow Dublin
Glasgow Ireland
UK
Olusegun S. Sholiyi
Xintong Ling National Space Research and Development
National Mobile Communications Agency, Obasanjo Space Centre
Research Laboratory, Southeast University Abuja
Nanjing Nigeria
China
Yao Sun
and James Watt School of Engineering
Purple Mountain Laboratories College of Science and Engineering
Nanjing, Jiangsu University of Glasgow
China Glasgow
UK
Weikang Liu
State Key Laboratory of Networking and Subhasis Thakur
Switching Technology, Beijing University National University of Ireland
of Posts and Telecommunications Galway
Beijing Ireland
China
Jiaheng Wang
Kumudu Munasighe National Mobile Communications
Faculty of Science and Technology Research Laboratory, Southeast University
University of Canberra Nanjing
Canberra China
Australia and
Asuquo A. Okon Purple Mountain Laboratories

Faculty of Science and Technology Nanjing, Jiangsu
University of Canberra China
Canberra
Australia Taotao Wang
College of Electronics and Information
Mugen Peng Engineering, Shenzhen University
State Key Laboratory of Networking and Shenzhen
Switching Technology, Beijing University China
of Posts and Telecommunications
Beijing
China
xvi List of Contributors
Qi Xia Lei Zhang

University of Electronic Science and James Watt School of Engineering
Technology of China College of Science and Engineering
Chengdu University of Glasgow
China Glasgow
UK
Hao Xu
James Watt School of Engineering Shengli Zhang
University of Glasgow College of Electronics and Information
Glasgow Engineering, Shenzhen University
UK Shenzhen
China
Bowen Yang
James Watt School of Engineering Zaixin Zhang
College of Science and Engineering James Watt School of Engineering
University of Glasgow University of Glasgow
Glasgow Glasgow
UK UK
xvii
Preface
Originally proposed as the backbone technology of Bitcoin, Ethereum, and many other
cryptocurrencies, blockchain has become a revolutionary decentralized data management
framework that establishes consensuses and agreements in trustless and distributed envi-
ronments. Thus, in addition to its soaring popularity in the finance sector, blockchain has
attracted much attention from many other major industrial sectors ranging from supply
chain, transportation, entertainment, retail, healthcare, information management to finan-
cial services, etc.
Essentially, blockchain is built on a physical network that relies on the communications,
computing, and caching, which serves the basis for blockchain functions such as incentive
mechanism or consensus. As such, blockchain systems can be depicted as a two-tier archi-
tecture: an infrastructure layer and a blockchain layer. The infrastructure layer is the under-
lying entity responsible for maintaining the P2P network, building connection through
wired/wireless communication, and computing and storing data. On the other hand, the
top layer is the blockchain that is responsible for trust and security functions based on
the underlying exchange of information. More specifically, blockchain features several key
components that are summarized as transactions, blocks, and the chain of blocks. Transac-
tions contain the information requested by the client and need to be recorded by the public
ledger; blocks securely record a number of transactions or other useful information; using
a consensus mechanism, blocks are linked orderly to constitute a chain of blocks, which
indicates logical relation among the blocks to construct the blockchain.
As a core function of the blockchain, the consensus mechanism (CM, also referred to as
consensus algorithm or consensus protocol) works in the blockchain layer in order to ensure
a clear sequence of transactions and the integrity and consistency of the blockchain across
geographically distributed nodes. The CM largely determines the blockchain system perfor-
mance in terms of security level (fault tolerance level), transaction throughput, delay, and
node scalability. Depending on application scenarios and performance requirements, differ-
ent CMs can be used. In a permissionless public chain, nodes are allowed to join/leave the
network without permission and authentication. Therefore, proof-based algorithms (PoX),
such as proof-of-work (PoW), proof-of-stake (PoS), and their variants, are commonly used
in many public blockchain applications (e.g. Bitcoin and Ethereum). PoX algorithms are
designed with excellent node scalability performance through node competition; however,
they could be very resource demanding. Also, these CMs have other limitations such as
long transaction confirmation latency and low throughput. Unlike public chains, private
xviii Preface
and consortium blockchains prefer to adopt lighter protocols such as Raft and practical
Byzantine fault tolerance (PBFT) to reduce computational power demand and improve
the transaction throughput. A well-known example of PBFT implementation is the Hyper-
Ledger Fabric, part of HyperLedger business blockchain frameworks. However, such CMs
may require heavy communication resources.
Today, most state-of-the-art blockchains are primarily designed in stable wired commu-
nication networks running in advanced devices with sufficient communication resource
provision. Hence, the blockchain performance degradation caused by communication is
negligible. Nevertheless, this is not the case for the highly dynamic wireless connected
digital society that is mainly composed of massive wireless devices encompassing finance,
supply chain, healthcare, transportation, and energy. Especially through the upcoming 5G
network, the majority of valuable information exchange may be through a wireless medium.
Thus, it is critically important to answer one question, how much communication resource
is needed to run a blockchain network (i.e. communication for blockchain).
From another equally important aspect when combining blockchain with communi-
cation (especially wireless communication), many works have focused on how to use
blockchain to improve the communication network performance (i.e. blockchain for
communication). This integration between wireless networks and blockchain allows the
network to monitor and manage communication resource utilization in a more efficient
manner, reducing its administration costs and improving the speed of communication
resource trading. In addition, because it is the blockchain’s inherit transparency, it can
also record real-time spectrum utilization and massively improve spectrum efficiency by
dynamically allocating spectrum bands according to the dynamic demands of devices.
Moreover, it can also provide the necessary incentive for spectrum and resource sharing
between devices, fully enabling new technologies and services that are bound to emerge.
The resource coordination and optimization between resource requesters and providers
can be automatically completed through smart contracts, thus improving the efficiency of
resource optimization. Furthermore, with future wireless networks shifting toward decen-
tralized solutions, with thousands of mobile cells deployed by operators and billions of
devices communicating with each other, fixed spectrum allocation and operator-controlled
resource sharing algorithms will not be scalable nor effective in future networks. As such,
by designing a communications network coupled with blockchain as its underlying infras-
tructure from the beginning, the networks can be more scalable and provide better and more
efficient solutions in terms of spectrum sharing and resource optimization, for example.
The book falls under a broad category of security and communication network and their
transformation and development, which itself is a very hot topic for research these years.
The book is written in such a way that it offers a wide range of benefits to the scientific
community: while beginners can learn about blockchain technologies, experienced
researchers and scientists can understand the extensive theoretical design and architecture
development of blockchain, and industrial experts can learn about various perspectives
of application-driven blockchains to facilitate different vertical sectors. Therefore, this
feature topic can attract graduate/undergraduate level students, as well as researchers and
leading experts from both academia and industry. In particular, some blockchain-enabled
use cases included in the book are suitable for audiences from healthcare, computer,
telecommunication, network, and automation societies.
Preface xix
In Chapter 1, the authors provide an overview of blockchain radio access network

(B-RAN), which is a decentralized and secure wireless access paradigm. It leverages
the principle of blockchain to integrate multiple trustless networks into a larger shared
network and benefits multiple parties from positive network effects. The authors start
from the block generation process and develop an analytical model to characterize B-RAN
behaviors. By defining the work flow of B-RAN and introducing an original queuing model
based on a time-homogeneous Markov chain, the steady state of B-RAN is characterized
and the average service latency is derived. The authors then use the probability of a
successful attack to define the safety property of B-RAN and evaluate potential factors
that influence its security. Based on the modeling and analysis, the authors uncover an
inherent trade-off relationship between security and latency and develop an in-depth
understanding regarding the achievable performance of B-RAN. Finally, the authors verify
the efficiency of the model through an innovative B-RAN prototype.
Chapter 2 theoretically and experimentally analyses different consensus algorithms
in blockchains. The chapter firstly analyses the PoW consensus algorithm. The authors
employ reinforcement learning (RL) to dynamically learn a mining strategy with the
performance approaching that of the optimal mining strategy. Because the mining Markov
decision process (MDP) problem has a non-linear objective function (rather than linear
functions of standard MDP problems), the authors design a new multi-dimensional RL
algorithm to solve the problem. Experimental results indicate that, without knowing
the parameter values of the mining MDP model, the proposed multi-dimensional RL
mining algorithm can still achieve optimal performance over time-varying blockchain
networks. Moreover, the chapter analyzes the Raft consensus algorithm that is usually
adopted in consortium/private blockchains. The authors investigate the performance of
Raft in networks with non-negligible packet loss rate. They propose a simple but accurate
analytical model to analyze the distributed network split probability. The authors conclude
the chapter by providing simulation results to validate the analysis.
Chapter 3 describes a PBFT-based blockchain system, which makes it possible to break
the communication complexity bottleneck of traditional PoW- or BFT-based systems. The
authors discuss a double-layer PBFT-based consensus mechanism, which re-distributes
nodes into two layers in groups. The analysis shows that this double-layer PBFT signifi-
cantly reduces communication complexity. The authors then prove that the complexity is
optimal when the nodes are evenly distributed in each group in the second layer. Further,
the security threshold is analyzed based on faulty probability-determined (FPD) and the
faulty number-determined (FND) models in the chapter. Finally, the chapter provides a
practical protocol for the proposed double-layer PBFT system with a review of how PBFT
is developed.
In Chapter 4, the authors start by introducing the basic concepts of blockchain and illus-
trating why a consensus mechanism plays an indispensable role in a blockchain-enabled
Internet of Things (IoT) system. Then, the authors discuss the main ideas of two famous
consensus mechanisms, PoW and PoS, and list their limitations in IoT. After that, the
authors introduce PBFT and direct acyclic graph (DAG)-based consensus mechanisms as
an effective solution. Next, several classic scenarios of blockchain applications in the IoT
are introduced. Finally, the chapter is concluded with the discussion of potential issues
and challenges of blockchain in IoT to be addressed in the future.
xx Preface
Chapter 5 addresses the issues associated with centralized marketplaces in 5G networks.

The authors firstly study how a distributed alternative based on blockchain and smart
contract technology could replace the costly and inefficient third-party-based trust
intermediaries. Next, the authors propose a smart contract based on a sealed-bid double
auction to allow resource providers and enterprise users to trade resources on a distributed
marketplace. In addition, the authors explain the implementation of this marketplace
application on HyperLedger Fabric permissioned blockchain while deploying the network
using a pragmatic scenario over a public, commercial cloud. Finally, the authors evaluated
the distributed marketplace’s performance under different transaction loads.
In Chapter 6, the authors describe an integrated blockchain and software-defined net-
work (SDN) architecture for multi-operator support in 6G networks. They present a uni-
fied SDN and blockchain architecture with enhanced spectrum management features for
enabling seamless user roaming capabilities between mobile network operators (MNOs).
The authors employ the smart contract feature of blockchain to enable the creation of busi-
ness and technical agreements between MNOs for intelligent and efficient management
of spectrum assets (i.e. the radio access network). The study shows that by integrating
blockchain and SDN, the foundation for creating trusted interactions in a trustless envi-
ronment can be established, and users can experience no disruption in service with very
minimal delay as they traverse between operators.
Chapter 7 investigates and discusses the integration of blockchain and mobile edge
computing (MEC). The authors firstly provide an overview of the MEC, which sinks
computing power to the edge of networks and integrates mobile access networks and
Internet services in 5G and beyond. Next, the authors introduce the typical framework for
blockchain-enabled MEC and MEC-based blockchain, respectively. The authors further
show that blockchain can be employed to ensure the reliability and irreversibility of data
in MEC systems, and in turn, MEC can also solve the major challenge in the development
of blockchain in IoT applications.
Chapter 8 establishes an analytical model for PoW-based blockchain-enabled wireless
IoT systems by modeling their spatial and temporal characteristics as Poisson point pro-
cesses (PPP). The authors derive the distribution of signal-to-interference-plus-noise ratio
(SINR), blockchain transaction successful rate, as well as its overall throughput. Based on
this performance analysis, the authors design an algorithm to determine the optimal full
function node deployment for blockchain systems under the criterion of maximizing trans-
action throughput. In addition, the security performance of the proposed system is analyzed
in the chapter considering three different types of malicious attacks. The chapter ends with
a series of numerical results to validate the accuracy of the theoretical analysis and optimal
node deployment algorithm.
In Chapter 9, the authors examine the factors governing successful deployment of
blockchain-based distributed energy trading (DET) applications and their technical
challenges. The chapter walks through the fundamentals of “citizen-utilities,” primarily
assessing its impact on efforts to manage distributed generation, storage, and consumption
on the consumer side of the distribution network, while intelligently coordinating DET
without relying on trusted third parties. Additionally, the chapter highlights some of the
open research challenges including scalability, interoperability, and privacy that hinder
the mainstream adoption of “citizen-utilities” in the energy sector. Then, to address these
Preface xxi
research challenges, the authors propose a scalable citizen-utility that supports interoper-
ability and a Privacy-preserving Data Clearing House (PDCH), which is a blockchain-based
data management tool for preserving on-ledger and off-ledger transactions data privacy.
The chapter is finished with outlines of future research directions of PDCH.
In Chapter 10, the authors introduce a blockchain-enabled COVID-19 contact tracing
solution named BeepTrace. This novel technology inherits the advantages of digital con-
tract tracing (DCT) and blockchain, ensuring the privacy of users and eliminating the con-
cerns about the third-party trust while protecting the population’s health. Then, based on
different sensing technologies, i.e. Bluetooth and GPS, the authors categorize BeepTrace
into BeepTrace-active mode and BeepTrace-passive mode, respectively. In addition, the
authors summarize and compare the two BeepTrace modes and indicate their working prin-
ciples and privacy preservation mechanisms in detail. After that, the authors demonstrate
a preliminary approach of BeepTrace to prove the feasibility of the scheme. At last, further
development prospects of BeepTrace or other decentralized contact tracing applications are
discussed, and potential challenges are pointed out.
Chapter 11 looks at the infusion of blockchain technology into medical data sharing. The
chapter provides an overview of medical data sharing and defines the challenges in this
filed. The authors revisit some already established angles of blockchain medical data shar-
ing in order to properly contextualize it and to highlight new perspectives on the logical
outworking of blockchain-enabled sharing arrangements. Then, the authors present three
cases that are especially suited to blockchain medical data sharing. They also present an
architecture to support each paradigm presented and analyze medical data sharing to high-
light privacy and security benefits to data owners. Finally, the authors highlight some new
and emerging services that can benefit from the security, privacy, data control, granular data
access, and trust blockchain medical data sharing infuses into healthcare.
In Chapter 12, the authors propose a blockchain-based decentralized content vetting for
social networks. The authors use Bitcoin as the underlying blockchain model and develop
an unidirectional channel model to execute the vetting procedure. In this vetting procedure,
all users get a chance to vote for and against a content. Content with sufficient positive votes
is considered as vetted content. The authors then optimize the offline channel network
topology to reduce computation overhead because of using blockchains. At last, the authors
prove the efficiency of the vetting procedure with experiments using simulations of content
propagation in social network.
Bin Cao, Lei Zhang, Mugen Peng, Muhammad Ali Imran

August 2021
xxiii
Abbreviations
3GPP third-Generation Partnership Project

4G fourth generation
5G fifth generation
6G sixth generation
AAS Authentication Agent Server
ABAC Attribute-Based Access Control
ABIs Application Binary Interfaces
ABM Adaptive Blockchain Module
Abstract Abortable Byzantine faulT toleRant stAte maChine replicaTion
ACC Access Control Contract
AI artificial intelligence
API application programming interface
APs access points
BaaS Blockchain as a Service
BAS Blockchain Agent Server
BASA Blockchain-assisted Secure Authentication
BFT Byzantine fault tolerance
BLE Bluetooth low energy
BMap Bandwidth Map
BN blockchain network
BPL building penetration losses
BPM Business Process Management
bps bits per second
B-RAN Blockchain radio access network
BTC bitcoin
CA Certification Authority
CAGR Compound annual growth rate
CAPEX Capital expenditure
CBRS Citizens Broadband Radio Services
CDC Center for Diseases Control
CDF cumulative distribution function
CFT crash fault tolerance
CM consensus mechanism
xxiv Abbreviations
CoAP Constrained application protocol

COVID-19 Coronavirus Disease 2019
CPU Central Processing Unit
DAG direct acyclic graph
DAS distributed antenna systems
DCT digital contact tracing
DDoS Distributed Denial of Service
DEPs Distributed Energy Prosumers
DER Distributed Energy Resources
DET Distributed Energy Trading
DIS Data integrity verification systems
DLT Distributed ledger technology
DoS Denial of Service
DPoS Delegate Proof-of-Stake
DR demand response
DS Directory Service
DSM Demand Side Management
DSO distributed system operator
dTAM data Tagging and Anonymization Module
DTLS Datagram Transport Layer Security
E2E end to end
ECO Energy Company Obligation
EE energy efficiency
EMR electronic medical record
eNBs eNodeBs
ESPs Edge computing service providers
ESS energy storage systems
EV electric vehicles
EVN Electric Vehicle Networks
FAPs femtocell access points
FCC Federal Communications Commission
FDI false data injection
FeGW Femtocell gateways
FiT Feedin Tariff
FL Federated learning
FND faulty number determined
FNs function nodes
FPD faulty probability determined
FSC food supply chain
FSCD fast smart contract deployment
FTTH Fiber-to-the-Home
G2V grid-to-vehicle
Gb/s gigabyte per second
GDPR General Data Protection Regulation
Geth go-Ethereum
Abbreviations xxv
GPS Global Position System

GTP GPRS tunneling protocol
HARB Hypergraph-based Adaptive Consortium Blockchain
HARQ Hybrid Automatic Repeat Request
HeNB home eNB
HLF Hyperledger fabric
HSS home subscriber server
HTLC hash time-locked contract
HVAC heating, ventilation, cooling, and air conditioning
IaaS Infrastructure as a Service
IBC Identity-based Cryptography
IBS Identity-based Signature
IDC International Data Corporation
IDE Integrated development environment
IFA Dentifier for advertisers
IIoT Industrial Internet of things
IMDs Internet of things/mobile devices
IMEI International mobile equipment identity
IMT International Mobile Telecommunications
InPs Infrastructure Providers
IoT Internet of things
IoVs Internet of vehicles
KGC Key Generation Center
KPIs Key performance indicators
LAN Local area network
LRSig Linkable Ring Signatures
LSA Licensed shared access
LTE long-term evolution
MAC Media access control
MadIoT Manipulation of demand via IoT
MBS Macrocell base station
MCMC Markov Chain Monte Carlo
MDP Markov decision process
MEC mobile edge computing
MIMO multiple-input, multiple-output
MME mobility management entity
MNOs mobile network operators
MOCN multi-operator core network
μOs micro-operators
MSP Membership Service Providers
MSP multi-sided platform
MTT maximum transaction throughput
MVNO Mobile Virtual Network Operator
MW megawatts
Naas Network as a Service
xxvi Abbreviations
NAT nucleic acid testing

NFV Network Function Virtualization
NGN next-generation network
NHS National Health Service
NPI Non-pharmaceutical intervention
ns-3 Network simulator 3
OAMC Object Attribute Management Contract
ODN Optical Distribution Network
OFSwitch open-flow switch
OPEX Operating expenditure
OSN Online Social Network
OTP one time programmable
OTT over-the-top
P2P peer-to-peer
PaaS Platform as a Service
PBFT Practical Byzantine Fault Tolerance
PBN public blockchain network
PCG Prosumer Community Groups
PCRF Policy Charging and Rules Function
PDCH Privacy-preserving Data Clearing House
PDF probability density function
P-GW packet data network gateway
PHY physical
PKI Public key infrastructure
PMC Policy Management Contract
PoD proof-of-device
PONs Passive optical networks
PoO proof-of-object
PoS proof of stake
PoW proof of work
PPP Poisson point processes
QoS Quality of Service
RAN Radio access network
RES renewable energy sources
RL reinforcement learning
RMG relative mining gain
RSRP reference signal received power
RSRQ reference signal received quality
SaaS software as a Service
SAMC Subject Attribute Management Contract
SARS-CoV-2 Severe Acute Respiratory Syndrome Coronavirus 2
SBCs single-board computers
SBSs small base stations
SDN software-defined network
SEMC Smart Energy Management Controller
Abbreviations xxvii
S-GW Serving gateway

SHeNB Serving HeNB
SINR signal-to-interference-plus-noise ratio
SLA service-level agreement
SM supermassive
SPF single point of failure
SPs service providers
SR spinning reserve
SUTs System under tests
Tb/s terabyte per second
TDP transaction data packet
THeNB target HeNB
TNs transaction nodes
TPA Third Party Auditor
TPS transactions per second
TTI transmission time interval
TTP Trusted Third Party
TTT time to trigger
UE user equipment
UE RRC UE radio resource control
URI Uniform Resource Identifier
URL Uniform Resource Locator
UUID Universally Unique Identifier
V2G vehicle-to-grid
V2V vehicle-to-vehicle
vCPUs Virtual Central Processing Units
VEN Vehicular energy networks
VLC visible light communications
VM virtual machine
VNO Virtual Network Operator
VPP Virtual Power Plants
WAN Wide Area Network
WHO World Health Organization
ZKP zero-knowledge-proofs
1
What is Blockchain Radio Access Network?*

Xintong Ling 1,2 , Yuwei Le 1 , Jiaheng Wang 1,2 , Zhi Ding 3 , and Xiqi Gao 1,2
1
National Mobile Communications Research Laboratory, Southeast University, Nanjing, 210096, China
2
Purple Mountain Laboratories, Nanjing, Jiangsu, 211111, China
3
Department of Electrical and Computer Engineering, University of California, 95616, Davis, CA, USA
1.1 Introduction
The past decade has witnessed tremendous growth in emerging wireless technologies
geared toward diverse applications [1]. Radio access networks (RANs) are becoming
more heterogeneous and highly complex. Without well-designed inter-operation, mobile
network operators (MNOs) must rely on their independent infrastructures and spectra to
deliver data, often leading to duplication, redundancy, and inefficiency. A huge number of
currently deployed business or individual access points (APs) have not been coordinated
in the existing architecture of RANs and are therefore under-utilized. Meanwhile, user
equipments (UEs) are not granted to access to APs of operators other than their own,
even though some of them may provide better link quality and economically sensible.
The present state of rising traffic demands coupled with the under-utilization of existing
spectra and infrastructures motivates the development of a novel network architecture
to integrate multiple parties of service providers (SPs) and clients to transform the rigid
network access paradigm that we face today.
Recently, blockchain has been recognized as a disruptive innovation shockwave [2–4].
Federal Communications Commission (FCC) has been suggested that blockchain may be
integrated into wireless communications for the next-generation network (NGN) in the
Mobile World Congress 2018. Along the same line, the new concept of blockchain radio
access network (B-RAN) was formally proposed and defined in [5, 6]. In a nutshell, B-RAN
is a decentralized and secure wireless access paradigm that leverages the principle of
blockchain to integrate multiple trustless networks into a larger shared network and bene-
fits multiple parties from positive network effects [6]. It is a new architecture that integrates
both characteristics of wireless networks and distributed ledger technologies. As revealed
in [5, 6], B-RAN can improve the overall throughput through simplified inter-operator
* X. Ling, Y. Le, J. Wang, Z. Ding, and X. Gao. Practical modeling and analysis of blockchain radio access
network. IEEE Transactions on Communications, 69(2): 1021–1037, Feb. 2021.
Wireless Blockchain: Principles, Technologies and Applications, First Edition.

Edited by Bin Cao, Lei Zhang, Mugen Peng and Muhammad Ali Imran.
© 2022 John Wiley & Sons Ltd. Published 2022 by John Wiley & Sons Ltd.
2 1 What is Blockchain Radio Access Network?
cooperation in the network layer (rather than increasing the channel capacity in the
physical layer). B-RAN can enhance the data delivery capability by connecting these RANs
into a big network and leveraging the power of multi-sided platform (MSP). The positive
network effect can help B-RAN recruit and attract more players, including network
operators, spectral owners, infrastructure manufacturers, and service clients alike [6].
The subsequent expansion of such a shared network platform would make the network
platform more valuable, thereby generating a positive feedback loop. In time, a vast
number of individual APs can be organized into B-RAN and commodified to form a sizable
and ubiquitous wireless network, which can largely improve the utility of spectra and
infrastructures. In practice, rights, responsibilities, and obligations of each participant in
B-RAN can be flexibly codified as smart contracts executed by blockchain.
Among the existing studies on leveraging blockchain in networks, most have focused on
Internet of Things (IoT) [7–11], cloud/edge computing [12–15], wireless sensor networks
[16], and consensus mechanisms [17–19]. Only a few considered the future integration of
blockchain in wireless communications [20–26]. Weiss et al. [20] discussed several poten-
tials of blockchain in spectrum management. Kuo et al. [21] summarized some critical
issues when applying blockchain to wireless networks and pointed out the versatility of
blockchain. Pascale et al. [22] adopted smart contracts as an enabler to achieve service level
agreement (SLA) for access. Kotobi and Bilen [23] proposed a secure blockchain verifica-
tion protocol associated with virtual currency to enable spectrum sharing. Le et al. [27]
developed an early prototype to demonstrate the functionality of B-RAN.
Despite the growing number of papers and heightened interests to blockchain-based
networking, works including fundamental analysis are rather limited. A number of
critical difficulties remain unsolved. (i) Existing works have not assessed the impact of
decentralization on RANs after introducing blockchain. Decentralization always comes
with a price that should be characterized and quantified. (ii) Very few papers have noticed
that service latency will be a crucial debacle for B-RAN as a price of decentralization [22].
Unfortunately, the length of such delay and its controllability are still open issues.
(iii) Security is yet another critical aspect of blockchain-based protocols. In particular,
alternative history attack, as an inherent risk of decentralized databases, is always possible
and must be assessed. (iv) A proper model is urgently needed to exploit the characteristics
of B-RAN (such as latency and security) and to further provide insights and guidelines for
real-world implementations.
To address the aforementioned open issues, this chapter establishes a framework to con-
cretely model and evaluate B-RAN. We start from the block generation process and develop
an analytical model to characterize B-RAN behaviors. We shall evaluate the performance
in terms of latency and security in order to present a more comprehensive view of B-RAN.
We further verify the efficacy of our model through an innovative B-RAN prototype. The
key contributions are summarized as follows:
● We define the workflow of B-RAN and introduce an original queuing model based on a
time-homogeneous Markov chain, the first known analytical model for B-RAN.
● From the queuing model, we analytically characterize the steady state of B-RAN and fur-
ther derive the average service latency.
● We use the probability of a successful attack to define the safety property of B-RAN and
evaluate potential factors that influence the security.
1.2 What is B-RAN 3
Table 1.1 Important variables in the modeling and analysis
Symbols Explanations Symbols Explanations
𝜏kc Required service time of request k tka Arrival epoch of request k

𝜆a Request arrival rate T a = 1∕𝜆a Average inter-arrival time
𝜆 b
Block generation rate b
T = 1∕𝜆 b
Average block time
𝜆c Service rate T c = 1∕𝜆c Average service time
N Number of required confirmations s Number of access links
𝜆a
𝛽 Relative mining rate of an attacker 𝜌= c Traffic intensity
s𝜆
{ }
Φ = 𝜆a , 𝜆b , 𝜆c , s Basic configuration of B-RAN
● Based on the modeling and analysis, we uncover an inherent trade-off relationship

between security and latency, and develop an in-depth understanding regarding the
achievable performance of B-RAN.
● Finally, we build a B-RAN prototype that can be used in comprehensive experiments to
validate the accuracy of our analytical model and results.
We organize this chapter as follows. Section 1.2 presents the B-RAN framework and the
prototype. Section 1.3 provides the mining model to describe the block generation pro-
cess. In Section 1.4, we establish the B-RAN queuing model, with which we analyze and
evaluate the B-RAN performance concerning latency and security in Sections 1.5 and 1.6,
respectively. We demonstrate the latency-security trade-off in Section 1.7 and provide some
in-depth insights into B-RAN. Section 1.8 concludes this chapter. Given the large number
of symbols to be used, we summarize the important variables in Table 1.1.
1.2 What is B-RAN
1.2.1 B-RAN Framework

B-RAN offers a decentralized and secure wireless access paradigm for large-scale, hetero-
geneous, and trustworthy wireless networks [5, 6]. B-RAN unites multilateral inherently
trustless network entities without any trusted middleman and manages network access,
authentication, authorization, and accounting via direct interactions. As an open unified
framework for diverse applications to achieve resource pooling and sharing across sectors,
B-RAN presents an attractive solution for future 6G networking.
With the help of blockchain, B-RAN can form an expansive cooperative network includ-
ing not only telecommunication giants but also small contract holders or individual MNOs
to deliver excellent quality services at high spectrum efficiency. B-RAN can integrate mul-
tiple networks across SPs for diverse applications. As illustrated in Figure 1.1, B-RAN is
self-organized by APs belonging to multiple SPs, massive UEs, and a blockchain maintained
by miners. In B-RAN, a confederacy of SPs (organizations or individuals) act as a virtual SP
(VSP) to provide public wireless access under shared control. These SPs in B-RAN allow
Massive UEs supporting diverse applications
Request
Authorize
Blockchain
APs belonging to different SPs
Figure 1.1 Conceptual illustration of self-organized B-RAN.
the greater pool of UEs to access their APs and networks by receiving payment or credit for
reciprocal services. Blockchain acts as a public ledger in B-RAN for recording, confirming,
and enforcing digital actions in smart contracts to protect the interests of all participants.
B-RAN is envisioned to be broadly inter-operative and to support diverse advanced wire-
less services and standards. In this chapter, we focus on the fundamental access approach
for which the procedure is shown in Figure 1.2.
● In preparation for access, UEs and SPs should first enter an SLA containing the details
including service types, compensation rates, among other terms. (For example, SPs can
first publish their service quality and charge standard, and UEs select suitable SPs accord-
ing to the expenditure and quality of service.) The service terms and fees will be explicitly
recorded in a smart contract authorized by the digital signatures of both sides.
● In phase 1, the smart contract with the access request is committed to the mining network
and is then verified by miners. The verified contracts are assembled into a new block,
which is then added at the end of the chain.
● In phase 2, the block is accepted into the main chain after sufficient blocks as confirma-
tions built on top of it.
● In phase 3, the request is waiting for service in the service queue.
● In phase 4, the access service is delivered according to the smart contract.
The above procedure can be viewed as a process of trust establishment between clients
and SPs, similar to negotiating and signing monthly contracts between users and MNOs.
Thus, in B-RAN, clients can obtain access services more conveniently through the above
process instead of signing contracts with a specific MNO in advance. The service duration
in B-RAN is flexible and can be as short as a few minutes or hours, which is different from
typical long-term plans (e.g. monthly plans). UEs can prolong access services by renewing
the contract earlier before the previous one expires in order to continue the connection
status. Therefore, service latency in this context refers to the delay when a UE accesses
an unknown network for the first time and can be viewed as the time establishing trust
between two trustless parties, which is significantly different from the transmission delay
in the physical layer.
1.2 What is B-RAN 5
Blockchain Block h Block h + 1 Block h + N – 1
Commit
Smart AP
Service AP
contract
Queue
Block Access
REQk(t ak , τ ck ) Wait for N Blocks
UE
τ ck
UE AP t ak
Phase 1 Phase 2 Phase 3 Phase 4
Wait to be included Wait for confirmations Wait for service In service
into a block
Figure 1.2 Four stages of the access workflow in B-RAN.

( )
Mathematically, we can describe the request structure by REQk tka , 𝜏kc shown in
Figure 1.2, where tka and 𝜏kc are the arrival epoch and the service duration of request k,
respectively. Assume that the access requests are mutually independent and arrive as a
Poisson process with rate 𝜆a . Equivalently, the inter-arrival time between two requests
U a follows exponential distribution with mean T a = 1∕𝜆a . Based on well-known studies
such as [28], the random service time 𝜏kc is also expected to be exponential with mean
T c = 1∕𝜆c . Note that in this chapter, we consider a tract covered by multiple trustless SPs
(organizations or individuals). Usually, the block size limit is much larger than the request
rate of a single tract and thus can be ignored in this case.
In the context of B-RAN, we introduce the concept of “virtual link” instead of the physical
channel, where one virtual link represents a tunnel providing access services to one client
at a time.1 Hence, the number of links means the maximum number of UEs that can receive
access services simultaneously from the APs belonging to these SPs in the tract and reflects
the access capability of a network. In the considered tract, we assume the maximum number
of links to be s.
It is worth pointing out that the efficiency improvement of B-RAN roots in the network
pooling principle, which requires a flexible offloading and sharing between subnetworks.
Inspired by the Schengen Agreement, B-RAN adopts the mechanism that, if a client estab-
lishes trust with specific SPs via the procedure in Figure 1.2, the client may use resources
pooled by SPs in B-RAN, e.g. a frequency band belonging to another SP. The miners can
use some intelligent algorithms2 to allocate and distribute the pooled resources for higher
network efficiency [6]. As a result, mobile devices may access suitable APs belonging to
the SPs, which likely provide higher quality coverage for the UEs in their current locations.
The trading among SPs caused by roaming would be calculated and settled via blockchain
1 Usually, the physical channel may provide services to many clients via proper multiple access
techniques. We use the concept of “virtual link” to avoid ambiguity.
2 Note that, in this chapter, the miners still follow the basic first-in-first-out principle to put the requests
into the queue.
periodically. Based on the above mechanism, B-RAN can take advantage of pooling and
sharing across subnetworks. Please refer to Figure 1.9 in this chapter and [6, 10] for more
details and evidence for the pooling effect in B-RAN.
1.2.2 Consensus Mechanism

B-RAN, as a decentralized system, requires proper consensus mechanisms for consistency
[18]. Proof-of-Work (PoW) has been widely used in practice and proven to be secure in cryp-
tocurrencies such as Bitcoin. In PoW, network maintainers, also known as miners, need to
obtain a hash value below a given target by repeatedly guessing a random variable named
nonce. However, PoW-based consensus mechanisms consume a tremendous amount of
energy, which is likely unbearable for energy-limited mobile devices.
Consequently, proof-of-device (PoD) is proposed for B-RAN as a low- cost alternative [5].
PoD utilizes the fact that wireless access usually depends on a hardware device associated
with a unique identifier in order to elevate the cost of creating new identities. To be more
specific, PoD can create new identifiers or rely on some existing ones, such as the inter-
national mobile equipment identity (IMEI) and identifier for advertisers (IFAs), in one
time programmable (OTP) memory to distinguish different network entities and prevent
identity fraud. Also, because of variations during manufacture, every device has multiple
hardware-dependent features, which could constitute a unique RF fingerprinting for each
device and can be identified from the transmitted RF signal [29, 30]. Forging an identity of a
device is often costly in the real world, whereas creating multiple identities is almost costless
in cryptocurrencies. Therefore, PoD can safeguard the security of B-RAN without expend-
ing immense computing power and is thus suitable for wireless networks. Notably, PoW,
PoD, and other alternatives can be put in the same class since all of them are based on hash
puzzles. We will further discuss and model the block generation process of a hash-based
consensus mechanism in Section 1.3.
1.2.3 Implementation
In order to evaluate our established model, we will provide demonstrative experimental
results from a home-built prototype throughout the whole article. We implement this ver-
sion of B-RAN prototype on four single board computers (SBCs) and use a workstation with
Intel Core CPU I7-8700K and 32GB RAM in order to provide sufficient computing power.
Our prototype consists of a standard file system for data storage, a key-value database for
file index, and the core modules written in Python. The prototype supports both PoW and
PoD consensus mechanisms as two available options and can adopt an appropriate one
according to the specific environment and requirements. We configure different SBCs as
UEs and APs and set up the integrated development environment (IDE), wherein UEs pro-
pose access requests according to the input configurations, and APs provide services based
on the workflow given in Section 1.2.1. During tests, the prototype can track running statis-
tics and provide them as output results.
The B-RAN prototype is a hierarchical architecture with a number of modules and com-
ponents [6]. For example, the fast smart contract deployment (FSCD) was proposed in [27]
to accelerate the service deployment, and the hash time-locked contract (HTLC) is designed
1.3 Mining Model 7
in [6] to enforce the contract and secure the trading process. This chapter is important by
modeling and assessing B-RAN, and thus, we cannot include all the technical details here.
Please see these citations for more details. Note that, in this chapter, the average service
time T c as time unit is set to unity without loss of generality, so time is measured as relative
variables in terms of time unit T c .
In Sections 1.5–1.7, we will assess the performance of B-RAN from different points of view
and verify our model step by step through prototype verifications. Although these verifica-
tions focus on different aspects, all of them are obtained from the same B-RAN prototype
described above.
1.3 Mining Model
1.3.1 Hash-Based Mining

In this section, we will present a general model for hash-based puzzles to describe the block
generation process, also known as mining. Usually, block propagation in a network is much
faster than block generation. We thus ignore the block spreading delay in the modeling and
verify the assumption by using the real data of Bitcoin and Ethereum.
Generally, a hash puzzle can be modeled as a problem to find a suitable answer to satisfy
the following conditions:
Hash(HP + DP + TS + OF) < GT. (1.1)
Here, “HP” stands for the hash pointer to a previous block, “DP” means the data payload,
“TS” is the current timestamp, “GT” represents a given target, and “OF” is the optional field
depending on the specific type of the hash puzzle. For instance, in the PoW protocol, the
optional field can be any random number, whereas in the PoD protocol, the optional field is
given by the hardware identifier. In PoW, the range of the optional field is unlimited. Hence,
a miner can guess many times to find a correct nonce, and hence, the number of trials is only
restricted by the mining rigs. In PoD, the optional field is given by the tamper-proof identi-
fier such that each device can perform the hash computation only once for each timestamp,
thereby largely reducing the power consumption. The premise behind that is that the enti-
ties in real RANs cannot be effortlessly forged or created. The characteristics, e.g. security
and power consumption, of different hash-based consensus mechanisms can be traded off
by properly choosing the optional field.
1.3.2 Modeling of Hash Trials

For a general hash-based mining process, each hash trial can be regarded as an indepen-
dent Bernoulli experiment with success probability p as the timestamp keeps changing. In
a sequence of independent Bernoulli trials, the probability that the first block is generated
after m failures is (1 − p)m p. Let W b be the number of failures preceding the first success.
Then, W b follows geometric distribution:
{ }
Pr W b = m = (1 − p)m p, m = 0, 1, … . (1.2)
Easily, we can obtain

{ }
Pr W b ≥ m = (1 − p)m , m = 0, 1, … . (1.3)
Note that W b can be viewed as the waiting period before a block is successfully generated,
and its distribution can be described by (1.2) and (1.3). The average number of successes in
m independent trials is mp.
Hence, if m hash trials are conducted in a time interval of length 𝜏, then the success rate
defined as the mean number of successes per unit time is 𝜆b = mp∕𝜏. Now, let p → 0 and
m → ∞ in the way that keeps 𝜆b constant. We can visualize an experiment with infinite
hash trials performed within interval 𝜏. Successive trials are infinitesimally close with van-
ishingly small probability of success, but the mean number of successes remains a non-zero
constant 𝜆b 𝜏. By using the fact that the geometric distribution approaches the exponential
distribution in the limit, we have
( )m
𝜆b 𝜏
lim (1 − p)m ∣p= 𝜆b 𝜏 = lim 1 − = exp(−𝜆b 𝜏). (1.4)
m→∞ m m→∞ m
Letting a random variable U b be the continuous time before preceding the first success, we
have
{ } { }
Pr U b > 𝜏 = Pr W b > m = exp(−𝜆b 𝜏). (1.5)
We name U b as the block time in the context of blockchain. According to (1.5), the average
[ ]
block time 𝔼 U b , denoted by T b , is equal to 1∕𝜆b , where 𝜆b is thus called as the mining
rate representing how fast blocks generate.
Equations (1.4) and (1.5) imply that, if the number of hash computations of the whole
network in a unit time tends to infinity, then the length of time between two successive
blocks, i.e. U b ’s, would follow the exponential distribution. Interestingly, most of mature
PoW blockchain networks indeed perform a huge number of hash computations every
moment. For example, the minimum hash rates of bitcoin and Ethereum during 2018 are
14 891 and 159TH/s.3 . These are humongous numbers in real world and practically support
the limiting condition that the number of trials tends to infinity. When massive hardware
devices are participating in mining, the condition also holds for PoD. Remark that, even
if the number of trials is finite and the exponential approximation no longer holds, the
block times U b ’s are still mutually independent and identically distributed because of the
memoryless property of geometric distribution.
Now, we further prove that block generation forms a Poisson process. Let B(n, t, t + h)
denote the event that n blocks are generated in interval (t, t + h). As h → 0, the probability
that at least one block is generated in (t, t + h) is
Pr {B(n ≥ 1, t, t + h)} = 1 − Pr {B(n = 0, t, t + h)}

{ }
= 1 − Pr U b > h
= 1 − exp(−𝜆b h)
= 𝜆b h + o(h), h → 0.
3 Source: bitinfocharts.com/, accessed February 2021.

1.3 Mining Model 9
We note that o(h) is an infinitesimal of higher order such that limh→0 o(h)∕h = 0. Exactly,
one block is generated in (t, t + h), i.e. B(n = 1, t, t + h), if and only if the first block occurs
within (t, t + h) and the second one occurs after the epoch t + h. Hence, the probability of
Pr {B(n = 1, t, t + h)} is
Pr {B(n = 1, t, t + h)}
h { } { }
= Pr U1b = h − u Pr U2b > u du
∫0
h
= 𝜆b exp(−𝜆b (h − u)) exp(−𝜆b u)du
∫0
= 𝜆b h exp(−𝜆b h) = 𝜆b h + o(h), h → 0,
where U1b and U2b represent the generation times of first and second blocks since t, respec-
tively. As h → 0, the probability of occurrence of two or more blocks in (t, t + h) is
Pr {B(n ≥ 2, t, t + h)} = 1 − exp(−𝜆b h) − 𝜆b h exp(−𝜆b h) = o(h), h → 0.
In conclusion, the block generation can be modeled as a Poisson process with mining rate
𝜆b . Since the negligible propagation delay and infinity hash rate are assumed in the model-
ing, we shall carefully verify the validation of the Poisson model by real data.
In Prototype Verification A, we illustrate the validity of the Poisson model in Figure 1.3
by empirical data. In our self-built B-RAN prototype described in Figure 1.3a, we deploy five
miners with equal mining rates and set the block generation rate to 𝜆b = 1∕10 or equiva-
lently the average block time T b = 1∕𝜆b = 10. We measure the block time for 10 000 blocks
and plot the histogram to approximate the distribution of block time U b . One can see that
the Poisson model closely fits the data. Furthermore, we collected the empirical data from
Bitcoin (51 036 blocks starting at height 530 1144 ) and Ethereum (2 078 000 blocks start-
ing at height 5 806 2895 ) to demonstrate the validity of the model in practice. The collected
Bitcoin and Ethereum data are also consistent with the model in a real network where prop-
agation delay does exist. To be more specific, we calculate R2 as a goodness-of-fit indicator
× 10–3
0.1
1.5 R2 > 0.999 0.06 2
R > 0.995
10 (predefined) 576 (measured) 13.5 (measured)

1 0.04
PDF
PDF
PDF
600 (predefined) 10–15 (predefined)

0.05
0.5 0.02
0 0 0
0 20 40 60 0 1000 2000 3000 0 20 40 60
Block time (s) Block time (s) Block time (s)
(a) (b) (c)
Figure 1.3 The distribution of block time from real data and simulations. (Please see footnotes 4
and 5 for sources.) (a) Self-built B-RAN prototype. (b) Bitcoin. (c) Ethereum.
4 Source: www.blockchain.com/en/btc/blocks/, accessed February 2021.

5 Source: etherscan.io/block/, accessed February 2021.
and obtain R2 > 0.999 for Bitcoin and R2 > 0.995 for Ethereum. The above results strongly
support that the Poisson model perfectly fits the data, even in Ethereum where the block
propagation is in the same order of magnitude as the block time. Therefore, even though
the propagation delay is simply neglected, the Poisson model can still well characterize the
block generation in practice.
1.3.3 Threat Model

In this study, we consider an adversary who mounts an alternative history attack by attempt-
ing to generate a longer fraudulent chain. The adversary has to generate new blocks in the
same way as honest miners because the hash value can hardly be tampered. Hence, the
fraudulent blocks are also generated as Poisson. Now, assume that the mining rate of the
𝛽
attacker is 𝛽𝜆b , while the mining rate of honest miners is 𝜆b , i.e. the attacker controls 1+𝛽
fraction of hash power.6 Both the attacker and the benign network are mining indepen-
dently. From the additive property of Poisson processes, the sum generation rate of both
benign and fraudulent block is (𝛽 + 1) 𝜆b . Note that, in an alternative history attack, the
attacker will not publish the fraudulent chain until it creates a more extended branch. Con-
sequently, the benign participants of B-RAN are unaware of the existence of the attacker and
can only observe the blocks generated by honest miners with the generation rate 𝜆b until
a fraud succeeds. We will provide a detailed procedure of the alternative history attack in
Section 1.7.
1.4 B-RAN Queuing Model

According to the B-RAN framework in Section 1.2, we divide the service process of a valid
request into four stages: (i) waiting to be included into a block, (ii) waiting for confirma-
tions, (iii) waiting for service, and (iv) in service. Naturally, we can model the process using
several queues in tandem based on the four phases. However, it is worth pointing out that,
in the third stage, the requests in the same block arrive simultaneously, and the number
of requests is related to the block generation time U b . Hence, this queue is non-Markovian
because some previous events (e.g. U b ) beside the current state of the queue may affect its
future state. Usually, a non-Markovian queue is challenging to tackle. Therefore, we should
carefully select the state space of B-RAN for further analysis.
Let in be the number of pending requests that already have n confirmations. A pending
∑+∞
request is confirmed after receiving N confirmations. Then, j = n=N in denotes the num-
ber of confirmed requests that have not been served yet. In this approach, B-RAN can be
fully identified by state E(i0 , i1 , … , iN−1 , j) belonging to the (N + 1)-dimensional state space
ℤN+1
+ , where ℤ+
N+1
represents the set of all (N + 1)-tuples of non-negative integers. Formally,
the queuing process of B-RAN is completely described by a vector stochastic process as:
+ , t ≥ 0}.
{X(t) ∈ ℤN+1
6 Usually, an attacker can hardly amass more hash power than the sum of other honest miners, i.e. 𝛽 < 1;
Otherwise, the attacker already dominates the mining network.
1.4 B-RAN Queuing Model 11
B-RAN is said to be in state E(i0 , i1 , … , iN−1 , j) at time t if X(t) = E(i0 , i1 , … , iN−1 , j). Note
that the way to establish a queuing model is not unique. We define B-RAN by using the
queuing model {X(t), t ≥ 0} owing to two critical properties, as shown in Theorem 1.1.
Theorem 1.1 The queuing model {X(t), t ≥ 0} is a continuous process with Markov prop-
erty and time homogeneity or mathematically:
(a) Pr{X(t + h) = E|X(t) = E′ , X(u) for 0 ≤ u ≤ t} (1.6)
{ }
= Pr X(t + h) = E|X(t) = E′ ;
{ }
(b) Pr{X(t + h) = E|X(t) = E′ } = Pr X(t) = E|X(0) = E′ . (1.7)
Proof: Recall that, as claimed in Section 1.2, requests arrive according to a Poisson process
with rate 𝜆a , and the service times are exponentially distributed with mean 1∕𝜆c , indepen-
dently of each other. According to the mining model in Section 1.3, the block generation
is a Poisson process with rate 𝜆b . In a nutshell, request inter-arrival times U a , block times
U b , and service times U c are exponential with mean 1∕𝜆a , 1∕𝜆b , and 1∕𝜆c , respectively, and
independent of each other.
The exponential service times U c imply that if j UEs are in service, the rate at which
service completions occur is j𝜆c . To show this, suppose that U1c , U2c , … , Ujc are the duration
c
of j i.i.d. exponential simultaneously running time intervals with mean 1∕𝜆c . Let Umin =
{ }
min U1 , U2 , … , Uj be the minimum service time. Observe that Umin will exceed u if and
c c c c
only if all Ujc exceed u. Hence,

{ c } { } { } { } ( )
Pr Umin > u = Pr U1c > u ⋅ Pr U2c > u · · · Pr Ujc > u = exp −j𝜆c u .
Only those UEs that are in service can possibly leave. Hence, the service completion rate
with j simultaneous services is j𝜆c for 0 ≤ j ≤ s. Since at most s UEs can be in service simul-
taneously, obviously the service completion rate is at most s𝜆c . Hence, we denote
𝜆cj = min (j, s) ⋅ 𝜆c , (1.8)
to represent the service completion rate of B-RAN compactly. Similarly, the time until
the next event (a request arrival, a block generation, or a service completion) is also
exponentially distributed with rate (𝜆a + 𝜆b + 𝜆cj ). The probability that an event occurs in
( ( ) ) ( )
(t, t + h) is 1 − exp − 𝜆a + 𝜆b + 𝜆cj h , which tends to 𝜆a + 𝜆b + 𝜆cj h + o(h) as h → 0.
The length of time required for the event to occur and the type of the event are independent.
There are several possible changes to a state. If a new request arrives, the state of B-RAN
( ) ( )
will switch from E′ i0 , i1 , … , iN−1 , j to E i0 + 1, i1 , … , iN−1 , j . The probability that a new
request arrives in (t, t + h) is given by
( ) ( )}
Pr {X(t + h) = E i0 + 1, i1 , … , iN−1 , j |X(t) = E′ i0 , i1 , … , iN−1 , j (1.9)
(( ) )
𝜆 a
= a 𝜆a + 𝜆b + 𝜆cj h + o(h) = 𝜆a h + o(h), (h → 0).
𝜆 + 𝜆b + 𝜆cj
Similarly, if a new block is generated, all the existing blocks will get one more confirmation,
( ) ( )
and the state of B-RAN will move from E′ i0 , i1 , … , iN−1 , j to E 0, i0 , i1 , … , iN−1 + j . The
E(i0, i1, ..., iN–1, j – 1)
λccjh
λah
1 – (λa + λb + λcj )h E΄(i0, i1, ..., iN–1, j) E(i0 + 1, i1, ..., iN–1, j)
λbh
E(0, i0, i1, ..., iN–1 + j)
( )
Figure 1.4 State transition graph of E i0 , i1 , … , iN−1 , j . (j ≥ 1.).
probability that a block is generated in (t, t + h) is 𝜆b h + o(h) as h → 0. If an access service

( )
is ended (j ≥ 1 at this instant), the state of B-RAN will switch from E′ i0 , i1 , … , iN−1 , j to
( )
E i0 , i1 , … , iN−1 , j − 1 with probability 𝜆cj h + o(h) as h → 0. The probability that no event
occurs in (t, t + h) is 1 − (𝜆a + 𝜆b + 𝜆cj )h + o(h), and the probability that more than one event
occurs in (t, t + h) is o(h) as h → 0.
{ }
Now, we have obtained the transition probabilities Pr X(t + h) = E|X(t) = E′ for any
E and E′ in the state space. Observe that the transition probabilities are irrelevant to the
starting time t, which indicates that {X(t), t ≥ 0} is homogeneous in time. Moreover, the
transition probabilities are independent of the states of previous moments, which implies
the Markov property. Therefore, we have proven that {X(t), t ≥ 0} is a time-homogeneous
Markov process. ◾
Although the queuing model established by Theorem 1.1 is a vector stochastic process
with possibly high dimensions, we would like to emphasize that such a queuing model
is more tractable than the original non-Markovian process. According to the proof of
{ }
Theorem 1.1, we can directly obtain the transition probabilities Pr X(h) = E|X(0) = E′
and characterize the queuing model {X(t), t ≥ 0} completely.
Figure 1.4 visualizes the transition relationships according to the proof of Theorem 1.1.
{ }
One can see that the basic elements 𝜆a , 𝜆b , 𝜆c , s and the number of confirmations N
are enough to determine the transition probabilities and thus characterize the behaviors
{ }
of B-RAN. Hence, we introduce a four-tuple Φ = 𝜆a , 𝜆b , 𝜆c , s as basic configurations to
describe B-RAN. In Sections 1.5 and 1.6, we will analyze B-RAN from a deeper view in
more dimensions.
1.5 Latency Analysis of B-RAN

1.5.1 Steady-State Analysis
Now, we have thus far modeled the B-RAN as a time-homogeneous Markov process
with (N + 1) dimensions. However, the dimensionality of the state space results in a
complex probability transition graph and is difficult to analyze in general. In this section,
we will analyze the access service latency of B-RAN by starting from a relatively simple
one-confirmation case (i.e. only one confirmation is required to confirm a request). In other
words, a request is confirmed, as long as it is assembled into a block. We must stress that
the service latency here is significantly different from the physical-layer transmission delay.
In the one-confirmation case, the queuing model is presented by {X(t) = E(i, j), t ≥ 0},
where we drop the subscript of i0 for notational simplicity. State E(i, j) means that i pending
requests are waiting for assembling into a block and j confirmed requests are waiting for
service. Define 𝑤i,j (t) = Pr {X(t) = E(i, j)} as the probability of the queue in state E(i, j) at
time t. Now, let us investigate the transition probabilities during time h with the help of
Theorem 1.1. By comparing the state of B-RAN at time t + h with that at time t, for all
i = 1, 2, … and j = 0, 1, 2, …, we have
( )
𝑤i,j (t + h) − 𝑤i,j (t) = 𝑤i−1,j (t)𝜆a h + 𝑤i,j+1 (t)𝜆cj+1 h − 𝑤i,j (t) 𝜆a + 𝜆b + 𝜆cj h,
where 𝜆cj is defined by (1.8). By letting h → 0, we get the following differential-difference

equation:
( )
d
𝑤i,j (t) = 𝑤i−1,j (t)𝜆a + 𝑤i,j+1 (t)𝜆cj+1 − 𝑤i,j (t) 𝜆a + 𝜆b + 𝜆cj .
dt
{ }
Let 𝑤i,j be the steady-state distribution of B-RAN. The equilibrium condition dtd 𝑤i,j (t) = 0
yields
( )
𝑤i−1,j 𝜆a + 𝑤i,j+1 𝜆cj+1 − 𝑤i,j 𝜆a + 𝜆b + 𝜆cj = 0. (1.10)
For the boundary cases (i = 0), we have

( j )
∑ ( )
𝑤𝓁,j−𝓁 𝜆b + 𝑤0,j+1 𝜆cj+1 − 𝑤0,j 𝜆a + 𝜆cj = 0, ∀j = 0, 1, 2, … (1.11)
𝓁=1
𝑤0,1 𝜆c1 − 𝑤0,0 𝜆a = 0. (1.12)
The differential-difference equations (1.10)–(1.12) are known as the forward Kolmogorov

equations [31]. We illustrate the state transition relationships with one confirmation in
Figure 1.5.
In order to present the forward Kolmogorov equations in a compact form, we rearrange
{ }
the two-dimension states 𝑤i,j by a particular order as shown in Figure 1.6, captured by
the probability vector:
[ ]T
w = 𝑤0,0 𝑤1,0 𝑤0,1 𝑤2,0 𝑤1,1 𝑤0,2 · · · .
λc λc2 λc3 λc4

0, 0 0, 1 0, 2 0, 3 ...
λb λb λb
λa λa
λa λa
λc λc2 λc3
1, 0 1, 1 1, 2 ...
λa λa λa
λc λc2
2, 0 2, 1 ...
λa λa
3, 0
λc ...
λa
...
Figure 1.5 State space diagram in B-RAN.

0, 0 0, 1 0, 2 0, 3
w0, 0
w1, 0
w0, 1
1, 0 1, 1 1, 2 w2, 0
w1, 1
w0, 2
2, 0 2, 1 w3, 0
w2, 1
w1, 2
3, 0
w0, 3
…
𝓌
…
Figure 1.6 States E(i, j) rearrangement into a row.
Now, the forward Kolmogorov equations can be rewritten in a matrix form:

Qw = 𝟎, (1.13)
where Q is the infinitesimal generator, or transition rate matrix, given by
⎡ −𝜆
a 𝜆c1 ··· ⎤
( )
⎢ 𝜆 a − 𝜆a + 𝜆b 𝜆c1 ··· ⎥
⎢ ( a ) ⎥
⎢ 𝜆 b − 𝜆 + 𝜆 c
𝜆c2 ··· ⎥
⎢
1 ( a )
Q= 𝜆 a − 𝜆 +𝜆 b ··· ⎥
⎢ ( ) ⎥
⎢ 𝜆a − 𝜆a + 𝜆b + 𝜆c1 ··· ⎥
( )
⎢ 𝜆b 𝜆b − 𝜆a + 𝜆c2 ··· ⎥
⎢ ⎥
⎣ ⋮ ⋮ ⋮ ⋮ ⋮ ⋮ ⋱ ⎦
(1.14)
d
The entry in Q equals to the corresponding transition rate given by dh Pr{X(h) =
{ }
E|X(0) = E′ } only depending on the B-RAN configuration tuple Φ = 𝜆a , 𝜆b , 𝜆c , s . We can
numerically solve the matrix equation by combining with the sum probability condition of
𝟏T w = 1, i.e.
[ ] [ ]
Q 𝟎
w = . (1.15)
𝟏T 1
From (1.15), the steady-state distribution w(Φ) can be expressed as an implicit function of
Φ. Note that the waiting space of B-RAN has no maximum limit. The number of states, i.e.
the dimensions of the vector w, should be infinite. In numerical calculations, we can use
the solution with large enough but finite dimensions to approximate the infinite-dimension
one. However, in practice, the number of UEs in a tract cannot be infinite, either. The aggres-
sive load 𝜆a is required to be less than 𝜆c by the stable condition.
We can obtain the steady-state distribution of B-RAN via (1.15). In Prototype
Verification B, we use our self-built prototype to measure the sojourn time of each state
and estimate the probability of a state. The results show that analytical steady distributions
are highly consistent with experimental outcomes, thereby validating our established
queuing model. We illustrate the steady-state distributions of B-RAN with different T b
𝜆a
and different traffic intensities 𝜌 = s𝜆 c
in Figure 1.7. The low, medium, and high traffic
(0,0) (0,1) (0,2) (0,3) (0,4) (0,5) (0,0) (0,1) (0,2) (0,3) (0,4) (0,5)
0.66 0.26 0.053 0.0071 ≈0 ≈0 0.62 0.25 0.051 0.0072 ≈0 ≈0
(1,0) (1,1) (1,2) (1,3) (1,4) (1,0) (1,1) (1,2) (1,3) (1,4)
0.011 0.0041 ≈0 ≈0 ≈0 0.049 0.016 0.0029 ≈0 ≈0
(2,0) (2,1) (2,2) (2,3) (2,0) (2,1) (2,2) (2,3)

≈0 ≈0 ≈0 ≈0 0.0038 0.0011 ≈0 ≈0
(3,0) (3,1) (3,2) (3,0) (3,1) (3,2)

≈0 ≈0 ≈0 ≈0 ≈0 ≈0
(4,0) (4,1) (4,0) (4,1)

≈0 ≈0 Rest < 9e–06 ≈0 ≈0 Rest < 2e–05
(5,0) (5,0)
0 T b = 0.04 T c, ρ = 0.1 ≈0 T b = 0.2 T c, ρ = 0.1
(a) (d)
(0,0) (0,1) (0,2) (0,3) (0,4) (0,5) (0,0) (0,1) (0,2) (0,3) (0,4) (0,5)
0.19 0.3 0.24 0.13 0.051 0.021 0.15 0.23 0.19 0.11 0.046 0.02
(1,0) (1,1) (1,2) (1,3) (1,4) (1,0) (1,1) (1,2) (1,3) (1,4)
0.012 0.018 0.014 0.0073 0.0028 0.044 0.06 0.043 0.021 0.0083
(2,0) (2,1) (2,2) (2,3) (2,0) (2,1) (2,2) (2,3)

≈0 0.0011 ≈0 ≈0 0.013 0.015 0.0095 0.0042
(3,0) (3,1) (3,2) (3,0) (3,1) (3,2)

≈0 ≈0 ≈0 0.0037 0.0037 0.002
(4,0) (4,1) (4,0) (4,1)

≈0 ≈0 Rest < 0.02 0.001 ≈0 Rest < 0.03
(5,0) (5,0)
≈0 T b = 0.04 T c, ρ = 0.4 ≈0 T b = 0.2 T c, ρ = 0.4
(b) (e)
(0,0) (0,1) (0,2) (0,3) (0,4) (0,5) (0,0) (0,1) (0,2) (0,3) (0,4) (0,5)
0.045 0.13 0.18 0.16 0.12 0.082 0.029 0.08 0.12 0.11 0.084 0.063
(1,0) (1,1) (1,2) (1,3) (1,4) (1,0) (1,1) (1,2) (1,3) (1,4)
0.005 0.013 0.018 0.016 0.011 0.015 0.036 0.045 0.039 0.027
(2,0) (2,1) (2,2) (2,3) (2,0) (2,1) (2,2) (2,3)

≈0 0.0014 0.0019 0.0016 0.0073 0.015 0.017 0.013
(3,0) (3,1) (3,2) (3,0) (3,1) (3,2)

≈0 ≈0 ≈0 0.0034 0.0063 0.0062
(4,0) (4,1) (4,0) (4,1)

≈0 ≈0 Rest < 0.2 0.0015 0.0025 Rest < 0.3
(5,0) (5,0)
≈0 T b = 0.04 T c, ρ = 0.7 ≈0 T b = 0.2 T c, ρ = 0.7
(c) (f)
Figure 1.7 The distribution of steady states under different traffic intensities and block time with
s = 4 links. (a) T b = 0.04T c under low traffic intensity 𝜌 = 0.1. (b) T b = 0.04T c under medium traffic
intensity 𝜌 = 0.4. (c) T b = 0.04T c under high traffic intensity 𝜌 = 0.7. (d) T b = 0.2T c under low traffic
intensity 𝜌 = 0.1. (e) T b = 0.2T c under medium traffic intensity 𝜌 = 0.4. (f) T b = 0.2T c under high
traffic intensity 𝜌 = 0.7.
Another random document with
no related content on Scribd:
must be humorous, or he could not have written the chapters on ‘A
New England Squire’ and ‘The Country Church,’ to say nothing of
the account of the loves of Clarence and Jenny. He must be
sentimental, or the chapter entitled ‘A Good Wife’ had been an
impossibility.
At every point the book betrays its Puritan origin. ‘Ik Marvel’ is a
moralist. He makes a direct and constant appeal to the ethical
sentiment. In one of his prefaces he mentions the fact—doubtless an
amused smile played about his lips as he wrote the lines—that
Dream Life has sometimes insinuated itself into Sunday-school
libraries. He hopes it has ‘worked no blight there.’ At all events,
‘there are six days in the week ... on which its perusal could do no
mischief.’ Doubtless the moral lessons are commonplace enough,
but their triteness is relieved by the literary quality. Puritanism
without its narrowness, and sentimentalism controlled by humor and
good sense, lie at the basis of Reveries of a Bachelor and Dream
Life. The character of their author is to be plainly if not completely
read in these two books.
The distinctive flavor of ‘Ik Marvel’s’ literary style may be got in
the pleasing volume entitled Fresh Gleanings. Limpidity, grace, ease,
are among the virtues of his prose. The fabric of words is light, airy,
richly colored at times, but not over colored. With due recognition of
his individuality it may be said that ‘Ik Marvel’ was a literary son of
‘Geoffrey Crayon.’ The sweetness, the leisurely flow of the narrative,
the unobtrusiveness of manner, all suggest Irving. Perhaps Mitchell
meant to acknowledge his literary paternity when he dedicated
Dream Life to the author of The Sketch Book. But while we
recognize this debt to Irving it is most important that we do not
exaggerate it.
One marked exception must be made. There is no hint of Irving
in Battle Summer, an account of the Revolution of 1848, every page
of which echoes more or less distinctly the voice of Carlyle. So close
is the imitation at times as to awaken a doubt whether Battle
Summer was not intended for a ‘serious parody.’ At all events, it is
one of many proofs of the strong hold the History of the French
Revolution had on the minds of young men.
III
THE WRITINGS
Fresh Gleanings is a volume of travel, written in a way to persuade
one of the uselessness of pictorial illustrations. Its manner
occasionally suggests Sterne’s Sentimental Journey, which the
young traveller may have been reading of late. Sentiment and humor
are agreeably blended. Under ‘Ik Marvel’s’ guidance one visits Paris,
Limoges, Arles, Nîmes, Montpellier, Rouen, carefully avoiding the
‘objects of interest’ and learning much about the life. A less
courageous writer would have told us more and shown us less.
Books like this always contain interpolated stories, told around
the inn fire, or over the half-cup at the café. The ‘Story of Le Merle,’
‘An Old Chronicle of the City,’ ‘Hinzelmann,’ and ‘Boldo’s Story’ are
graceful, but so brief as to seem mere anecdotes.
The Lorgnette, consisting of the lucubrations of one ‘John
Timon,’ is an amusing and instructive periodical. Not its least
entertaining feature is the account of the literary distempers of the
day, the Tupper fever, the Festus outbreak, the Jane Eyre malady,
and the Typee disorder, together with other literary epidemics.
Neither The Lorgnette nor Fudge Doings is now much read. But if
the modern cynic, who takes, possibly, a condescending attitude
towards these old satires on fashionable life, will but pick up a copy
of Fudge Doings and try a few chapters, he will be forced to admit
that if we should not to-day think of writing satire in this manner, it
may have been a good way in 1855. Perchance in opening the
volume at random he comes on the account of the adventure of
Wash. Fudge with the black domino. In which case he will find
himself betrayed into reading two chapters at least, for he must
needs take the trouble to learn how the affair ended.
Fudge Doings and The Lorgnette may be looked on as a
contribution to the history of manners. By their aid one reconstructs
the drama of fashionable life in the mid-century, sees what was then
thought monstrous, and incidentally learns how simple the vices of
the grandfathers were.
Reveries of a Bachelor ushers one into a quaint and delightful
world. The reveries are of love—whether, in the words of Robert
Burton quoting Plotinus, ‘it be a God, or a divell, or passion of the
minde.’ The book is by no means compounded exclusively of
moonshine and roses. Some of the pictures are calculated to give a
bachelor pause. Here is Peggy who loves you, or at least swears it,
with her hand on the Sorrows of Werther. She is not bad looking,
Peggy, ‘save a bit too much of forehead.’ But she is ‘such a sad blue’
who will spend her money on the ‘Literary World’ and the Friends in
Council.
By the severer standards of our day Peggy was not so much of a
‘blue.’ None the less she is distinctly literary. She reads Dante and
‘funny Goldoni’ and leaves spots of baby-gruel on a Tasso of 1680.
She adores La Bruyère; even reads him while nurse gets dinner and
‘you are holding the baby.’
The vision presently becomes terrific and can only be dispelled
by a vicious kick at the forestick. Revery, misnamed idleness, has its
uses. Whatever else comes true, the Bachelor will not marry a young
woman who consoles her husband for an ill-cooked dinner by
quotations from the Greek Anthology.
Dream Life is also a collection of ‘reveries.’ Under the similitude
of the seasons, the author has pencilled little sketches of boyhood,
youth, manhood, and age. The temptation to the obvious in morals
and sentiment must have been great; but again Mitchell’s literary skill
and his humor carry him through successfully.
Seven Stories with Basement and Attic is a group of narratives
drawn from the author’s ‘plethoric little note books of travel.’ The
‘Basement’ is the introduction, the ‘Attic’ the conclusion. The first
story, ‘Wet Day at an Irish Inn,’ shows how, if he be observant, a man
may have adventures without taking the trouble to cross the street in
search of them. Three of the stories are French (‘Le Petit Soulier,’
‘The Cabriolet,’ and ‘Emile Roque’); another is Swiss (the ‘Bride of
the Ice King’); yet another is Italian (‘Count Pesaro’), and all are
exquisite, written in a style which for sweetness and unaffected ease
is, if not a lost art, at all events a neglected one. It has been said that
our young men would not care to write in this fashion to-day; it is a
question whether our young men would be able to do so.
One novel stands to ‘Ik Marvel’s’ credit, Doctor Johns, a story of
a New England country parsonage, well written because its author
could not write otherwise, faithful and exact because he knew the
life, yet going no deeper than other attempts to explain the New
England character, the externals of which are so easy to portray and
the real essence so baffling.
Among the best of ‘Ik Marvel’s’ books are those dealing with
rural life. My Farm of Edgewood sets forth the author’s adventures in
buying a country home, and his subsequent adventures in settling
therein and making life variously profitable. It is a successful attempt
to magnify the office of gentleman-farmer. The attractiveness of the
life is not over-emphasized, nor is it pretended that that is legitimate
farming which produces big crops regardless of expense.
The picture as a whole is seductive in ways not to be referred to
the literary skill of the artist. It is odd enough how a lay-reader,
unused to carrots and cabbages, will follow every detail of Mitchell’s
experiment. Here must be some outcroppings of the primitive
instinct. Moreover, the book relates to home-making, a subject
perennially dear to the American heart. Our restlessness has never
unsettled us in that regard.
Wet Days at Edgewood is a companion volume. The days here
celebrated, nine in number, were made bright by readings about ‘old
farmers, old gardeners, and old pastorals.’ Rejoicing in the strong
common sense of ancient writers on husbandry, and in the quaint
flavor of their style, ‘Ik Marvel’ chats of Roman farm and villa life,
recalling what Varro and Columella had to say about the art of tilling
the soil. He takes pleasure in the reflection that ‘yon open furrow ...
carries trace of the ridging in the “Works and Days;” that the brown
field of half-broken clods is the fallow (Νεός) of Xenophon,’ and that
‘Cato gives orders for the asparagus.’
Then he comes to modern times, to the days of Thomas Tusser,
Sir Hugh Platt, Gervase Markham, Samuel Hartlib, Jethro Tull, and
William Shenstone, men who farmed practically, or theoretically, or
even poetically. ‘Ik Marvel’ loves them all, even those whose
enthusiasm was in the ratio of their helplessness. No less dear to
him is Goldsmith, who wrote what passes for a rural tale and is not
rural at all, but comically urban, and Charles Lamb, who hated the
country and gladly avowed it.
These are Mitchell’s principal works. Having read thus far, it were
a pity to overlook the two volumes on English Lands, Letters, and
Kings, and a greater pity to overlook the instructive and entertaining
American Lands and Letters. In brief, the reader who insists on
knowing ‘Ik Marvel’ only by Reveries of a Bachelor does his author
an injustice and robs himself of many hours of literary delight.
Sentimentalism will always manifest itself in literature in one form
or another. That there will be a return to the manner which we
associate with ‘Ik Marvel’ is not likely, yet it was sentimentalism in its
manliest form. The continued popularity of Reveries of a Bachelor
suggests that Americans of to-day are not quite as cynical and
irreverent as they are sometimes painted, or as they love to paint
themselves.
FOOTNOTES:
61
There were to have been two volumes of Battle Summer,
called respectively the ‘Reign of the Blouse’ and the ‘Reign of
the Bourgeoisie.’ Only the first was published.
62
Reprinted under the title Out-of-Town Places, 1884.
XVIII
JAMES RUSSELL LOWELL
REFERENCES:
F. H. Underwood: The Poet and the Man:

Recollections and Appreciations of James
Russell Lowell, 1893.
E. E. Hale: James Russell Lowell and his Friends,
1899.
H. E. Scudder: James Russell Lowell, a Biography,
1901.
Ferris Greenslet: James Russell Lowell, his Life and
Work, 1905.
I
HIS LIFE
The Lowells of New England are descendants of Percival Lowell, a

prosperous Bristol merchant who came to America in 1639 and settled
at Newbury, Massachusetts. The family has been distinguished through
its various representatives for public spirit and business acumen as well
as for a devotion to letters. The grandfather of the poet, Judge John
Lowell, was author of the clause in the Bill of Rights abolishing slavery
in Massachusetts. One of his sons was founder of the great
manufacturing city on the Merrimac which bears his name. A grandson
established the Lowell Institute, a system of popular instruction by free
courses of lectures,—a system unique, in that it aims to bring to its
audiences representative scholars, chosen less for their skill in the
graceful but often specious art of public speaking than for solid
attainments.
James Russell Lowell, the youngest son of the Reverend Charles
Lowell, minister of the West Church in Boston, was born at Cambridge,
Massachusetts, in the colonial mansion known as ‘Elmwood,’ on
February 22, 1819. His mother, Harriet (Spence) Lowell, was a
63
daughter of Keith Spence, of Portsmouth, New Hampshire.
Under William Wells (an English pedagogue of the old school)
Lowell prepared for college, entered Harvard, and after some
disciplinary tribulations was graduated with his class (1838). He studied
law and was admitted to the bar (August, 1840), but remained briefless
during the few months of his efforts to begin a practice.
While waiting for clients, he busied himself with literature. He was
early a rhymer. At twelve years of age his skill in making verse had
astonished his schoolfellows, one of whom rushed home in great
excitement to announce that ‘Jemmy Lowell thought he was going to
be a poet.’
With the fearlessness of youth and in the hope of bettering himself
financially, Lowell, aided by his friend Robert Carter, started a
magazine, ‘The Pioneer.’ According to the prospectus, dated October
15, 1842, the editors proposed to supply ‘the intelligent and reflecting
portion of the Reading Public with a substitute for the enormous
quantity of thrice diluted trash, in the shape of namby-pamby love tales
and sketches, which is monthly poured out to them....’ Only three
64
numbers of ‘The Pioneer’ were issued. The ‘Reading Public’ was
joined to its idols and declined to encourage ‘a healthy and manly
Periodical Literature.’
In 1841 was published A Year’s Life, Lowell’s first volume of verse;
it was followed by Poems (1844), by a volume of prose, Conversations
on Some of the Old Poets (1845), and by Poems, ‘second series’
(1848).
The ‘Ianthe’ of A Year’s Life was easily identified with Maria White,
the gifted and beautiful girl who, in December, 1844, became the poet’s
wife. The first year of their married life was passed in Philadelphia,
whither Lowell had taken his bride to protect her from the harsh New
England winter. Their financial resources were few, but of gayety and
courage there was no lack. Lowell aspired to live by his pen. What with
the small sums paid him (rather against his will) for editorial work on
‘The Pennsylvania Freeman,’ what with the hardly larger sums for
contributions to ‘Graham’s Magazine’ and ‘The Broadway Journal,’ he
managed to subsist.
Nevertheless, it seemed best for a number of reasons that the
young people return to Cambridge and make a common home at
‘Elmwood’ with Lowell’s parents. In June of this year (1846) appeared
‘A Letter from Mr. Ezekiel Biglow of Jaalam to the Hon. Joseph T.
Buckingham, editor of the Boston Courier, inclosing a poem of his son,
Mr. Hosea Biglow.’ This was the first of The Biglow Papers, the initial
attack of many attacks Lowell was to make on slavery with the
weapons of satire and ridicule. During 1847 three more ‘papers’ were
printed in the ‘Courier;’ the remaining five appeared in ‘The National
Anti-Slavery Standard.’
When the ‘Standard’ passed from the control of a board of editors
into the hands of Sydney Howard Gay, Lowell became a salaried
contributor, and for a time his name appeared as corresponding editor.
He was allowed a free hand. Abolitionist though he was, his
abolitionism was tempered with a deal of sympathy for slaveholders.
And he had interests which most reformers of the time lacked, a
passionate love of letters, for example. Hence it was that in the midst of
leader-writing he was penning A Fable for Critics and The Vision of Sir
Launfal.
The winter of 1851–52 Lowell spent with his family in Italy, and the
following spring and summer in journeyings through France, England,
Scotland, and Wales. In October he sailed for home, having as ship
companions Thackeray and Arthur Hugh Clough. Just a year later Mrs.
Lowell died (October 27, 1853). For months afterward Lowell was in
‘great agony of mind, and he had to force himself into those laborious
65
hours which one instinctively feels contain a wise restorative.’
He abounded in literary plans, some of which (and among them a
novel) were never carried out, whereas others, his papers in ‘Putnam’s
Magazine’ and his lectures on English Poetry, before the Lowell
Institute, were in a high degree successful. Each lecture of the Institute
course had to be given twice, so great was the demand for tickets.
Lowell was very nervous over his first platform experience, and not a
little pleased when he found that he could hold the audience an hour
and a quarter (‘they are in the habit of going out at the end of the
hour’). The singular merit of the lectures led to his being appointed to
the chair of belles-lettres at Harvard, just resigned by Longfellow. After
a year’s study abroad the new professor entered on his academic
duties (September, 1856).
In 1857 Lowell married Miss Frances Dunlap, of Portland, Maine.
She was a woman of reserved though gracious manners and rare
beauty, who through her serene temper and fine critical sagacity,
together with a keen sense of the humorous, exerted a most beneficent
influence on Lowell’s life.
The burdens of college work were not so heavy as to prevent
Lowell’s assuming the editorship of ‘The Atlantic Monthly,’ a new
literary magazine with an anti-slavery bias. He held this post from 1857
to 1861, and proved to be one of the best of editors, though routine
was irksome to him, and the vagaries of contributors called for more
patience than he could at all times command. Two years after leaving
the ‘Atlantic’ he undertook to edit the ‘North American Review’ in
company with Charles Eliot Norton, on whom fell the chief
responsibilities. Lowell, for his part, contributed to the ‘Review’ many
notable papers on politics and literature.
The Civil War called out much of Lowell’s most spirited prose and
not a little of his best poetry. A second series of Biglow Papers
appeared in the ‘Atlantic,’ and for the commemoration of sons of
Harvard who had fought for the Union, Lowell wrote his magnificent
Commemoration Ode. This noble performance was literally an
improvisation, written in a single night.
At this point we may take note of Lowell’s publications, subsequent
to the Poems, ‘second series.’ They are: A Fable for Critics, 1848; The
Biglow Papers, 1848; Fireside Travels, 1864; The Biglow Papers,
‘second series,’ 1866; Under the Willows and Other Poems, 1869; The
Cathedral, 1870; Among My Books, 1870; My Study Windows, 1871;
Among My Books, ‘second series,’ 1876; Three Memorial Poems,
1877; Democracy and Other Addresses, 1887; Political Addresses,
1888; Heartsease and Rue, 1888.
There appeared posthumously Latest Literary Essays, 1891; The
Old English Dramatists, 1892; Letters of James Russell Lowell, edited
by C. E. Norton, 1893; Last Poems, 1895; The Anti-Slavery Papers of
James Russell Lowell, 1902.
Lowell resigned his professorship in 1872 and went abroad for two
years. Oxford conferred on him the degree of D. C. L. and Cambridge
that of LL. D.; it pleased him to regard the Cambridge degree ‘as in a
measure a friendly recognition of the University’s daughter in the
American Cambridge.’ In 1874 he returned home, and on the opening
of college was persuaded to resume his lectures.
During the presidential campaign of 1876 Lowell became politically
active in ways new to him. He was a delegate to the Republican
National convention and a presidential elector. His fellow-townsmen
had wished him to accept a nomination for representative in Congress;
but Lowell refused, believing himself unqualified for the post.
Not long after his inauguration President Hayes, at the instance of
W. D. Howells, offered Lowell the Austrian mission, an honor the poet
felt impelled to decline; when, however, it was learned that he would be
very willing to go to Spain, the appointment was made. He arrived in
Madrid on August 14, 1878. Two years later he was transferred to
England. Reappointed by President Garfield, he held this important
charge until the close of President Arthur’s administration.
Few ministers have been as popular as he. And not the least factor
of his popularity in England was his sturdy patriotism. Lowell was the
author of the essay ‘On a Certain Condescension in Foreigners,’ an
66
essay which an ingratiating Anglican clergyman says was meant to
be ‘overheard’ in England. It were more exact to say that the essay was
meant to be heard, and heard distinctly. ‘They honor stoutness in each
other,’ said Emerson, noting the traits of the English people. And it is
not unreasonable to believe that they also admire the same virtue in
others.
The summer of 1885 Lowell passed at Southborough, forty miles
from Boston, the home of his daughter, Mrs. Burnett. He made a
number of public addresses, gave a Lowell Institute course of lectures
on the ‘Old English Dramatists,’ argued the question of International
Copyright before a committee of the Senate, and is believed to have
had real influence in persuading representatives of this great country
that stealing is a sin. He found himself inveigled into an author’s
reading, and humorously bewailed his weakness in ever having written
a line of poetry. The demands upon him were enormous. It was now an
effort for him to do things, and if the grasshopper had not yet become a
burden, public occasions had, and more than once he was obliged to
beg off from keeping a promise inconsiderately made.
He enjoyed being in England for the summer, and usually divided
his time between London and Whitby. The last of these visits took place
in 1889. The ensuing winter he gave to a careful revision of his writings.
In the spring of 1890 he was ill for six weeks, and though he recovered
enough to be able to move about a little and to welcome his friends,
serious work was out of the question. He wrote two or three short
papers, and had strong inducements held out to him to write more, but
the time for writing was past, and he knew it.
His sufferings during his last illness were great, but he bore them
like the man he was. Lowell died at ‘Elmwood,’ Cambridge, on August
12, 1891.
II
LOWELL’S CHARACTER
‘I am a kind of twins myself, divided between grave and gay,’ said
Lowell, in one of those rare moments when he condescended to self-
analysis. The duality of temperament here pointed at is one secret of
the fascination he exerted on all who were privileged to know him
intimately. The fascination was certainly great and the tributes to it
numerous. Lowell’s personality was so winning, and the man was so
genuine, human, and lovable, that it is difficult to speak of him in terms
having even the semblance of impartiality. Although strong-willed and
positive, not indisposed now and then to indulge himself in the luxury of
stubbornness, he was open-minded, wholly unselfish, kind-hearted,
affectionate, and gentle; and while he had his reserves he was
democratic in all the best senses of the word, for his democracy sprang
from the depths of his nature. Changeable in his moods, he could be
teasing, whimsical, irritating; but when he was most mocking and
perverse he was most delightful.
There is something very attractive in Lowell’s attitude toward
literature and literary fame. Books were an essential part of his life. He
had mastered that difficult art of reading as few men have mastered it.
He was rarely endowed as a poet and prose-writer. And yet Lowell, the
most complete illustration we have of the literary man, showed no
inclination to magnify the importance of letters.
As to his individual achievements, he not only never thought of
himself more highly than he ought to think, but was the rather inclined
to place too low an estimate on the value of his work. Self-distrust
increased with years. Nevertheless, Lowell indulged himself in no
philosophy of despair. He had had much to be grateful for. ‘I have
always believed that a man’s fate is born with him, and that he cannot
escape from it nor greatly modify it’ (Lowell once wrote to his friend
Charles Eliot Norton) ‘and that consequently every one gets in the long
run exactly what he deserves, neither more nor less.’ Lowell goes on to
say that the creed is a ‘cheerful’ one; he might have added that it is no
less sensible and manly than it is cheerful.
Whether he found his creed satisfactory at all times or was always
conscious that he had a creed, we cannot know, but he could be the
blithest of fatalists when it pleased him to be.
III
POET AND PROSE WRITER
Lowell’s prose is manly, direct, varied, flexible, generally harmonious,
abounding in passages marked by grace, beauty, and sweetness, and
capable of rising to genuine eloquence. In its overflowing vitality and
human warmth it is an adequate expression of the man, imaging his
mocking and humorous moods no less than his deep sincerity, his
strength of purpose, and his passion. Much of it has the confidence and
ease that go with successful improvisation. If Lowell was ‘willing to risk
the prosperity of a verse upon a lucky throw of words,’ he was even
more willing to take like chances with his prose.
His thought ran easily into figurative form, and the making of
metaphor was as natural to him as breathing. He would even amuse
himself with conceits, for he loved to play with language, to force words
into shapes he might perchance have condemned had he found them
in the work of another. But if style is to be representative, this
playfulness, however annoying to Lowell’s critics, is a virtue. A Lowell
chastened in his English and wholly academic would not be the Lowell
we rejoice in.
He practised the art of poetry in many forms and always with
success. Of everything he wrote you might say that it had been his
study, though you might refrain from saying that ‘it had been all in all his
study.’ In other words, as we read Lowell the question never arises
whether or not the poet is working in unfamiliar materials, but whether
he might not have given his product a higher finish, the materials and
the form remaining the same. He was no aspirant after flawless beauty.
He wrote spontaneously and was for the time wholly possessed by his
theme. But what he had written he had written; and if never content
with the result he at least compelled himself to be philosophical. He
made a few changes, to be sure, but (as was said of a far greater poet)
he would correct with an afterglow of poetic inspiration, not with a
painful tinkering of the verse.
It is by tinkering with the verse, however (the ‘higher’ tinkering), that
perfection is attained. And he who wrote with evident ease so many
lovely and felicitous lines could as easily have bettered lines that are
wanting in finish. It was not Lowell’s way. Too much may not be
required of a man who often felt the utmost repugnance to reading his
own writings, once they were in print.
IV
POEMS, THE BIGLOW PAPERS, FABLE FOR
CRITICS, VISION OF SIR LAUNFAL
Lowell’s first poetic flights were strong-winged. ‘Threnodia,’ ‘The
Sirens,’ ‘Summer Storm,’ ‘To Perdita, Singing,’ whatever their faults,
have a richness, a melody, a freedom of structure, an almost careless
grace, that are captivating. Here was no painful effort in production with
the inevitable result of frigidity and hardness.
The poet’s gift matured rapidly. There is strength in such poems as
‘Prometheus,’ ‘Columbus,’ ‘A Glance behind the Curtain,’ rare beauty in
‘A Legend of Brittany,’ ‘Hebe,’ and ‘Rhœcus,’ a mystical power in the
haunting lines of ‘The Sower,’ passion and uplift in ‘The Present Crisis,’
‘Anti-Apis,’ the lines ‘To W. L. Garrison,’ and the ‘Ode to France,’ while
in ‘An Interview with Miles Standish’ is a promise of that satirical power
which was presently to find complete expression in The Biglow Papers.
Early in his career Lowell announced his theory of the poet’s office,
which is to inspire to high thought and noble action, not merely to
please with pretty fancies and melodious verse. The ‘Ode,’ written in
1841, is an expression of his poetic faith. The ethical and reforming
bent in Lowell’s character was so strong as to make it difficult for him,
true bard though he was, to look on poetry as an art to be cultivated for
itself alone.
Inspiriting as were stanzas like ‘The Present Crisis,’ Lowell’s power
became most effective in the anti-slavery struggle when the outbreak of
the Mexican War led to the writing of The Biglow Papers. Printed
anonymously in a journal, copied into other newspapers, the question
of their authorship much debated, these satires were at last adjudicated
to the man who wrote them, but not until he himself had heard it
demonstrated ‘in the pauses of a concert’ that he was wholly incapable
of such a performance.
Of the characters of the little drama, Hosea Biglow, the country
youth, stands for the plain common-sense of New England, opposed to
the extension of slavery whatever the means employed, and above all
by legalized murder with an accompaniment of drums and fifes. The
Reverend Homer Wilbur acts as ‘chorus,’ and by his learned comments
surrounds the productions of the country muse with an atmosphere of
scholarship. Birdofredom Sawin is the clown of the little show.
Many finer touches have become obscure by the lapse of time, and
The Biglow Papers is now provided with historical notes; but the
energy, the spirit, and the unfailing humor of the work are perennial.
Lowell was most fortunate in his verbal felicities. Who could have
foreseen that so much danger lurked in a middle initial, or that a plain
name of the sort borne by the former senator from Middlesex contained
such comic potentialities?
We were gittin’ on nicely up here to our village,

With good old idees o’ wut’s right an’ wut aint,
We kind o’ thought Christ went agin war an’ pillage,
An’ thet eppyletts worn’t the best mark of a saint;
But John P.
Robinson he
Sez this kind o’ thing’s an exploded idee.
Lowell was surprised at his own success. What he at first thought ‘a

mere fencing stick’ proved to be a weapon. The blade was two-edged,
and the Yankees did well to fall back a little when he lifted it against the
enemy. For in writing The Biglow Papers Lowell took real delight in
noting the oddities and laughing at the foibles of his own New
Englanders, a people whom he loved with all tenderness, but to whose
faults he was not in the least blind.
In 1861 the little puppets were taken out of the box where they had
lain for fifteen years and furbished up for a new tragi-comedy. The
second series of The Biglow Papers was read no less eagerly than the
first had been. Quite as brilliant as their predecessors, the later poems
are more impassioned, and in those touching on English hostility to the
North the satire is bitterly stinging.
While the numbers of the first series were in course of publication
Lowell produced a rhymed primer of contemporary American literature
under the title of A Fable for Critics. It was an improvisation, and
therefore the buoyancy, the jovial off-hand manner, the impudence
even, were a matter of course and all in its favor. Often penetrating and
just in his criticisms, Lowell was invariably amusing, and in the
cleverness of the rhyme and word play quite inimitable.
Two months after the appearance of the Fable the popular Vision of
Sir Launfal was published. Though undoubtedly read more for the sake
of the preludes than for the slight but touching story, it is by no means
certain that the preludes, brought out as independent poems, could
have won the number of readers they now have. In other words, The
Vision of Sir Launfal has a unity which it seems on first acquaintance to
lack.
V
UNDER THE WILLOWS, THE CATHEDRAL,
COMMEMORATION ODE, THREE MEMORIAL
POEMS, HEARTSEASE AND RUE
‘Under the Willows’ is a poem of Nature in which the poet at no time
loses sight either of the world of books or of the world of men. If he be
driven indoors by the rigors of May, he is content to sit by his wood-fire
and read what the poets have said in praise of that inclement month. Or
if June has come and he can dream under his favorite willows, his
reveries gain a zest from the interruptions of the tramp, ‘lavish
summer’s bedesman,’ the scissors-grinder, that grimy Ulysses of New
England, the school-children, and the road-menders,
Vexing Macadam’s ghost with pounded slate.
It is a poem of thanksgiving in which the poet voices his gratitude
for the benediction of the higher mood and the human kindness of the
lower.
The volume to which ‘Under the Willows’ gives its name is typical.
He who prizes Lowell’s verse will hardly be content with any selection
which does not include ‘Al Fresco,’ ‘A Winter-Evening Hymn to my
Fire,’ ‘Invita Minerva,’ ‘The Dead House,’ ‘The Parting of the Ways,’
‘The Fountain of Youth,’ and ‘The Nightingale in the Study.’
Its manner of contrasting To-Day with Yesterday, the genius that
creates with the spirit that analyzes, makes The Cathedral an
essentially American poem. The minster in its ‘vast repose,’
Silent and gray as forest-leaguered cliff,
must always seem a marvel to a dweller among temples of ‘deal and
paint.’ The poem is the meditation of a New-World conservative,
altogether catholic of sympathies, who holds no less firmly to the past
because, under the fascination of democracy, he breathes in the
presence of the ‘backwoods Charlemagne’ a braver air and is
conscious of an ‘ampler manhood.’ And what, he asks, will be the faith
of this new avatar of the Goth, what temples will the creature build?
Very beautiful, very suggestive, and in its shifting moods entirely
representative of the poet who wrote it must this fine work always
seem.
The Ode recited at the Harvard Commemoration (July 21, 1865) is
Lowell’s supreme achievement in verse. It breathes the most exalted
patriotism, a love of native land that is intense, fiery, consuming.
Though written in honor of sons of the University who had gone to the
war, the spirit of the Ode is not local and particular. The poet celebrates
not individual deeds alone but the sum of those deeds, not man but
manhood:—
That leap of heart whereby a people rise

Up to a noble anger’s height,
And, flamed on by the Fates, not shrink, but grow more bright,
That swift validity in noble veins,
Of choosing danger and disdaining shame,
Of being set on flame
By the pure fire that flies all contact base,
But wraps its chosen with angelic might,
These are imperishable gains,
Sure as the sun, medicinal as light,
These hold great futures in their lusty reins
And certify to earth a new imperial race.
The mingling of proud humility, tenderness, and reverence, the
throbbing passion and the exultant fervor of the concluding verses, lift
this ode to a high place in American poetry, it may be to the highest
place. To the many, however, the chief value of The Commemoration
Ode lies in the stanza on Lincoln. So just as an estimate of character,
so restrained in its accents of praise, American in all finer meanings of
the word, splendid in its imagery and poignant in the note of grief, this
beautiful tribute to the great president is final and satisfying.
The first of the Three Memorial Poems is an ‘Ode, read at the One
Hundredth Anniversary of the Fight at Concord.’
In the opening stanzas on Freedom the poet strikes the notes of
exultation fitting the time and the place, then passes to those inevitable
allusions which appeal to local pride (and Lowell handles this passage
with utmost skill), draws the lesson that must of necessity be drawn
from the ‘home-spun deeds’ of the men of old, makes Freedom utter
her warning to the men of the present, and, no prophet of evil, closes in
the triumphant spirit in which he began.
‘Under the Old Elm’ is a magnificent tribute to a man so great that
there is need of odes like this to help us comprehend his greatness.
After calling up the scene when Washington, ‘a stranger among
strangers,’ stood beneath that legendary tree to take command of his
army, ‘all of captains,’ a motley rout, valorous deacons, selectmen, and
village heroes among others, more skilled in debating their orders than
obeying them, good fighters all, but ‘serious drill’s despair,’—the poet
chants those beautiful lines in which is drawn the distinction between
‘Nation’ and ‘Country.’ The one is fashioned of computable things, good
each in its kind and important in its place:—
But Country is a shape of each man’s mind
Sacred from definition, unconfined
By the cramped walls where daily drudgeries grind;
An inward vision, yet an outward birth
Of sweet familiar heaven and earth;
A brooding Presence that stirs motions blind
Of wings within our embryo being’s shell
That wait but her completer spell
To make us eagle-natured, fit to dare
Life’s nobler spaces and untarnished air.
You who hold dear this self-conceived ideal,

Whose faith and works alone can make it real,
Bring all your fairest gifts to deck her shrine
Who lifts our lives away from Thine and Mine
And feeds the lamp of manhood more divine
With fragrant oils of quenchless constancy.
When all have done their utmost, surely he
Hath given the best who gives a character
Erect and constant, which nor any shock
Of loosened elements, nor the forceful sea
Of flowing or of ebbing fates, can stir
From its deep bases in the living rock
Of ancient manhood’s sweet security....
And the poet longs for skill to praise him fitly whom he does fitly praise
in the stanzas that follow. It is a thoughtful, nobly eloquent, and
poetically beautiful characterization of the great Virginian, and
appropriately closes with a fine apostrophe to the historic
Commonwealth from which Washington sprang.
The ‘Ode for the Fourth of July, 1876,’ though not lacking in forceful
lines and fine imagery, is the least happy of the three poems. The
questioning and critical mood is prominent. But the spirit of confidence
prevails and is voiced in the invocation with which the ode concludes.
Various notes are touched in the collection of eighty-eight poems to
which its author gave the title of Heartsease and Rue. Here are verses
new and old, grave and gay, satirical, humorous, sentimental, and
elegiac, epigrams, inscriptions, lyrics, poems of occasion, sonnets,
epistles, and, chief among them, the ode written on hearing the news of
the death of Agassiz. Whether, as has been asserted, ‘this poem takes
its place with the few great elegies in our language, gives a hand to
“Lycidas” and to “Thyrsis,”’ is a question to be decided by the suffrages
of many good critics, rather than by the dictum of one. There is no
doubt, however, that by virtue of its human quality, depth of personal
feeling, sincerity in the accent of bereavement, and felicity of phrase,
the ‘Agassiz’ will always stand in the first rank of Lowell’s greater verse.
VI
FIRESIDE TRAVELS, MY STUDY WINDOWS,
AMONG MY BOOKS, LATEST LITERARY
ESSAYS
Fireside Travels is so entertaining a book as to make one wish that
Lowell had chronicled more of his journeyings at home and abroad in
the same amusing style. Two of the six essays—‘Cambridge Thirty
Years Ago’ and ‘A Moosehead Journal’—take the form of letters
addressed to the author’s friend, ‘the Edelmann Storg’ (W. W. Story).
The others are grouped under the general title of ‘Leaves from my
Journal in Italy and Elsewhere.’
One spirit animates the pages of this book,—a love of plain people,
homely adventures, everyday sights and sounds. In a half-serious way
(as if to show that he knows how to ‘do’ a tempest in the mountains or
an illumination of St. Peter’s) Lowell throws in a number of
unconventional passages on entirely conventional themes. But the
strength of the book lies in the sympathetic and humorous accounts of
that protean animal Man, who, whether he showed himself in the guise
of a denizen of Old Cambridge, or of Uncle Zeb, who had been ‘to the
‘Roostick war,’ or of the Chief Mate of the packet ship, or of Leopoldo,
the Italian guide, was more interesting to Lowell than any other object
of his study.

Wireless Blockchain Principles Technologies and Applications 1st Edition Bin Cao Lei Zhang Mugen Peng Muhammad Ali Imran Eds

Uploaded by

Copyright:

Available Formats

You might also like

Wireless Blockchain Principles Technologies and Applications 1st Edition Bin Cao Lei Zhang Mugen Peng Muhammad Ali Imran Eds

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Wireless Blockchain Principles Technologies and Applications 1st Edition Bin Cao Lei Zhang Mugen Peng Muhammad Ali Imran Eds

Uploaded by

Copyright:

Available Formats

Wireless Blockchain Principles

Technologies and Applications 1st

Wireless Blockchain Principles Technologies and

Autonomous Airborne Wireless Networks (IEEE Press) 1st

Eco-Friendly Corrosion Inhibitors: Principles,

Networked Microgrids 1st Edition Peng Zhang

Wireless Communication with Artificial Intelligence:

3-D Computer Vision: Principles, Algorithms and

Molecular Characterization of Polymers: A Fundamental

Wireless Communication: Advancements and Challenges

Principles, Technologies and Applications

Muhammad Ali Imran

Published by John Wiley & Sons Ltd., Chichester, United Kingdom.

Library of Congress Cataloging-in-Publication Data

Cover Design: Wiley

Set in 9.5/12.5pt STIXTwoText by Straive, Chennai, India

List of Contributors xiii

1 What is Blockchain Radio Access Network? 1

2 Consensus Algorithm Analysis in Blockchain: PoW and Raft 27

2.2.1 Blockchain Preliminaries 30

3 A Low Communication Complexity Double-layer PBFT Consensus 73

3.1.4.3 PBFT Protocol 76

4 Blockchain-Driven Internet of Things 93

4.3.1.4 Security Analysis 105

5 Hyperledger Blockchain-Based Distributed Marketplaces for

6 Blockchain for Spectrum Management in 6G Networks 137

7 Integration of MEC and Blockchain 161

8 Performance Analysis on Wireless Blockchain IoT System 179

9 Utilizing Blockchain as a Citizen-Utility for Future Smart Grids 201

10 Blockchain-enabled COVID-19 Contact Tracing Solutions 225

11 Blockchain Medical Data Sharing 245

11.1.4.1 Precision Medicine 261

12 Decentralized Content Vetting in Social Network with

Nima Afraz Bin Cao

Jaafar M. H. Elmirghani Raja Jurdak

Wenyu Li Marco Rufﬁni

Asuquo A. Okon Purple Mountain Laboratories

Qi Xia Lei Zhang

In Chapter 1, the authors provide an overview of blockchain radio access network

Chapter 5 addresses the issues associated with centralized marketplaces in 5G networks.

Bin Cao, Lei Zhang, Mugen Peng, Muhammad Ali Imran

3GPP third-Generation Partnership Project

CoAP Constrained application protocol

GPS Global Position System

NAT nucleic acid testing

S-GW Serving gateway

What is Blockchain Radio Access Network?*

Wireless Blockchain: Principles, Technologies and Applications, First Edition.

Table 1.1 Important variables in the modeling and analysis

Symbols Explanations Symbols Explanations

𝜏kc Required service time of request k tka Arrival epoch of request k

● Based on the modeling and analysis, we uncover an inherent trade-off relationship

1.2 What is B-RAN

1.2.1 B-RAN Framework

Massive UEs supporting diverse applications

APs belonging to different SPs

Figure 1.1 Conceptual illustration of self-organized B-RAN.

Blockchain Block h Block h + 1 Block h + N – 1