5G and Beyond Wireless Communication Networks

Haijian Sun, PhD, is an Assistant Professor in the School of Electrical and Computer Engineering at the University of Georgia in Athens, USA. His research interests include wireless communications for 5G and beyond, efficient edge computing, wireless security, and wireless for distributed learning. Rose Qingyang Hu, PhD, is a Professor in the Department of Electrical and Computer Engineering at Utah State University in Logan, USA. Her research interests include next-generation wireless communications, wireless network design and optimization, and more. Yi Qian, PhD, is a Professor in the Department of Electrical and Computer Engineering at the University of Nebraska-Lincoln in Omaha, USA. His research interests include cyber security and communication network security, computer networks, and wireless networks.

• About the Authors xiPreface xiiiAcknowledgments xv1 Introduction to 5G and Beyond Network 11.1 5G and Beyond System Requirements 11.1.1 Technical Challenges 21.2 Enabling Technologies 31.2.1 5G New Radio 31.2.1.1 Non-orthogonal Multiple Access (NOMA) 31.2.1.2 Channel Codes 51.2.1.3 Massive MIMO 51.2.1.4 Other 5G NR Techniques 61.2.2 Mobile Edge Computing (MEC) 61.2.3 Hybrid and Heterogeneous Communication Architecture for Pervasive IoTs 71.3 Book Outline 82 5G Wireless Networks with Underlaid D2D Communications 112.1 Background 112.1.1 MU-MIMO 112.1.2 D2D Communication 112.1.3 MU-MIMO and D2D in 5G 122.2 NOMA-Aided Network with Underlaid D2D 122.3 NOMA with SIC and Problem Formation 142.3.1 NOMA with SIC 142.3.2 Problem Formation 152.4 Precoding and User Grouping Algorithm 152.4.1 Zero-Forcing Beamforming 162.4.1.1 First ZF Precoding 162.4.1.2 Second ZF Precoding 162.4.2 User Grouping and Optimal Power Allocation 162.4.2.1 First ZF Precoding 172.4.2.2 Second ZF Precoding 182.5 Numerical Results 182.6 Summary 193 5G NOMA-Enabled Wireless Networks 213.1 Background 213.2 Error Propagation in NOMA 223.3 SIC and Problem Formulation 223.3.1 SIC with Error Propagation 233.3.2 Problem Formation 243.4 Precoding and Power Allocation 253.4.1 Precoding Design 253.4.2 Case Studies for Power Allocation 263.4.2.1 Case I 263.4.2.2 Case II 273.5 Numerical Results 273.6 Summary 304 NOMA in Relay and IoT for 5G Wireless Networks 314.1 Outage Probability Study in a NOMA Relay System 314.1.1 Background 314.1.2 System Model 324.1.2.1 NOMA Cooperative Scheme 324.1.2.2 NOMA TDMA Scheme 344.1.3 Outage Probability Analysis 354.1.3.1 Outage Probability in NOMA Cooperative Scheme 354.1.4 Outage Probability in NOMA TDMA Scheme 364.1.5 Outage Probability with Error Propagation in SIC 374.1.5.1 Outage Probability in NOMA Cooperative Scheme with EP 384.1.5.2 Outage Probability in NOMA TDMA Scheme with EP 384.1.6 Numerical Results 394.2 NOMA in a mmWave-Based IoT Wireless System with SWIPT 414.2.1 Introduction 414.2.2 System Model 414.2.2.1 Phase 1 Transmission 424.2.2.2 Phase 2 Transmission 444.2.3 Outage Analysis 454.2.3.1 UE 1 Outage Probability 454.2.3.2 UE 2 Outage Probability 454.2.3.3 Outage at High SNR 474.2.3.4 Diversity Analysis for UE 2 474.2.4 Numerical Results 474.2.5 Summary 485 Robust Beamforming in NOMA Cognitive Radio Networks: Bounded CSI 515.1 Background 515.1.1 RelatedWork and Motivation 525.1.1.1 Linear EH Model 525.1.1.2 Non-linear EH Model 535.1.2 Contributions 535.2 System and Energy Harvesting Models 545.2.1 System Model 545.2.2 Non-linear EH Model 555.2.3 Bounded CSI Error Model 555.2.3.1 NOMA Transmission 565.3 Power Minimization-Based Problem Formulation 565.3.1 Problem Formulation 575.3.2 Matrix Decomposition 595.4 Maximum Harvested Energy Problem Formulation 605.4.1 Complexity Analysis 615.5 Numerical Results 625.5.1 Power Minimization Problem 625.5.2 Energy Harvesting Maximization Problem 645.6 Summary 676 Robust Beamforming in NOMA Cognitive Radio Networks: Gaussian CSI 696.1 Gaussian CSI Error Model 696.2 Power Minimization-Based Problem Formulation 696.2.1 Bernstein-Type Inequality I 706.2.2 Bernstein-Type Inequality II 716.3 Maximum Harvested Energy Problem Formulation 726.3.1 Complexity Analysis 736.4 Numerical Results 736.4.1 Power Minimization Problem 746.4.2 Energy Harvesting Maximization Problem 766.5 Summary 797 Mobile Edge Computing in 5G Wireless Networks 817.1 Background 817.2 System Model 827.2.1 Data Offloading 837.2.2 Local Computing 837.3 Problem Formulation 837.3.1 Update pk, tk, and fk 857.3.2 Update Lagrange Multipliers 867.3.3 Update Auxiliary Variables 867.3.4 Complexity Analysis 877.4 Numerical Results 877.5 Summary 908 Toward Green MEC Offloading with Security Enhancement 918.1 Background 918.2 System Model 928.2.1 Secure Offloading 928.2.2 Local Computing 938.2.3 Receiving Computed Results 938.2.4 Computation Efficiency in MEC Systems 938.3 Computation Efficiency Maximization with Active Eavesdropper 948.3.1 SCA-Based Optimization Algorithm 948.3.2 Objective Function 958.3.3 Proposed Solution to P4 with given (λΚ, βΚ) 968.3.4 Update (λΚ, βΚ) 978.4 Numerical Results 978.5 Summary 1009 Wireless Systems for Distributed Machine Learning 1019.1 Background 1019.2 System Model 1029.2.1 FL Model Update 1029.2.2 Gradient Quantization 1049.2.3 Gradient Sparsification 1049.3 FL Model Update with Adaptive NOMA Transmission 1049.3.1 Uplink NOMA Transmission 1049.3.2 NOMA Scheduling 1059.3.3 Adaptive Transmission 1069.4 Scheduling and Power Optimization 1079.4.1 Problem Formulation 1079.5 Scheduling Algorithm and Power Allocation 1089.5.1 Scheduling Graph Construction 1089.5.2 Optimal scheduling Pattern 1099.5.3 Power Allocation 1109.6 Numerical Results 1119.7 Summary 11410 Secure Spectrum Sharing with Machine Learning: An Overview 11510.1 Background 11510.1.1 SS: A Brief History 11610.1.2 Security Issues in SS 11810.2 ML-Based Methodologies for SS 11910.2.1 ML-Based CRN 11910.2.1.1 Spectrum Sensing 12010.2.1.2 Spectrum Selection 12210.2.1.3 Spectrum Access 12310.2.1.4 Spectrum Handoff 12510.2.2 Database-Assisted SS 12510.2.2.1 ML-Based EZ Optimization 12610.2.2.2 Incumbent Detection 12610.2.2.3 Channel Selection and Transaction 12710.2.3 ML-Based LTE-U/LTE-LAA 12710.2.3.1 ML-Based LBT Methods 12810.2.3.2 ML-Based Duty Cycle Methods 12910.2.3.3 Game-Theory-Based Methods 12910.2.3.4 Distributed-Algorithm-Based Methods 13010.2.4 Ambient Backscatter Networks 13110.2.4.1 Information Extraction 13110.2.4.2 Operating Mode Selection and User Coordination 13210.2.4.3 AmBC-CR Methods 13310.3 Summary 13411 Secure Spectrum Sharing with Machine Learning: Methodologies 13511.1 Security Concerns in SS 13511.1.1 Primary User Emulation Attack 13511.1.2 Spectrum Sensing Data Falsification Attack 13511.1.3 Jamming Attacks 13611.1.4 Intercept/Eavesdrop 13711.1.5 Privacy Issues in Database-Assisted SS Systems 13711.2 ML-Assisted Secure SS 13811.2.1 State-of-the-Art Methods of Defense Against PUE Attack 13811.2.1.1 ML-Based Detection Methods 13811.2.1.2 Robust Detection Methods 14011.2.1.3 ML-Based Attack Methods 14111.2.2 State-of-the-Art Methods of Defense Against SSDF Attack 14211.2.2.1 Outlier Detection Methods 14311.2.2.2 Reputation-Based Detection Methods 14311.2.2.3 SSDF and PUE Combination Attacks 14411.2.3 State-of-the-Art Methods of Defense Against Jamming Attacks 14411.2.3.1 ML-Based Anti-Jamming Methods 14511.2.3.2 Attacker Enhanced Anti-Jamming Methods 14611.2.3.3 AmBC Empowered Anti-Jamming Methods 14811.2.4 State-of-the-Art Methods of Defense Against Intercept/Eavesdrop 14911.2.4.1 RL-Based Anti-Eavesdropping Methods 14911.2.5 State-of-the-Art ML-Based Privacy Protection Methods 15011.2.5.1 Privacy Protection for PUs in SS Networks 15011.2.5.2 Privacy Protection for SUs in SS Networks 15111.2.5.3 Privacy Protection for ML Algorithms 15111.3 Summary 15312 Open Issues and Future Directions for 5G and Beyond Wireless Networks 15512.1 Joint Communication and Sensing 15512.2 Space-Air-Ground Communication 15512.3 Semantic Communication 15612.4 Data-Driven Communication System Design 156Appendix A Proof of Theorem 5.1 157Bibliography 161Index 181