Unmanned Aerial Vehicle Applications over Cellular Networks for 5G and Beyond (eBook)

Preface 6
Contents 8
Acronyms 10
1 Overview of 5G and Beyond Communications 12
1.1 Background and Requirements 12
1.2 UAV Applications 13
1.2.1 Flying Infrastructure 14
1.2.2 Aerial Internet-of-Things 15
1.3 Current State-of-the-art 17
1.3.1 Channel Model 17
1.3.1.1 Elevation Angle-Based Model 17
1.3.1.2 3GPP Model 18
1.3.2 Aerial Access Networks 20
1.3.3 Aerial IoT Networks 24
1.3.4 Propulsion and Mobility Model 32
References 36
2 Basic Theoretical Background 37
2.1 Brief Introductions to Optimization Theory 37
2.1.1 Continuous Optimization 38
2.1.1.1 Convex Optimization Problem 38
2.1.1.2 Non-convex Optimization Problem 40
2.1.2 Integer Optimization 41
2.1.2.1 Branch-and-Bound Method 42
2.1.2.2 Bound Calculation 43
2.2 Basics of Game Theory 44
2.2.1 Basic Concepts 44
2.2.1.1 Definition of a Game 44
2.2.1.2 Nash Equilibrium 45
2.2.1.3 Examples of Game Theory 46
2.2.2 Contract Theory 47
2.2.2.1 Classification 47
2.2.2.2 Models and Reward Design 50
2.3 Related Machine Learning Technologies 52
2.3.1 Classical Machine Learning 52
2.3.1.1 Supervised Learning 53
2.3.1.2 Unsupervised Learning 54
2.3.1.3 Machine Learning Algorithm Design 55
2.3.2 Deep Learning 55
2.3.2.1 Basics of Neural Networks 56
2.3.2.2 Back-Propagation Algorithm 59
2.3.3 Reinforcement Learning 62
2.3.3.1 Markov Decision Processes 62
2.3.3.2 Reinforcement Learning Methods 65
References 70
3 UAV Assisted Cellular Communications 71
3.1 UAVs Serving as Base Stations 71
3.1.1 System Model 73
3.1.1.1 Mobility and Energy Consumption 74
3.1.1.2 Wireless Downlink Model 74
3.1.1.3 The Utility of the UAV Operators 75
3.1.1.4 Cost of the MBS Manager 76
3.1.1.5 Contract Formulation 77
3.1.2 Optimal Contract Design 78
3.1.2.1 Basic Properties 79
3.1.2.2 Optimal Pricing Strategy 81
3.1.2.3 Optimal Quality Assignment Problem 84
3.1.2.4 Algorithm for the MBS Optimal Contract 86
3.1.2.5 Socially Optimal Contract 89
3.1.3 Theoretical Analysis and Discussions 90
3.1.3.1 The Impact of the Height on the UAV Types 90
3.1.3.2 The Impact of the UAV Types on the Optimal Revenue 92
3.1.4 Simulation Results 92
3.1.4.1 Simulation Setups 93
3.1.4.2 Simulation Results and Discussions 93
3.1.5 Summary 98
3.2 UAVs Serving as Relays 99
3.2.1 System Model and Problem Formulation 99
3.2.2 Power and Trajectory Optimization 103
3.2.2.1 Trajectory Design 104
3.2.2.2 Power Control 105
3.2.3 Simulation Results 106
3.2.4 Summary 108
References 108
4 Cellular Assisted UAV Sensing 111
4.1 Cellular Internet of UAVs 111
4.1.1 System Model 112
4.1.1.1 UAV Sensing 112
4.1.1.2 UAV Transmission 113
4.1.2 Problem Formulation 114
4.1.2.1 Energy Consumption 114
4.1.2.2 Problem Description 115
4.1.3 Energy Efficiency Maximization Algorithm 115
4.1.3.1 UAV Sensing Optimization 116
4.1.3.2 UAV Transmission Optimization 117
4.1.3.3 Overall Algorithm 119
4.1.4 Simulation Results 119
4.1.5 Summary 121
4.2 Cooperative Cellular Internet of UAVs 121
4.2.1 System Model 122
4.2.1.1 UAV Sensing 123
4.2.1.2 UAV Transmission 124
4.2.1.3 Task Completion Time 125
4.2.2 Sense-and-Send Protocol 125
4.2.3 Problem Formulation 128
4.2.3.1 Problem Description 128
4.2.3.2 Problem Decomposition 129
4.2.3.3 Iterative Algorithm Description 130
4.2.4 Iterative Trajectory, Sensing, and Scheduling Optimization Algorithm 131
4.2.4.1 Trajectory Optimization 131
4.2.4.2 Sensing Location Optimization 134
4.2.4.3 UAV Scheduling 137
4.2.4.4 Performance Analysis 138
4.2.5 Simulation Results 141
4.2.6 Summary 147
4.3 UAV-to-X Communications 148
4.3.1 System Model 149
4.3.1.1 Scenario Description 149
4.3.1.2 Data Transmission 151
4.3.1.3 Channel Model 152
4.3.2 Cooperative UAV Sense-and-Send Protocol 154
4.3.3 Problem Formulation 156
4.3.3.1 Joint Subchannel Allocation and UAV Speed Optimization Problem Formulation 156
4.3.3.2 Problem Decomposition 158
4.3.4 Joint Subchannel Allocation and UAV SpeedOptimization 159
4.3.4.1 U2N and CU Subchannel Allocation Algorithm 159
4.3.4.2 U2U Subchannel Allocation Algorithm 161
4.3.4.3 UAV Speed Optimization Algorithm 165
4.3.4.4 Iterative Subchannel Allocation and UAV Speed Optimization Algorithm 168
4.3.5 Simulation Results 170
4.3.6 Summary 175
4.4 Reinforcement Learning for the Cellular Internet of UAVs 175
4.4.1 System Model 176
4.4.1.1 UAV Sensing 177
4.4.1.2 UAV Transmission 177
4.4.2 Decentralized Sense-and-Send Protocol 178
4.4.2.1 Sense-and-Send Cycle 178
4.4.2.2 Uplink Subchannel Allocation Mechanism 180
4.4.3 Sense-and-Send Protocol Analysis 181
4.4.3.1 Outer Markov Chain of UAV Sensing 181
4.4.3.2 Inner Markov Chain of UAV Transmission 182
4.4.3.3 Analysis on the Data Rate 185
4.4.4 Decentralized Trajectory Design 186
4.4.4.1 UAV Trajectory Design Problem 186
4.4.4.2 Reinforcement Learning Framework 188
4.4.4.3 Enhanced Multi-UAV Q-Learning Algorithm for UAV Trajectory Design 191
4.4.4.4 Analysis of Reinforcement Learning Algorithms 194
4.4.5 Simulation Results 196
4.4.6 Summary 201
4.5 Applications of the Cellular Internet of UAVs 201
4.5.1 Preliminaries of UAV Sensing System 203
4.5.1.1 System Overview 203
4.5.1.2 Dataset Description 204
4.5.1.3 Data Reliability 205
4.5.1.4 Selection of Model Parameters 206
4.5.2 Fine-Grained AQI Distribution Model 206
4.5.2.1 Physical Particle Dispersion Model 206
4.5.2.2 Neural Network Model 207
4.5.2.3 GPM-NN Model 208
4.5.3 Adaptive AQI Monitoring Algorithm 212
4.5.3.1 Complete Monitoring 213
4.5.3.2 Selective Monitoring 213
4.5.3.3 Trajectory Optimization 215
4.5.4 Application Scenario I: Performance Analysis in Horizontal Open Space 216
4.5.4.1 Scenario Description 216
4.5.4.2 Performance Analysis 217
4.5.5 Application Scenario II: Performance Analysis in Vertical Enclosed Space 222
4.5.5.1 Scenario Description 222
4.5.5.2 Performance Analysis 223
4.5.6 Summary 226
References 227