Driving 5G Mobile Communications with Artificial Intelligence towards 6G

Cover
Half Title
Title Page
Copyright Page
Table of Contents
Foreword
Preface
Editors
Contributors
Part 1 Advanced 5G Communication
	Chapter 1 5G-Advanced Mobile Communication: New Concepts and Research Challenges
		1.1 Introduction
			1.1.1 IMT-2020 Submission and Evaluation Process
			1.1.2 3GPP Standardization Activities
		1.2 5G-Advanced Transformational Phase
			1.2.1 5G System Architecture Options and Standardization Process
				1.2.1.1 5G System Reference Architecture
				1.2.1.2 5G NR Architecture
				1.2.1.3 5G Core Network
				1.2.1.4 Separation of Control Plane and User Plane
				1.2.1.5 RAN Protocol Architecture
				1.2.1.6 IAB Protocol and Physical Layer
				1.2.1.7 Physical Layer (PHY)
				1.2.1.8 Network Slicing (NS)
				1.2.1.9 Quality of Service (QoS)
				1.2.1.10 5G Security
			1.2.2 5G New Radio and Core Network Enhancement and Vertical Expansion
				1.2.2.1 Support for Industrial IoT Applications
				1.2.2.2 Ultra-Reliable Low-Latency Communication
				1.2.2.3 Support for V2X Connections
				1.2.2.4 Non-Public Networks
			1.2.3 Advanced Service Requirements and Performance Indicators
				1.2.3.1 Phase 1 and Phase 2
				1.2.3.2 Foundation for the Next Phase
				1.2.3.3 5G-Advanced
			1.2.4 Novel System AI&ML Paradigm
				1.2.4.1 AI-Enabled RAN Architecture
				1.2.4.2 Network Automation and Data Analytics Function
				1.2.4.3 5G Advanced Architecture and Technical Trends
		1.3 6G Concept, Research, and Transition Technologies
			1.3.1 Extreme System Performance and Network Evolution
				1.3.1.1 IMT Vision for 2030 and Beyond
			1.3.2 Technology Enablers and Research Programs
				1.3.2.1 Review of Global Activities and Research Programs
		1.4 Open Issues and Concluding Remarks
		References
	Chapter 2 5G Advanced Mobile Broadband: New Multimedia Delivery Platform
		2.1 Introduction
		2.2 Extensions of 5G Architecture for Common Media Delivery Platform
			2.2.1 Enhanced Mobile Broadband eMBB and Media Streaming 5G MS
				2.2.1.1 Downlink Data Transfer and Control Operation
				2.2.1.2 Uplink Data Transfer and Control Operation
				2.2.1.3 Media Streaming Principles and Architecture
			2.2.2 Evolution of Multimedia Multicast and Broadcast Services
				2.2.2.1 5G Broadcast Over Dedicated Terrestrial Network
				2.2.2.2 5G MBS Multicast and Broadcast Services
			2.2.3 Multimedia Production and Distribution Ecosystem
		2.3 Immersive Communication
			2.3.1 Immersive Media Over 5G
				2.3.1.1 Architectural Enhancements for Immersive Media Support
				2.3.1.2 Levels of Immersion and Technological Complexity
			2.3.2 Extremely Interactive Communication and Low Latency
				2.3.2.1 Immersive XR Communication
				2.3.2.2 Holographic Communication
			2.3.3 AI-Based Multimedia Tools
				2.3.3.1 DNNVC Video Codec
		2.4 Concluding Remarks
		Bibliography
	Chapter 3 5G Ultrareliable and Low-Latency Communication in Vertical Domain Expansion
		3.1 Introduction
		3.2 5G Vertical Domain Expansion
			3.2.1 uRLLC Services
			3.2.2 Key Characteristics of mMTC
			3.2.3 5G Private Networks, Energy Efficiency, and AI/ML Support
				3.2.3.1 Energy Efficiency
				3.2.3.2 AI&ML Support
		3.3 5G Critical and mMTC Connections
			3.3.1 Performance Trade-Offs for uRLLC in Industrial IoT
			3.3.2 V2X High-Bandwidth, Low-Latency, and Highly Reliable Communication
			3.3.3 AI-Enabled Massive IoT Toward 6G
		3.4 Concluding Remarks
		Bibliography
	Chapter 4 Vehicular Systems for 5G and Beyond 5G: Channel Modeling for Performance Evaluation
		4.1 Introduction
		4.2 Direct V2V Communications
			4.2.1 Direct V2V Communications Over Double-Scattered Fading Channels
			4.2.2 Performance Measures of Direct V2V Communications
			4.2.3 Numerical Results of Direct V2V Communications
		4.3 Relay-Assisted Dual-Hop V2V Communications
			4.3.1 Relay-Assisted V2V Communications Over DSc Fading Channels
			4.3.2 Performance Measures of Relay-Assisted V2V Communications
			4.3.3 Numerical Results for Relay-Assisted V2V Communications
		4.4 V2V Cooperative Dual-Hop Communications
			4.4.1 V2V Cooperative Dual-Hop Communications Over Double-Scattered Fading Channels
			4.4.2 Performance Measures of V2V Cooperative Dual-Hop Communications
			4.4.3 Numerical Results for V2V Cooperative Dual-Hop Communications
		4.5 V2V Cooperative Communications with Direct V2V Link
			4.5.1 Cooperative Communications with Direct V2V Link Over DSc Fading Channels
			4.5.2 Performance Measures of Cooperative Communications with Direct V2V Link
			4.5.3 Numerical Results for Cooperative Communications with Direct V2V Link
		4.6 V2V Mixed RF-FSO Communications
			4.6.1 Mixed V2V RF-FSO Communications Over Double-Scattered And TF Channels
			4.6.2 Performance Measures of Mixed V2V RF-FSO Communications
			4.6.3 Numerical Results for Mixed V2V RF-FSO Communications
		4.7 V2V Cooperative Mixed RF-FSO Communications with Direct and Relay-Assisted RF Links
			4.7.1 V2V Cooperative Mixed RF-FSO Communications with Direct and Relay-Assisted RF Links Over DSc and TF Channels
			4.7.2 Performance Measures of V2V Cooperative Mixed RF-FSO Communications with Direct and Relay-Assisted RF Links
			4.7.3 Numerical Results for V2V Cooperative Mixed RF-FSO Communications with Direct and Relay-Assisted RF Links
		4.8 Conclusions
		Acknowledgements
		References
	Chapter 5 Distribution of NFV Infrastructure Providing Efficient Edge Computing Architecture for 5G Environments
		5.1 Introduction
		5.2 Related Work
		5.3 Analytical Models
			5.3.1 Assumptions
			5.3.2 Centralized NFV Model
			5.3.3 NFV Model with Distributed Data Plane and Central MANO
			5.3.4 Fully Distributed NFV Model
		5.4 Performance Evaluation of the Analytical Models
		5.5 Experimental Evaluation of a Distributed NFV
			5.5.1 Experimental Testbed
			5.5.2 Experimental Results
		5.6 Concluding Remarks
		References
Part 2 Machine Learning-Based Communication and Network Automation
	Chapter 6 5G-AIoT Artificial Intelligence of Things: Opportunity and Challenges
		6.1 Introduction
		6.2 5G for Smart IoT Connectivity
			6.2.1 Co-Desing of AI and 5G Network Technology
			6.2.2 Intelligent Connectivity in Complex Use Cases
		6.3 5G Artificial Intelligence IoT
			6.3.1 IIoT Framework
			6.3.2 DT Models and Application Scenarios
				6.3.2.1 5G-AIoT Initiative
				6.3.2.2 5G Network Digital Twin
				6.3.2.3 Manufacturing Digital Twin
		6.4 Concluding Remarks
		References
	Chapter 7 Machine Learning-Based Scheduling in 5G/6G Communication Systems
		7.1 Overview of 5G/6G
		7.2 Definition and Importance of Scheduling
		7.3 Scheduling Schemes in 5G/6G
		7.4 Machine Learning Techniques Used in 5G/6G
			7.4.1 Supervised Learning
				7.4.1.1 Classification
				7.4.1.2 Regression
				7.4.1.3 Decision Trees
				7.4.1.4 Forecasting
			7.4.2 Semi-Supervised Learning
			7.4.3 Unsupervised Learning
				7.4.3.1 K-Means Algorithm
				7.4.3.2 Hidden Markov Model
				7.4.3.3 Deep Learning
			7.4.4 Reinforcement Learning
		7.5 Overview of Previous Works Using Machine Learning for Scheduling in 5G/6G Systems
			7.5.1 Reinforcement Learning-Based Scheduling for Data Traffic Management
			7.5.2 Reinforcement Learning for 5G Scheduling Parameter Optimization
			7.5.3 Knowledge-Assisted Deep Reinforcement Learning in 5G Scheduler Design
			7.5.4 Deep Reinforcement Learning for Radio Resource Scheduling in 5G MAC Layer
			7.5.5 Intelligent Resource Scheduling for 5G Radio Access Network Slicing
			7.5.6 Delay-Aware Cellular Traffic Scheduling with DRL
			7.5.7 A Fairness-Oriented Scheduler Using Multiagent RL
			7.5.8 Deep Learning-Based User Scheduling for Massive MIMO Downlink System
			7.5.9 Scheduling Based on DRL Model for UAV Video
			7.5.10 Reinforcement Learning Algorithms in Fairness-Oriented OFDMA Schedulers
		7.6 Conclusion
		References
	Chapter 8 Application of Deep Learning Techniques to Modulation and Detection for 5G and Beyond Wireless Systems
		8.1 Introduction
		8.2 Deep Learning with 6G Technology
		8.3 Deep Learning Classifications
			8.3.1 Supervised Learning
			8.3.2 Unsupervised Learning
			8.3.3 Reinforcement Learning
		8.4 Deep Learning and Modulation in Beyond 5G Networks
			8.4.1 CLDNN-Based Modulation
				8.4.1.1 Architecture of the CLDNN
				8.4.1.2 Principal Component Analysis
				8.4.1.3 CLDNN Architecture
				8.4.1.4 Error Performance of CLDNN Architecture
			8.4.2 Deep Neural Network Architectures for Modulation Classifications
		8.5 Concluding Remarks
		References
	Chapter 9 AI-Based Channel Coding for 5G/6G
		9.1 Introduction
		9.2 Application of ML in 5G NR LDPC Codes
			9.2.1 5G NR LDPC Codes
			9.2.2 LDPC Codes with Neural Networks
				9.2.2.1 Linear Approximation Min-Sum-Based ML for Optimizing LDPC Decoding
			9.2.3 Neural MS Decoder for Protograph-Based LDPC Codes
			9.2.4 Blind Recognition Method Using Convolutional Neural Networks
			9.2.5 Deep Learning-Based Unified Polar-LDPC Decoder
			9.2.6 Neural Normalized Min-Sum LDPC Decoding Network
			9.2.7 Neural 2D NMS (N-2D-NMS) Decoders
			9.2.8 Neural Layered MS Decoder for Protograph-Based LDPC Codes
		9.3 LDPC Codes with Reinforcement Learning
			9.3.1 Multiarmed Bandit-Based Node-Wise Scheduling (MAB-NS) Scheme
			9.3.2 Reinforcement Learning and Monte Carlo Tree Search
		9.4 Application of ML in 5G NR Polar Codes
			9.4.1 5G NR Polar Codes
			9.4.2 Polar Codes with Neural Networks
				9.4.2.1 Neural Network-Based Frame Error Rate Prediction of Polar Codes
				9.4.2.2 Transfer Learning-Based Decoder Training Method
				9.4.2.3 A CNN-Based Polar Decoder
				9.4.2.4 A Residual NND for Polar Codes
				9.4.2.5 A Differentiable Neural Computer-Aided Flip Decoding Algorithm
				9.4.2.6 Neural Network-Based Bit Flipping
				9.4.2.7 Stacked Denoising Autoencoder for Polar Codes
				9.4.2.8 A Machine Learning-Based Multi-Flips SC Decoding Scheme
				9.4.2.9 Double Long–Short-Term Memory-Based SC Flipping Decoder
			9.4.3 Polar Codes with Reinforcement Learning
				9.4.3.1 Reinforcement Learning for Polar Codes Construction
				9.4.3.2 A Reinforcement Learning-Aided CRC-Aided BP Decoder
		9.5 Gaps in Previous Research
		9.6 Summary
		Bibliography
Part 3 Artificial Intelligence towards 6G
	Chapter 10 Enabling Technologies and Applications of 5G/6G-Powered Intelligent Connectivity
		10.1 Introduction
		10.2 Enabling Technologies of Intelligent Connectivity
			10.2.1 Internet of Things
				10.2.1.1 Cellular IoT
				10.2.1.2 Massive IoT
				10.2.1.3 Broadband IoT
				10.2.1.4 Critical IoT
				10.2.1.5 Industrial Automation
			10.2.2 5G Mobile Networks
				10.2.2.1 5G-IoT Requirements
				10.2.2.2 5G-IoT-Enabling Technologies
				10.2.2.3 Wireless Network Function Virtualization
				10.2.2.4 Heterogeneous Networks
				10.2.2.5 Device-to-Device Communications (D2D)
				10.2.2.6 Advanced Spectrum Sharing and Interference Management
			10.2.3 Artificial Intelligence
				10.2.3.1 Intelligent Networks
				10.2.3.2 AI-Enabled Autonomous Systems and Human Interaction
			10.2.4 Cloud Computing and Networking
				10.2.4.1 Cloud Deployment Models and Service Classes
				10.2.4.2 Intelligent CC and Networks
			10.2.5 Blockchain
				10.2.5.1 Blockchain in 5G
				10.2.5.2 Blockchain in IoT
				10.2.5.3 Blockchain in AI
		10.3 Applications of Intelligent Connectivity
			10.3.1 Transportation
			10.3.2 Industry 4.0 and Manufacturing Operations
			10.3.3 Smart Cities
			10.3.4 Health Care
			10.3.5 Education
		10.3 Summary
		References
	Chapter 11 AI-Assisted Extended Reality Toward the 6G Era: Challenges and Prospective Solutions
		11.1 Introduction
		11.2 Background: The Convergence of XR, AI, and Cellular Data Networks
			11.2.1 Augmented Reality
			11.2.2 Virtual Reality
			11.2.3 Mixed Reality
			11.2.4 Artificial Intelligence
			11.2.5 Machine Learning
			11.2.6 Deep Learning
		11.3 AI-Assisted XR: Opportunities in the 5G and 6G Era
			11.3.1 AI-Assisted AR
				11.3.1.1 Camera Calibration and Pose Estimation
				11.3.1.2 Detection and Tracking of Real Objects
				11.3.1.3 Creation of Virtual Objects
				11.3.1.4 Displaying Virtual Objects
			11.3.2 AI-Assisted Virtual Reality
				11.3.2.1 Content Creation
				11.3.2.2 Optimization and Rendering
				11.3.2.3 Interaction in the Virtual World
			11.3.3 AI-Assisted Mixed Reality
		11.4 Challenges and Prospective Solutions
			11.4.1 Portability and Compatibility Issues
			11.4.2 Skepticism in Adoption of XR Solutions
			11.4.3 Issues with Visual Interfaces
			11.4.4 Processing Requirements and Resource Constraints
			11.4.5 Technological Limitations
			11.4.6 Power and Thermal Efficiency
			11.4.7 Challenges Related to the Creation of Contents
			11.4.8 Lack of Skills and Competencies
			11.4.9 Challenges Related to Application of AI
			11.4.10 Cellular Data Network Challenges and Deployment Issues
			11.4.11 Privacy and Security Issues
		11.5 Conclusions
		References
	Chapter 12 An Integrated 5G-IoT Architecture in Smart Grid Wide-Area Monitoring, Protection, and Control: Requirements, Opportunities, and Challenges
		12.1 Introduction
		12.2 Requirements
		12.3 5G-IoT Standardization and Interoperability
		12.4 Use Cases
			12.4.1 SmartZone
			12.4.2 VISOR
			12.4.3 EFCC and MIGRATE
		12.5 Integration of Advanced IoT Architectures
		12.6 Concluding Remarks
		References
	Chapter 13 Privacy Requirements in a Hyper-Connected World: Data Innovation vs. Data Protection
		13.1 Introduction
		13.2 Evolution of Security and Privacy Issues in Wireless Systems
			13.2.1 What Is Classified as Personal Data?
			13.2.2 Principles for Data Processing
		13.3 Data Privacy by Design
		13.4 AI Regulation and Accountability
		References
	Chapter 14 Evaluation of Representative 6G Use Cases: Identification of Functional Requirements and Technology Trends
		14.1 Introduction
		14.2 Emerging Use Cases and Applications
			14.2.1 Immersive Multimedia and Holographic Communication
			14.2.2 Tactile/Haptic-Based Communication
			14.2.3 Space–Terrestrial Integrated Networks
		14.3 Evolution of Usage Scenarios and Technology Trends
			14.3.1 Requirements on Network Operation
			14.3.2 Definition of System Capabilities
			14.3.3 Focus Areas for Further Study
		14.4 Concluding Remarks
		Bibliography
Index