AI/ML Integration in 5G and Beyond
Architecture and Prototypes
Presenters
University of Novi Sad
Jozef Stefan Institute
About the Tutorial
Machine Learning (ML) and Artificial Intelligence (AI) concepts are gradually being integrated in modern 3GPP standardized cellular networks. The process is initiated by introducing AI/ML functions in the 5G core network (5G CN) and from there, it is expanding towards AI-enabled 5G radio access network (RAN). AI/ML is currently at the early stage of design within 3GPP standardisation and many pathways for its impact are still open. In this talk, we present an overview of AI/ML integration in 5G and Beyond. Following the 3GPP perspective, we start from 5G CN and move towards 5G RAN, focusing on AI/ML impact on 5G and B5G physical layer design. We also cover complementary work done by Open RAN (O-RAN) Alliance. The tutorial is extended with demonstration of several use cases and applications of AI/ML integration in 5G/B5G.
Topics:
- Integration of AI/ML in 5G Core Network
- Integration of AI/ML in 5G Radio Access Network
- AI/ML for the 5G Physical Layer
- AI/ML in 5G Services and Applications
- AI/ML for 5G Network Management and Orchestration
- Examples of AI/ML in 5G/B5G Use Cases Throughout the Tutorial
Reading list
đź“ť Key Papers
AI/ML in the Core Network
- Private 5G: The future of industrial wireless (IEEE Industrial Electronics Magazine)
- Towards supporting intelligence in 5G/6G core networks: NWDAF implementation and initial analysis (IWCMC 2022)
- AI-Driven Provisioning in the 5G Core (IEEE Internet Computing)
- Scaling Network Slices with a 5G Testbed: A Resource Consumption Study (WCNC)
- Design and Implementation of Network Data Analytics Function in 5G (ICTC)
- A distributed collaborative learning approach in 5G+ core networks (IEEE Network)
- Network Architecture for Machine Learning: A Network Operator’s Perspective (IEEE Communications Magazine)
- Mobility prediction for 5G core networks (IEEE Communications Standards Magazine)
- End-to-end data analytics framework for 5G architecture (IEEE Access)
- TS 23.501, “System architecture for the 5G System (5GS)”
- TS 29.520, “5G System; Network Data Analytics Services; Stage 3”
- TR23.700-80 Study on 5G System Support for AI/ML-based Services
- TR23.700-81 Study of Enablers for Network Automation for the 5G System (5GS)
AI/ML in the RAN
- Democratizing the Network Edge (ACM SIGCOMM Computer Communication Review)
- An Overview and Solution for Democratizing AI Workflows at the Network Edge (Journal of Network and Computer Applications)
- A Representation Learning Approach to Feature Drift Detection in Wireless Networks​
- The carbon impact of artificial intelligence (Nature Machine Intelligence)
- Energy consumption in data centres and broadband communication networks in the EU (Publications Office of the European Union)
- The Energy Cost of Artificial Intelligence Lifecycle in Communication Networks
- MRM3: Machine Readable ML Model Metadata
- Understanding O-RAN: Architecture, interfaces, algorithms, security, and research challenges (IEEE Communications Surveys & Tutorials)
- Intelligent O-RAN beyond 5g: Architecture, use cases, challenges, and opportunities (IEEE Access)
- Enabling Real-Time AI-Based Open RAN Control
- Actor-Critic Network for O-RAN Resource Allocation: xApp Design, Deployment, and Analysis (IEEE Globecom Workshops)
- On the Implementation of a Reinforcement Learning-based Capacity Sharing Algorithm in O-RAN (IEEE Globecom Workshops)
- Prototyping next-generation O-RAN research testbeds with SDRs
- ColO-RAN: Developing Machine Learning-based xApps for Open RAN Closed-loop Control on Programmable Experimental Platforms (IEEE Trans. Mobile Computing)
- AI-RAN Alliance Vision and Mission White Paper
- TS 38.401, “NG-RAN Architecture Description”
- TR 37.817, “Study on enhancement for data collection for NR and EN-DC”
- TR 38.843, “Study on artificial intelligence (AI)/machine learning (ML) for NR air interface”
AI/ML for the PHY
- A survey of wireless path loss prediction and coverage mapping methods (IEEE Communications Surveys & Tutorials)
- Scientific discovery in the age of artificial intelligence (Nature)
- Towards Automated and Interpretable Pathloss Approximation Methods (AI4WCN)
- Automatic detection of wireless transmissions (IEEE Access)
- Learning approximate neural estimators for wireless channel state information (MLSP)
- Detection of Impaired OFDM Waveforms Using Deep Learning Receiver (IEEE SPAWC 2022)
- Deep learning-based packet detection and carrier frequency offset estimation in IEEE 802.11 ah (IEEE Access)
- Power of deep learning for channel estimation and signal detection in OFDM systems (IEEE Wireless Communications Letters)
- Deep learning-based channel estimation (IEEE Communications Letters)
- Deep learning for channel estimation: Interpretation, performance, and comparison (IEEE Transactions on Wireless Communications)
- Deep Neural Network Augmented Wireless Channel Estimation for Preamble-Based OFDM PHY on Zynq System on Chip (IEEE Transactions on VLSI)
- An introduction to deep learning for the physical layer (IEEE Transactions on Cognitive Communications and Networking)
- Deep learning based communication over the air (IEEE Journal of Selected Topics in Signal Processing)
- Autoencoder-Based Unequal Error Protection Codes (IEEE Communications Letters)
- Rateless Autoencoder Codes: Trading off Decoding Delay and Reliability (IEEE ICC 2023)
- Deep learning methods for improved decoding of linear codes (IEEE Journal of Selected Topics in Signal Processing)
- Model-based deep learning
- Graph neural networks for channel decoding (GC Wkshps)
- Decoding Quantum LDPC Codes Using Graph Neural Networks
- Learning Linear Block Error Correction Codes
- Toward a 6G AI-native air interface (IEEE Communications Magazine)
- Machine learning for beam alignment in millimeter wave massive MIMO (IEEE Wireless Communications Letters)
- Deep active learning approach to adaptive beamforming for mmWave initial alignment (IEEE Journal on Selected Areas in Communications)
- Learning site-specific probing beams for fast mmWave beam alignment (IEEE Transactions on Wireless Communications)
- A Review of the State of the Art and Future Challenges of Deep Learning-Based Beamforming (IEEE Access)
- Machine Learning for Millimeter Wave and Terahertz Beam Management: A Survey and Open Challenges (IEEE Access)
AI/ML at the Application Layer
- COMSPLIT: A Communication-Aware SPLIT Learning for Heterogeneous IoT Platforms (IEEE Internet of Things Journal)
- Artificial Intelligence in 3GPP 5G-Advanced: A Survey
- TR22.874, “Study on traffic characteristics and performance requirements for AI/ML model transfer”
- TS22.261, “Service requirements for the 5G system”
AI/ML-driven Network Management and Orchestration
- Demonstrating Smart Scaling of AI-Services for Future Networks
- Enabling mobile AI agent in 6G era: Architecture and key technologies (IEEE Network)
- Intent-based management of next-generation networks: An LLM-centric approach (IEEE Network)
- Mobile-llama: Instruction fine-tuning open-source llm for network analysis in 5g networks (IEEE Network)