The wide variety in enabling technologies, operating scenarios, environments, and use cases required for the fifth generation (5G) communication system and beyond 5G (B5G) entails the availability of descriptive channel models. In this chapter, we revise the key requirements of efficient channel modeling for 5G and beyond, highlight the outstanding and expected challenges, and present the main efforts made in this domain by leading industrial and research entities. We review channel modeling through machine learning (ML) as a promising channel modeling approach and revise the compressed sensing (CS) -based channel modeling and estimation framework.

Chapter Contents:

• 11.1 Introduction
• 11.1.1 What defines a good channel model for 5G and B5G?
• 11.2 Evolution of radio frequency channel models before 5G
• 11.2.1 Analytical channel models
• 11.2.1.1 Correlation-based models
• 11.2.1.2 Propagation-motivated models
• 11.2.2 Physical channel models
• 11.2.2.1 Geometry-based stochastic models
• 11.2.2.2 Non-geometry-based stochastic models
• 11.2.2.3 Deterministic models
• 11.2.3 Standardized channel models
• 11.2.3.1 The COST channel models (259 and 273)
• 11.2.3.2 The multidimensional parametric channel model
• 11.2.3.3 The 3GPP spatial channel model
• 11.2.3.4 The WINNER channel model
• 11.2.3.5 IMT-advanced channel models from ITU
• 11.3 Channel models for 5G and beyond
• 11.3.1 Enhanced 3GPP channel models
• 11.3.2 The MiWEBA channel model
• 11.3.3 METIS channel models
• 11.3.3.1 The METIS map-based model
• 11.3.3.2 The METIS stochastic model
• 11.3.3.3 The METIS hybrid model
• 11.3.4 The QuaDRiGa/mmMAGIC channel model
• 11.3.5 The IEEE 802.11ay channel model
• 11.3.6 The IMT-2020 channel model
• 11.3.7 The NYUSIM channel model
• 11.4 Machine learning-based channel modeling for 5G and B5G
• 11.5 Channel sparsity and compressed modeling in 5G and B5G
• 11.5.1 Pilot reduction through compressive channel sampling
• 11.5.2 Channel sparsity aspects in 5G and B5G
• 11.5.3 Outstanding challenges and questions
• 11.6 Conclusion
• Acknowledgment
• References

Inspec keywords: compressed sensing; 5G mobile communication; learning (artificial intelligence); telecommunication computing; channel estimation

Other keywords: CS-based channel modeling; descriptive channel models; compressed sensing; fifth generation communication system; machine learning; 5G communication system; estimation framework; beyond 5G

Subjects: Communications computing; Knowledge engineering techniques; Signal processing and detection; Mobile radio systems; Communication channel equalisation and identification; Signal processing theory