6G networks are set to exceed the performance of 5G by delivering unprecedented data rates, sub-millisecond latencies, and ubiquitous connectivity. However, meeting these ambitious targets introduces immense complexity in network management and optimization. Artificial Intelligence and Machine Learning (AI/ML) have emerged as essential enablers in this context, equipping 6G with intelligent automation to tackle challenges beyond the scope of conventional techniques. This survey provides a comprehensive overview of the pivotal role AI/ML plays in several key 6G communication technologies. In particular, we examine how AI/ML techniques enhance and empower Massive MIMO (M-MIMO) systems, Integrated Sensing and Communication (ISAC) paradigms, Reconfigurable Intelligent Surfaces (RIS), Non-Terrestrial Networks (NTNs), Visible Light Communications (VLC), Cell-Free MIMO, Terahertz communications, ML-enhanced RedCap for scalable IoT connectivity and AI native Open-RAN architecture. We outline the synergy between AI algorithms and these technologies, highlighting improvements in capacity, efficiency, and adaptability across the wireless network. We define 6G and the necessity of AI-driven approaches, then survey AI/ML applications in each of the targeted domains, and finally discuss open challenges and promising future research directions. We introduce a comprehensive taxonomy of AI integration levels and demonstrate how AI-native design principles enable capabilities that conventional approaches cannot achieve.
Tok et al. (Sat,) studied this question.