Towards Intelligent 6G Networks: A Comprehensive Review of AI-Driven Control and Optimization
The transition from fifth-generation (5G) to sixth-generation (6G) communication networks represents a fundamental shift from conventional model-driven architectures toward AI-native, intelligence-driven network ecosystems capable of autonomous control, optimization, and decision-making. Although artificial intelligence (AI) has demonstrated significant potential in enhancing network performance, existing research remains fragmented, with most studies focusing on isolated network functions rather than integrated system-level intelligence. This study presents a systematic literature review (SLR) combined with a critical synthesis to examine the current state of AI-driven control and optimization in 6G communication networks. The review systematically analyzes 87 peer-reviewed studies published between 2020 and 2024, retrieved from IEEE Xplore, ScienceDirect, SpringerLink, and Wiley Online Library using predefined inclusion and exclusion criteria. The selected studies are critically evaluated with respect to machine learning, deep learning, and reinforcement learning techniques, emphasizing their architectural roles, operational capabilities, deployment feasibility, and system-level implications. The findings indicate that while AI techniques substantially improve network adaptability, resource management, and autonomous operation, significant challenges remain regarding scalability, computational complexity, data dependency, interoperability, explainability, and deployment in real-world environments. Furthermore, the review identifies a considerable gap between algorithmic advances and practical implementation, highlighting the need for integrated AI frameworks and architecture-aware design strategies capable of supporting scalable, trustworthy, and autonomous 6G communication systems. By providing a comprehensive synthesis of recent research, comparative analysis of major AI paradigms, and future research directions, this review contributes to bridging the gap between theoretical developments and practical deployment of AI-native communication networks.