Distributed AI for Emerging Spectrum Management Technologies in 6G Networks

Jialin Xing, Ning Yang, Haijun Zhang, Zijun Wu, Chen Sun, Haojin Li

Abstract: In 6G communication networks, real-time and efficient spectrum management is the key to achieving low latency, high reliability and high energy efficiency. Distributed artificial intelligence provides a new solution for spectrum management in 6G networks by extending intelligent computing capabilities to multiple nodes. This approach can address the challenges of spectrum sensing and resource allocation in complex and dynamic environments, thereby significantly optimizing wireless communication performance. This paper systematically summarizes the characteristics and requirements of spectrum management in scenarios such as large-scale satellite networks, drone swarms, industrial Internet of Things, Internet of Vehicles and military communications, and sorts out the core communication issues of distributed AI technology in 6G spectrum management from the two aspects of spectrum sensing and allocation. In response to these issues, this paper explores the key technologies supported by distributed AI from the perspective of technical optimization, including efficient communication perception, data security, collaborative scheduling, convergence, real-world deployment, and technology comparison. Finally, this paper proposes innovative research directions such as subtask chain decomposition, model quantization and general spectrum management model, and puts forward specific suggestions for future research and practical applications.

External IDs:doi:10.1109/mnet.2025.3580622