Artificial intelligence in 6G communication networks: Physical-layer integration, intelligent scheduling, and deployment challenges
Zekai Li
Fuzhou University of International Studies and Trade
DOI: https://doi.org/10.59429/esta.v13i3.15556
Keywords: 6G; artificial intelligence; physical-layer integration; resource scheduling; semantic communications; standardization
Abstract
As sixth-generation (6G) networks evolve toward integrated space–air–ground connectivity and the convergence of communication, sensing, and computing, artificial intelligence (AI) is shifting from an auxiliary tool to a native network capability. This paper adopts a three-part analytical framework that covers protocol-layer integration, scheduling optimization, and deployment constraints. At the physical layer, it compares the operating boundaries of several deep-learning models for channel estimation and identifies the main obstacles to their practical use. At the network layer, it examines intelligent scheduling through multi-agent reinforcement learning, network-slice orchestration, federated learning, and an endogenous-intelligence architecture with closed-loop network autonomy. It then discusses three barriers to industrial deployment: The AI-native redesign of semantic communications, cross-vendor standardization, and the combined requirements of trustworthiness, security, and environmental sustainability. The analysis provides a structured account of how AI may be integrated into 6G networks and where further technical and institutional work is required.
References
[1] J. Sun, W. Sheng, J. Qu, K. Yan, and B. Zhou, "Research and analysis of signal transmission mechanisms for 6G UAV communication networks," Radio Communications Technology, vol. 51, no. 1, pp. 86–95, 2025 (in Chinese).
[2] M. Xu, "Standardization of satellite-based mobile communication networks for 5G-Advanced and 6G," Mobile Communications, vol. 48, no. 9, pp. 31–39, 2024 (in Chinese).
[3] H. Tong, Key Technologies for Efficient Semantic-Based Wireless Communications. Beijing, China: Beijing University of Posts and Telecommunications, 2025, dissertation (in Chinese).