Environmental-Aware Scheduler for Trustworthy 6G Communication

Trustworthiness is emerging as a critical design objective for 6G networks, ensuring consistent and reliable performance under dynamic conditions. While several studies have addressed its evaluation, enhancing trustworthiness through intelligent network optimization remains largely unexplored. This paper proposes a deep reinforcement learning-based scheduler that integrates environmental-aware knowledge from channel knowledge maps to optimize reliability, availability, and fairness aspects of trustworthiness. Compared to the round-robin scheduler, the proposed method improves reliability by over 300 it improves availability and fairness by 61.40.6 respectively, with just a 4.5aware scheduling solution for trustworthy 6G.