基于强化学习天地一体化网络无线资源管理研究

Wireless Resource Management in Space-Ground Integrated Networks based on Reinforcement Learning

有限的能耗和频谱资源并不能满足日益增长的通信需求,以网络系统能耗最小和网络资源利用率最大化为目标,联合用户时延、数据速率、计算资源需求提出基于强化学习的无线资源优化管理策略。仿真表明,提出的方案有效降低接入点切换次数和降低在高业务到达率情况下的平均接入时延,从而降低切换拥塞率,提升网络的平均负载,从而提升网络无线接入性能。

Limited energy and spectrum resources are insufficient to meet the ever-growing communication demands.To minimize network energy consumption and maximize resource utilization,this paper proposes a reinforcement learning-based wireless resource optimization management strategy that jointly considers user latency,data rate,and computation resource requirements.Simulation results demonstrate the effectiveness of the proposed approach in reducing access point switching and average access latency under high traffic arrival rates,thereby decreasing switch congestion and improving network average load,consequently enhancing the wireless access performance of the network.