基于强化学习的无线带宽与功率联合资源分配方法研究OA

Wireless resource scheduling is a fundamental issue in mobile communication network research.Efficient and reliable wireless resource scheduling plays a pivotal role in service quality assurance and system efficiency.Towards the development of 6G intelligent endogenous networks,we propose a unified scheduling method for PRB allocation and power allocation based on Twin Delayed Deep Deterministic Policy Gradient(TD3)reinforcement learning in this paper.In contrast to traditional resource scheduling and single-user resource scheduling based on reinforcement learning,this paper's algorithm considers the user's demand and network resources in a unified manner throughout the scheduling cycle.Furthermore,it is more efficient and reliable than bandwidth polling allocation and power averaging.In comparison to bandwidth polling allocation and power average allocation,the system throughput is enhanced by over 10％,thereby effectively improving the utilisation efficiency of wireless resources and system throughput.