基于干扰识别的低轨卫星网络多维资源调度研究

his paper presents an interference identification-based multi-dimensional resource scheduling algorithm aimed at addressing the problems of severe uplink interference and low transmission rates in low Earth orbit (LEO) mega-constellation satellite networks, caused by wide user antenna patterns and high density of satellites and beams. Firstly, the model of the studied system is constructed and theoretically analyzed. Then, an interference identification-based multi-dimensional resource scheduling algorithm and its implementation are introduced.This algorithm leverages the centralized processing capability of satellite clusters, the perception of wireless environments based on interference identification, and the decision-making capability of double deep Q neural networks for complex problems. By dynamically scheduling multi-dimensional resources such as beam, frequency, and time for satellite clusters through the decisions made by the proposed algorithm, interference between links can be minimized, and the total throughput of the network can be maximized. Finally, the performance of the proposed algorithm is evaluated through simulation results.