基于用户数估计的MIMO无源多址接收优化OA

To solve the problems of severe interference between user signals and low decoding success rate in Passive Multiple Access(UMA)systems under high-density user scenarios,it is necessary to improve signal separation and decoding performance of the system in such scenarios.A multi-level iterative reception framework is constructed based on Multiple-Input Multiple-Output(MIMO)technology,and performance optimization is achieved by combining user number estimation,multi-beam interpretation scheduling,and iterative decoding technologies.A beam user number estimation model based on deep learning is designed to dynamically schedule low-conflict beams for priority decoding,while eliminating interference from decoded users through cross-beam Successive Interference Cancellation(SIC).Simulation results show that the beam decoding success rate in high-density user scenarios is improved by approximately 28.09%compared with the traditional Compressed Sensing-Orthogonal Matching Pursuit(CS-OMP)algorithm,by about 18.3%and 10.4%compared with the Approximate Message Passing(AMP)and Sparse Low-Rank Matrix Algorithm(SPARC)schemes,respectively,and remains stable when the number of active users reaches 600.Through the closed-loop optimization of spatial resource allocation and cooperative interference suppression,this multi-level iterative reception framework significantly improves the capacity and robustness of the system in massive user scenarios,providing an innovative solution for 6G ultra-dense internet of things access.