南京邮电大学学报（自然科学版）2025，Vol.45Issue(2)：20-29,10.DOI:10.14132/j.cnki.1673-5439.2025.02.003
基于块压缩感知的大规模免授权多址接入方案
A massive grant-free multiple access scheme based on block compressed sensing
摘要
Abstract
A massive grant-free multiple access(GFMA)scheme based on block compressed sensing(CS)and adaptive matching pursuit is proposed for sporadic burst transmission scenarios of large-scale machine terminals.First,the uplink multiple access signals are modeled as a CS equation with a sparse structure frame by frame.Second,the issue of reconstructing uplink multiple access signals is trans-formed into a block sparse structure CS multiuser detection.Third,a block sparse model adaptive match-ing pursuit algorithm is proposed to detect the multiple access signals,in which three adaptive strate-gies,the dynamic step,the dynamic pruning,and dynamic iteration,are introduced to improve the per-formance of multi-user signal detection and reconstruction.Simulation results show that the proposed scheme significantly reduces the error rate of uplink multiple access signals and effectively improves the performance of massive GFMA.关键词
大规模机器类型通信/免授权多址接入/压缩感知多用户检测/块稀疏模型/自适应匹配追踪Key words
massive machine type communication(mMTC)/grant-free multiple access(GFMA)/com-pressive sensing(CS)multiuser detection/block sparse model/adaptive matching pursuit分类
电子信息工程引用本文复制引用
张晶,马林,何艳..基于块压缩感知的大规模免授权多址接入方案[J].南京邮电大学学报（自然科学版）,2025,45(2):20-29,10.基金项目
国家重点研发计划(2020YFB1807202)、国家自然科学基金(92067201)和江苏省重点研发计划(BE2020084-1)资助项目 （2020YFB1807202）
