摘要
M2M通信具有零星性,为了缓解大规模M2M通信带来的大量控制信号开销问题,CS-MUD被广泛研究。讨论了在帧传输结构下利用压缩感知的MMV模型来进一步提高多用户检测中的行为检测性能。为了使SOMP算法更具实际价值,提出CV-SOMP算法,通过利用交叉验证残差在迭代次数等于稀疏度时具有极小值来估计稀疏度。一方面,MMV模型通过利用源信号具有相同的稀疏结构,使得SOMP算法每次迭代中通过利用残差矩阵来更新支撑集,大幅提高了用户行为检测性能;另一方面,用户行为检测性能的提高导致用户数据检测性能也得到了大幅度改善。仿真结果表明CV-SOMP算法在稀疏度未知情况下不仅能够100%进行稀疏度估计而且能够大幅度提高多用户检测性能,另外在保证检测性能的条件下能够支持过载系统,提高频谱利用率。
M2M communication is sporadic,in order to alleviate control signal overhead problem lead by massive compressive sensing-based multi-user detection（CS-MUD）is widely studied.This paper discusses the use of compressed sensing multiple measurement vector model（MMV-CS）to further improve the performance of multiuser detection in frame transmission structure.In order to make the SOMP algorithm more practical,this paper proposed CV-SOMP algorithm which uses cross validation to estimate sparsity by using the residual value reaching the minimum value when the number of iterations equals the sparsity.on the one hand,MMV-CS model greatly improves the detection performance of user behavior by taking advantage of source signal has same sparse structure which makes SOMP algorithm updating the support set by using the residual matrix in each iteration,On the other hand,the improvement of user behavior detection results in a significant improvement in the performance of the user′s data detection.The simulation results show that the CV-SOMP algorithm in the case of unknown sparsity can not only100% estimates the sparsity but also greatly improves the performance of multiuser detection,in addition,CV-SOMP can support the overload system on the premise of ensuring the detection performance,improving the spectrum efficiency.
作者
张男
龚磊
翟旭平
Zhang Nan;Gong Lei;Zhai Xuping(Key Laboratory of Specialty Fiber Optics and Optical Access Networks,Joint International Research Laboratory of Specialty Fiber Optics and Advanced Communication,Shanghai Institute for Advanced Communication and Data Science Shanghai University,Shanghai 200444,China)
出处
《电子测量技术》
2018年第15期72-77,共6页
Electronic Measurement Technology
关键词
机器通信
压缩感知
多用户检测
MMV模型
交叉验证
M2M
compressive sensing
muhi-user detection
muhiple measurement vector
cross validation
