论文部分内容阅读
运用多通道盲最小均方(MBLMS)算法对2×2模分复用(MDM)系统进行解复用,阐述了基于非高斯性最大化的MBLMS算法原理,分析了此算法的解复用性能,并与基于数据辅助的最小均方(LMS)算法进行了比较。仿真结果表明:MBLMS算法在无需数据辅助的条件下能够实现卷积混合信号的盲分离,其解复用性能与基于数据辅助的LMS算法相当,且收敛速度比基于数据辅助的LMS算法提高了33.3%。
The multi-channel blind minimum mean square (MBLMS) algorithm is used to demultiplex the 2 × 2 MDM system. The principle of MBLMS algorithm based on maximization of non-Gaussianity is described, and the performance of demultiplexing , And compared with data-aided least-mean-square (LMS) algorithm. The simulation results show that MBLMS algorithm can blindly decompose convolutional mixed signal without data aiding, whose performance of demultiplexing is equivalent to LMS algorithm based on data aiding, and its convergence rate is 33.3 higher than that of LMS algorithm based on data aiding %.