基于PAST盲信号分离方法的仿真结果分析

Simulation Analysis of Blind Source Separation Based on PAST Algorithm

盲信号分离问题在信号处理领域中是一个值得深入研究的很重要的方向。因此关于这个问题已经提出了很多分离算法,能够在一定程度上解决越来越复杂的实际问题。投影逼近子空间跟踪(PAST)是一种很重要的盲信号分离方法。本文就PAST方法用于盲信号分离给出仿真结果并进行性能分析。本文的目的是验证非线性PCA方法用于盲信号分离的有效性,并分析参数取值对分离结果的影响。衡量分离效果的指标是串音误差,程序运行的结果给出了串音误差随算法参数改变的曲线图。

The blind source separation is very important in the signal processing area, which can solve many practical problems. Recently, many separation methods have been presented. In this paper, we aim to present several simulation results to analyze the performance of the separation method based on the projection approximation subspace tracking （PAST）. We aim to show the validity of the nonlinear principal component analysis （PCA） method and analyze the relation between the parameter and the separation result, The cross talking error is used to evaluate the performance of the PAST algorithm.