独立分量分析盲信号分离方法研究

盲源分离是指从多个相互独立的源信号的混合信号中分离出源信号来。独立分量分析法是盲源分离的一种新方法,由于其在语音信号处理、阵列信号处理、生物医学信号处理、移动通信及图象处理等领域的应用前景,越来越引起人们的关注,成为研究的热点。介绍一种基于四阶累积量的非高斯性最大化的ICA算法解决盲源分离的问题,并给出了该算法分离通信信号的计算机仿真结果,验证了算法的有效性。

Researchon Independent Component Analysis of Blind Signal Separation

BSS means separating the statistically independent source signals from their mixture. Independent component analysis （ICA） is a new method of blind source separation. It has attracted great attention and has been an attractive trend because of its poten-tial application in signal processing such as speech signal processing, array signal processing,biological and medical signal processing, and wireless communication and image processing. A non-Gaussian maximization ICA algorithm based on fourth-order cumulants is introduced. And the simulation results of communication signal separation are given. The results show that the algorithm is effective.