一种适用于微弱信号盲提取的白化方法

摘要

独立分量分析(ICA)算法是解决盲信号分离(BSS)问题的最有效方法之一.ICA中,对观测信号预白化处理的作用至关重要.通常采用主分量分析(PCA)来进行预白化处理.实际中,在利用广播、电视等作为照射源的被动雷达系统中,观测信号通常被强噪声和干扰严重污染,这很大程度上降低了BSS方法的性能.然而,传统的BSS方法中没有考虑这一问题.本文,我们关注这一问题并提出一种新的含噪BSS的白化框架,其主要思想是在白化之前从观测信号的协防矩阵中减去噪声方差.仿真结果验证了所提出方法能够明显提高BSS的性能.%Independent component analysis (ICA) is one of the most important methods for blind source separation (BSS),in which the pre-whitening procedure of the observed signals plays an important role. Usually,principle component analysis (PCA) is employed for this preprocessing task. In practice,the observed signals of a passive radar system are usually corrupted by strong noise and outliers,which greatly reduces the performance of BSS methods. However,this problem is rarely taken into account in the whitening step of traditional BSS methods. We focus on this problem and propose a new whitening framework for noisy BSS. The idea is that the noise variance is removed from the covariance matrix of the observed signals before whitening. The experiments show that the BSS performance is greatly improved using the proposed whitening framework.