任意概率密度信号的盲分离

Blind Separation of Independent Sources with Any Probability Density Function

独立分量分析(ICA)是一种把多维随机矢量转换为尽可能统计独立的分量的统计方法,被广泛用于非高斯信号处理领域。本文给出了一种基于峰度的盲源分离(BSS)算法,也可看作是最大似然方法的扩展,解决了最大似然方法限制过多的缺陷,且与用Comon的方法求解Givens矩阵相比,结构清晰、实现简单。仿真证明了算法的有效性。

Independent component analysis （ICA）is a statistical method for transforming an observed multidimensional random vector into components that are statistically as independent from each other as possible. It is widely used in nonGaussian signals processing. An algorithm of ICA based on kurtosis is presented in this paper; it can also be regarded as Extended Maximum Likelihood （ML）; the problem of too many constraints in ML is solved ; it is simpler and easier to implement in solving Givens Matrix compared with Comon＂ s. The simulation justifies its effectiveness.