A Novel Method for Blind Source Separation
M. Kumar, Ashok Kumar · 2011
A novel blind source separation algorithm is proposed which can separate the mixture of super- and sub-Gaussian sources. one novel method is proposed to estimate the probability density function of super and sub-Gaussian sources respectively. In the framework of natural gradient, the parameters of two models are adaptively manipulated by online kurtosis learning. Applied to the speeches, experiments give good performance of proposed algorithm.