Blind separation of non-stationary and non-Gaussian independent sources

In this paper, the problem of blind separation of an instantaneous mixture of independent sources by exploiting their non-stationarity and/or non-Gaussianity is addressed. We show that non-stationarity and non-Gaussianity can be exploited by modeling the distribution of the sources using Gaussian Mixture Model. The Maximum Likelihood estimator is utilized in order to derive two novel source separation techniques. Both methods are based on estimation of the sensors distribution parameters via the Expectation-Maximization algorithm for GMM parameter estimation. In the first method, the separation matrix is estimated by applying simultaneous joint diagonalization of the estimated GMM covariance matrices. In the second proposed method, the separation matrix is estimated by applying singular value decomposition of a weighted sum of the estimated GMM covariance matrices. The performances of the two proposed methods are evaluated and compared to existing blind source separation techniques. The results show superior performances of the proposed methods in terms of Interference-to-Signal Ratio.