Blind Sources Separation Based On Estimation Of Probability Density Function Of Mixture Signals

Blind Sources Separation(BSS) have made rapid progress and have been used in many fields. Independent Component Analysis is a efficient approach to be used for BSS.An algorithm for blind source separation based on estimation of probability density function of mixture signals is proposed. By using Gram-Charlier expansion, we directly estimate probability density function of the signals, and therefore the score function used for demixing. The advantages of the proposed algorithm over many existing blind source separation algorithms are its ability to separate hybrid mixtures that contain both super-Gaussian and sub-Gaussian sources and its rapid convergence to desired solutions. The method can be applied to all the blind source separation algorithms where the score function is obtained by a nonlinear function. Instead of the traditional biomedicine methods, the method above can be used to do the Partial Volume Correction for getting the pure Microarray data. The method is easy in operation and reduces the biomedicine experiment cost as well.The determination of the number of unknown source signals when there are fewer sources than the mixtures in bind source separation (BSS) has always been an unsolved problem. In order to improve the application of BSS better, the approach by estimating the eigenvalue of covariance of observation vector is presented. Finally, the experiment results show that the approach is efficient.