Blind source separation for noisy mixtures with non-Gaussianity and nonlinear autocorrelation

One often solve the BSS problem by using the statistical properties of original sources,e.g.,non-Gaussianity or time-structure information.Nevertheless,real-life mixtures are likely to contain both non-Gaussianity and time-structure information,rendering the algorithm using only one statistical property fail.The BSS algorithms are often limited to noise-free mixtures,which are not realistic.Therefore,this paper address the separation of the noisy model based on non-Gaussianity and nonlinear autocorrelation of sources.An objective function which based on the two statistical characteristics of sources is proposed.Maximizing this objective function,we present a blind source separation algorithm for noisy mixtures.The validity of the proposed algorithm is demonstrated by computer simulation.Moreover,comparisons with the existing algorithm for noisy mixtures based on non-Gaussianity and nonlinear autocorrelation indicate the better performance.