Convolutive BSS using a penalized mutual information criterion

The Blind Separation problem of linear time dependent mixtures is addressed in this paper. We have developed a novel algorithm based on the minimization of the mutual information plus a penalized term which ensures an a priori normalization of the estimated sources (outputs). The criterion minimization is done using a well known gradient approach. Finally, some numerical results are presented to illustrate the performance of the penalized algorithm comparing to the Babaie-Zadeh approach presented in [4].