Title :
Mutual information approach to blind separation-deconvolution
Author_Institution :
Lab. of Modeling & Comput., UJF, Grenoble, France
Abstract :
In this paper, we propose an approach to the blind separation-deconvolution problem, based on the mutual information criterion. Formulas for the quasi Newton algorithm are provided. More interesting results have been obtained in the pure deconvolution case. By a clever parameterization of the deconvoluting filter, the quasi Newton algorithm also become particularly simple. Simulation results showing the good performance of this algorithm are provided.
Keywords :
Newton method; blind source separation; deconvolution; filtering theory; blind separation-deconvolution problem; deconvoluting filter; mutual information approach; quasiNewton algorithm; Approximation methods; Deconvolution; Entropy; Equations; Mutual information; Signal processing algorithms; Vectors;
Conference_Titel :
Signal Processing Conference, 2005 13th European
Conference_Location :
Antalya
Print_ISBN :
978-160-4238-21-1