Title :
Exploiting source non stationary and coloration in blind source separation
Author_Institution :
Lab. of Modeling & Comput., Grenoble Univ., France
Abstract :
A new method for blind source separation of instantaneous mixtures is developed. It exploits both the spectral and time diversity of the sources and is based on Gaussian mutual information. As a result, it uses only second order statistics and can be efficiently implemented through a joint diagonalization algorithm. Simulation results illustrate the good performance of the method.
Keywords :
blind source separation; matrix algebra; statistical analysis; Gaussian mutual information; blind source separation; coloration; instantaneous mixtures; joint diagonalization algorithm; matrices; nonstationarity; second order statistics; spectral diversity; time diversity; Blind source separation; Computational modeling; Entropy; Gaussian processes; Laboratories; Mutual information; Source separation; Statistical distributions; Statistics; Tin;
Conference_Titel :
Digital Signal Processing, 2002. DSP 2002. 2002 14th International Conference on
Print_ISBN :
0-7803-7503-3
DOI :
10.1109/ICDSP.2002.1027859
