DocumentCode :
395089
Title :
A probabilistic approach for blind source separation of underdetermined convolutive mixtures
Author :
Peterson, J. Michael ; Kadambe, Shubha
Author_Institution :
Signal & Image Process. Inst., Univ. of Southern California, Los Angeles, CA, USA
Volume :
6
fYear :
2003
fDate :
6-10 April 2003
Abstract :
There are very few techniques that can separate signals from the convolutive mixture in the underdetermined case. We have developed a method that uses overcomplete expansion of the signal created with a time-frequency transform and that also uses the property of sparseness and a Laplacian source density model to obtain the source signals from the instantaneously mixed signals in the underdetermined case. This technique has been extended here to separate signals (a) in the case of underdetermined convolutive mixtures, and (b) in the general case of more than 2 mixtures. Here, we also propose a geometric constrained based search approach to significantly reduce the computational time of our original "dual update" algorithm. Several examples are provided. The results of signal separation from the convolutive mixtures indicate that an average signal to noise ratio improvement of 5.3 dB can be obtained.
Keywords :
Laplace transforms; blind source separation; convolution; probability; search problems; time-frequency analysis; Laplacian source density model; blind source separation; dual update algorithm; geometric constrained based search; instantaneously mixed signals; overcomplete expansion; probabilistic approach; signal separation; sparseness; time-frequency transform; underdetermined case; underdetermined convolutive mixtures; Blind source separation; Delay; Density functional theory; Iterative algorithms; Laplace equations; Robustness; Signal to noise ratio; Source separation; Time frequency analysis; Transfer functions;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03). 2003 IEEE International Conference on
ISSN :
1520-6149
Print_ISBN :
0-7803-7663-3
Type :
conf
DOI :
10.1109/ICASSP.2003.1201748
Filename :
1201748
Link To Document :
بازگشت