Abstract

This paper presents an unified approach to the problem of separation of sources, based on the consideration of mutual information. The basic setup is that the sources are independent stationary random processes which are mixed either instantaneously or through a convolution, to produce the observed records. We define the entropy of stationary processes and then the mutual information between them as a measure of their independence. This provides us with a contrast for the separation of source problem. For practical implementation, we introduce several degraded forms of this contrast, which can be computed from a finite dimensional distribution of the reconstructed source processes only. From them, we derive several sets of estimating equations generalising those considered earlier. 1 Introduction Blind separation of sources is a topic which have received much attention recently, as it has many important applications (speech analysis, radar, sonar, : : : ). Basically, one observes seve...

Citations

1034	Independent component analysis: A new concept - COMON - 1994
368	Blind separation of sources, part i: An adaptive algorithm based on neuromimetic architecture - Jutten, Herault - 1991
139	Multiple Time Series - Hannan - 1970
87	Indeterminacy and identifiability of blind identification - Tong - 1991
78	Blind separation of mixture of independent sources through a quasimaximum likelihood approach - Pham, Garrat - 1997
58	Waveform preserving blind estimation of multiple independent sources - Tong, Inouye, et al. - 1993
53	Source separation without a priori knowledge: The maximum likelihood solution - Gaeta, Lacoume - 1990
49	Trigonometric Series Vol. I - Zygmund - 1959
45	separation of instantaneous mixture of sources via an independent component analysis - Pham, “Blind - 1996
33	Iterative techniques for blind source separation using only fourth-order cumulants - Cardoso - 1992
25	An efficient technique for blind separation of complex sources - Cardoso, Souloumiac - 1993
16	A direct solution for blind separation of sources - Mansour, Jutten - 1996
9	Sources identification: A solution based on cumulants - Lacoume, Ruiz - 1988
6	Conditional likelihood and unconditional optimum estimating equations. Biometrika 63 - Godambe - 1976
5	S'eparation adaptive de sources ind'ependantes par une approch de d'eflation - Loubaton, Delfosse - 1993
2	Source separation using higher order moments", proc - Cardoso - 1989
2	Criteria for multichanel Signal Separation - Yellin, Weinstein
1	Second order blind identification of convolutive mixture with temporally correlated sources: a subspace based approach - Gorokov, Loubaton - 1996