Sequential blind signal extraction in order specified by stochastic properties

A new neural-network adaptive algorithm is proposed for performing extraction of independent source signals from a linear mixture of them. Using a suitable nonlinear Hebbian learning rule and a new deflation technique, the developed neural network is able to extract the source signals (sub-Gaussian and/or super-Gaussian) one-by-one with specified order according to their stochastic properties, namely, in decreasing order of absolute normalised kurtosis. The validity and performance of the algorithm are confirmed through extensive computer simulations.

[1]  Christian Jutten,et al.  Blind separation of sources, part I: An adaptive algorithm based on neuromimetic architecture , 1991, Signal Process..

[2]  Nathalie Delfosse,et al.  Adaptive blind separation of independent sources: A deflation approach , 1995, Signal Process..

[3]  Andrzej Cichocki,et al.  A New Learning Algorithm for Blind Signal Separation , 1995, NIPS.

[4]  Shun-ichi Amari,et al.  Self-adaptive neural networks for blind separation of sources , 1996, 1996 IEEE International Symposium on Circuits and Systems. Circuits and Systems Connecting the World. ISCAS 96.

[5]  Jean-François Cardoso,et al.  Equivariant adaptive source separation , 1996, IEEE Trans. Signal Process..

[6]  Aapo Hyvärinen,et al.  A neuron that learns to separate one signal from a mixture of independent sources , 1996, Proceedings of International Conference on Neural Networks (ICNN'96).

[7]  Zied Malouche,et al.  Extended anti-Hebbian adaptation for unsupervised source extraction , 1996, 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing Conference Proceedings.