On Coefficient Delay in Natural Gradient Blind Deconvolution and Source Separation Algorithms

In this paper, we study the performance effects caused by coefficient delays in natural gradient blind deconvolution and source separation algorithms. We present a statistical analysis of the effect of coefficient delays within such algorithms, quantifying the relative loss in performance caused by such coefficient delays with respect to delayless algorithm updates. We then propose a simple change to one such algorithm to improve its convergence performance.

[1]  Scott C. Douglas,et al.  Self-stabilized gradient algorithms for blind source separation with orthogonality constraints , 2000, IEEE Trans. Neural Networks Learn. Syst..

[2]  Lucas C. Parra,et al.  Convolutive blind separation of non-stationary sources , 2000, IEEE Trans. Speech Audio Process..

[3]  B. Widrow,et al.  Adaptive inverse control , 1987, Proceedings of 8th IEEE International Symposium on Intelligent Control.

[4]  Russell H. Lambert,et al.  Blind separation of multiple speakers in a multipath environment , 1997, 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[5]  S.C. Douglas,et al.  Multichannel blind deconvolution and equalization using the natural gradient , 1997, First IEEE Signal Processing Workshop on Signal Processing Advances in Wireless Communications.

[6]  Bernard Widrow.  Adaptive inverse control , 1990, Defense, Security, and Sensing.

[7]  K. Matsuoka,et al.  Minimal distortion principle for blind source separation , 2002, Proceedings of the 41st SICE Annual Conference. SICE 2002..

[8]  Teresa H. Y. Meng,et al.  Stochastic gradient adaptation under general error criteria , 1994, IEEE Trans. Signal Process..