Decorrelation of multiple non-stationary sources using a multivariable crosstalk-resistant adaptive noise canceller
Article first published online: 6 JUN 2012
Copyright © 2012 John Wiley & Sons, Ltd.
International Journal of Adaptive Control and Signal Processing
Volume 27, Issue 5, pages 349–367, May 2013
How to Cite
Moir, T. J. and Harris, J. I. (2013), Decorrelation of multiple non-stationary sources using a multivariable crosstalk-resistant adaptive noise canceller. Int. J. Adapt. Control Signal Process., 27: 349–367. doi: 10.1002/acs.2303
- Issue published online: 15 APR 2013
- Article first published online: 6 JUN 2012
- Manuscript Accepted: 1 MAY 2012
- Manuscript Revised: 29 APR 2012
- Manuscript Received: 15 SEP 2011
- blind source separation (BSS);
- least mean squares (LMS);
- speech enhancement
A new structured approach to adaptive noise cancelling of non-stationary stochastic signals is given. The divide and conquer method subdivides the problem into a number of sources with a power of 2 and subdivides the solution to smaller problems in a similar fashion to that of an FFT. Hence, with the number of sources 2p, where p is a positive integer, the problem is successively reduced to simpler solutions at each stage of order 2p − 1. In its basic form, the method uses only multivariable least mean squares (MLMS), ordinary LMS, and second-order statistics. Copyright © 2012 John Wiley & Sons, Ltd.