A New Orthogonal Projected Natural Gradient BSS Algorithm with a Dynamically Changing Source Number under Over-Determined Mode

Abstract:

Article Preview

Blind source separation (BSS) attempts to recover unknown independent sources from a given set of observed mixtures. Algorithm based on natural gradient is one of the main methods in BSS. An analysis has been done on the problem that the old algorithm goes to diverging under over-determined mode. A new improved algorithm based on orthogonal projected natural gradient is studied in the paper. The simulated result using crosstalk error proves the capability to perform the BSS under over-determined mode and the better convergence stability of the new algorithm. It is also effective with a dynamically changing source number.

Info:

Periodical:

Edited by:

Yanwen Wu

Pages:

768-773

Citation:

P. Wang et al., "A New Orthogonal Projected Natural Gradient BSS Algorithm with a Dynamically Changing Source Number under Over-Determined Mode", Advanced Materials Research, Vol. 267, pp. 768-773, 2011

Online since:

June 2011

Export:

Price:

$38.00

[1] Zhang Xianda, Bao zheng: Blind source separation, Chinese Journal of Electronics. Dec 2001, Vol 29(12). 1766-1771.

[2] Yang Xiaoniu, Fu Weihong: Blind source separation--Theory, Application and Outlook. Communication Countermeasures . 2006, Vol 3. 3-10.

[3] Bell A J, Sejnowski T J : An information-maximization approach to blind separation and blind deconvolution. Neural Computation. 1995, Vol. 7(6). 1129-1159.

DOI: https://doi.org/10.1162/neco.1995.7.6.1129

[4] Xi-Lin Li, Xian-Da Zhang : Nonorthogonal Joint Diagonalization Free of Degenerate Solution. IEEE Transactions on Signal Processing. May 2007, Vol 55(5). 1805-1808.

DOI: https://doi.org/10.1109/tsp.2006.889983

[5] Zhang L Q, Cichocki A, Amari S: Natural gradient algorithm for blind separation of overdetermined mixtures with additive noise. IEEE Signal Processing Letters. 1999, Vol 6(11). 293-295.

DOI: https://doi.org/10.1109/97.796292

[6] Amari S: Natural Gradient Learning for Over and Under Complete Bases in ICA. Neural Computation . 1999, Vol 11(3). 1875-1883.

DOI: https://doi.org/10.1162/089976699300015990

[7] Ye J M, Zhu X L, Zhang X D: Adaptive Blind Separation with an Unknown Number of Sources. Neural Computation. 2004, Vol 16(8). 1641-1660.

DOI: https://doi.org/10.1162/089976604774201622

[8] Ali MANSOUR, Mitsuru KAWAMOTO, Noboru QHNISHI: A Survey of The Performance Indexes of ICA Algorithms. Proceedings IASTED International Conference MODELING, IDENTIFICATION, AND CONTROL, Februray 18-21, 2002, Innsbruck, Austria.