Research on Algorithm for Blind Source Separation Based on Negentropy Maximization

Article Preview

Abstract:

To detect week signal of underwater magnetic fields, we designed a negentropy maximization blind source separation algorithm based on the stochastic gradient descent algorithm by bring in a penalty factor. The simulation study demonstrates that the new algorithm is effective to separate the mixed-signal with high precision, to decrease the iterative calculation and to enhance the convergence rate.

You might also be interested in these eBooks

Info:

Periodical:

Pages:

1911-1916

Citation:

Online since:

September 2013

Export:

Price:

Permissions CCC:

Permissions PLS:

Сopyright:

© 2013 Trans Tech Publications Ltd. All Rights Reserved

Share:

Citation:

* - Corresponding Author

[1] Ju Liu and Zhenya He. Blind Source Separation and Blind Deconvolution [J]. Chinese Journal of Electronics. 2002, 30(4): 570-576P.

Google Scholar

[2] P. Comon. Independent Component analysis, a new concept Signal Processing. 1994, 36(3): 287-314P.

DOI: 10.1016/0165-1684(94)90029-9

Google Scholar

[3] Aapo Hyvarinen and E. Oja. A Fastfixed-Point Algorithm for Independent Component Analysis [J]. Neural Computation. 1997, 9(7): 1483-1492P.

Google Scholar

[4] Joho M, Rahbar K. Joint Diagonalization of Correlation Matrices by Using Newton Methods with Application to Blind Signal Separation, SAM 2002, 2002. available http: /www. phonak. uiuc. edu.

DOI: 10.1109/sam.2002.1191070

Google Scholar

[5] Aapo Hyvarinen. Fast and robust fixed-point algorithm for independent component analysis [J]. IEEE Transactions on Neural Networks. 1999, 10(3): 626-634P.

DOI: 10.1109/72.761722

Google Scholar