Mixing Matrix Estimation in Blind Source Separation Based on Generalized Gaussian Mixture Modal

Abstract:

The accurate estimation of mixing matrix is critical for blind separation, for solving the problems of traditional methods such as bad robustness and low accuracy, a method based on statistical modal is proposed. The generalized Gaussian mixture modal is used to fit the distribution of single-source-points, a new objective function for clustering is obtained from the view of maximum likelihood estimation. Constrained particle swarm optimization is used to optimize the objective function, by which the mixing matrix is estimated. This method is applicable to determined and underdetermined blind source separation. The simulation shows that the proposed method has higher estimation accuracy and is more robust than traditional methods.

• Periodical: Advanced Engineering Forum (Volume 4)
• Pages: 217-221
• DOI: https://doi.org/10.4028/www.scientific.net/AEF.4.217
• Online since: June 2012
• Authors: Yong Qiang Chen, Jun Liu
• Keywords: Constrained Particle Swarm Optimization, Generalized Gaussian Mixture Modal, Mixing Matrix Estimation, Underdetermined Blind Source Separation

References

[1] S Choi, A Cichocki, H. M. Park, et al. Blind Source Separation and Independent Component Analysis: A Review [J]. Neural Information Processing, 2005, 6(1): 1-57.

[2] Chen Y Q, Tian Y F. Blind signal separation based on ppso-grads algorithm[J]. Journal of Electronic& Information Technology, 2007, 29(8): 1817-1820.

[3] A Belouchrani, M.G. Amin. Blind source separation based on time–frequency signal representations [J]. IEEE Trans. on Signal Processing , 1998, 46 (11): 2888–2897.

DOI: 10.1109/78.726803

[4] S Rickard, O Yilmaz. On the approximate W-disjoint orthogonality of speech[C]/ Proceedings of ICASSP, 2002: 529–532.

[5] O Yilmaz, S Rickard. Blind separation of speech mixtures via time–frequency masking [J]. IEEE Trans. on Signal Processing, 2004, 52 (7): 1830–1847.

DOI: 10.1109/tsp.2004.828896

[6] M Pedersen, T Lehn-Schioler, J Larsen. Blues from music: Blind underdetermined extraction of Sources from music[C]/ Proceedings of Independent Component Analysis and Blind Signal Separation, 2006: 392–399.

DOI: 10.1007/11679363_49

[7] N Linh-Trung, A Belouchrani, Abed-Meraimet K, et al. Separating more sources than sensors using time–frequency distributions[C]/Proceedings of the International Symposium on Signal Processing and its Applications, 2001: 583–586.

DOI: 10.1109/isspa.2001.950212

[8] A Aissa-El-Bey, N Linh-Trung, Abed-Meraimet K, et al. Underdetermined blind separation of nondisjoint sources in the time- frequency domain [J]. IEEE Trans. on Signal Processing, 2007, 55(3): 897–907.

DOI: 10.1109/tsp.2006.888877

[9] S. G. Kim, C. D. Yoo. Underdetermined Blind Source Separation Based on Subspace Representation [J]. IEEE Trans. on Signal Processing, 2009, 57(7): 2604-2614.

DOI: 10.1109/tsp.2009.2017570

[10] S Arberet, A Ozerov, R Gribonval, et al. Blind spectral-GMM estimation for underdetermined instantaneous audio source separation[C]/ Proceedings of 8th International Conference on Independent Component Analysis and Signal Separation, 2009: 751-758.

DOI: 10.1007/978-3-642-00599-2_94

[11] G. T. Pulido, C. A. C. Coello. A Constraint-Handling Mechanism for Particle Swarm Optimization[C]/ Proceedings of Evolutionary Computation, 2004, 2: 1396 – 1403.

DOI: 10.1109/cec.2004.1331060

[12] S Zhang, S.X. Gong. A modified PSO for low sidelobe concentric ring arrays synthesis with multiple constraints [J]. Journal of Electromagnetic Waves and Applications, 2009, 23: 1535-1544.

DOI: 10.1163/156939309789476239