An ICA Algorithm Based on Symmetric and Asymmetric Generalized Gaussian Model

Abstract:

Maximum likelihood estimation is a very popular method to estimate the independent component analysis model because of good performance. Independent component analysis algorithm (the natural gradient method) based on this method is widely used in the field of blind signal separation. It potentially assumes that the source signal was symmetrical distribution, in fact in practical applications, source signals may be asymmetric. This article by distinguishing that the source signal is symmetrical or asymmetrical, proposes an improved natural gradient method based on symmetric generalized Gaussian model (People usually call generalized Gaussian model) and asymmetric generalized Gaussian model. The random mixed-signal simulation results show that the improved algorithm is better than the natural gradient separation method.

• Periodical: Advanced Materials Research (Volumes 204-210)
• Pages: 470-475
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.204-210.470
• Online since: February 2011
• Authors: Feng Zhao, Yun Jie Zhang, Min Cai
• Keywords: Asymmetric Generalized Gaussian Model, Natural Gradient Method, Symmetric Generalized Gaussian Model

References

[1] Potanutis L. et al. Independent component analysis applied to feature extraction for robust automatic speech recognition. Electronics Letters, vol. 36(23), (2000), p.1977- (1978).

DOI: 10.1049/el:20001365

[2] Li S. Z. et al. Learing multiview face subspaces and facial pose estimation using independent component analysis. IEEE Trans. Image Processing, vol. 14(6), (2005), pp.705-712.

DOI: 10.1109/tip.2005.847295

[3] Dagher I., Nachar R. Face recognition using IPCA-ICA algorithm. IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 28(6), (2006), pp.996-1000.

DOI: 10.1109/tpami.2006.118

[4] Cardoso J. F., Souloumiac A. Blind beamforming for non Gaussian signals. Radar and Signal Processing, IEE Proceedings F, vol. 140(6), (1993), pp.362-370.

DOI: 10.1049/ip-f-2.1993.0054

[5] Cichocki A., Unbehanen R., Rummert E. Robust learning algorithm for blind separation of signals. Electronics Letters, vol. 30(17), (1994) , pp.1386-1387.

DOI: 10.1049/el:19940956

[6] Cardoso J. F. Blind signal separation: statistical principles. Proceedings of the IEEE, vol. 86(10), (1998), p.2009-(2025).

DOI: 10.1109/5.720250

[7] Hyvärinen A., Oja E. Independent component analysis: algorithms and applications. Neural Networks, vol. 13(4-5), (2000), pp.411-430.

DOI: 10.1016/s0893-6080(00)00026-5

[8] Hyvärinen A., Karhunen J., Oja E. Independent component analysis. New York: John Wiley & Sons Inc., (2001).

[9] Ying Tang and Jianping Li, Normalized natural gradient in independent component analysis, Signal Processing, vol. 90(9), (2010), pp.2773-2777.

DOI: 10.1016/j.sigpro.2010.03.015

[10] Li Z. M., Wang T.Y., Li H.W. Application to Locally Optimum Detection Based on Asymmetric Generalized Gaussian distribution. Natural Science Journal of HaiNan University, vol. 27(1), (2000) , pp.73-77.