Research on Blind Source Separation with Noise Based on Multi Factor Analysis

Abstract:

The pre-processing method of blind source separation based on multifactor analysis is proposed to solve the blind source with noise. Firstly, the shortcomings of existing methods of blind source separation are point out after analyzing their principles. The multifactor analysis is introduced in blind source separation and the maximum likelihood estimate based on expectation maximum is used to estimate the common factor and random error. Finally the FastICA algorithm is used to separate BSS result. The validity and the advantage of this method are illustrated by an example.

• Periodical: Applied Mechanics and Materials (Volumes 644-650)
• Pages: 3947-3950
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.644-650.3947
• Online since: September 2014
• Authors: Guo Hua Liu*, Zeng Shan Yin, Zheng Wei Wang, Xiao Song Yao, Wen Zhe Yang
• Keywords: Blind Source Separation, Independent Component Analysis, Maximum Likelihood Estimate, Multi Factor Analysis

References

[1] Zhizhou Li, GuohuaLiu, RuiZhang, Zhencai Zhu, Fault detection, identification and reconstruction for gyroscope in satellite based on independent component analysis, Acta Astronautica vol68 (2011) , p.1015–1023.

DOI: 10.1016/j.actaastro.2010.09.010

[2] A.S. Willsky, A survey of design method for failure detection in dynamic systems, Automatica vol 12 (1976), p.601–611.

DOI: 10.1016/0005-1098(76)90041-8

[3] F.N. Pirmoradi, F. Sassani, C.W. deSilva, Fault detection and diagnosis in a spacecraft attitude determination system, Acta Astronautica vol 65 (2009), p.710–729.

DOI: 10.1016/j.actaastro.2009.03.002

[4] T. Wei, Y. Huang, C.L.P. Chen, Adaptive sensor fault detection and identification using particle filter algorithms, IEEE Transactions on Systems, Man, and Cybernetics–Part C: Applications and Reviews vol 39 (2) (2009) pp.510-514.

DOI: 10.1109/tsmcc.2008.2006759

[5] N. Venkateswaran, M.S. Siva, P.S. Goel, Analytical redundancy based fault detection of gyroscopes in spacecraft applications, Acta Astronautica vol 50 (9) (2002), p.535–545.

DOI: 10.1016/s0094-5765(01)00209-0

[6] C. Lee, S.W. Choi, I. -B. Leem, Sensor fault identification based on time-lagged PCA in dynamic processes, Chemometrics and Intelli- gent Laboratory Systems vol 70 (2004), p.165–178.

DOI: 10.1016/j.chemolab.2003.10.011

[7] J. Lin, A. Zhang, Fault feature separation using wavelet-ICA filter, NDT&E International 38 (2005), p.421–427.

DOI: 10.1016/j.ndteint.2004.11.005

[8] P.O. Hoyer and A. Hyvärinen. Independent Component Analysis Applied to Feature Extraction from Colour and Stereo Images. Network: Computation in Neural Systems, vol 11 (2000), pp.191-210.

DOI: 10.1088/0954-898x_11_3_302

[9] A. Hyvarinen and P. O. Hoyer. Emergence of phase and shift invariant features by decomposition of natural images into independent feature subspaces. Neural Computation, vol 12(2000) pp.1705-1720.

DOI: 10.1162/089976600300015312