Kybernetika 47 no. 3, 426-438, 2011

Fast and accurate methods of independent component analysis: A Survey

Petr Tichavský and Zbyněk Koldovský

Abstract:

This paper presents a survey of recent successful algorithms for blind separation of determined instantaneous linear mixtures of independent sources such as natural speech or biomedical signals. These algorithms rely either on non-Gaussianity, nonstationarity, spectral diversity, or on a combination of them. Performance of the algorithms will be demonstrated on separation of a linear instantaneous mixture of audio signals (music, speech) and on artifact removal in electroencephalogram (EEG).

Keywords:

probability distribution, Blind source separation, score function, autoregressive random processes, audio signal processing, electroencephalogram, artifact rejection

Classification:

94A12, 92-02, 92-04, 92-08

paper.pdf

References:

  1. A. Belouchrani, K. Abed-Meraim, J.-F. Cardoso and E. Moulines: A blind source separation technique using second-order statistics. IEEE Trans. Signal Processing 45 (1997), 434-444.   CrossRef
  2. R. Boscolo, H. Pan and V. P. Roychowdhury: Independent component analysis based on nonparametric density estimation. IEEE Trans. Neural Networks 15 (2004), 55-65.   CrossRef
  3. J.-F. Cardoso: Blind signal separation: statistical principles. Proc. IEEE 90 (1998), 2009-2026.   CrossRef
  4. J. Chambers, W. Cleveland, B. Kleiner and P. Tukey: Graphical Methods for Data Analysis. Wadsworth, 1983.   CrossRef
  5. S. Cruces, A. Cichocki and L. De Lathauwer: Thin QR and SVD factorizations for simultaneous blind signal extraction. In: Proc. European Signal Processing Conference (EUSIPCO), Vienna 2004, pp. 217-220.   CrossRef
  6. S. Dégerine and A. Za\"{\i}di: Separation of an instantaneous mixture of Gaussian autoregressive sources by the exact maximum likelihood approach. IEEE Trans. Signal Processing 52 (2004), 1492-1512.   CrossRef
  7. A. Delorme, T. Sejnowski and S. Makeig: Enhanced detection of artifacts in EEG data using higher-order statistics and independent component analysis. Neuroimage 34 (2007) 1443-1449.   CrossRef
  8. A. Hyvärinen and E. Oja: A fast fixed-point algorithm for independent component analysis. Neural Computation 9 (1997), 1483-1492.   CrossRef
  9. A. Hyvärinen, J. Karhunen and E. Oja: Independent Component Analysis. John Wiley \   CrossRef
  10. C. J. James and C. W. Hesse: Independent component analysis for biomedical signals. Physiol. Meas. 26 (2005), R15-R39.   CrossRef
  11. Z. Koldovský, P. Tichavský and E. Oja: Efficient variant of algorithm FastICA for independent component analysis attaining the Cramér-Rao lower bound. IEEE Trans. Neural Networks 17 (2006), 1265-1277.   CrossRef
  12. Z. Koldovský, J. Málek, P. Tichavský, Y. Deville and S. Hosseini: Blind separation of piecewise stationary nonGaussian Sources. Signal Process. 89 (2009), 2570-2584.   CrossRef
  13. Z. Koldovský and P. Tichavský: A comparison of independent component and independent subspace analysis algorithms. In: Proc. European Signal Processing Conference (EUSIPCO), Glasgow 2009, pp. 1447-1451.   CrossRef
  14. Z. Koldovský and P. Tichavský: Time-domain blind separation of audio sources based on a complete ICA decomposition of an observation space. IEEE Trans. Audio, Speech and Language Processing 19 (2011), 406-416.   CrossRef
  15. E. G. Learned-Miller and J. W. Fisher III: ICA using spacings estimates of entropy. J. Machine Learning Research 4 (2004), 1271-1295.   CrossRef
  16. Te-Won Lee: Independent Component Analysis, Theory and Applications. Kluwer Academic Publishers, 1998.   CrossRef
  17. D.-T. Pham: Joint approximate diagonalization of positive definite Hermitian matrices. SIAM J. Matrix Anal. Appl. 22 (2001), 1136-1152.   CrossRef
  18. D.-T. Pham: Blind separation of instantaneous mixture of sources via the Gaussian mutual information criterion. Signal Process. 81 (2001), 855-870.   CrossRef
  19. D.-T. Pham and J.-F. Cardoso: Blind separation of instantaneous mixtures of nonstationary sources. IEEE Trans. Signal Process. 49 (2001), 1837-1848.   CrossRef
  20. P. Tichavský, Z. Koldovský and E. Oja: Performance analysis of the FastICA algorithm and Cramér-Rao bounds for linear independent component analysis. IEEE Trans. Signal Process. 54 (2006), 1189-1203.   CrossRef
  21. P. Tichavský, Z. Koldovský and E. Oja: Corrections to ``Performance analysis of the FastICA algorithm and Cramér-Rao Bounds for linear independent component analysis, TSP 04/06". IEEE Tran. Signal Process. 56 (2008), 1715-1716.   CrossRef
  22. P. Tichavský, Z. Koldovský and E. Oja: Speed and accuracy enhancement of linear ICA techniques using rational nonlinear functions. Lecture Notes in Comput. Sci. {\mi 4666} (2007), 285-292.   CrossRef
  23. P. Tichavský, Z. Koldovský, A. Yeredor, G. Gomez-Herrero and E. Doron: A hybrid technique for blind non-Gaussian and time-correlated sources using a multicomponent approach. IEEE Trans. Neural Networks 19 (2008), 421-430.   CrossRef
  24. P. Tichavský and A. Yeredor: Fast approximate joint diagonalization incorporating weight matrices. IEEE Trans. Signal Process. 57 (2009), 878-891.   CrossRef