Performance analysis of the FastICA algorithm and Cramér-Rao bounds for linear independent component analysis

IEEE Transactions on Signal Processing

Volumn 54, Issue 4, 2006, Pages 1189-1203

  Tichavský, Petr ^a   Koldovský, Zbyněk ^a,b   Oja, Erkki ^c  

^a INSTITUTE OF INFORMATION THEORY AND AUTOMATION   (Czech Republic)

^b TECHNICAL UNIVERSITY OF LIBEREC   (Czech Republic)

^c AALTO UNIVERSITY   (Finland)

Author keywords

Blind source separation; Cram Rao lower bound; Independent component analysis (ICA)

Indexed keywords

ALGORITHMS; COMPUTATIONAL COMPLEXITY; COMPUTER SIMULATION; INDEPENDENT COMPONENT ANALYSIS;

FIXED-POINT ALGORITHM; LINEAR INDEPENDENT COMPONENT ANALYSIS; LOWER BOUND; ONE-UNIT ALGORITHM; SADDLE POINTS; SYMMETRIC ALGORITHM;

BLIND SOURCE SEPARATION;

EID: 33645288306     PISSN: 1053587X     EISSN: None     Source Type: Journal    
DOI: 10.1109/TSP.2006.870561     Document Type: Article

References (33)

1. S.-I. Amari and A. Cichocki, Adaptive Blind Signal and Image Processing. New York: Wiley, 2002.
2. P. Comon, "Independent components analysis: A new concept?," Signal Process., vol. 36, no. 3, pp. 287-314, 1994.
3. A. Hyvärinen, J. Karhunen, and E. Oja, Independent Component Analysis. New York: Wiley-Interscience, 2001.
4. D. Donoho, "On minimum entropy deconvolution," in Applied Time-Series Analysis II. New York: Academic, 1981, pp. 565-609.
5. P. Comon, "Contrasts for multichannel blind deconvolution," IEEE Signal Process. Lett., vol. 3, no. 7, pp. 209-211, Jul. 1996.
6. O. Shalvi and E. Weinstein, "New criteria for blind deconvolution of nonminimum phase systems (channels)," IEEE Trans. Inf. Theory, vol. 36, no. 2, pp. 312-321, Mar. 1990.
7. A. Hyvärinen and E. Oja, "A fast fixed-point algorithm for independent component analysis," Neural Comput., vol. 9, pp. 1483-1492, 1997.
8. A. Hyvärinen, "Fast and robust fixed-point algorithms for independent component analysis," IEEE Trans. Neural Netw., vol. 10, no. 3, pp. 626-634, May 1999.
9. N. Delfosse and P. Loubaton, "Adaptive blind separation of independent sources: A deflation approach," Signal Process., vol. 45, pp. 59-83, 1995.
10. J.-F. Cardoso and B. H. Laheld, "Equivariant adaptive source separation," IEEE Trans Signal Process., vol. 44, no. 12, pp. 3017-3030, Dec. 1996.
11. C. B. Papadias, "Globally convergent blind source separation based on a multiuser kurtosis maximization criterion," IEEE Trans. Signal Process., vol. 48, no. 12, pp. 3508-3519, Dec. 2000.
12. A. Hyvärinen and E. Bingham, "A fast fixed-point algorithm for independent component analysis of complex-valued signals," Int. J. Neural Syst., vol. 10, no. 1, pp. 1-8, 2000.
13. P. A. Regalia and E. Kofidis, "Monotonic convergence of fixed-point algorithms for ICA," IEEE Trans. Neural Netw., vol. 14, no. 4, pp. 943-949, Jul. 2003.
14. E. Oja, "Convergence of the symmetrical FastICA algorithm," in 9th Int. Conf. Neural Information Processing (ICONIP), Singapore, Nov. 18-22, 2002.
15. S. Douglas, "On the convergence behavior of the FastICA algorithm," in Proc. 4th Symp. Independent Component Analysis Blind Source Separation, Nara, Japan, Apr. 2003, pp. 409-414.
16. X. Giannakopoulos, J. Karhunen, and E. Oja, "Experimental comparison of neural algorithms for independent component analysis and blind separation," Int. J. Neural Syst., vol. 9, pp. 651-656, 1999.
17. A. Hyvärinen, "One-unit contrast functions for independent component analysis: A statistical analysis," in Proc. IEEE Neural Networks for Signal Processing (NNSP) Workshop, Amelia Island, FL, 1997, pp. 388-397.
18. R. C. Rao, Linear Statistical Inference and Its Applications, 2nd ed. New York: Wiley, 1973.
19. K. Waheed and F. M. Salam, "Blind source recovery using an adaptive generalized Gaussian score function," in Proc. 45th Midwest Symp. Circuits Systems (MWSCAS), vol. 3, Aug. 4-7, 2002, pp. 656-659.
20. P. Tichavský and Z. Koldovský, "Optimal pairing of signal components separated by blind techniques," IEEE Signal Process. Lett., vol. 11, no. 2, pp. 119-122, Feb. 2004.
21. Z. Koldovský, P. Tichavský, and E. Oja, "Cramér-Rao lower bound for linear independent component analysis," in Proc. IEEE Int. Conf. Acoustics, Speech, Signal Processing (ICASSP), vol. III, Philadelphia, PA, Mar. 2005, pp. 581-584.
22. E. G. Learned-Miller and J. W. Fisher III, "ICA using spacings estimates of entropy," J. Mach. Learning Res., vol. 4, pp. 1271-1295, 2003.
23. R. Boscolo, H. Pan, and V. P. Roychowdhury, "Independent component analysis based on nonparametric density estimation," IEEE Tran. Neural Netw., vol. 15, no. 1, pp. 55-65, Jan. 2004.
24. J.-F. Cardoso, "Blind signal separation: Statistical principles," Proc. IEEE, vol. 90, no. 8, pp. 2009-2026, Oct. 98.
25. J.-F. Cardoso, "On the performance of orthogonal source separation algorithms," in Proc. EUSIPCO, Edinburgh, U.K., Sep. 1994, pp. 776-779.
26. D. T. Pham and P. Garat, "Blind separation of mixture of independent sources through a quasi-maximum likelihood approach," IEEE Trans. Signal Process., vol. 45, no. 7, pp. 1712-1725, Jul. 1997.
27. D. Yellin and B. Friedlander, "Multichannel system identification and deconvolution: Performance bounds," IEEE Trans. Signal Process., vol. 47, no. 5, pp. 1410-1414, May 1999.
28. O. Shalvi and E. Weinstein, "Maximum likelihood and lower bounds in system identification with non-Gaussian inputs," IEEE Trans. Inf. Theory, vol. 40, no. 2, pp. 328-339, Mar. 1994.
29. V. Vigneron and Ch. Jutten, "Fisher information in source separation problems," in Proc. 5th Int. Conf. Independent Component Analysis (ICA 2004), C. G. Puntonet and A. Prieto, Eds., Granada, Spain, Sep. 2004, pp. 168-176.
30. A. M. Bronstein, M. M. Bronstein, M. Zibulevsky, and Y. Y. Zeevi, "QML, blind deconvolution: Asymptotic analysis," in Proc. 5th Int. Conf. Independent Component Analysis (ICA 2004), C. G. Puntonet and A. Prieto, Eds., Granada, Spain, Sep. 2004, pp. 677-684.
31. B. Porat, Digital Processing of Random Signals: Theory and Methods. Englewood Cliffs, NJ: Prentice-Hall, 1993.
32. Y. Lomnitz and A. Yeredor, "A blind-ML scheme for blind source separation," presented at the IEEE Workshop on Statistical Signal Processing (SSP'03), St. Louis, MO, Sep. 2003.
33. Z. Koldovský, P. Tichavský, and E. Oja, "Efficient variant of algorithm FastICA for independent component analysis attaining the Cramer-Rao lower bound," IEEE Trans. Neural Netw., to be published.