Abstract
A rigorous convergence analysis for the fixed point ICA algorithm of Hyvärinen and Oja is provided and a generalization of it involving cumulants of an arbitrary order is presented. We consider a specific optimization problem OP(p), p>3, integer, arising from a Blind Source Extraction problem (BSE) and prove that every local maximum of OP(p) is a solution of (BSE) in sense that it extracts one source signal from a linear mixture of unknown statistically independent signals. An algorithm for solving OP(p) is constructed, which has a rate of convergence p−1.
Similar content being viewed by others
References
Alekseev, V.M., Tihomirov, V.M., Fomin, S.V.: Optimal Control. Nauka, Moscow (1979) (in Russian)
Amari, S., Cichocki, A.: Adaptive blind signal processing—neural network approaches. Proc. IEEE 86(10), 2016–2048 (1998) (invited paper)
Cichocki, A., Amari, S.: Adaptive Blind Signal and Image Processing. Wiley, Chichester (2002)
Delfosse, N., Loubaton, P.: Adaptive blind separation of independent sources: a deflation approach. Signal Process. 45, 59–83 (1995)
Hyvarinen, A., Karhunen, J., Oja, E.: Independent Component Analysis. Wiley, New York (2001)
Hyvarinen, A., Oja, E.: A fast fixed-point algorithm for independent component analysis. Neural Comput. 9, 1483–1492 (1997)
Hyvarinen, A.: Blind source separation by nonstationarity of variance: a cumulant based approach. IEEE Trans. Neural Netw. 12(6), 1471–1474 (2001)
Shiryaev, A.N.: Probability. Graduate Texts in Mathematics. Springer, New York (1996)
Author information
Authors and Affiliations
Corresponding author
Additional information
Research partially supported by NSF and NIH grants.
Rights and permissions
About this article
Cite this article
Georgiev, P., Pardalos, P. & Cichocki, A. Algorithms with high order convergence speed for blind source extraction. Comput Optim Appl 38, 123–131 (2007). https://doi.org/10.1007/s10589-007-9031-2
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10589-007-9031-2