Advertisement

Temporal Decorrelation as Preprocessing for Linear and Post-nonlinear ICA

  • Juha Karvanen
  • Toshihisa Tanaka
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3195)

Abstract

We present a straightforward way to use temporal decorrelation as preprocessing in linear and post-nonlinear independent component analysis (ICA) with higher order statistics (HOS). Contrary to the separation methods using second order statistics (SOS), the proposed method can be applied when the sources have similar temporal structure. The main idea is that componentwise decorrelation increases non-Gaussianity and therefore makes it easier to separate sources with HOS ICA. Conceptually, the non-Gaussianizing filtering matches very well with the Gaussianization used to cancel the post-nonlinear distortions. Examples demonstrating the consistent improvement in the separation quality are provided for the both linear and post-linear cases.

Keywords

Independent Component Analysis Independent Component Analysis Linear Prediction Second Order Statistic Blind Signal 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Cichocki, A., Amari, S.I.: Adaptive Blind Signal and Image Processing: Learning Algorithms and Applications. Wiley, Chichester (2002)CrossRefGoogle Scholar
  2. 2.
    Lee, T.W., Ziehe, A., Orglmeister, R., Sejnowski, T.: Combining time-delayed decorrelation and ICA: towards solving the cocktail party problem. In: Proc. ICASSP 1998, vol. 2, pp. 1249–1252 (1998)Google Scholar
  3. 3.
    Kokkinakis, K., Zarzoso, V., Nandi, A.K.: Blind separation of acoustic mixtures based on linear prediction analysis. In: Proc. Fourth International Symposium on Independent Component Analysis and Blind Signal Separation (ICA 2003), pp. 343–348 (2003)Google Scholar
  4. 4.
    Nishikawa, T., Sarauwatari, H., Shikano, K.: Stable learning algorithm for blind separation of temporally correlated signals combining multistage ICA and linear prediction. In: Proc. Fourth International Symposium on Independent Component Analysis and Blind Signal Separation (ICA 2003), pp. 337–342 (2003)Google Scholar
  5. 5.
    Cichocki, A., Rutkowski, T., Siwek, K.: Blind signal extraction of signals with specified frequency band. In: Proc. International Workshop on Neural Networks for Signal Processing (2002)Google Scholar
  6. 6.
    Mandic, D.P., Cichocki, A.: An online algorithm for blind extraction of sources with different dynamical structures. In: Proc. Fourth International Symposium on Independent Component Analysis and Blind Signal Separation (ICA 2003), pp. 645–650 (2003)Google Scholar
  7. 7.
    Tanaka, T., Cichocki, A.: Subband decomposition independent component analysis and new performance criteria. In: Proc. of 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2004 (2004)Google Scholar
  8. 8.
    Cichocki, A., Amari, S., Siwek, K., Tanaka, T., et al.: ICALAB Toolboxes (2002), http://www.bsp.brain.riken.jp/ICALAB
  9. 9.
    Jung, A., Kaiser, A.: Considering temporal structures in independent component analysis. In: Proc. Fourth International Symposium on Independent Component Analysis and Blind Signal Separation (ICA 2003), pp. 95–100 (2003)Google Scholar
  10. 10.
    Hyvärinen, A.: Independent component analysis for time-dependent stochastic processes. In: Proc. Int. Conf. on Artificial Neural Networks (ICANN 1998), pp. 541–546 (1998)Google Scholar
  11. 11.
    Jutten, C., Karhunen, J.: Advances in nonlinear blind source separation. In: Proc. Fourth International Symposium on Independent Component Analysis and Blind Signal Separation (ICA 2003), pp. 245–256 (2003)Google Scholar
  12. 12.
    Ziehe, A., Kawanabe, M., Harmeling, S., Müller, K.R.: Blind separation of postnonlinear mixtures using Gaussianizing transformations and temporal decorrelation. In: Proc. Fourth International Symposium on Independent Component Analysis and Blind Signal Separation (ICA 2003), pp. 269–274 (2003)Google Scholar
  13. 13.
    Stuart, A., Ord, J.K.: Kendall’s Advanced Theory of Statistics: Distribution Theory, 6th edn., vol. 1. Edward Arnold, London (1994)Google Scholar
  14. 14.
    Jackson, L.: Digital Filters and Signal Processing, 2nd edn. Kluwer Academic Publishers, Dordrecht (1989)Google Scholar
  15. 15.
    Karvanen, J., Koivunen, V.: Blind separation methods based on Pearson system its extensions. Signal Processing 82, 663–673 (2002)MATHCrossRefGoogle Scholar
  16. 16.
    Cardoso, J., Souloumiac, A.: Blind beamforming for non Gaussian signals. IEEProceedings- F 140, 362–370 (1993)Google Scholar
  17. 17.
    Belouchrani, A., Meraim, K.A., Cardoso, J.F., Moulines, E.: A blind source separation technique based on second order statistics. IEEE Transactions on Signal Processing 45, 434–444 (1997)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Juha Karvanen
    • 1
  • Toshihisa Tanaka
    • 1
  1. 1.Laboratory for Advanced Brain Signal ProcessingBrain Science Institute, RIKENSaitamaJapan

Personalised recommendations