Multi-level Independent Component Analysis

  • Woong Myung Kim
  • Chan Ho Park
  • Hyon Soo Lee
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3971)


This paper presents a new method which uses multi-level density estimation technique to generate score function in ICA (independent Component Analysis). Score function is very closely related with density function in information theoretic ICA. We tried to solve mismatch of marginal densities by controlling the number of kernels. Also, we insert a constraint that can satisfy sufficient condition to guarantee asymptotic stability. Multi-level ICA uses kernel density estimation method in order to derive differential equation of source adaptively score function by original signals. To increase speed of kernel density estimation, we used FFT algorithm after changing density formula to convolution form. Proposed multi-level score function generation method reduces estimate error which is density difference between recovered signals and original signals. We estimate density function more similar to original signals compared with existent other algorithms in blind source separation problem and get improved performance in the SNR measurement.


Score Function Original Signal Kernel Density Estimation Independent Component Analysis Blind Signal 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Comon, P.: Independent Component Analysis, A New Concept? Signal Processing 36(3), 287–314 (1994)MATHCrossRefGoogle Scholar
  2. 2.
    Bell, A.J., Sejnowski, T.J.: An Information Maximization Approach to Blind Separation and Blind Deconvolution. Neural Computation 7(6), 1129–1159 (1995)CrossRefGoogle Scholar
  3. 3.
    Amari, S., Cichocki, A., Yang, H.H.: A New Learning Algorithm for Blind Signal Separation. In: Touretzky, D., Mozer, M. (eds.) Advances in Neural Information Processing systems, vol. 8, pp. 757–763 (1996)Google Scholar
  4. 4.
    Cardoso, J.F.: Blind Signal Separation, Statistical Principles. Proc. IEEE Special Issue on Blind Identification and Estimation 9(10), 2009–2025 (1998)Google Scholar
  5. 5.
    Lee, T.-W., Girloami, M., Sejnowski, T.J.: Independent Component Analysis Using Extended Infomax Algorithm for Mixed SubGausssian and SuperGaussian Sources. Neural Computation 1(2), 417–441 (1999)CrossRefGoogle Scholar
  6. 6.
    Hyvarinen, A.: Survey on Independent Component Analysis. Neural Computing Surveys 2, 94–128 (1999)Google Scholar
  7. 7.
    Silverman, B.W.: Density Estimation for Statistics and Data Analysis. Chapman and Hall, New York (1995)Google Scholar
  8. 8.
    Vlassis, N., Motomura., Y.: Efficient Source Adaptivity in Independent Component Analysis. IEEE Trans. Neural Networks 12(3), 559–566 (2001)CrossRefGoogle Scholar
  9. 9.
    Fiori, S.: Blind Signal Processing by the Adaptive Activation Function Neurons. Neural Networks 13(6), 597–611 (2000)CrossRefGoogle Scholar
  10. 10.
    Boscolo, R., Pan, H.: Independent Component Analysis Based on Nonparametric Density Estimation. IEEE Trans. on Neural Networks 15(1), 55–65 (2004)CrossRefGoogle Scholar
  11. 11.
    Kim, W.-M., Lee, H.-S.: An Efficient Score Function Generation Algorithm with Information Maximization. In: Wang, L., Chen, K., S. Ong, Y. (eds.) ICNC 2005. LNCS, vol. 3610, pp. 760–768. Springer, Heidelberg (2005)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Woong Myung Kim
    • 1
  • Chan Ho Park
    • 2
  • Hyon Soo Lee
    • 1
  1. 1.Dept. of Computer EngineeringKyunghee UniversityGyeonggi-doRepublic of Korea
  2. 2.Dept. of Internet Information ScienceBucheon CollegeGyeonggi-doRepublic of Korea

Personalised recommendations