A New Improved Method to Permutation Ambiguity in BSS with Strong Reverberation

  • Huxiong Li
  • Gu Fan
Conference paper
Part of the Communications in Computer and Information Science book series (CCIS, volume 143)


The major problem of blind source separation in frequency domain is the permutation ambiguity between different frequency bins, which is the key factor to recover the original sources correctly. A new idea is to consider the frequency components from the same source as a multivariate vector with a certain probability density function, and the vectors from different sources are independent each other. An algorithm based on this idea is proposed to solve the permutation ambiguity problem of BSS in frequency domain, and some approximate cost functions are compared with the existing algorithm in frequency domain. The computer simulations to two true speeches with strong reverberation are shown to verify the efficiency of the proposed algorithm.


Blind Source Separation (BSS) Strong Reverberant Environment Multivariate PDF KL Divergence 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Hyvarinen, A., et al.: Independent component analysis. John Wiley and Sons, Chichester (2001)CrossRefzbMATHGoogle Scholar
  2. 2.
    Benesty, J., Sonhi, M.M., Huang, Y., et al.: Handbook of Speech Processing. Springer, Heidelberg (2008)CrossRefGoogle Scholar
  3. 3.
    Saruwatari, H., Kurita, S., Takeda, K.: Blind source separation combining frequency-domain ICA and beamforming. ICASSP 5(7), 2733–2736 (2001)Google Scholar
  4. 4.
    Murata, N., Ikeda, S., Ziehe, A.: An approach to blind source separation based on temporal structure of speech signals. Neurocomputing 41(1-4), 1–24 (2001)CrossRefzbMATHGoogle Scholar
  5. 5.
    Zhu, J., Wang, H., Li, H.: Joint algorithm for permutation problem in frequency-domain in blind speech source separation. Computer Applications 28(6), 1552–1554 (2008) (in chinese)CrossRefzbMATHGoogle Scholar
  6. 6.
    Lee, I., Kim, T., Lee, T.-w.: Fast fixed-point independent vector analysis algorithms for convolutive blind source separation. Signal Processing 87, 1859–1871 (2007)CrossRefzbMATHGoogle Scholar
  7. 7.
    Matsuoka, K., Nakashima, S.: Minimal distortion principle for blind source separation. In: Proceeding of the 41st SICE Annual (SICE 2002), Washington, vol. 4, pp. 2138–2143 (2002)Google Scholar
  8. 8.
    Davies, M.: Audio Source Separation. Mathematics in Signal Processing, vol. 5, pp. 57–68. Oxford University Press, Oxford (2002)zbMATHGoogle Scholar
  9. 9.
    Lee, I., Lee, T.-W.: On the Assumption of Spherical Symmetry and Sparseness for the Frequency-Domain Speech Model. IEEE Trans. on Speech, Audio and Language Processing 15(5), 1521–1528 (2007)CrossRefGoogle Scholar
  10. 10.
    Lehmann, E., Johansson, A.: Diffuse Reverberation Model for Efficient Image-Source Simulation of Room Impulse Responses. IEEE Trans. on Audio, Speech and Language Processing 18(6), 1429–1439 (2010)CrossRefGoogle Scholar
  11. 11.
    Hiroe, A.: Solution of permutation problem in frequency domain ICA, using multivariate probability density functions. In: Rosca, J.P., Erdogmus, D., Príncipe, J.C., Haykin, S. (eds.) ICA 2006. LNCS, vol. 3889, pp. 601–608. Springer, Heidelberg (2006)CrossRefGoogle Scholar
  12. 12.
    Araki, S., Mukai, R., Makino, S., Nishikawa, T., et al.: The fundamental limitation of frequency domain blind source separation for convolutive mixtures of speech. IEEE Trans. Speech and Audio Processing 11, 109–116 (2003)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Huxiong Li
    • 1
  • Gu Fan
    • 2
  1. 1.School of Computer EngineeringWenzhou UniversityWenzhouChina
  2. 2.College of MarineNorthwestern Polytechnic UniversityXi’anChina

Personalised recommendations