Blind Source Separation Algorithm using the Second-Order Statistics

이차 통계치를 이용한 블라인드 신호분리 알고리즘

  • 김천수 (텔슨전자주식회사) ;
  • 양완철 (한국항공대학교 전자·정보통신·컴퓨터공학부) ;
  • 이병섭 (한국항공대학교 전자·정보통신·컴퓨터공학부)
  • Published : 2002.02.01

Abstract

The problem of blind signal separation of independent sources consist in retrieving the source from the observation of unknown mixtures of unknown sources. In this paper, we propose a technique for blind signal separation that can extract original signals from their non-stationary mixtures observed in a ordinary room. The proposed method implements blind signal separation by minimizing a non-negative cost function that achieves the minimum when the second-order cross-correlation value of the observed signals becomes zero. The validity of the proposed method has been verified by a computer simulation and experiment that extracts two source signals from their mixtures observed in a normal room.

미지의 신호원들의z 합성으로부터 관측된 신호만을 이용하여 통계적으로 독립인 원신호를 추출하는 문제를 블라인드 신호분리라 한다. 본 논문에서는 보통의 실내에서 얻어진 비정상(non-stationary) 합성신호로부터 원신호론 추출해내는 블라인드 신호분리 기법을 제안한다. 제안된 기법은 관측 신호들 간의 이타 상호상관 값이 제로가 될 때만 최소값을 가지는 비용함수를 최소화시키는 방식으로 블라인드 신호분리를 구현한다. 제안된 기법의 유효성을 컴퓨터 시뮬레이션과 보통의 실내에서 관측된 2개의 합성신호로부터 2개의 원신호를 추출해내는 실험을 통하여 증명한다.

Keywords

References

  1. Proc. Xeme colloque GRETSI Detection de grandeurs primitives dans un message composite par une architecture de calcul neurominetique en apprentissage non supervise J. Herault;C. Jutten;B. Ans
  2. AIP Conf. Proc. v.151 Space and time adaptive signal processing by neural network model J. Herault;C. Jutten
  3. Signal Processing v.36 Independednt Component Analysis, A New Concept? P. Comon https://doi.org/10.1016/0165-1684(94)90029-9
  4. Signal Processing v.24 Blind separation of sources. Part I: An adaptive algorithm based on neuromimetic architecture C. Jutten;J. Herault https://doi.org/10.1016/0165-1684(91)90079-X
  5. Neural Computation v.11 no.2 Independent Component Analysis using an Extended Infomax Algorithm for Mixed Sub-Gaussian and Super-Gaussian Sources T.-W. Lee;M. Girolami;T.J. Sejnowski https://doi.org/10.1162/089976699300016719
  6. Proc. Int. Symp. on Nonlinear Theory and Its Application Blind separation of delayed and convolutive signals with self-adaptive learning rate A. Cichoki;S. Amari;J. Cao
  7. Proc. 1st IEEE Workshop on Signal Processing App. Wireless Comm. Multichannel blind deconvolution using the natural gradient S. Amari;S.C. Douglas;A. Cichocki
  8. Advances in neural information processing system v.9 Blind separation of delayed and convolved sources T. W. Lee;A. J. Bell; R. H. Lambert
  9. IEEE Workshop on Neural Networks for Signal Processing Ⅵ Kyoto Blind separation of convolved sources based on information maximazation K.Tokkola
  10. Independent Component Analysis Aapo Hyvarien;Juha Karhunen;Erkki Oja
  11. IEEE Trans Acoustic Speech Signal Process v.36 Inverse filtering of room acoustics M. Miyoshi;Y. Kaneda https://doi.org/10.1109/29.1509
  12. IEICE Technical Report 97 A new algorithm for blind separation of convolved sources K. Matsuoka;T. Tokunari