간섭 및 반향신호 제거를 위한 다단계 구조의 다채널 암묵 디콘볼루션 = Multichannel Blind Deconvolution of Multistage Structure to Eliminate Interference and Reverberation Signals

국문 초록 (Abstract)

다채널 암묵 디콘볼루션을 자기상관 값이 큰 신호에 적용할 경우 분리필터행렬의 주대각 성분에 의해서 분리신호의 시간백색화가 발생한다. 이러한 왜곡을 줄이기 위해 분리필터 행렬의 주대각 성분을 강제하거나 선형예측 잔여신호를 이용하여 분리필터 행렬을 구하는 방법들이 제안되었지만 신호자신의 반향성분이나 간섭신호 분리에 있어서 문제점이 발생된다. 본 논문에서는 이러한 문제들을 해결하기 위해서 간섭신호의 분리를 위한 단계와 신호자신의 반향을 감소시키기 위한 단계를 분리하여 처리하는 구조의 다채널 암묵 디콘볼루션 방법을 제안한다. 모의실험 결과 혼합신호에서 간섭신호를 분리해낼 수 있을 뿐만아니라 신호 자신의 반향 또한 감소됨을 확인하였다.

다국어 초록 (Multilingual Abstract)

In case that multichannel blind deconvolution (MBD) applies to signals of which autocorrelation has a high level, separated signals are temporally whitened by diagonal elements of a separation filter matrix. In order to reduce this distortion, the algorithms, which are based on either constraining diagonal elements of a separation filter matrix or estimating a separation filter matrix by using linear prediction residual signals, are presented. Still, some problems are generated in these methods, when we separate reverberation of signals themselves or interference signals from mixed signals. To solve these problems, this paper proposes the multichannel blind deconvolution method which divides processing procedure into the stage to separate interference signals and the stage to eliminate a reverberation of signals themselves. In simulation results, we confirm that the proposed algorithm can solve the problems.

참고문헌 (Reference)

1 "Nonholonomic Orthogonal Learning Algorithms for Blind Source Separation" MIT Press 2 : 1463-1484, June,2000

2 "Multichannel blind deconvolution for source separation in convolutive mixtures of speech" 14 : 200-212, January,2006

3 "Multichannel blind deconvolution and equalization using the natural gradient" 101-104, April,1997

4 "Independent Component Anlalysis" Wiley Interscience 2001

5 "Independent Component Analysis : Theory and Applications" 1998

6 "Efficient blind separation of convolved sound mixtures" October,1997

7 "Convolutive blind separation of nonstationary sources" 8 (8): 320327-, May,2000

8 "Convolutive Blind Source Separation by Minimizing Mutual Imformation Between Segment of Signals" 52 (52): 2188-2196, October,2005

9 "Coding for Low Bit Rate Communication Systems" 2004

10 "Blind source separation of real world signals" 1997

11 "Blind separation of convolved sources based on information maximization" 423-432, September,1996

12 "Blind Source Separation Exploiting Higher- Order Frequency Dependencies" 2006

13 "A natural gradient convolutive blind source separation algorithm for speech mixtures" ICA and BSS, San Diego, CA December,2001

14 "A New Learning Algorithm for Blind Signal Separation" MIT Press 1996

15 "A Frequency- Domain Normalized MBD Algorithm with Undirectional Filter for Blind Speech Separation The Journal of The Acoustical Society of Korea" 54-60, 2005