SPECTRAL SMOOTHING FOR FREQUENCY-DOMAIN BLIND SOURCE SEPARATION

This paper describes the circularity problem of frequencydomain blind source separation (BSS), and presents a new method for solving it. Frequency-domain BSS performs independent component analysis (ICA) in each frequency bin. It is more efficient than time-domain BSS where ICA is applied to convolutive mixtures. However, frequency-domain BSS has two problems. The first is the permutation problem, for which we have recently proposed a method. It provides a robust and precise solution for the permutation problem and reveals the influence of the second problem, namely the circularity problem. Our solution for this second problem is based on spectral smoothing by windowing. However, the direct application of windowing changes the frequency responses for separation obtained by ICA and causes an error. Therefore, we adjust the frequency responses before windowing so that the error is minimized. The effectiveness of the method is shown by experimental results for the separation of up to four sources.

[1]  Shoko Araki,et al.  ARRAY GEOMETRY ARRANGEMENT FOR FREQUENCY DOMAIN BLIND SOURCE SEPARATION , 2003 .

[2]  Daniel W. E. Schobben,et al.  A frequency domain blind signal separation method based on decorrelation , 2002, IEEE Trans. Signal Process..

[3]  Arthur H. M. van Roermund,et al.  Unsupervised adaptive filtering, volume I: blind source separation [Book Review] , 2002, IEEE Circuits and Devices Magazine.

[4]  Kazuya Takeda,et al.  Blind source separation combining frequency-domain ICA and beamforming , 2001, 2001 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.01CH37221).

[5]  Walter Kellermann,et al.  A GENERALIZATION OF A CLASS OF BLIND SOURCE SEPARATION ALGORITHMS FOR CONVOLUTIVE MIXTURES , 2003 .

[6]  Hiroshi Sawada,et al.  A robust and precise method for solving the permutation problem of frequency-domain blind source separation , 2004, IEEE Transactions on Speech and Audio Processing.

[7]  Paris Smaragdis,et al.  Blind separation of convolved mixtures in the frequency domain , 1998, Neurocomputing.

[8]  Lucas C. Parra,et al.  Convolutive blind separation of non-stationary sources , 2000, IEEE Trans. Speech Audio Process..

[9]  Shiro Ikeda,et al.  A METHOD OF ICA IN TIME-FREQUENCY DOMAIN , 2003 .

[10]  J. Shynk.  Frequency-domain and multirate adaptive filtering , 1992, IEEE Signal Processing Magazine.

[11]  Erkki Oja,et al.  Independent Component Analysis , 2001 .

[12]  Philip Schniter,et al.  FREQUENCY DOMAIN REALIZATION OF A MULTICHANNEL BLIND DECONVOLUTION ALGORITHM BASED ON THE NATURAL GRADIENT , 2003 .

[13]  K. Matsuoka,et al.  Minimal distortion principle for blind source separation , 2002, Proceedings of the 41st SICE Annual Conference. SICE 2002..