Audio Signal Blind Deconvolution Based on the Quotient Space Hierarchical Theory

The Time-Frequency Domain Blind Deconvolution is discussed based on the quotient space theory. The domain transformation [Xw] → [Xn]G → [X′] and the corresponding granular computing method is introduced to describe the hierarchical structure of deconvolution process, which converts the convolutive mixture of original time-domain signals into instantaneous mixtures in the quotient space. The experimental results show that the algorithm proposed in this paper achieves a relatively high separation quality.

[1]  E. Oja,et al.  Independent Component Analysis , 2001 .

[2]  J. Cardoso,et al.  Blind beamforming for non-gaussian signals , 1993 .

[3]  Hualiang Li,et al.  Gradient and Fixed-Point Complex ICA Algorithms Based on Kurtosis Maximization , 2006, 2006 16th IEEE Signal Processing Society Workshop on Machine Learning for Signal Processing.

[4]  Pierre Comon,et al.  Blind Separation of Independent Sources from Convolutive Mixtures , 2003, IEICE Trans. Fundam. Electron. Commun. Comput. Sci..

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

[6]  Bo Zhang,et al.  A Quotient Space Approximation Model of Multiresolution Signal Analysis , 2005, Journal of Computer Science and Technology.

[7]  Aapo Hyvärinen,et al.  A Fast Fixed-Point Algorithm for Independent Component Analysis of Complex Valued Signals , 2000, Int. J. Neural Syst..

[8]  Bo Zhang,et al.  Theory and Applications of Problem Solving , 1992 .