MAP-Based Underdetermined Blind Source Separation of Convolutive Mixtures by Hierarchical Clustering and -Norm Minimization

We address the problem of underdetermined BSS. While most previous approaches are designed for instantaneous mixtures, we propose a time-frequency-domain algorithm for convolutive mixtures. We adopt a two-step method based on a general maximum a posteriori (MAP) approach. In the first step, we estimate the mixing matrix based on hierarchical clustering, assuming that the source signals are sufficiently sparse. The algorithm works directly on the complex-valued data in the time-frequency domain and shows better convergence than algorithms based on self-organizing maps. The assumption of Laplacian priors for the source signals in the second step leads to an algorithm for estimating the source signals. It involves the-norm minimization of complex numbers because of the use of the time-frequency-domain approach. We compare a combinatorial approach initially designed for real numbers with a second-order cone programming (SOCP) approach designed for complex numbers. We found that although the former approach is not theoretically justified for complex numbers, its results are comparable to, or even better than, the SOCP solution. The advantage is a lower computational cost for problems with low input/output dimensions.

[1]  David B. Dunson,et al.  Bayesian Data Analysis , 2010 .

[2]  B. De Moor,et al.  ICA techniques for more sources than sensors , 1999, Proceedings of the IEEE Signal Processing Workshop on Higher-Order Statistics. SPW-HOS '99.

[3]  Rémi Gribonval,et al.  BSS_EVAL Toolbox User Guide -- Revision 2.0 , 2005 .

[4]  Vamsi K. Potluru,et al.  Sparse separation: principles and tricks , 2003, SPIE Defense + Commercial Sensing.

[5]  Özgür Yilmaz,et al.  On the approximate W-disjoint orthogonality of speech , 2002, 2002 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[6]  Jan Larsen,et al.  BLUES from Music: BLind Underdetermined Extraction of Sources from Music , 2006, ICA.

[7]  Hiroshi Sawada,et al.  Overcomplete BSS for Convolutive Mixtures Based on Hierarchical Clustering , 2004, ICA.

[8]  Terrence J. Sejnowski,et al.  Learning Overcomplete Representations , 2000, Neural Computation.

[9]  Rémi Gribonval,et al.  Performance measurement in blind audio source separation , 2006, IEEE Transactions on Audio, Speech, and Language Processing.

[10]  Daniel Patrick Whittlesey Ellis,et al.  Prediction-driven computational auditory scene analysis , 1996 .

[11]  D. Ruppert The Elements of Statistical Learning: Data Mining, Inference, and Prediction , 2004 .

[12]  Athanasios Papoulis,et al.  Probability, Random Variables and Stochastic Processes , 1965 .

[13]  Hiroshi Sawada,et al.  Blind extraction of a dominant source signal from mixtures of many sources [audio source separation applications] , 2005, Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005..

[14]  T. Sikora,et al.  On the Use of Auditory Representations for Sparsity-Based Sound Source Separation , 2005, 2005 5th International Conference on Information Communications & Signal Processing.

[15]  Mineichi Kudo,et al.  Performance analysis of minimum /spl lscr//sub 1/-norm solutions for underdetermined source separation , 2004, IEEE Transactions on Signal Processing.

[16]  Deniz Erdogmus,et al.  Estimation of the mixing matrix for underdetermined blind source separation using spectral estimation techniques , 2002, 2002 11th European Signal Processing Conference.

[17]  Barak A. Pearlmutter,et al.  Survey of sparse and non‐sparse methods in source separation , 2005, Int. J. Imaging Syst. Technol..

[18]  Lars Kai Hansen,et al.  Blind Separation of More Sources than Sensors in Convolutive Mixtures , 2006, 2006 IEEE International Conference on Acoustics Speech and Signal Processing Proceedings.

[19]  Jos F. Sturm,et al.  A Matlab toolbox for optimization over symmetric cones , 1999 .

[20]  Danielle Nuzillard,et al.  Sparse Deflations in Blind Signal Separation , 2006, ICA.

[21]  Laurent Albera,et al.  Blind identification of underdetermined mixtures based on the hexacovariance and higher-order cyclostationarity , 2009, 2009 IEEE/SP 15th Workshop on Statistical Signal Processing.

[22]  Lieven De Lathauwer,et al.  Second-Order Blind Identification of Underdetermined Mixtures , 2006, ICA.

[23]  Barak A. Pearlmutter,et al.  Blind Source Separation by Sparse Decomposition in a Signal Dictionary , 2001, Neural Computation.

[24]  Keikichi Hirose,et al.  Separation of Mixed Audio Signals by Source Localization and Binary Masking with Hilbert Spectrum , 2006, ICA.

[25]  Pierre Comon,et al.  Blind channel identification and extraction of more sources than sensors , 1998, Optics & Photonics.

[26]  Deniz Erdogmus,et al.  Underdetermined blind source separation in a time-varying environment , 2002, 2002 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[27]  Stephen P. Boyd,et al.  Convex Optimization , 2004, Algorithms and Theory of Computation Handbook.

[28]  Wai Lok Woo,et al.  Non-sparse approach to underdetermined blind signal estimation , 2005, Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005..

[29]  Bhaskar D. Rao,et al.  Sparse signal reconstruction from limited data using FOCUSS: a re-weighted minimum norm algorithm , 1997, IEEE Trans. Signal Process..

[30]  Yannick Deville,et al.  Differential Fast Fixed-Point BSS for Underdetermined Linear Instantaneous Mixtures , 2006, ICA.

[31]  Hiroshi Sawada,et al.  Geometrical Interpretation of the PCA Subspace Approach for Overdetermined Blind Source Separation , 2006, EURASIP J. Adv. Signal Process..

[32]  Yeung Sam Hung,et al.  On a Sparse Component Analysis Approach to Blind Source Separation , 2006, ICA.

[33]  Yannick Deville,et al.  A time-frequency blind signal separation method applicable to underdetermined mixtures of dependent sources , 2005, Signal Process..

[34]  H. Sawada,et al.  On real and complex valued /spl lscr//sub 1/-norm minimization for overcomplete blind source separation , 2005, IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, 2005..

[35]  Pierre Comon,et al.  Blind identification and source separation in 2×3 under-determined mixtures , 2004, IEEE Trans. Signal Process..

[36]  Rémi Gribonval,et al.  Audio source separation with a single sensor , 2006, IEEE Transactions on Audio, Speech, and Language Processing.

[37]  DeLiang Wang,et al.  Separating Underdetermined Convolutive Speech Mixtures , 2006, ICA.

[38]  Enric Monte-Moreno,et al.  Underdetermined Convoluted Source Reconstruction Using LP and SOCP, and a Neural Approximator of the Optimizer , 2006, ICA.

[39]  Özgür Yilmaz,et al.  Blind separation of disjoint orthogonal signals: demixing N sources from 2 mixtures , 2000, 2000 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.00CH37100).

[40]  John R. Hershey,et al.  Single microphone source separation using high resolution signal reconstruction , 2004, 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[41]  Hiroshi Sawada,et al.  A NOVEL BLIND SOURCE SEPARATION METHOD WITH OBSERVATION VECTOR CLUSTERING , 2005 .

[42]  Donald Goldfarb,et al.  Second-order cone programming , 2003, Math. Program..

[43]  Boualem Boashash,et al.  Separating More Sources Than Sensors Using Time-Frequency Distributions , 2005, EURASIP J. Adv. Signal Process..

[44]  K. Matsuoka,et al.  Independent Component Analysis and Its Applications to Sound Signal Separation , 2003 .

[45]  Fabian J. Theis,et al.  Mathematics in independent component analysis , 2003, Seventh International Symposium on Signal Processing and Its Applications, 2003. Proceedings..

[46]  Hiroshi Sawada,et al.  Underdetermined blind separation for speech in real environments with sparseness and ICA , 2004, 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[47]  Pierre Comon,et al.  Blind identification of under-determined mixtures based on the characteristic function , 2006, Signal Process..

[48]  Michael Elad,et al.  Optimally sparse representation in general (nonorthogonal) dictionaries via ℓ1 minimization , 2003, Proceedings of the National Academy of Sciences of the United States of America.

[49]  Jun Wang,et al.  Blind source extraction from convolutive mixtures in ill-conditioned multi-input multi-output channels , 2004, IEEE Transactions on Circuits and Systems I: Regular Papers.

[50]  Sacha Krstulovic,et al.  Under-Determined Source Separation: Comparison of Two Approaches Based on Sparse Decompositions , 2006, ICA.

[51]  Stephen P. Boyd,et al.  Applications of second-order cone programming , 1998 .

[52]  Chen Wei,et al.  Post-nonlinear Underdetermined ICA by Bayesian Statistics , 2006, ICA.

[53]  Christopher M. Bishop,et al.  Neural networks for pattern recognition , 1995 .

[54]  Michael A. Saunders,et al.  Atomic Decomposition by Basis Pursuit , 1998, SIAM J. Sci. Comput..

[55]  Ole Winther,et al.  Low complexity Bayesian single channel source separation , 2004, 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[56]  Simon J. Godsill,et al.  Bayesian separation and recovery of convolutively mixed autoregressive sources , 1999, 1999 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings. ICASSP99 (Cat. No.99CH36258).

[57]  Shoko Araki,et al.  Underdetermined Blind Separation of Convolutive Mixtures of Speech Using Time-Frequency Mask and Mixing Matrix Estimation , 2005, IEICE Trans. Fundam. Electron. Commun. Comput. Sci..

[58]  Jean-Francois Cardoso,et al.  Super-symmetric decomposition of the fourth-order cumulant tensor. Blind identification of more sources than sensors , 1991, [Proceedings] ICASSP 91: 1991 International Conference on Acoustics, Speech, and Signal Processing.

[59]  Dmitry M. Malioutov,et al.  Optimal sparse representations in general overcomplete bases , 2004, 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[60]  Barak A. Pearlmutter,et al.  Blind source separation by sparse decomposition , 2000, SPIE Defense + Commercial Sensing.

[61]  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.

[62]  Laurent Albera,et al.  Fourth-order blind identification of underdetermined mixtures of sources (FOBIUM) , 2005, IEEE Transactions on Signal Processing.

[63]  Joos Vandewalle,et al.  Independent component analysis of largely underdetermined mixtures , 2003 .

[64]  B. De Moor,et al.  ICA algorithms for 3 sources and 2 sensors , 1999, Proceedings of the IEEE Signal Processing Workshop on Higher-Order Statistics. SPW-HOS '99.

[65]  Fionn Murtagh,et al.  Comments on 'Parallel Algorithms for Hierarchical Clustering and Cluster Validity' , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[66]  Pau Bofill,et al.  Underdetermined blind separation of delayed sound sources in the frequency domain , 2003, Neurocomputing.

[67]  Mark D. Plumbley,et al.  Source extraction from two-channel mixtures by joint cosine packet analysis , 2006, 2006 14th European Signal Processing Conference.

[68]  Fathi M. Salem,et al.  ALGEBRAIC OVERCOMPLETE INDEPENDENT COMPONENT ANALYSIS , 2003 .

[69]  Michael R. Bussieck,et al.  Conic Programming in GAMS , 2003 .

[70]  Simon J. Godsill,et al.  Blind Separation of Sparse Sources Using Jeffrey's Inverse Prior and the EM Algorithm , 2006, ICA.

[71]  Lieven De Lathauwer,et al.  Simultaneous Matrix Diagonalization : the Overcomplete Case , 2003 .

[72]  L. Vielva,et al.  UNDERDETERMINED BLIND SOURCE SEPARATION USING A PROBABILISTIC SOURCE SPARSITY MODEL , 2001 .

[73]  P. Comon,et al.  Blind Identification of Overcomplete MixturEs of sources (BIOME) , 2004 .

[74]  Ian K. Proudler,et al.  Exploitation of source nonstationarity in underdetermined blind source separation with advanced clustering techniques , 2006, IEEE Transactions on Signal Processing.

[75]  Scott Rickard,et al.  Blind separation of speech mixtures via time-frequency masking , 2004, IEEE Transactions on Signal Processing.

[76]  Fabian J Theis,et al.  Formalization of the Two-Step Approach to Overcomplete BSS , 2002 .

[77]  Nikolaos Mitianoudis,et al.  Overcomplete source separation using Laplacian mixture models , 2005, IEEE Signal Processing Letters.

[78]  E. Oja,et al.  Independent Component Analysis , 2013 .

[79]  Trevor Hastie,et al.  The Elements of Statistical Learning , 2001 .

[80]  W. Kellermann,et al.  Wideband algorithms versus narrowband algorithms for adaptive filtering in the DFT domain , 2003, The Thrity-Seventh Asilomar Conference on Signals, Systems & Computers, 2003.