Nonorthogonal Joint Diagonalization by Combining Givens and Hyperbolic Rotations

A new algorithm for computing the nonorthogonal joint diagonalization of a set of matrices is proposed for independent component analysis and blind source separation applications. This algorithm is an extension of the Jacobi-like algorithm first proposed in the joint approximate diagonalization of eigenmatrices (JADE) method for orthogonal joint diagonalization. The improvement consists mainly in computing a mixing matrix of determinant one and columns of equal norm instead of an orthogonal mixing matrix. This target matrix is constructed iteratively by successive multiplications of not only Givens rotations but also hyperbolic rotations and diagonal matrices. The algorithm performance, evaluated on synthetic data, compares favorably with existing methods in terms of speed of convergence and complexity.

[1]  Bijan Afsari,et al.  Sensitivity Analysis for the Problem of Matrix Joint Diagonalization , 2008, SIAM J. Matrix Anal. Appl..

[2]  Arie Yeredor,et al.  Non-orthogonal joint diagonalization in the least-squares sense with application in blind source separation , 2002, IEEE Trans. Signal Process..

[3]  Bijan Afsari.  What Can Make Joint Diagonalization Difficult? , 2007, 2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07.

[4]  Antoine Souloumiac,et al.  Jacobi Angles for Simultaneous Diagonalization , 1996, SIAM J. Matrix Anal. Appl..

[5]  Bijan Afsari,et al.  Simple LU and QR Based Non-orthogonal Matrix Joint Diagonalization , 2006, ICA.

[6]  A. Bunse-Gerstner,et al.  Numerical Methods for Simultaneous Diagonalization , 1993, SIAM J. Matrix Anal. Appl..

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

[8]  Bijan Afsari,et al.  Some Gradient Based Joint Diagonalization Methods for ICA , 2004, ICA.

[9]  El Mostafa Fadaili,et al.  Nonorthogonal Joint Diagonalization/Zero Diagonalization for Source Separation Based on Time-Frequency Distributions , 2007, IEEE Transactions on Signal Processing.

[10]  Arie Yeredor,et al.  A fast approximate joint diagonalization algorithm using a criterion with a block diagonal weight matrix , 2008, 2008 IEEE International Conference on Acoustics, Speech and Signal Processing.

[11]  S. Dégerine,et al.  Sur la diagonalisation conjointe approchée par un critère des moindres carrés , 2001 .

[12]  Dinh Tuan Pham,et al.  Joint Approximate Diagonalization of Positive Definite Hermitian Matrices , 2000, SIAM J. Matrix Anal. Appl..

[13]  Andreas Ziehe,et al.  A Fast Algorithm for Joint Diagonalization with Non-orthogonal Transformations and its Application to Blind Source Separation , 2004, J. Mach. Learn. Res..

[14]  Kui Liu,et al.  Dual-state systolic architectures for up/downdating RLS adaptive filtering , 1992 .

[15]  Jun Zhang,et al.  Nonorthogonal Joint Diagonalization Algorithm Based on Trigonometric Parameterization , 2007, IEEE Transactions on Signal Processing.

[16]  Eric Moulines,et al.  A blind source separation technique using second-order statistics , 1997, IEEE Trans. Signal Process..

[17]  J. Cardoso,et al.  COMPARAISONS DE METHODES DE SEPARATION DE SOURCES , 1991 .

[18]  Robert H. Halstead,et al.  Matrix Computations , 2011, Encyclopedia of Parallel Computing.

[19]  Klaus Obermayer,et al.  Quadratic optimization for simultaneous matrix diagonalization , 2006, IEEE Transactions on Signal Processing.

[20]  Gene H. Golub,et al.  Matrix computations (3rd ed.) , 1996 .