The Iby and Aladar Fleischman
Faculty of Engineering

School of Electrical Engineering

               Home               

               Teaching               

             Publications             

             Invited Talks             

     Activities and Awards     

Former Graduate Students

     Downloads and Links     

Downloads and Links:

Matlab code for computation of the sample-charm and sample-charrelation, as well as their (asymptotic) covariance matrix, and demonstration of their use. See:
Slapak, A and Yeredor, A., “Charrelation and Charm: Generic Statistics Incorporating Higher-Order Information”, to appear in IEEE Trans. on Signal Processing, 2012.

Matlab code for the “AC-DC” algorithm for non-orthogonal approximate joint diagonalization, as described in:
Yeredor, A., “Non-Orthogonal Joint Diagonalization in the Least-Squares Sense with Application in Blind Source Separation”, IEEE Trans. on Signal Processing, vol. 50 no. 7 pp. 1545-1553, July 2002.

The following are available at Petr Tichavsky’s web-site:

o       A toolbox containing Matlab code for the “WEDGE”, “UWEDGE” and “BG-WEDGE” algorithms for weighted (non-orthogonal) approximate joint diagonalization. These algorithms are described in:
Tichavsky, P. and Yeredor, A., "Fast Approximate Joint Diagonalization Incorporating Weight Matrices", IEEE Trans. on Signal Processing, vol. 57 no.3 pp. 878-891, March 2009.

o       Matlab code for “WASOBI” - A BSS algorithm based on the spectral diversity (for asymptotically optimal separation of Gaussian autoregressive sources). Described in
Tichavský, P., Doron, E., Yeredor, A. and Nielsen, J.,
A Computationally Affordable Implementation of an Asymptotically Optimal BSS Algorithm for AR Sources - Proceedings of The 2006 European Signal Processing Conference (EUSIPCO'2006), September 2006.
and in
Tichavsky, P. and Yeredor, A., "Fast Approximate Joint Diagonalization Incorporating Weight Matrices", IEEE Trans. on Signal Processing, vol. 57 no.3 pp. 878-891, March 2009.

o       Matlab code for “BARBI” - A BSS algorithm for Gaussian nonstationary sources with Block-AR structures. Described in
Tichavský, P., Yeredor, A. and Koldovsky, Z.
A fast asymptotically efficient algorithm for blind separation of a linear mixture of block-wise stationary autoregressive processes - Proceedings of The 2009 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP'2009), April 2009.

o       Matlab code for “COMBI” and “MULTI-COMBI” - BSS algorithms combining the exploitation of both spatial and temporal diversity, each using an asymptotically optimal approach. Described in
Tichavský, P., Koldovsky, Z., Yeredor, A., Gomez-Herrero G. and Doron, E.,
A Hybrid Technique for blind Separation of Non-Gaussian and Time-Correlated Sources using a Multicomponent Approach, IEEE Trans. on Neural Networks, vol. 19 no. 3, pp. 421-430, March 2008.

Image and video separation demo for the paper
Be'ery, E. and Yeredor, A., Blind Separation of Superimposed Shifted Images Unsing Parameterized Joint Diagonalization, IEEE Trans. on Image Processing, vol. 17 no. 3, pp. 340-353, March 2008.

The TAU Teaching Center is another place to find me, among the recipients of the Rector’s Excellence in Teaching Prize for 2001, 2003, 2005, and also as a sample-case (in Hebrew).

Get my Children's Book: "Is it a visit?" (in Hebrew)…

Get my other Children's Book: "Electricity" (in Hebrew and also in Arabic)…