Publications of Marcel Joho

PhD thesis

[1]  M. Joho
       A Systematic Approach to Adaptive Algorithms for Multichannel System Identification, Inverse Modeling, and Blind Identification
       Dissertation ETH Zurich, , No 13783, Hartung-Gorre, Konstanz, Series in Signal and Information Processing, Vol. 6,
       ISBN 3-89649-632-8, ISSN 1616-671X, December 2000.
       ( ps-file , pdf-file or two sided ps-file , pdf-file )
       ( FDBDeconv.m : Matlab implementation of the Frequency-Domain Blind Deconvolution algorithm from Appendix F )

Adaptive Beamforming

[2] W. Knecht, R. Steiner, M. Joho, and G.S. Moschytz,
       Cancelling Spatial Interference with Nonlinear Filters
       in European Conference on Circuit Theory and Design, H. Dedieu, Ed. August/September 1993, vol. 1, pp. 537-542, Elsevier.

[3] M. Joho and G. S. Moschytz,
       Adaptive beamforming with partitioned frequency-domain filters
       in IEEE Workshop on Applications of Signal Processing to Audio and Acoustics, New Paltz, NY, USA, Oct 19-22, 1997.
       ( ps-file , pdf-file )

[4] M. Joho and G. S. Moschytz,
       On the design of the target-signal filter in adaptive beamforming
       in IEEE International Symposium on Circuits and Systems, Monterey, CA, June, 1998, vol. 5, pp. 166-169.
       ( ps-file )

[5] M.Joho and G.S. Moschytz,
     On the design of the target-signal filter in adaptive beamforming
     IEEE Transactions on Circuits and Systems-II, vol. 46, no. 7, pp. 963-966, July 1999.
     ( ps-file , pdf-file )

Adaptive Filtering

[6]  M. Joho and G. S. Moschytz
        Connecting partitioned frequency-domain filters in parallel or in cascade
        IEEE Transactions on Circuits and Systems-II,  vol. 47, no. 8, pp. 685 - 698, August 2000.
       ( ps-file, pdf-file )

Blind Signal Separation, Blind Deconvolution, Blind Identification

[7] M. Joho and H. Mathis,
       ``Performance comparison of combined blind/non-blind source separation algorithms,''
       in Proc. International Conference on Independent Component Analysis and Blind Source Separation,
       Aussois, France, January 11-15, 1999, pp. 139-142.
       ( ps-file , pdf-file )

[8] M. Joho and H. Mathis,
       An FFT-based algorithm for multichannel blind deconvolution
       in IEEE International Symposium on Circuits and Systems, Orlando, FL, May 30 - June 2, 1999, pp. III 203-206.
       ( ps-file ) (filter coefficients )

[9] H. Mathis, M. Joho, and G. S. Moschytz,
       ``A simple threshold nonlinearity for blind signal separation'',
       IEEE International Symposium on Circuits and Systems, Geneva, Switzerland, May 28-31, 2000, pp. IV 489-492.
       ( ps-file , pdf-file )

[10] M. Joho, H. Mathis, and R. H. Lambert,
        Overdetermined blind source separation: Using more sensors than source signals in a noisy mixture'
        International Conference on Independent Component Analysis and Blind Signal Separation, Helsinki, Finland, June 19-22, 2000, pp. 81-86.
        ( ps-file , pdf-file )

[11]  H. Mathis, T. P. von Hoff, and M. Joho,
         Blind separation of mixed-kurtosis signals using an adaptive threshold nonlinearity
         Proc. International Conference on Independent Component Analysis and Blind Signal Separation, Helsinki, Finland, June 19-22, 2000, pp. 221-226.
         ( ps-file , pdf-file )

[12] H. Mathis and M. Joho,
       Unbiased Blind Separation Using the Threshold Nonlinearity'
       Proc. Signal Processing Advances in Wireless Communications SPAWC 2001,
       Taoyuan, Taiwan, March 20-23, 2001, pp. 239-242.
        ( ps-file , pdf-file )

[13] M. Joho, R. H. Lambert, and H. Mathis,
        Elementary Cost Functions for Blind Separation of Non-Stationary Source Signals
        ICASSP 2001, Salt Lake City, UT, May 7-11, 2001, Vol 5, pp. 2793-2796.
       ( ps-file , pdf-file )

[14]  H. Mathis, T. P. von Hoff, and M. Joho,
        Blind Separation of Signals with Mixed Kurtosis Signs Using Threshold Activation Functions
        IEEE Transactions on Neural Networks, 2001, Vol. 12, No. 3, May 2001, pp. 618-624.
      ( pdf-file )

[15] M. Joho, H. Mathis, and G. S. Moschytz,
       Combined Blind/non-Blind Source Separation Based on the Natural Gradient
       IEEE Signal Processing Letters, Vol. 8, No. 8, August 2001, pp. 236-238.
      ( pdf-file )

[16] R. H. Lambert, M. Joho, H. Mathis,
        Polynomial Singular Values for Number of Wideband Sources Estimation and Principal Component Analysis'
        International Conference on Independent Component Analysis and Blind Signal Separation ICA, San Diego, CA,
        December 9-12, 2001, pp. 379-383.
        ( pdf-file )

[17] M. Joho and H. Mathis,
       Joint Diagonalization of Correlation Matrices by Using Gradient Methods with Application to Blind Signal Separation
       IEEE Sensor Array and Multichannel Signal Processing Workshop SAM,  Rosslyn, VA, August 4-6, 2002, pp. 273-277.
      ( ps-file , pdf-file )

[18] M. Joho and K. Rahbar,
       Joint Diagonalization of Correlation Matrices by Using Newton Methods with Application to Blind Signal Separation
       IEEE Sensor Array and Multichannel Signal Processing Workshop SAM,  Rosslyn, VA, August 4-6, 2002, pp. 403-407.
      ( ps-file , pdf-file )

[19]  H. Mathis, and M. Joho,
        Blind signal separation in noisy environments using a three-step quantizer
        Neurocomputing, Vol. 49,  No. 1-4, 2002, pp. 61-78.

[20] M. Joho and P. Schniter,
       On frequency-domain implementations of filtered-gradient blind deconvolution algorithms'
       36th Asilomar Conference on Signals, Systems, and Computers, Pacific Grove, CA, USA, Nov. 3-6, 2002, Vol. II, pp. 1653-1658.
      ( ps-file , pdf-file )         Additional Notes  ( ps-file , pdf-file )

[21] M. Joho and P. Schniter,
       Frequency-domain realization of a multichannel blind deconvolution algorithm based on the natural gradient'
       International Conference on Independent Component Analysis and Blind Signal Separation, ICA 2003, Nara, Japan,
        April 1-4, 2003, pp. 543-548.
      ( ps-file , pdf-file

[22] M. Joho,
       Blind Signal Separation of Convolutive Mixtures: A Time-Domain Joint-Diagonalization Approach
       International Conference on Independent Component Analysis and Blind Signal Separation, ICA 2004, Granada, Spain,
        September 22-24, 2004, pp. 577-584.
      ( ps-file , pdf-file )

[23] M. Joho,
       Convolutive blind signal separation in acoustics by joint approximate diagonalization of spatiotemporal correlation matrices
       Proc. Asilomar Conference on Signals, Systems, and  Computers, Pacific Grove, CA, USA, November 9-12, 2004, Vol.1, pp. 983-988.
      ( ps-file , pdf-file )

[24] M. Joho,
       Newton Method for Joint Approximate Diagonalization of Positive Definite Hermitian Matrices

       SIAM Journal on Matrix Analysis and Applications, Special Issue on Tensor Decompositions and Applications,
       vol. 30, no. 3, pp. 1205 - 1218,
2008.
      ( ps-file (draft) , pdf-file (draft) )


back to main page