AN ALGORITHM EVALUATION TEST SUITE FOR BLIND SOURCE SEPARATION PROBLEM

Marina Charwath, Imke Hahn, Sascha Hauke, Martin Pyka, Slawi Stesny, Dietmar Lammers, Steffen Wachenfeld, Markus Borschbach

Abstract

To ensure the integration and a systematic performance evaluation of the CubICA, the JADE and the EfICA algorithm, a previously developed testsuite for BSS-problems is used. All the steps to perform a competition of methods for source separation are part of a testsuite that supports the development and evaluation of blind source separation (BSS) algorithms in a highly automated way. The concept of the testsuite is presented and it is shown how the testsuite can be used to apply a selection of BSS-algorithms to four standard sub-problems. To compare the performance of arbitrary algorithms on given problems the testsuite allows the integration of new algorithms and testing problems using well defined interfaces. A brief example is given by the integration of the FlexICA, EVD, EVD24 and the FastICA. Also the integration of CubICA, JADE and the EfICA algorithm and the results achieved from automated tests and parameter optimizations will be described in comparison.

References

  1. Cichocki, A. and Amari, S.-I. 2002. Adaptive Blind Signal and Image Processing, (Wiley).
  2. Hyvärinen, A., Karhunen, J. and Oja E. 2001. Independent Component Analysis, (Wiley).
  3. Giannakopoulos, X. 1998. Comparison of Adaptive Independent Component Analysis Algorithms, (Helsinki University: Master Thesis).
  4. Calhoun, V., Adali, T., Larsen, J., Miller, D. and Douglas, S. 2005. Proceedings of IEEE Machine Learning for Signal Processing Workshop XV.
  5. Cardoso, J.-F. 1999. High-Order Contrasts for Independent Component Analysis, Neural Computation, 11(1): 157-192.
  6. Blaschke, T. and Wiskott, L. 2003. CubICA: Independent Component Analysis by Simultaneous Thirdand Forth-Order Cumulant Diagonalization, Computer Science Preprint Server (CSPS): Computational Intelligence/0304002.
  7. Koldovsky, Z. and Tichavsky, P. 2005. Efficient Variant of Algorithm FASTICA for Independent Component Analysis Attaining the CRAMER-RAO LOWER BOUND, IEEE Statistical Signal Processing Workshop, Bordeaux..
  8. Borschbach, M. and Schulte, M. 1999. Performance Analysis of Learning Rules for the Blind Separation of Magnetencephalography Signals, Proc. of ICA'99, First Int. Workshop on Independent Component Analysis and Signal Separation, pp. 341-346.
  9. Hagen, C. 1997. Neural Networks and Multivariate Statistical Data Analysis (in German), (University of Darmstadt: Ph.D. Thesis.)
  10. Testengine and -suite. 2006. http://cs.uni-muenster.de/ICA/ ;-last date of access 03.05.2006.
  11. Schobben, D., Torkkola, K., and Smaragdis, P. 1999. Evaluation Blind Signal Separation Methods”, Proc. of ICA'99, First Int. Workshop on Independent Component Analysis and Signal Separation, pp. 261-266.
  12. Giannakopoulos, X., Karhunen, J. and Oja, E. 1999. An experimental comparison of neural algorithms for independent component analysis and blind separation, Int. J. of Neural Systems, Vol. 9, No. 2.
Download


Paper Citation


in Harvard Style

Charwath M., Hahn I., Hauke S., Pyka M., Stesny S., Lammers D., Wachenfeld S. and Borschbach M. (2006). AN ALGORITHM EVALUATION TEST SUITE FOR BLIND SOURCE SEPARATION PROBLEM . In Proceedings of the Third International Conference on Informatics in Control, Automation and Robotics - Volume 3: ICINCO, ISBN 978-972-8865-61-0, pages 186-192. DOI: 10.5220/0001211301860192


in Bibtex Style

@conference{icinco06,
author={Marina Charwath and Imke Hahn and Sascha Hauke and Martin Pyka and Slawi Stesny and Dietmar Lammers and Steffen Wachenfeld and Markus Borschbach},
title={AN ALGORITHM EVALUATION TEST SUITE FOR BLIND SOURCE SEPARATION PROBLEM},
booktitle={Proceedings of the Third International Conference on Informatics in Control, Automation and Robotics - Volume 3: ICINCO,},
year={2006},
pages={186-192},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0001211301860192},
isbn={978-972-8865-61-0},
}


in EndNote Style

TY - CONF
JO - Proceedings of the Third International Conference on Informatics in Control, Automation and Robotics - Volume 3: ICINCO,
TI - AN ALGORITHM EVALUATION TEST SUITE FOR BLIND SOURCE SEPARATION PROBLEM
SN - 978-972-8865-61-0
AU - Charwath M.
AU - Hahn I.
AU - Hauke S.
AU - Pyka M.
AU - Stesny S.
AU - Lammers D.
AU - Wachenfeld S.
AU - Borschbach M.
PY - 2006
SP - 186
EP - 192
DO - 10.5220/0001211301860192