Abstract
We propose a new algorithm for Independent Component Extraction that extracts one non-Gaussian component and is capable to exploit the non-Gaussianity of background signals without decomposing them into independent components. The algorithm is suitable for situations when the signal to be extracted is determined through initialization; it shows an extra stable convergence when the target component is dominant. In simulations, the proposed method is compared with Natural Gradient and One-unit FastICA, and it yields improved results in terms of the Signal-to-Interference ratio and the number of successful extractions.
This work was supported by The Czech Science Foundation through Project No. 17-00902S and partially supported by JSPS KAKENHI Grant Number 16H01735.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
A particular variant of these algorithms (OGICE\(_\mathbf{w}\)) coincides with a method proposed earlier by Pham in [12], which was derived based on a simplified form of mutual information that is valid for Gaussian background.
- 2.
Note that the separated sources by NG are not reordered after the separation, because the BSE problem is assumed to be resolved correctly only if the SOI appears in the assumed output channel.
References
Amari, S., Cichocki, A., Yang, H.H.: A new learning algorithm for blind signal separation. In: Proceedings of Neural Information Processing Systems, pp. 757–763 (1996)
Cardoso, J.F.: On the performance of orthogonal source separation algorithms. In: Proceedings of European Signal Processing Conference, pp. 776–779 (1994)
Comon, P.: Independent component analysis, a new concept? Sign. Process. 36, 287–314 (1994)
Huber, P.J.: Projection pursuit. Ann. Statist. 13(2), 435–475 (1985)
Hyvärinen, A.: One-unit contrast functions for independent component analysis: a statistical analysis. In: Proceedings of the 1997 IEEE Signal Processing Society Workshop on Neural Networks for Signal Processing VII, pp. 388–397 (1997)
Hyvärinen, A., Oja, E.: A fast fixed-point algorithm for independent component analysis. Neural Comput. 9(7), 1483–1492 (1997)
Javidi, S., Mandic, D.P., Cichocki, A.: Complex blind source extraction from noisy mixtures using second-order statistics. IEEE Trans. Circuits Syst. I: Regul. Pap. 57(7), 1404–1416 (2010)
Kautský, V., Koldovský, Z., Tichavský, P.: Cramér-Rao-induced bound for interference-to-signal ratio achievable through non-gaussian independent component extraction. In: 2017 IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP), pp. 94–97 (2017)
Kim, T., Attias, H.T., Lee, S.Y., Lee, T.W.: Blind source separation exploiting higher-order frequency dependencies. IEEE Trans. Audio Speech Lang. Process. 15(1), 70–79 (2007)
Koldovský, Z., Tichavský, P.: Gradient algorithms for complex non-gaussian independent component/vector extraction (2018). ArXiv1803.10108
Koldovský, Z., Tichavský, P., Kautský, V.: Orthogonally constrained independent component extraction: Blind MPDR beamforming. In: Proceedings of European Signal Processing Conference, pp. 1195–1199 (2017)
Pham DTA: Blind partial separation of instantaneous mixtures of sources. In: Proceedings of International Conference on Independent Component Analysis and Signal Separation, pp. 868–875. Springer, Heidelberg (2006)
Sawada, H., Mukai, R., Araki, S., Makino, S.: A robust and precise method for solving the permutation problem of frequency-domain blind source separation. IEEE Trans. Speech Audio Process. 12(5), 530–538 (2004)
Tichavský, P., Koldovský, Z., Oja, E.: Performance analysis of the FastICA algorithm and Cramér-Rao bounds for linear independent component analysis. IEEE Trans. Sign. Process. 54(4), 1189–1203 (2006)
Tichavský, P., Koldovský, Z., Oja, E.: Corrections to “Performance analysis of the FastICA algorithm and Cramér Rao bounds for linear independent component analysis". IEEE Trans. Sign. Process. 56(4), 1715–1716 (2008)
Van Trees, H.L.: Optimum Array Processing: Part IV of Detection, Estimation, and Modulation Theory. Wiley, New York (2002)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer International Publishing AG, part of Springer Nature
About this paper
Cite this paper
Koldovský, Z., Tichavský, P., Ono, N. (2018). Orthogonally-Constrained Extraction of Independent Non-Gaussian Component from Non-Gaussian Background Without ICA. In: Deville, Y., Gannot, S., Mason, R., Plumbley, M., Ward, D. (eds) Latent Variable Analysis and Signal Separation. LVA/ICA 2018. Lecture Notes in Computer Science(), vol 10891. Springer, Cham. https://doi.org/10.1007/978-3-319-93764-9_16
Download citation
DOI: https://doi.org/10.1007/978-3-319-93764-9_16
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-93763-2
Online ISBN: 978-3-319-93764-9
eBook Packages: Computer ScienceComputer Science (R0)