A Bayesian network view on linear and nonlinear acoustic echo cancellation

Maas R, Hümmer C, Schwarz A, Hofmann C, Kellermann W (2014)

Publication Language: English

Publication Status: Published

Publication Type: Conference contribution, Conference Contribution

Publication year: 2014

Pages Range: 495-499

Article Number: 6889292

Event location: Xi'an

ISBN: 978-1-4799-5403-2

DOI: 10.1109/ChinaSIP.2014.6889292

Abstract

In this contribution, we provide a new derivation of the normalized least mean square (NLMS) algorithm from a machine learning perspective. By applying the inference rules of Bayesian networks to a linear observation model, the NLMS can be shown to arise as a modification of the Kalman filter equations. Based on a nonlinear observationmodel, we exemplify the benefit of the Bayesian point of view by employing the technique of particle filtering to realize a tractable algorithm for nonlinear acoustic echo cancellation. Experiments carried out on real smartphone recordings reveal the remarkable performance of the new approach.

Authors with CRIS profile

How to cite

APA:

Maas, R., Hümmer, C., Schwarz, A., Hofmann, C., & Kellermann, W. (2014). A Bayesian network view on linear and nonlinear acoustic echo cancellation. In Proceedings of the 2nd IEEE China Summit and International Conference on Signal and Information Processing, IEEE ChinaSIP 2014 (pp. 495-499). Xi'an, CN.

MLA:

Maas, Roland, et al. "A Bayesian network view on linear and nonlinear acoustic echo cancellation." Proceedings of the 2nd IEEE China Summit and International Conference on Signal and Information Processing, IEEE ChinaSIP 2014, Xi'an 2014. 495-499.

BibTeX: Download