PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Combinations of Proportionate Adaptive Filters in Acoustics: An application to Active Noise Control
Jerónimo Arenas-Garcia, María de Diego, Luis A. Azpicueta-Ruiz, Miguel Ferrer and Alberto González
In: EUSIPCO 2011, Barcelona(2011).

Abstract

Proportionate adaptive schemes have been proposed to exploit spar- sity and accelerate filter convergence in acoustic echo cancellation. Recently, combinations of adaptive filters have been extended to op- erate with proportionate schemes, in order to achieve more robust operation when the actual degree of sparsity of the optimal solution is unknown. Furthermore, it is possible to exploit the asymmetric distribution of adaptation energy in proportionate schemes to reduce the overall steady-state misadjustment. In this contribution, we ex- plain how these novel adaptive filtering structures, which have been proposed and tested mainly for echo cancellation, can also be ef- fectively extended to improve the usual performance trade-offs that appear in active noise control scenarios by introducing some mi- nor modifications. Experimental results in realistic scenarios show that the proposed schemes provide an interesting alternative to the traditional use of a single adaptive filter.

PDF - PASCAL Members only - Requires Adobe Acrobat Reader or other PDF viewer.
EPrint Type:Conference or Workshop Item (Lecture)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Theory & Algorithms
ID Code:8831
Deposited By:Jerónimo Arenas-Garcia
Deposited On:21 February 2012