Title :
Sparsity-aware pseudo affine projection algorithm for active noise control
Author :
Albu, Felix ; Gully, Amelia ; de Lamare, Rodrigo
Author_Institution :
Dept. of Electron., Valahia Univ. of Targoviste, Targoviste, Romania
Abstract :
This paper describes new algorithms promoting sparsity in modified filtered-x algorithms for active noise control. The proposed algorithms are based on recent techniques incorporating approximations to the l0-norm in the cost function used. An approximation to the affine projection algorithm leads to new zero-attracting and reweighted zero-attracting modified filtered-x pseudo affine projection algorithms. The results of simulations indicate that the proposed techniques demonstrate good performance and reduce numerical complexity compared to the original affine projection based algorithms.
Keywords :
active noise control; filtering theory; active noise control; cost function; filtered-x algorithms; filtered-x pseudo affine projection algorithms; l0-norm; numerical complexity; reweighted zero-attracting; sparsity-aware pseudo affine projection algorithm; Approximation algorithms; Approximation methods; Decision support systems; Filtering algorithms; Noise; Speech; Speech processing;
Conference_Titel :
Asia-Pacific Signal and Information Processing Association, 2014 Annual Summit and Conference (APSIPA)
Conference_Location :
Siem Reap
DOI :
10.1109/APSIPA.2014.7041652