Sparsity Regularized Affine Projection Adaptive Filtering for System Identification

Young-Seok CHOI

doi:10.1587/transinf.E97.D.964

Sparsity Regularized Affine Projection Adaptive Filtering for System Identification

Young-Seok CHOI

Full Text Views

0

Share
Cite this

Summary :

A new type of the affine projection (AP) algorithms which incorporates the sparsity condition of a system is presented. To exploit the sparsity of the system, a weighted l₁-norm regularization is imposed on the cost function of the AP algorithm. Minimizing the cost function with a subgradient calculus and choosing two distinct weightings for l₁-norm, two stochastic gradient based sparsity regularized AP (SR-AP) algorithms are developed. Experimental results show that the SR-AP algorithms outperform the typical AP counterparts for identifying sparse systems.

Publication: IEICE TRANSACTIONS on Information Vol.E97-D No.4 pp.964-967

Publication Date: 2014/04/01

Publicized

Online ISSN: 1745-1361

DOI: 10.1587/transinf.E97.D.964

Type of Manuscript: LETTER

Category: Fundamentals of Information Systems

Authors

Young-Seok CHOI
Gangneung-Wonju National University

Keyword

system identification, adaptive filter, affine projection, sparsity, sparse system

Cite this

Copy

Young-Seok CHOI, "Sparsity Regularized Affine Projection Adaptive Filtering for System Identification" in IEICE TRANSACTIONS on Information, vol. E97-D, no. 4, pp. 964-967, April 2014, doi: 10.1587/transinf.E97.D.964.
Abstract: A new type of the affine projection (AP) algorithms which incorporates the sparsity condition of a system is presented. To exploit the sparsity of the system, a weighted l₁-norm regularization is imposed on the cost function of the AP algorithm. Minimizing the cost function with a subgradient calculus and choosing two distinct weightings for l₁-norm, two stochastic gradient based sparsity regularized AP (SR-AP) algorithms are developed. Experimental results show that the SR-AP algorithms outperform the typical AP counterparts for identifying sparse systems.
URL: https://globals.ieice.org/en_transactions/information/10.1587/transinf.E97.D.964/_p

Copy

@ARTICLE{e97-d_4_964,
author={Young-Seok CHOI, },
journal={IEICE TRANSACTIONS on Information},
title={Sparsity Regularized Affine Projection Adaptive Filtering for System Identification},
year={2014},
volume={E97-D},
number={4},
pages={964-967},
abstract={A new type of the affine projection (AP) algorithms which incorporates the sparsity condition of a system is presented. To exploit the sparsity of the system, a weighted l₁-norm regularization is imposed on the cost function of the AP algorithm. Minimizing the cost function with a subgradient calculus and choosing two distinct weightings for l₁-norm, two stochastic gradient based sparsity regularized AP (SR-AP) algorithms are developed. Experimental results show that the SR-AP algorithms outperform the typical AP counterparts for identifying sparse systems.},
keywords={},
doi={10.1587/transinf.E97.D.964},
ISSN={1745-1361},
month={April},}

Copy

TY - JOUR
TI - Sparsity Regularized Affine Projection Adaptive Filtering for System Identification
T2 - IEICE TRANSACTIONS on Information
SP - 964
EP - 967
AU - Young-Seok CHOI
PY - 2014
DO - 10.1587/transinf.E97.D.964
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E97-D
IS - 4
JA - IEICE TRANSACTIONS on Information
Y1 - April 2014
AB - A new type of the affine projection (AP) algorithms which incorporates the sparsity condition of a system is presented. To exploit the sparsity of the system, a weighted l₁-norm regularization is imposed on the cost function of the AP algorithm. Minimizing the cost function with a subgradient calculus and choosing two distinct weightings for l₁-norm, two stochastic gradient based sparsity regularized AP (SR-AP) algorithms are developed. Experimental results show that the SR-AP algorithms outperform the typical AP counterparts for identifying sparse systems.
ER -