Transient performance analysis of zero-attracting LMS

Jie Chen; Cedric Richard; Yingying Song; David Brie

doi:10.1109/LSP.2016.2616890

Transient performance analysis of zero-attracting LMS

Jie Chen, Cedric Richard, Yingying Song, David Brie

School of Marine Science and Technology

Research output: Contribution to journal › Article › peer-review

34 Scopus citations

Abstract

Zero-attracting least-mean-square (ZA-LMS) algorithm has been widely used for online sparse system identification. It combines the LMS framework and ℓ₁-norm regularization to promote sparsity, and relies on subgradient iterations. Despite the significant interest in ZA-LMS, few works analyzed its transient behavior. The main difficulty lies in the nonlinearity of the update rule. In this study, a detailed analysis in the mean and mean-square sense is carried out in order to examine the behavior of the algorithm. Simulation results illustrate the accuracy of the model and highlight its performance through comparisons with an existing model.

Original language	English
Article number	7589076
Pages (from-to)	1786-1790
Number of pages	5
Journal	IEEE Signal Processing Letters
Volume	23
Issue number	12
DOIs	https://doi.org/10.1109/LSP.2016.2616890
State	Published - Dec 2016

Keywords

Performance analysis
sparse system identification
transient behavior
zero-attracting least-mean square (ZA-LMS)

Access to Document

10.1109/LSP.2016.2616890

Cite this

@article{64f6d5e071f94e3f9da3fbef814f3f29,

title = "Transient performance analysis of zero-attracting LMS",

abstract = "Zero-attracting least-mean-square (ZA-LMS) algorithm has been widely used for online sparse system identification. It combines the LMS framework and ℓ1-norm regularization to promote sparsity, and relies on subgradient iterations. Despite the significant interest in ZA-LMS, few works analyzed its transient behavior. The main difficulty lies in the nonlinearity of the update rule. In this study, a detailed analysis in the mean and mean-square sense is carried out in order to examine the behavior of the algorithm. Simulation results illustrate the accuracy of the model and highlight its performance through comparisons with an existing model.",

keywords = "Performance analysis, sparse system identification, transient behavior, zero-attracting least-mean square (ZA-LMS)",

author = "Jie Chen and Cedric Richard and Yingying Song and David Brie",

note = "Publisher Copyright: {\textcopyright} 1994-2012 IEEE.",

year = "2016",

month = dec,

doi = "10.1109/LSP.2016.2616890",

language = "英语",

volume = "23",

pages = "1786--1790",

journal = "IEEE Signal Processing Letters",

issn = "1070-9908",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "12",

}

TY - JOUR

T1 - Transient performance analysis of zero-attracting LMS

AU - Chen, Jie

AU - Richard, Cedric

AU - Song, Yingying

AU - Brie, David

PY - 2016/12

Y1 - 2016/12

N2 - Zero-attracting least-mean-square (ZA-LMS) algorithm has been widely used for online sparse system identification. It combines the LMS framework and ℓ1-norm regularization to promote sparsity, and relies on subgradient iterations. Despite the significant interest in ZA-LMS, few works analyzed its transient behavior. The main difficulty lies in the nonlinearity of the update rule. In this study, a detailed analysis in the mean and mean-square sense is carried out in order to examine the behavior of the algorithm. Simulation results illustrate the accuracy of the model and highlight its performance through comparisons with an existing model.

AB - Zero-attracting least-mean-square (ZA-LMS) algorithm has been widely used for online sparse system identification. It combines the LMS framework and ℓ1-norm regularization to promote sparsity, and relies on subgradient iterations. Despite the significant interest in ZA-LMS, few works analyzed its transient behavior. The main difficulty lies in the nonlinearity of the update rule. In this study, a detailed analysis in the mean and mean-square sense is carried out in order to examine the behavior of the algorithm. Simulation results illustrate the accuracy of the model and highlight its performance through comparisons with an existing model.

KW - Performance analysis

KW - sparse system identification

KW - transient behavior

KW - zero-attracting least-mean square (ZA-LMS)

UR - http://www.scopus.com/inward/record.url?scp=85012012694&partnerID=8YFLogxK

U2 - 10.1109/LSP.2016.2616890

DO - 10.1109/LSP.2016.2616890

M3 - 文章

AN - SCOPUS:85012012694

SN - 1070-9908

VL - 23

SP - 1786

EP - 1790

JO - IEEE Signal Processing Letters

JF - IEEE Signal Processing Letters

IS - 12

M1 - 7589076

ER -

Transient performance analysis of zero-attracting LMS

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this