Variable step-size LMS algorithm using regulated least square error criterion

Sin-Chun Ng; Chi-Chung Cheung; Chi-Yin Chung; Shu-Hung Leung

doi:10.1049/el:20030113

Variable step-size LMS algorithm using regulated least square error criterion

Sin-Chun Ng, Chi-Chung Cheung, Chi-Yin Chung, Shu-Hung Leung

Department of Electrical Engineering

Research output: Journal Publications and Reviews › RGC 21 - Publication in refereed journal › peer-review

4 Citations (Scopus)

Abstract

A new variable step-size algorithm is proposed using an evolutionary approach for the least mean square algorithm. The step-size candidates are evaluated by calculating their square errors, which are composed of a priori and a posteriori errors. The composition of the square error measure is regulated according to different properties of errors. The new algorithm always outperforms other traditional variable step-size methods to provide fast converging and good tracking capability.

Original language	English
Pages (from-to)	160-162
Journal	Electronics Letters
Volume	39
Issue number	1
DOIs	https://doi.org/10.1049/el:20030113
Publication status	Published - 9 Jan 2003

Access Output

10.1049/el:20030113

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@article{9b4dceb07ff446ee80631ba8921b603d,

title = "Variable step-size LMS algorithm using regulated least square error criterion",

abstract = "A new variable step-size algorithm is proposed using an evolutionary approach for the least mean square algorithm. The step-size candidates are evaluated by calculating their square errors, which are composed of a priori and a posteriori errors. The composition of the square error measure is regulated according to different properties of errors. The new algorithm always outperforms other traditional variable step-size methods to provide fast converging and good tracking capability.",

author = "Sin-Chun Ng and Chi-Chung Cheung and Chi-Yin Chung and Shu-Hung Leung",

year = "2003",

month = jan,

day = "9",

doi = "10.1049/el:20030113",

language = "English",

volume = "39",

pages = "160--162",

journal = "Electronics Letters",

issn = "0013-5194",

publisher = "John Wiley & Sons",

number = "1",

}

TY - JOUR

T1 - Variable step-size LMS algorithm using regulated least square error criterion

AU - Ng, Sin-Chun

AU - Cheung, Chi-Chung

AU - Chung, Chi-Yin

AU - Leung, Shu-Hung

PY - 2003/1/9

Y1 - 2003/1/9

N2 - A new variable step-size algorithm is proposed using an evolutionary approach for the least mean square algorithm. The step-size candidates are evaluated by calculating their square errors, which are composed of a priori and a posteriori errors. The composition of the square error measure is regulated according to different properties of errors. The new algorithm always outperforms other traditional variable step-size methods to provide fast converging and good tracking capability.

AB - A new variable step-size algorithm is proposed using an evolutionary approach for the least mean square algorithm. The step-size candidates are evaluated by calculating their square errors, which are composed of a priori and a posteriori errors. The composition of the square error measure is regulated according to different properties of errors. The new algorithm always outperforms other traditional variable step-size methods to provide fast converging and good tracking capability.

UR - http://www.scopus.com/inward/record.url?scp=0037427017&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-0037427017&origin=recordpage

U2 - 10.1049/el:20030113

DO - 10.1049/el:20030113

M3 - RGC 21 - Publication in refereed journal

SN - 0013-5194

VL - 39

SP - 160

EP - 162

JO - Electronics Letters

JF - Electronics Letters

IS - 1

ER -

Variable step-size LMS algorithm using regulated least square error criterion

Abstract

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this