Transient Performance Analysis of the ℓ1-RLS

Wei Gao; Jie Chen; Cedric Richard; Shi Wentao; Qunfei Zhang

doi:10.1109/LSP.2021.3127467

Transient Performance Analysis of the ℓ₁-RLS

Wei Gao, Jie Chen, Cedric Richard, Shi Wentao, Qunfei Zhang

School of Marine Science and Technology

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

7 Scopus citations

Abstract

The recursive least-squares algorithm with ℓ1-norm regularization (ℓ1-RLS) exhibits excellent performance in terms of convergence rate and steady-state error in identification of sparse systems. Nevertheless few works have studied its stochastic behavior, in particular its transient performance. In this letter, we derive analytical models of the transient behavior of the ℓ1-RLS in the mean and mean-square error sense. Simulation results illustrate the accuracy of these models.

Original language	English
Pages (from-to)	90-94
Number of pages	5
Journal	IEEE Signal Processing Letters
Volume	29
DOIs	https://doi.org/10.1109/LSP.2021.3127467
State	Published - 2022

Keywords

online identification
sparse system
Transient analysis
ℓ-RLS

Access to Document

10.1109/LSP.2021.3127467

Cite this

@article{ff4f65e6a1c54a7bbb75dc84f0b72f8c,

title = "Transient Performance Analysis of the ℓ1-RLS",

abstract = "The recursive least-squares algorithm with ℓ1-norm regularization (ℓ1-RLS) exhibits excellent performance in terms of convergence rate and steady-state error in identification of sparse systems. Nevertheless few works have studied its stochastic behavior, in particular its transient performance. In this letter, we derive analytical models of the transient behavior of the ℓ1-RLS in the mean and mean-square error sense. Simulation results illustrate the accuracy of these models.",

keywords = "online identification, sparse system, Transient analysis, ℓ-RLS",

author = "Wei Gao and Jie Chen and Cedric Richard and Shi Wentao and Qunfei Zhang",

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

year = "2022",

doi = "10.1109/LSP.2021.3127467",

language = "英语",

volume = "29",

pages = "90--94",

journal = "IEEE Signal Processing Letters",

issn = "1070-9908",

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

}

TY - JOUR

T1 - Transient Performance Analysis of the ℓ1-RLS

AU - Gao, Wei

AU - Chen, Jie

AU - Richard, Cedric

AU - Wentao, Shi

AU - Zhang, Qunfei

PY - 2022

Y1 - 2022

N2 - The recursive least-squares algorithm with ℓ1-norm regularization (ℓ1-RLS) exhibits excellent performance in terms of convergence rate and steady-state error in identification of sparse systems. Nevertheless few works have studied its stochastic behavior, in particular its transient performance. In this letter, we derive analytical models of the transient behavior of the ℓ1-RLS in the mean and mean-square error sense. Simulation results illustrate the accuracy of these models.

AB - The recursive least-squares algorithm with ℓ1-norm regularization (ℓ1-RLS) exhibits excellent performance in terms of convergence rate and steady-state error in identification of sparse systems. Nevertheless few works have studied its stochastic behavior, in particular its transient performance. In this letter, we derive analytical models of the transient behavior of the ℓ1-RLS in the mean and mean-square error sense. Simulation results illustrate the accuracy of these models.

KW - online identification

KW - sparse system

KW - Transient analysis

KW - ℓ-RLS

UR - http://www.scopus.com/inward/record.url?scp=85119401425&partnerID=8YFLogxK

U2 - 10.1109/LSP.2021.3127467

DO - 10.1109/LSP.2021.3127467

M3 - 文章

AN - SCOPUS:85119401425

SN - 1070-9908

VL - 29

SP - 90

EP - 94

JO - IEEE Signal Processing Letters

JF - IEEE Signal Processing Letters

ER -

Transient Performance Analysis of the ℓ₁-RLS

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this