A sparse adaptive filtering using time-varying soft-thresholding techniques

Yukihiro Murakami, Masao Yamagishi, Masahiro Yukawa, Isao Yamada

Research output: Chapter in Book/Report/Conference proceedingConference contribution

75 Citations (Scopus)

Abstract

In this paper, we propose a novel adaptive filtering algorithm based on an iterative use of (i) the proximity operator and (ii) the parallel variable-metric projection. Our time-varying cost function is a weighted sum of squared distances (in a variable-metric sense) plus a possibly nonsmooth penalty term, and the proposed algorithm is derived along the idea of proximal forward-backward splitting in convex analysis. For application to sparse-system identification problems, we employ the (weighted) ℓ1 norm as the penalty term, leading to a time-varying soft-thresholding operator. As the simple example of the proposed algorithm, we present the variable-metric affine projection algorithm composed with the time-varying soft-thresholding operator. Numerical examples demonstrate that the proposed algorithms notably outperform their counterparts without soft-thresholding both in convergence speed and steady-state mismatch, while the extra computational complexity due to the additional soft-thresholding is negligibly low.

Original languageEnglish
Title of host publicationICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Pages3734-3737
Number of pages4
DOIs
Publication statusPublished - 2010
Externally publishedYes
Event2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010 - Dallas, TX, United States
Duration: 2010 Mar 142010 Mar 19

Other

Other2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010
CountryUnited States
CityDallas, TX
Period10/3/1410/3/19

Fingerprint

Adaptive filtering
Cost functions
Mathematical operators
Computational complexity
Identification (control systems)

Keywords

  • Parallel projection
  • Proximal forward-backward splitting
  • Soft-thresholding
  • Sparse adaptive filtering
  • Variable metric

ASJC Scopus subject areas

  • Signal Processing
  • Software
  • Electrical and Electronic Engineering

Cite this

Murakami, Y., Yamagishi, M., Yukawa, M., & Yamada, I. (2010). A sparse adaptive filtering using time-varying soft-thresholding techniques. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings (pp. 3734-3737). [5495870] https://doi.org/10.1109/ICASSP.2010.5495870

A sparse adaptive filtering using time-varying soft-thresholding techniques. / Murakami, Yukihiro; Yamagishi, Masao; Yukawa, Masahiro; Yamada, Isao.

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2010. p. 3734-3737 5495870.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Murakami, Y, Yamagishi, M, Yukawa, M & Yamada, I 2010, A sparse adaptive filtering using time-varying soft-thresholding techniques. in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings., 5495870, pp. 3734-3737, 2010 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2010, Dallas, TX, United States, 10/3/14. https://doi.org/10.1109/ICASSP.2010.5495870
Murakami Y, Yamagishi M, Yukawa M, Yamada I. A sparse adaptive filtering using time-varying soft-thresholding techniques. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2010. p. 3734-3737. 5495870 https://doi.org/10.1109/ICASSP.2010.5495870
Murakami, Yukihiro ; Yamagishi, Masao ; Yukawa, Masahiro ; Yamada, Isao. / A sparse adaptive filtering using time-varying soft-thresholding techniques. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2010. pp. 3734-3737
@inproceedings{53973409ecae4b2c9b812a225531cf41,
title = "A sparse adaptive filtering using time-varying soft-thresholding techniques",
abstract = "In this paper, we propose a novel adaptive filtering algorithm based on an iterative use of (i) the proximity operator and (ii) the parallel variable-metric projection. Our time-varying cost function is a weighted sum of squared distances (in a variable-metric sense) plus a possibly nonsmooth penalty term, and the proposed algorithm is derived along the idea of proximal forward-backward splitting in convex analysis. For application to sparse-system identification problems, we employ the (weighted) ℓ1 norm as the penalty term, leading to a time-varying soft-thresholding operator. As the simple example of the proposed algorithm, we present the variable-metric affine projection algorithm composed with the time-varying soft-thresholding operator. Numerical examples demonstrate that the proposed algorithms notably outperform their counterparts without soft-thresholding both in convergence speed and steady-state mismatch, while the extra computational complexity due to the additional soft-thresholding is negligibly low.",
keywords = "Parallel projection, Proximal forward-backward splitting, Soft-thresholding, Sparse adaptive filtering, Variable metric",
author = "Yukihiro Murakami and Masao Yamagishi and Masahiro Yukawa and Isao Yamada",
year = "2010",
doi = "10.1109/ICASSP.2010.5495870",
language = "English",
isbn = "9781424442966",
pages = "3734--3737",
booktitle = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

}

TY - GEN

T1 - A sparse adaptive filtering using time-varying soft-thresholding techniques

AU - Murakami, Yukihiro

AU - Yamagishi, Masao

AU - Yukawa, Masahiro

AU - Yamada, Isao

PY - 2010

Y1 - 2010

N2 - In this paper, we propose a novel adaptive filtering algorithm based on an iterative use of (i) the proximity operator and (ii) the parallel variable-metric projection. Our time-varying cost function is a weighted sum of squared distances (in a variable-metric sense) plus a possibly nonsmooth penalty term, and the proposed algorithm is derived along the idea of proximal forward-backward splitting in convex analysis. For application to sparse-system identification problems, we employ the (weighted) ℓ1 norm as the penalty term, leading to a time-varying soft-thresholding operator. As the simple example of the proposed algorithm, we present the variable-metric affine projection algorithm composed with the time-varying soft-thresholding operator. Numerical examples demonstrate that the proposed algorithms notably outperform their counterparts without soft-thresholding both in convergence speed and steady-state mismatch, while the extra computational complexity due to the additional soft-thresholding is negligibly low.

AB - In this paper, we propose a novel adaptive filtering algorithm based on an iterative use of (i) the proximity operator and (ii) the parallel variable-metric projection. Our time-varying cost function is a weighted sum of squared distances (in a variable-metric sense) plus a possibly nonsmooth penalty term, and the proposed algorithm is derived along the idea of proximal forward-backward splitting in convex analysis. For application to sparse-system identification problems, we employ the (weighted) ℓ1 norm as the penalty term, leading to a time-varying soft-thresholding operator. As the simple example of the proposed algorithm, we present the variable-metric affine projection algorithm composed with the time-varying soft-thresholding operator. Numerical examples demonstrate that the proposed algorithms notably outperform their counterparts without soft-thresholding both in convergence speed and steady-state mismatch, while the extra computational complexity due to the additional soft-thresholding is negligibly low.

KW - Parallel projection

KW - Proximal forward-backward splitting

KW - Soft-thresholding

KW - Sparse adaptive filtering

KW - Variable metric

UR - http://www.scopus.com/inward/record.url?scp=77954901517&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77954901517&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2010.5495870

DO - 10.1109/ICASSP.2010.5495870

M3 - Conference contribution

AN - SCOPUS:77954901517

SN - 9781424442966

SP - 3734

EP - 3737

BT - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

ER -