Robust reduced-rank adaptive algorithm based on parallel subgradient projection and krylov subspace

Masahiro Yukawa, Rodrigo C. De Lamare, Isao Yamada

Research output: Contribution to journalArticle

17 Citations (Scopus)

Abstract

In this paper, we propose a novel reduced-rank adaptive filtering algorithm exploiting the Krylov subspace associated with estimates of certain statistics of input and output signals. We point out that, when the estimated statistics are erroneous (e.g., due to sudden changes of environments), the existing Krylov-subspace-based reduced-rank methods compute the point that minimizes a wrong mean-square error (MSE) in the subspace. The proposed algorithm exploits the set-theoretic adaptive filtering framework for tracking efficiently the optimal point in the sense of minimizing the true MSE in the subspace. Therefore, compared with the existing methods, the proposed algorithm is more suited to adaptive filtering applications. A convergence analysis of the algorithm is performed by extending the adaptive projected subgradient method (APSM). Numerical examples demonstrate that the proposed algorithm enjoys better tracking performance than the existing methods for system identification problems.

Original languageEnglish
Article number5164903
Pages (from-to)4660-4674
Number of pages15
JournalIEEE Transactions on Signal Processing
Volume57
Issue number12
DOIs
Publication statusPublished - 2009 Dec
Externally publishedYes

Fingerprint

Adaptive algorithms
Adaptive filtering
Mean square error
Statistics
Identification (control systems)

Keywords

  • Krylov subspace
  • Reduced-rank adaptive filtering
  • Set theory
  • Subgradient methods

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Signal Processing

Cite this

Robust reduced-rank adaptive algorithm based on parallel subgradient projection and krylov subspace. / Yukawa, Masahiro; De Lamare, Rodrigo C.; Yamada, Isao.

In: IEEE Transactions on Signal Processing, Vol. 57, No. 12, 5164903, 12.2009, p. 4660-4674.

Research output: Contribution to journalArticle

@article{b332992f695a49ddbfd3ca0b239f8d9f,
title = "Robust reduced-rank adaptive algorithm based on parallel subgradient projection and krylov subspace",
abstract = "In this paper, we propose a novel reduced-rank adaptive filtering algorithm exploiting the Krylov subspace associated with estimates of certain statistics of input and output signals. We point out that, when the estimated statistics are erroneous (e.g., due to sudden changes of environments), the existing Krylov-subspace-based reduced-rank methods compute the point that minimizes a wrong mean-square error (MSE) in the subspace. The proposed algorithm exploits the set-theoretic adaptive filtering framework for tracking efficiently the optimal point in the sense of minimizing the true MSE in the subspace. Therefore, compared with the existing methods, the proposed algorithm is more suited to adaptive filtering applications. A convergence analysis of the algorithm is performed by extending the adaptive projected subgradient method (APSM). Numerical examples demonstrate that the proposed algorithm enjoys better tracking performance than the existing methods for system identification problems.",
keywords = "Krylov subspace, Reduced-rank adaptive filtering, Set theory, Subgradient methods",
author = "Masahiro Yukawa and {De Lamare}, {Rodrigo C.} and Isao Yamada",
year = "2009",
month = "12",
doi = "10.1109/TSP.2009.2027397",
language = "English",
volume = "57",
pages = "4660--4674",
journal = "IEEE Transactions on Signal Processing",
issn = "1053-587X",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "12",

}

TY - JOUR

T1 - Robust reduced-rank adaptive algorithm based on parallel subgradient projection and krylov subspace

AU - Yukawa, Masahiro

AU - De Lamare, Rodrigo C.

AU - Yamada, Isao

PY - 2009/12

Y1 - 2009/12

N2 - In this paper, we propose a novel reduced-rank adaptive filtering algorithm exploiting the Krylov subspace associated with estimates of certain statistics of input and output signals. We point out that, when the estimated statistics are erroneous (e.g., due to sudden changes of environments), the existing Krylov-subspace-based reduced-rank methods compute the point that minimizes a wrong mean-square error (MSE) in the subspace. The proposed algorithm exploits the set-theoretic adaptive filtering framework for tracking efficiently the optimal point in the sense of minimizing the true MSE in the subspace. Therefore, compared with the existing methods, the proposed algorithm is more suited to adaptive filtering applications. A convergence analysis of the algorithm is performed by extending the adaptive projected subgradient method (APSM). Numerical examples demonstrate that the proposed algorithm enjoys better tracking performance than the existing methods for system identification problems.

AB - In this paper, we propose a novel reduced-rank adaptive filtering algorithm exploiting the Krylov subspace associated with estimates of certain statistics of input and output signals. We point out that, when the estimated statistics are erroneous (e.g., due to sudden changes of environments), the existing Krylov-subspace-based reduced-rank methods compute the point that minimizes a wrong mean-square error (MSE) in the subspace. The proposed algorithm exploits the set-theoretic adaptive filtering framework for tracking efficiently the optimal point in the sense of minimizing the true MSE in the subspace. Therefore, compared with the existing methods, the proposed algorithm is more suited to adaptive filtering applications. A convergence analysis of the algorithm is performed by extending the adaptive projected subgradient method (APSM). Numerical examples demonstrate that the proposed algorithm enjoys better tracking performance than the existing methods for system identification problems.

KW - Krylov subspace

KW - Reduced-rank adaptive filtering

KW - Set theory

KW - Subgradient methods

UR - http://www.scopus.com/inward/record.url?scp=70549113152&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70549113152&partnerID=8YFLogxK

U2 - 10.1109/TSP.2009.2027397

DO - 10.1109/TSP.2009.2027397

M3 - Article

VL - 57

SP - 4660

EP - 4674

JO - IEEE Transactions on Signal Processing

JF - IEEE Transactions on Signal Processing

SN - 1053-587X

IS - 12

M1 - 5164903

ER -