An estimator for the eigenvalues of the system matrix of a periodic-reference LMS algorithm

Tuomas Haarnoja, Kari Tammi, Kai Zenger

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

Abstract

The convergence analysis of the Least Mean Square (LMS) algorithm has been conventionally based on stochastic signals and describes thus only the average behavior of the algorithm. It has been shown previously that a periodic-reference LMS system can be regarded as a linear time-periodic system whose stability can be determined from the monodromy matrix. Generally, the monodromy matrix can only be solved numerically and does not thus reveal the actual factors behind the dynamics of the system. This paper derives an estimator for the eigenvalues of the monodromy matrix. The estimator is easy to calculate, and it also reveals the underlying reason for the bad convergence of the LMS algorithm in some special cases. The estimator is confirmed by comparing it to the precise eigenvalues of the monodromy matrix. The estimator is found to be accurate for the eigenvalues close to unity
Original languageEnglish
Title of host publicationProceedings of 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012
PublisherInstitute of Electrical and Electronic Engineers IEEE
Pages3777-3780
ISBN (Electronic)978-1-4673-0046-9
ISBN (Print)978-1-4673-0045-2, 978-1-4673-0044-5
DOIs
Publication statusPublished - 2012
MoE publication typeNot Eligible
EventIEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 - Kyoto, Japan
Duration: 25 Mar 201230 Mar 2012

Conference

ConferenceIEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012
Abbreviated titleICASSP 2012
CountryJapan
CityKyoto
Period25/03/1230/03/12

Fingerprint

Time varying systems
System stability

Keywords

  • LMS algorithm
  • monodromy matrix
  • estimator for the eigenvalues
  • convergence rate

Cite this

Haarnoja, T., Tammi, K., & Zenger, K. (2012). An estimator for the eigenvalues of the system matrix of a periodic-reference LMS algorithm. In Proceedings of 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012 (pp. 3777-3780). Institute of Electrical and Electronic Engineers IEEE. https://doi.org/10.1109/ICASSP.2012.6288739
Haarnoja, Tuomas ; Tammi, Kari ; Zenger, Kai. / An estimator for the eigenvalues of the system matrix of a periodic-reference LMS algorithm. Proceedings of 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012. Institute of Electrical and Electronic Engineers IEEE, 2012. pp. 3777-3780
@inproceedings{191aa43d610f454c928d875646a41717,
title = "An estimator for the eigenvalues of the system matrix of a periodic-reference LMS algorithm",
abstract = "The convergence analysis of the Least Mean Square (LMS) algorithm has been conventionally based on stochastic signals and describes thus only the average behavior of the algorithm. It has been shown previously that a periodic-reference LMS system can be regarded as a linear time-periodic system whose stability can be determined from the monodromy matrix. Generally, the monodromy matrix can only be solved numerically and does not thus reveal the actual factors behind the dynamics of the system. This paper derives an estimator for the eigenvalues of the monodromy matrix. The estimator is easy to calculate, and it also reveals the underlying reason for the bad convergence of the LMS algorithm in some special cases. The estimator is confirmed by comparing it to the precise eigenvalues of the monodromy matrix. The estimator is found to be accurate for the eigenvalues close to unity",
keywords = "LMS algorithm, monodromy matrix, estimator for the eigenvalues, convergence rate",
author = "Tuomas Haarnoja and Kari Tammi and Kai Zenger",
note = "Project code: 26153",
year = "2012",
doi = "10.1109/ICASSP.2012.6288739",
language = "English",
isbn = "978-1-4673-0045-2",
pages = "3777--3780",
booktitle = "Proceedings of 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012",
publisher = "Institute of Electrical and Electronic Engineers IEEE",
address = "United States",

}

Haarnoja, T, Tammi, K & Zenger, K 2012, An estimator for the eigenvalues of the system matrix of a periodic-reference LMS algorithm. in Proceedings of 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012. Institute of Electrical and Electronic Engineers IEEE, pp. 3777-3780, IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012, Kyoto, Japan, 25/03/12. https://doi.org/10.1109/ICASSP.2012.6288739

An estimator for the eigenvalues of the system matrix of a periodic-reference LMS algorithm. / Haarnoja, Tuomas; Tammi, Kari; Zenger, Kai.

Proceedings of 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012. Institute of Electrical and Electronic Engineers IEEE, 2012. p. 3777-3780.

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

TY - GEN

T1 - An estimator for the eigenvalues of the system matrix of a periodic-reference LMS algorithm

AU - Haarnoja, Tuomas

AU - Tammi, Kari

AU - Zenger, Kai

N1 - Project code: 26153

PY - 2012

Y1 - 2012

N2 - The convergence analysis of the Least Mean Square (LMS) algorithm has been conventionally based on stochastic signals and describes thus only the average behavior of the algorithm. It has been shown previously that a periodic-reference LMS system can be regarded as a linear time-periodic system whose stability can be determined from the monodromy matrix. Generally, the monodromy matrix can only be solved numerically and does not thus reveal the actual factors behind the dynamics of the system. This paper derives an estimator for the eigenvalues of the monodromy matrix. The estimator is easy to calculate, and it also reveals the underlying reason for the bad convergence of the LMS algorithm in some special cases. The estimator is confirmed by comparing it to the precise eigenvalues of the monodromy matrix. The estimator is found to be accurate for the eigenvalues close to unity

AB - The convergence analysis of the Least Mean Square (LMS) algorithm has been conventionally based on stochastic signals and describes thus only the average behavior of the algorithm. It has been shown previously that a periodic-reference LMS system can be regarded as a linear time-periodic system whose stability can be determined from the monodromy matrix. Generally, the monodromy matrix can only be solved numerically and does not thus reveal the actual factors behind the dynamics of the system. This paper derives an estimator for the eigenvalues of the monodromy matrix. The estimator is easy to calculate, and it also reveals the underlying reason for the bad convergence of the LMS algorithm in some special cases. The estimator is confirmed by comparing it to the precise eigenvalues of the monodromy matrix. The estimator is found to be accurate for the eigenvalues close to unity

KW - LMS algorithm

KW - monodromy matrix

KW - estimator for the eigenvalues

KW - convergence rate

U2 - 10.1109/ICASSP.2012.6288739

DO - 10.1109/ICASSP.2012.6288739

M3 - Conference article in proceedings

SN - 978-1-4673-0045-2

SN - 978-1-4673-0044-5

SP - 3777

EP - 3780

BT - Proceedings of 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012

PB - Institute of Electrical and Electronic Engineers IEEE

ER -

Haarnoja T, Tammi K, Zenger K. An estimator for the eigenvalues of the system matrix of a periodic-reference LMS algorithm. In Proceedings of 2012 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2012. Institute of Electrical and Electronic Engineers IEEE. 2012. p. 3777-3780 https://doi.org/10.1109/ICASSP.2012.6288739