Unbiased parameter identification for noisy autoregressive signals

Wei Xing Zheng

doi:10.1109/ISCAS.2001.921021

Unbiased parameter identification for noisy autoregressive signals

Wei Xing Zheng

School of Computer, Data & Mathematical Sciences

Research output: Chapter in Book / Conference Paper › Conference Paper › peer-review

2 Citations (Scopus)

Abstract

A simple algorithm is developed for unbiased parameter identification of autoregressive (AR) signals subject to white measurement noise. It is shown that the corrupting noise variance, which determines the bias in the standard least-squares (LS) parameter estimator, can be estimated by simply using the expected LS errors when the ratio between the driving noise variance and the corrupting noise variance is known or obtainable in some way. Then an LS based algorithm is established via the principle of bias compensation. Compared with the other LS based algorithms recently developed, the introduced algorithm produces better parameter estimates, requires fewer computations and has a simpler algorithmic structure.

Original language	English
Title of host publication	ISCAS 2001 - 2001 IEEE International Symposium on Circuits and Systems, Conference Proceedings
Pages	121-124
Number of pages	4
DOIs	https://doi.org/10.1109/ISCAS.2001.921021
Publication status	Published - 2001
Event	2001 IEEE International Symposium on Circuits and Systems, ISCAS 2001 - Sydney, NSW, Australia Duration: 6 May 2001 → 9 May 2001

Publication series

Name	ISCAS 2001 - 2001 IEEE International Symposium on Circuits and Systems, Conference Proceedings
Volume	2

Conference

Conference	2001 IEEE International Symposium on Circuits and Systems, ISCAS 2001
Country/Territory	Australia
City	Sydney, NSW
Period	6/05/01 → 9/05/01

Access to Document

10.1109/ISCAS.2001.921021

Cite this

@inproceedings{df6986b432a949708d00cf111c2a151a,

title = "Unbiased parameter identification for noisy autoregressive signals",

abstract = "A simple algorithm is developed for unbiased parameter identification of autoregressive (AR) signals subject to white measurement noise. It is shown that the corrupting noise variance, which determines the bias in the standard least-squares (LS) parameter estimator, can be estimated by simply using the expected LS errors when the ratio between the driving noise variance and the corrupting noise variance is known or obtainable in some way. Then an LS based algorithm is established via the principle of bias compensation. Compared with the other LS based algorithms recently developed, the introduced algorithm produces better parameter estimates, requires fewer computations and has a simpler algorithmic structure.",

author = "Zheng, {Wei Xing}",

year = "2001",

doi = "10.1109/ISCAS.2001.921021",

language = "English",

isbn = "0780366859",

series = "ISCAS 2001 - 2001 IEEE International Symposium on Circuits and Systems, Conference Proceedings",

pages = "121--124",

booktitle = "ISCAS 2001 - 2001 IEEE International Symposium on Circuits and Systems, Conference Proceedings",

note = "2001 IEEE International Symposium on Circuits and Systems, ISCAS 2001 ; Conference date: 06-05-2001 Through 09-05-2001",

}

Zheng, WX 2001, Unbiased parameter identification for noisy autoregressive signals. in ISCAS 2001 - 2001 IEEE International Symposium on Circuits and Systems, Conference Proceedings., 921021, ISCAS 2001 - 2001 IEEE International Symposium on Circuits and Systems, Conference Proceedings, vol. 2, pp. 121-124, 2001 IEEE International Symposium on Circuits and Systems, ISCAS 2001, Sydney, NSW, Australia, 6/05/01. https://doi.org/10.1109/ISCAS.2001.921021

Unbiased parameter identification for noisy autoregressive signals. / Zheng, Wei Xing.
ISCAS 2001 - 2001 IEEE International Symposium on Circuits and Systems, Conference Proceedings. 2001. p. 121-124 921021 (ISCAS 2001 - 2001 IEEE International Symposium on Circuits and Systems, Conference Proceedings; Vol. 2).

Research output: Chapter in Book / Conference Paper › Conference Paper › peer-review

TY - GEN

T1 - Unbiased parameter identification for noisy autoregressive signals

AU - Zheng, Wei Xing

PY - 2001

Y1 - 2001

N2 - A simple algorithm is developed for unbiased parameter identification of autoregressive (AR) signals subject to white measurement noise. It is shown that the corrupting noise variance, which determines the bias in the standard least-squares (LS) parameter estimator, can be estimated by simply using the expected LS errors when the ratio between the driving noise variance and the corrupting noise variance is known or obtainable in some way. Then an LS based algorithm is established via the principle of bias compensation. Compared with the other LS based algorithms recently developed, the introduced algorithm produces better parameter estimates, requires fewer computations and has a simpler algorithmic structure.

AB - A simple algorithm is developed for unbiased parameter identification of autoregressive (AR) signals subject to white measurement noise. It is shown that the corrupting noise variance, which determines the bias in the standard least-squares (LS) parameter estimator, can be estimated by simply using the expected LS errors when the ratio between the driving noise variance and the corrupting noise variance is known or obtainable in some way. Then an LS based algorithm is established via the principle of bias compensation. Compared with the other LS based algorithms recently developed, the introduced algorithm produces better parameter estimates, requires fewer computations and has a simpler algorithmic structure.

UR - http://www.scopus.com/inward/record.url?scp=0035014638&partnerID=8YFLogxK

U2 - 10.1109/ISCAS.2001.921021

DO - 10.1109/ISCAS.2001.921021

M3 - Conference Paper

AN - SCOPUS:0035014638

SN - 0780366859

SN - 9780780366855

T3 - ISCAS 2001 - 2001 IEEE International Symposium on Circuits and Systems, Conference Proceedings

SP - 121

EP - 124

BT - ISCAS 2001 - 2001 IEEE International Symposium on Circuits and Systems, Conference Proceedings

T2 - 2001 IEEE International Symposium on Circuits and Systems, ISCAS 2001

Y2 - 6 May 2001 through 9 May 2001

ER -

Unbiased parameter identification for noisy autoregressive signals

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this