Fast adaptive identification of autoregressive signals subject to noise

Wei Xing Zheng

Fast adaptive identification of autoregressive signals subject to noise

Research output: Chapter in Book / Conference Paper › Conference Paper

Abstract

Adaptive identification of autoregressive (AR) signals subject to white measurement noise is studied. A fast adaptive algorithm, which is based on the recently proposed improved least-squares (LS) method, is developed. The variance of the white measurements noise, which specifies the source of the noise-induced bias in the standard LS estimate, is calculated by means of extra noisy measurements of the AR signal. With a good estimate of the measurement noise variance being attained more quickly, the convergence speed of the developed adaptive identification algorithm can be accelerated. Numerical results are presented to demonstrate the promising performance of the new fast adaptive identification algorithm.

Original language	English
Title of host publication	Proceedings of 37th IEEE International Symposium on Circuits and Systems, held in Vancouver, B.C., Canada, 23-26 May, 2004
Publisher	IEEE
Number of pages	1
ISBN (Print)	078038251X
Publication status	Published - 2004
Event	International Symposium on Circuits and Systems - Duration: 1 Jan 2004 → …

Conference

Conference	International Symposium on Circuits and Systems
Period	1/01/04 → …

Keywords

least squares
noise
autoregressive signals
noise measurement
algorithms
convergence speed

Cite this

@inproceedings{e1cee5c5894a43089c389c25486872d2,

title = "Fast adaptive identification of autoregressive signals subject to noise",

abstract = "Adaptive identification of autoregressive (AR) signals subject to white measurement noise is studied. A fast adaptive algorithm, which is based on the recently proposed improved least-squares (LS) method, is developed. The variance of the white measurements noise, which specifies the source of the noise-induced bias in the standard LS estimate, is calculated by means of extra noisy measurements of the AR signal. With a good estimate of the measurement noise variance being attained more quickly, the convergence speed of the developed adaptive identification algorithm can be accelerated. Numerical results are presented to demonstrate the promising performance of the new fast adaptive identification algorithm.",

keywords = "least squares, noise, autoregressive signals, noise measurement, algorithms, convergence speed",

author = "Zheng, {Wei Xing}",

year = "2004",

language = "English",

isbn = "078038251X",

booktitle = "Proceedings of 37th IEEE International Symposium on Circuits and Systems, held in Vancouver, B.C., Canada, 23-26 May, 2004",

publisher = "IEEE",

note = "International Symposium on Circuits and Systems ; Conference date: 01-01-2004",

}

TY - GEN

T1 - Fast adaptive identification of autoregressive signals subject to noise

AU - Zheng, Wei Xing

PY - 2004

Y1 - 2004

N2 - Adaptive identification of autoregressive (AR) signals subject to white measurement noise is studied. A fast adaptive algorithm, which is based on the recently proposed improved least-squares (LS) method, is developed. The variance of the white measurements noise, which specifies the source of the noise-induced bias in the standard LS estimate, is calculated by means of extra noisy measurements of the AR signal. With a good estimate of the measurement noise variance being attained more quickly, the convergence speed of the developed adaptive identification algorithm can be accelerated. Numerical results are presented to demonstrate the promising performance of the new fast adaptive identification algorithm.

AB - Adaptive identification of autoregressive (AR) signals subject to white measurement noise is studied. A fast adaptive algorithm, which is based on the recently proposed improved least-squares (LS) method, is developed. The variance of the white measurements noise, which specifies the source of the noise-induced bias in the standard LS estimate, is calculated by means of extra noisy measurements of the AR signal. With a good estimate of the measurement noise variance being attained more quickly, the convergence speed of the developed adaptive identification algorithm can be accelerated. Numerical results are presented to demonstrate the promising performance of the new fast adaptive identification algorithm.

KW - least squares

KW - noise

KW - autoregressive signals

KW - noise measurement

KW - algorithms

KW - convergence speed

UR - http://handle.uws.edu.au:8081/1959.7/44843

M3 - Conference Paper

SN - 078038251X

BT - Proceedings of 37th IEEE International Symposium on Circuits and Systems, held in Vancouver, B.C., Canada, 23-26 May, 2004

PB - IEEE

T2 - International Symposium on Circuits and Systems

Y2 - 1 January 2004

ER -

Fast adaptive identification of autoregressive signals subject to noise

Abstract

Conference

Keywords

Other files and links

Fingerprint

Cite this