Pseudorandom maximum length signal design with bias compensation least squares estimation for system identification

M F L Foo; Ai Hui Tan; Kartik Prasad Basu

doi:10.1109/IMTC.2007.378981

Pseudorandom maximum length signal design with bias compensation least squares estimation for system identification

M F L Foo, Ai Hui Tan, Kartik Prasad Basu

Multimedia University

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

1 Citation (Scopus)

Abstract

The effect of input and output noise towards the identification of the best linear approximation of a system is investigated. This leads to the problem of errors-in-variables (EIV). The effectiveness of one particular EIV method, namely the bias compensation least squares estimation method, is analyzed, with simulations carried out on a first order bilinear system. It is shown that the use of perturbation signals with carefully selected harmonic properties can lead to significant improvements in the estimation of the best linear approximation of the system. In particular, a spectrum that is sparser but having a larger magnitude at the nonzero harmonics is found to be more robust towards the effect of noise.

Original language	English
Title of host publication	IMTC 2007. IEEE Instrumentation and Measurement Technology Conference Proceedings, 2007
Place of Publication	New York
Publisher	IEEE
Pages	417-422
Number of pages	6
ISBN (Electronic)	1-4244-1080-0
ISBN (Print)	1-4244-0588-2
DOIs	https://doi.org/10.1109/IMTC.2007.378981
Publication status	Published - 25 Jun 2007
Event	IEEE Instrumentation and Measurement Technology Conference - Warsaw, Poland Duration: 1 May 2007 → 3 May 2007

Conference

Conference	IEEE Instrumentation and Measurement Technology Conference
Abbreviated title	IMTC 2007
Country/Territory	Poland
City	Warsaw
Period	1/05/07 → 3/05/07

Keywords

bias compensation
estimation methods
maximum length signals
perturbation signals
pseudorandom signals
system identification

Access to Document

10.1109/IMTC.2007.378981

Cite this

Pseudorandom maximum length signal design with bias compensation least squares estimation for system identification. / Foo, M F L; Tan, Ai Hui; Basu, Kartik Prasad.
IMTC 2007. IEEE Instrumentation and Measurement Technology Conference Proceedings, 2007. New York: IEEE, 2007. p. 417-422.

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

Foo, MFL, Tan, AH & Basu, KP 2007, Pseudorandom maximum length signal design with bias compensation least squares estimation for system identification. in IMTC 2007. IEEE Instrumentation and Measurement Technology Conference Proceedings, 2007. IEEE, New York, pp. 417-422, IEEE Instrumentation and Measurement Technology Conference, Warsaw, Poland, 1/05/07. https://doi.org/10.1109/IMTC.2007.378981

@inproceedings{d50875dd608a4d38aad3001712b1868a,

title = "Pseudorandom maximum length signal design with bias compensation least squares estimation for system identification",

abstract = "The effect of input and output noise towards the identification of the best linear approximation of a system is investigated. This leads to the problem of errors-in-variables (EIV). The effectiveness of one particular EIV method, namely the bias compensation least squares estimation method, is analyzed, with simulations carried out on a first order bilinear system. It is shown that the use of perturbation signals with carefully selected harmonic properties can lead to significant improvements in the estimation of the best linear approximation of the system. In particular, a spectrum that is sparser but having a larger magnitude at the nonzero harmonics is found to be more robust towards the effect of noise.",

keywords = "bias compensation, estimation methods, maximum length signals, perturbation signals, pseudorandom signals, system identification",

author = "Foo, {M F L} and Tan, {Ai Hui} and Basu, {Kartik Prasad}",

year = "2007",

month = jun,

day = "25",

doi = "10.1109/IMTC.2007.378981",

language = "English",

isbn = "1-4244-0588-2",

pages = "417--422",

booktitle = "IMTC 2007. IEEE Instrumentation and Measurement Technology Conference Proceedings, 2007",

publisher = "IEEE",

address = "United States",

note = "IEEE Instrumentation and Measurement Technology Conference, IMTC 2007 ; Conference date: 01-05-2007 Through 03-05-2007",

}

TY - GEN

T1 - Pseudorandom maximum length signal design with bias compensation least squares estimation for system identification

AU - Foo, M F L

AU - Tan, Ai Hui

AU - Basu, Kartik Prasad

PY - 2007/6/25

Y1 - 2007/6/25

N2 - The effect of input and output noise towards the identification of the best linear approximation of a system is investigated. This leads to the problem of errors-in-variables (EIV). The effectiveness of one particular EIV method, namely the bias compensation least squares estimation method, is analyzed, with simulations carried out on a first order bilinear system. It is shown that the use of perturbation signals with carefully selected harmonic properties can lead to significant improvements in the estimation of the best linear approximation of the system. In particular, a spectrum that is sparser but having a larger magnitude at the nonzero harmonics is found to be more robust towards the effect of noise.

AB - The effect of input and output noise towards the identification of the best linear approximation of a system is investigated. This leads to the problem of errors-in-variables (EIV). The effectiveness of one particular EIV method, namely the bias compensation least squares estimation method, is analyzed, with simulations carried out on a first order bilinear system. It is shown that the use of perturbation signals with carefully selected harmonic properties can lead to significant improvements in the estimation of the best linear approximation of the system. In particular, a spectrum that is sparser but having a larger magnitude at the nonzero harmonics is found to be more robust towards the effect of noise.

KW - bias compensation

KW - estimation methods

KW - maximum length signals

KW - perturbation signals

KW - pseudorandom signals

KW - system identification

U2 - 10.1109/IMTC.2007.378981

DO - 10.1109/IMTC.2007.378981

M3 - Conference proceeding

SN - 1-4244-0588-2

SP - 417

EP - 422

BT - IMTC 2007. IEEE Instrumentation and Measurement Technology Conference Proceedings, 2007

PB - IEEE

CY - New York

T2 - IEEE Instrumentation and Measurement Technology Conference

Y2 - 1 May 2007 through 3 May 2007

ER -

Pseudorandom maximum length signal design with bias compensation least squares estimation for system identification

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this