Brief New identification approaches for disturbed models

Authors:
Kueiming Lo;Wook-Hyun Kwon
Affiliations:
Department of Mathematical Science, Tsinghua University, Beijing 100084, China;School of Electrical Engineering, Seoul National University, Seoul 151-742, South Korea
Venue:
Automatica (Journal of IFAC)
Year:
2003

Citing 8
Cited 0

Integral constraints on the accuracy of least-squares estimation

Automatica (Journal of IFAC)
Subspace-based identification of infinite-dimensional multivariable systems from frequency-response data

Automatica (Journal of IFAC)
Shaping frequency-dependent time resolution when estimatingspectral properties with parametric methods

IEEE Transactions on Signal Processing
Frequency domain analysis of tracking and noise performance ofadaptive algorithms

IEEE Transactions on Signal Processing
A metric for ARMA processes

IEEE Transactions on Signal Processing
On designing stable allpass filters using AR modeling

IEEE Transactions on Signal Processing
Cepstral coefficients, covariance lags, and pole-zero models forfinite data strings

IEEE Transactions on Signal Processing
Brief Finite sample properties of system identification of ARX models under mixing conditions

Automatica (Journal of IFAC)

Quantified Score

Hi-index	22.14

Visualization

Abstract

Recently, much research has been conducted in the field of identification of the linear models. In general, these methods use a time-domain estimate or a frequency-domain estimate. In this paper, the time-domain estimate and the frequency-domain estimate were combined to identify the autoregressive exogenous noise (ARX) interference model. The concept of a general prediction error criterion is introduced for the time-domain estimate. An optimal frequency estimation is introduced for the frequency-domain estimate. A new identification method, called the empirical frequency-domain optimal parameter estimate, is proposed for disturbed systems. It is fully applied and developed for the output error model and a specific case or the ARX model. The algorithm theoretically provides the globally optimum frequency-domain estimate of the model. Some simulations are included to illustrate the new identification method.