Sequential generation of D-optimal input designs for linear dynamic systems

This paper considers the problem of sequentially generating test signals for parameter estimation in linear, single-input, single-output, discrete-time systems using a frequency-domain approach based on the theory of optimal experiments in regression analysis. The input signals are power constrained...

Full description

Saved in:
Bibliographic Details
Published in:Journal of optimization theory and applications 1981-10, Vol.35, p.277-291
Main Author: ZARROP, M B
Format: Article
Language:English
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper considers the problem of sequentially generating test signals for parameter estimation in linear, single-input, single-output, discrete-time systems using a frequency-domain approach based on the theory of optimal experiments in regression analysis. The input signals are power constrained and are optimal in the sense that system information is maximized, where the criterion employed is the determinant of the Fisher information matrix (D-optimality). A class of algorithms is investigated, each member of which generates a sequence of input designs converging to a D-optimum. A number of these algorithms are compared computationally.
ISSN:0022-3239
DOI:10.1007/BF00934581