A fast approach to identification using deconvolution

1 January 1983

conference paper
Published by Institute of Electrical and Electronics Engineers (IEEE)

Vol. 22, 1347-1352
https://doi.org/10.1109/cdc.1983.269747

Abstract

In this paper, we propose a fast approach to impulse response and noise-variance identification for a finite-order, linear, time-invariant, single-input/single-output system, whose input driving noise is white (stationary or nonstationary) and measurement noise is stationary, white and Gaussian. Our algorithm is an iterative block component method that includes two stages, deconvolution and prediction-error identification. Experiences with our method indicates that it works well and saves about an order of magnitude in computation. Analyses and examples are given in this paper to support this claim.

Keywords

This publication has 7 references indexed in Scilit:

Identifiability of nonminimum phase linear stochastic systems
IEEE Transactions on Automatic Control, 1983
Maximum-Likelihood Seismic Deconvolution
IEEE Transactions on Geoscience and Remote Sensing, 1983
Maximum likelihood detection and estimation of Bernoulli - Gaussian processes
IEEE Transactions on Information Theory, 1982
Minimum-Variance Deconvolution
IEEE Transactions on Geoscience and Remote Sensing, 1981
Identification of stochastic linear dynamic systems using Kalman filter representation
AIAA Journal, 1971
Recursive generalised-least-squares procedure for online identification of process parameters
Proceedings of the Institution of Electrical Engineers, 1969
Evaluation of likelihood functions for Gaussian signals
IEEE Transactions on Information Theory, 1965