Unbiased Discrete Linear System Identification in a Noisy Environment

Abstract

An optimal filter is proposed to compute a basis of a set of "noisy" filter input functions for the case where signal and noise statistics are specified. Procedures for determining the filter transfer function and the transfer function of a restoring filter are given. The filter is then applied to the problem of minimum-order mathematical modeling of discrete, multi-variable, dynamic systems in noisy environments. It is shown that the resulting estimates of system order and system parameters are unbiased.

Department(s)

Mechanical and Aerospace Engineering

Keywords and Phrases

Information Theory

Document Type

Article - Journal

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 1974 Adv Instrumparts, All rights reserved.

Publication Date

01 Jan 1974

This document is currently not available here.

Share

 
COinS