An order downdating algorithm for tracking system order and parameters in recursive least squares identification

Author

Apley, Daniel W. ; Shi, Jianjun

Author_Institution

Dept. of Ind. Eng., Texas A&M Univ., College Station, TX, USA

Volume

47

Issue

11

fYear

1999

fDate

11/1/1999 12:00:00 AM

Firstpage

3134

Lastpage

3137

Abstract

A new time and order recursive method for on-line tracking of system order and parameters using recursive least squares (RLS) is presented. The method consists of two parts: a time updating portion that uses existing RLS inverse QR decomposition algorithms and a new computationally efficient “order downdating” portion that calculates the model parameters and residual error energies for an entire set of models with order varying from one to some prespecified maximum model order

Keywords

error analysis; least squares approximations; matrix decomposition; matrix inversion; recursive estimation; spectral analysis; tracking; RLS inverse QR decomposition algorithms; computationally efficient order downdating; inverse QR factorization; model parameters; order downdating algorithm; recursive least squares identification; residual error energies; spectral estimation; system order tracking; system parameters tracking; time updating; Adaptive control; Least squares approximation; Least squares methods; Recursive estimation; Resonance light scattering; Robots; Signal processing algorithms; Solid modeling; System identification; Time varying systems;

fLanguage

English

Journal_Title

Signal Processing, IEEE Transactions on

Publisher

ieee

ISSN

1053-587X

Type

jour

DOI

10.1109/78.796448

Filename

796448