DocumentCode
337702
Title
Least squares identification using μ-Markov parameterizations
Author
Van Pelt, Tobin H. ; Bernstein, Dennis S.
Author_Institution
Dept. of Aerosp. Eng., Michigan Univ., MI, USA
Volume
1
fYear
1998
fDate
1998
Firstpage
618
Abstract
In this paper we introduce μ-Markov parameterizations for use in least squares estimation. These parameterizations explicitly contain the system impulse response parameters, or Markov parameters, and, under very general noise models, the least squares estimates of the Markov parameters are consistent regardless of model order choice when the input is white noise. A numerical example is given to illustrate this result
Keywords
Markov processes; least squares approximations; parameter estimation; μ-Markov parameterizations; general noise models; least squares estimation; least squares identification; model order choice; system impulse response parameters; white noise; Convergence; Finite impulse response filter; Least squares approximation; Least squares methods; Parameter estimation; Signal generators; Stochastic systems; Transfer functions; White noise;
fLanguage
English
Publisher
ieee
Conference_Titel
Decision and Control, 1998. Proceedings of the 37th IEEE Conference on
Conference_Location
Tampa, FL
ISSN
0191-2216
Print_ISBN
0-7803-4394-8
Type
conf
DOI
10.1109/CDC.1998.760748
Filename
760748
Link To Document