DocumentCode
816295
Title
A recursive (on-line) maximum likelihood identification method
Author
Gertler, Janos ; Banyasz, C.
Author_Institution
Institute for Computing and Automation, Hungarian Academy of Sciences, Budapest, Hungary
Volume
19
Issue
6
fYear
1974
fDate
12/1/1974 12:00:00 AM
Firstpage
816
Lastpage
820
Abstract
A recursive (on-line) identification algorithm is developed based upon the off-line maximum likelihood method by Åström and Bohlin. The basic idea of the algorithm consists in two modifications to the classical method. First an approximate noisemodel is applied to eliminate auto-regressive filtering in the computation of the noise-derivatives. Second, some approximations are introduced to make the direct recursive version of the iteration equations really on-line. The combination of the two modifications yields a compact on-line algorithm.
Keywords
Linear systems, time-invariant discrete-time; Parameter identification; Recursive estimation; maximum-likelihood (ML) estimation; Equations; Filtering; Helium; Maximum likelihood estimation; Measurement standards; Noise measurement; Parameter estimation; Polynomials; Read only memory; Transfer functions;
fLanguage
English
Journal_Title
Automatic Control, IEEE Transactions on
Publisher
ieee
ISSN
0018-9286
Type
jour
DOI
10.1109/TAC.1974.1100722
Filename
1100722
Link To Document