Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
Theory, implementation, and simulated examples of a Maximum-Likelihood Parameter-Estimation algorithm for state-space systems with coloured measurement noise
Thesis (M. Ing.) -- University of Stellenbosch, 1994.
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | English |
| Published: |
Stellenbosch : Stellenbosch University
2012
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867614029549142016 |
|---|---|
| access_status_str | Open Access |
| author | Eglington, Michael Lyle |
| author2 | Milne, G. W. |
| author_browse | Eglington, Michael Lyle Milne, G. W. |
| author_facet | Milne, G. W. Eglington, Michael Lyle |
| author_sort | Eglington, Michael Lyle |
| collection | Thesis |
| dc_rights_str_mv | Stellenbosch University |
| description | Thesis (M. Ing.) -- University of Stellenbosch, 1994. |
| format | Thesis |
| id | oai:scholar.sun.ac.za:10019.1/58600 |
| institution | Stellenbosch University (South Africa) |
| language | English |
| last_indexed | 2026-06-10T12:45:32.686Z |
| license_str | Other — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from SUNScholar — Stellenbosch University Repository |
| publishDate | 2012 |
| publishDateRange | 2012 |
| publishDateSort | 2012 |
| publisher | Stellenbosch : Stellenbosch University |
| publisherStr | Stellenbosch : Stellenbosch University |
| record_format | dspace |
| source_str | SUNScholar — Stellenbosch University Repository |
| spelling | oai:scholar.sun.ac.za:10019.1/58600 Theory, implementation, and simulated examples of a Maximum-Likelihood Parameter-Estimation algorithm for state-space systems with coloured measurement noise Eglington, Michael Lyle Milne, G. W. Stellenbosch University. Faculty of Engineering. Dept. of Electrical and Electronic Engineering. Estimation theory -- Asymptotic theory Parameter estimation Random noise theory System identification Estimation theory Mathematical models Gaussian processes Dissertations -- Electronic engineering Thesis (M. Ing.) -- University of Stellenbosch, 1994. Maine and Iliff's Modified-Maximum-Likelihood-Estimation (MMLE) algorithm estimates the parameters in state-space systems with process and measurement noise. The algorithm assumes white-Gaussian measurement noise, and for the measurement-noise-only case considered here, the MMLE estimator is an exact Maximum-Likelihood (ML) estimator. It is also asymptotically efficient, and therefore the covariance of the MMLE estimates is expected to be asymptotically equal to the Cramer-Rao lower-bound. Practical experience shows that the covariance is equal to the MMLE estimate of the Cramer-Rao bound for sufficiently-long data contaminated by white-Gaussian noise, but not for equally-long data contaminated by coloured noise. This casts doubt on the accuracy of the MMLE Cramer-Rao bound and the asymptotic efficiency of the MMLE estimator. This thesis investigates the effect of coloured measurement noise on the MMLE estimator. The aim is to correct either the MMLE estimator or the MMLE Cramer-Rao bound or both. The effect of coloured measurement noise on Ordinary-Least-Squares and Markov estimators can be analyzed in terms of the correlation between the noise sequence and the sensitivities of the predicted output to changes in the parameters. The analysis is extended to the MMLE estimator, which also minimizes a sum-of-weighted-squared-prediction-errors-type cost function. Masters 2012-08-27T11:39:03Z 2012-08-27T11:39:03Z 1994 Thesis http://hdl.handle.net/10019.1/58600 en Stellenbosch University 282 pages : ill. application/pdf Stellenbosch : Stellenbosch University |
| spellingShingle | Estimation theory -- Asymptotic theory Parameter estimation Random noise theory System identification Estimation theory Mathematical models Gaussian processes Dissertations -- Electronic engineering Eglington, Michael Lyle Theory, implementation, and simulated examples of a Maximum-Likelihood Parameter-Estimation algorithm for state-space systems with coloured measurement noise |
| title | Theory, implementation, and simulated examples of a Maximum-Likelihood Parameter-Estimation algorithm for state-space systems with coloured measurement noise |
| title_full | Theory, implementation, and simulated examples of a Maximum-Likelihood Parameter-Estimation algorithm for state-space systems with coloured measurement noise |
| title_fullStr | Theory, implementation, and simulated examples of a Maximum-Likelihood Parameter-Estimation algorithm for state-space systems with coloured measurement noise |
| title_full_unstemmed | Theory, implementation, and simulated examples of a Maximum-Likelihood Parameter-Estimation algorithm for state-space systems with coloured measurement noise |
| title_short | Theory, implementation, and simulated examples of a Maximum-Likelihood Parameter-Estimation algorithm for state-space systems with coloured measurement noise |
| title_sort | theory implementation and simulated examples of a maximum likelihood parameter estimation algorithm for state space systems with coloured measurement noise |
| topic | Estimation theory -- Asymptotic theory Parameter estimation Random noise theory System identification Estimation theory Mathematical models Gaussian processes Dissertations -- Electronic engineering |
| url | http://hdl.handle.net/10019.1/58600 |
| work_keys_str_mv | AT eglingtonmichaellyle theoryimplementationandsimulatedexamplesofamaximumlikelihoodparameterestimationalgorithmforstatespacesystemswithcolouredmeasurementnoise |