Feras M. Recurrent parameter estimation for nonlinear Hammershtain model

Українська версія

Thesis for the degree of Candidate of Sciences (CSc)

State registration number

0403U000518

Applicant for

Specialization

  • 05.13.03 - Системи та процеси керування

21-01-2003

Specialized Academic Board

Д 64.052.02

Kharkiv National University Of Radio Electronics

Essay

The thesis is devoted to the development and studying of recurrent identification algorithms of nonlinear Hammershtain model in the presence of noises and uncertainties in the parameter drift. Methods of research: the theory of adaptive systems, theory of an optimality, theory of probability and mathematical statistics, theory of stability, theory of matrixes and methods of calculus mathematics, imitating modeling.Parameter estimation of transfer function models in the case of colonred process noise is studied. A new bias correction based method is proposed with a view to attaining estimation irrespective of noise dynamics. The special structure of the new parameter vector enables calculation of the coloured-noise-induced bias, which can eventually result in unbiased parameter estimates via the bias correction scheme. A new factorization procedure for identification of nonstationary Hammershtain model is proposed

Files

Similar theses