Minimax estimation of random elements: theory and applications

Author

Siemenikhin, Konstantin V. ; Lebedev, Maxim V.

Author_Institution

Probability Theor. Dept. of Appl. Math. & Phys. Fac., Moscow Aviation Inst., Russia

Volume

4

fYear

2004

fDate

14-17 Dec. 2004

Firstpage

3581

Abstract

The problem of minimax estimation for the infinite-dimensional stochastic model is considered. The prior information about the random elements involved is formulated in term of second-order moment characteristics. The minimax estimation procedure is described and the corresponding numerical algorithm is presented. It is proved that the least favorable distribution of the model random elements is Gaussian. The efficiency of the proposed estimation algorithms is illustrated by means of the examples related to the signal and field robust processing.

Keywords

estimation theory; minimax techniques; multidimensional systems; stochastic systems; Gaussian distribution; infinite-dimensional stochastic model; minimax estimation; numerical algorithm; random elements; second-order moment characteristics; signal processing; Algorithm design and analysis; Hilbert space; Minimax techniques; Random processes; Recursive estimation; Robustness; Signal processing; Stochastic processes; Sufficient conditions; Uncertainty;

fLanguage

English

Publisher

ieee

Conference_Titel

Decision and Control, 2004. CDC. 43rd IEEE Conference on

ISSN

0191-2216

Print_ISBN

0-7803-8682-5

Type

conf

DOI

10.1109/CDC.2004.1429268

Filename

1429268