Asymptotical behavior of a version of fast filtering algorithms for complex valued process

Author

Hagström, Martin

Author_Institution

Dept. of Guidance & Control, Nat. Defence Res. Establ., Stockholm, Sweden

Volume

3

fYear

1996

fDate

11-13 Dec 1996

Firstpage

2533

Abstract

The purpose of this paper is to systematically study the asymptotic behavior of a version of fast filtering algorithms in discrete time for complex-valued process, and consequently for that of the Riccati equation of one dimension. It is shown that the trajectories generated by these iterations either escape to infinity in finite time, or converge to a fixed point, or are periodic/dense on the real line. Necessary and sufficient conditions for these dynamical behaviors are given in terms of sign definiteness (on the unit circle) of the corresponding dissipation polynomial

Keywords

Hermitian matrices; Kalman filters; Riccati equations; filtering theory; matrix algebra; stochastic systems; Riccati equation; asymptotical behavior; complex valued process; dissipation polynomial; dynamical behaviors; fast filtering algorithms; necessary and sufficient conditions; sign definiteness; Control systems; Filtering algorithms; H infinity control; Integrated circuit noise; Kalman filters; Polynomials; Riccati equations; Stochastic systems; Sufficient conditions; White noise;

fLanguage

English

Publisher

ieee

Conference_Titel

Decision and Control, 1996., Proceedings of the 35th IEEE Conference on

Conference_Location

Kobe

ISSN

0191-2216

Print_ISBN

0-7803-3590-2

Type

conf

DOI

10.1109/CDC.1996.573478

Filename

573478