A performance study of the extended Kalman algorithm as a HF channel estimator

Author

McLaughlin, S. ; Mulgrew, B. ; Cowan, C.F.N.

Author_Institution

Edinburgh Univ., UK

fYear

1988

fDate

11-14 April 1988

Firstpage

335

Lastpage

338

Abstract

A study is made of the tracking and convergence properties of an extended Kalman (EK) algorithm as a high frequency (HF) channel estimator. This study uses the EK algorithm to adaptively estimate not only the time varying taps of the system but the parameters in the system model which generate the nonstationarity. This approach makes use of the a-priori knowledge of the HF channel to construct the extended Kalman algorithm for this system. Simulations are presented of the performance of the EK algorithm as a HF channel estimator for both white and coloured input signal conditions and are compared with the performance of the least mean square (LMS) and minimum variance Kalman (MVK) algorithms.<>

Keywords

Kalman filters; filtering and prediction theory; radio systems; telecommunication channels; HF channel estimator; LMS algorithm; coloured input signal; convergence; extended Kalman algorithm; minimum variance Kalman algorithm; nonstationarity; performance study; radio systems; simulations; system model parameters; time varying taps; tracking; Communication channels; Filtering; Kalman filtering; Prediction methods;

fLanguage

English

Publisher

iet

Conference_Titel

HF Radio Systems and Techniques, 1988., Fourth International Conference on

Conference_Location

London, UK

Print_ISBN

0-85296-358-0

Type

conf

Filename

7049