Some observations about centralized linear prediction

Author

Therrien, Charles W.

Author_Institution

Dept. of Electr. & Comput. Eng., Naval Postgraduate Sch., Monterey, CA, USA

Volume

1

fYear

1999

fDate

24-27 Oct. 1999

Firstpage

466

Abstract

A new formula for the coefficients of the prediction error filter for noncausal symmetric (centralized) linear prediction is presented. It is shown that when the signal is AR, the centralized filter reduces to a scaled product of the optimal forward and backward prediction error filters for the process. The result appears to be unique for linear prediction. For example, the symmetric noncausal Wiener filter for estimating a signal in noise has no such realization in terms of optimal causal filters.

Keywords

autoregressive processes; circuit feedback; error analysis; feedforward; filtering theory; noise; prediction theory; AR signal; centralized filter; centralized linear prediction; filter coefficients; noise; noncausal symmetric linear prediction; optimal backward prediction error filter; optimal causal filters; optimal forward prediction error filter; scaled product; symmetric noncausal Wiener filter; Computer errors; Equations; Filtering; Nonlinear filters; Predictive models; Signal processing; Wiener filter;

fLanguage

English

Publisher

ieee

Conference_Titel

Signals, Systems, and Computers, 1999. Conference Record of the Thirty-Third Asilomar Conference on

Conference_Location

Pacific Grove, CA, USA

ISSN

1058-6393

Print_ISBN

0-7803-5700-0

Type

conf

DOI

10.1109/ACSSC.1999.832373

Filename

832373