Least squares predictive transform modeling

Author

Guerci, Joseph R. ; Feria, Erlan H.

Author_Institution

Grumman Aircraft Syst., Bethpage, NY, USA

fYear

1988

fDate

23-27 May 1988

Firstpage

47

Lastpage

51

Abstract

A least-squares error (LSE) linear predictive transform (LPT) method is offered to obtain adaptive time-varying signal models in filtering applications. The method is a direct extension of minimum mean-square error (MSE) LPT modeling to the adaptive case. The LSE LPT method is illustrated with a digital monochrome image filtering example which shows that the resulting time-varying signal model behaves, approximately, as a time-varying whitening filter of the Kalman type

Keywords

Kalman filters; filtering and prediction theory; information theory; least squares approximations; picture processing; Kalman type; LSE LPT method; adaptive time-varying signal models; digital monochrome image filtering; extension of minimum mean-square error; filtering applications; least-squares error; linear predictive transform; predictive transform modeling; time-varying signal model; time-varying whitening filter; Adaptive filters; Covariance matrix; Decoding; Educational institutions; Equations; Filtering; Least squares methods; Predictive models; Statistics; Technological innovation;

fLanguage

English

Publisher

ieee

Conference_Titel

Aerospace and Electronics Conference, 1988. NAECON 1988., Proceedings of the IEEE 1988 National

Conference_Location

Dayton, OH

Type

conf

DOI

10.1109/NAECON.1988.194993

Filename

194993