Conjugate gradient method for adaptive nonlinear filtering

Author

Rajan, Sreeraman ; Joo, Kyung Sub ; Bose, Tamal ; Chen, Mei-Qin

Author_Institution

Dept. of Electr. Eng., Colorado Univ., Denver, CO, USA

Volume

2

fYear

1994

fDate

3-5 Aug 1994

Firstpage

1327

Abstract

The application of the Conjugate Gradient (CG) method for the identification of a new class of non-linear discrete time systems, called “Generalized Bilinear System” is investigated. An algorithm based on the CG method is developed for adaptive generalized bilinear filtering with output error method. This algorithm outperforms the LMS and RLS methods in terms of speed of convergence. In this algorithm, the optimization is done over blocks of input and output data rather than a single pair of data. However, only one iteration and coefficient update is done for every sample of data. Some examples are given to illustrate the efficiency of the algorithm

Keywords

adaptive filters; bilinear systems; conjugate gradient methods; digital filters; discrete time filters; filtering theory; iterative methods; nonlinear filters; CG method; adaptive nonlinear filtering; coefficient update; conjugate gradient method; generalized bilinear system; iteration; nonlinear discrete time systems; output error method; Adaptive filters; Application software; Character generation; Computer science; Convergence; Discrete time systems; Equations; Filtering; Filtering algorithms; Finite impulse response filter; Gradient methods; Least squares approximation; Mathematics; Nonlinear systems; Resonance light scattering;

fLanguage

English

Publisher

ieee

Conference_Titel

Circuits and Systems, 1994., Proceedings of the 37th Midwest Symposium on

Conference_Location

Lafayette, LA

Print_ISBN

0-7803-2428-5

Type

conf

DOI

10.1109/MWSCAS.1994.519054

Filename

519054