3rd cumulant based MA parameter identification

Author

Xu, You ; Crebbin, Greg

Author_Institution

Dept. of Electr. & Electron. Eng., Western Australia Univ., Nedlands, WA, Australia

Volume

1

fYear

1996

fDate

14-18 Oct 1996

Firstpage

553

Abstract

In this paper, a novel optimization method, based on the conjugate gradient technique, is proposed for moving average (MA) system identification. The independent, identically distributed non-Gaussian input and Gaussian measurement noise are assumed. Experiments were performed to compare the proposed methods with some of current linear methods. The simulation results show that the proposed optimization method gives better performance with smaller bias and smaller standard deviation

Keywords

Gaussian noise; conjugate gradient methods; higher order statistics; moving average processes; optimisation; parameter estimation; signal processing; 3rd cumulant; Gaussian measurement noise; MA parameter identification; conjugate gradient technique; i.i.d. input; independent identically distributed non-Gaussian input; moving average system identification; optimization method; simulation results; Autocorrelation; Closed-form solution; Equations; Gaussian noise; Higher order statistics; Large scale integration; Noise measurement; Optimization methods; Parameter estimation; System identification;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing, 1996., 3rd International Conference on

Conference_Location

Beijing

Print_ISBN

0-7803-2912-0

Type

conf

DOI

10.1109/ICSIGP.1996.567324

Filename

567324