Title :
LMS Identification of CAR Models
Author :
Lahalle, Elisabeth ; Oksman, Jacques
Author_Institution :
Dept. of Signal Process. & Electron. Syst., Supelec, Gif sur Yvette
Abstract :
In this paper a new CAR LMS identification algorithm is proposed. It uses an implicit numerical integration formula to build an adaptive predictor from the stochastic differential equation of the CAR model. The performances of the proposed method are evaluated for implicit integration formulas of different orders. Several results are shown, including an application to electrical circuits
Keywords :
autoregressive processes; differential equations; integration; least mean squares methods; signal processing; CAR LMS identification algorithm; adaptive predictor; continuous autoregressive; electrical circuit; least mean square; numerical integration formula; stochastic differential equation; Adaptive signal processing; Differential equations; Least squares approximation; Performance evaluation; Predictive models; Signal processing; Signal processing algorithms; Signal sampling; State-space methods; Stochastic processes; CAR modeling; adaptive identification;
Conference_Titel :
Information, Communications and Signal Processing, 2005 Fifth International Conference on
Conference_Location :
Bangkok
Print_ISBN :
0-7803-9283-3
DOI :
10.1109/ICICS.2005.1689072
