WSEAS Transactions on Systems
Print ISSN: 1109-2777, E-ISSN: 2224-2678
Volume 19, 2020
Robust Recursive Least-Squares Finite Impulse Response Predictor in Linear Discrete-Time Stochastic Systems with Uncertain Parameters
Author:
Abstract: This paper newly proposes the robust RLS Wiener FIR prediction algorithm based on the innovation theory for the linear stochastic systems including with parameters. In the robust RLS Wiener predictor, the following information is used. (1) The system matrices for the signal and the degraded signal. (2) The observation matrices for the signal and the degraded signal. (3) The variance of the state for the degraded signal. (4) The cross-variance of the state for the signal with the state. (5) The variance of the observation noise. As a step to obtain the robust RLS Wiener FIR prediction algorithm, this paper presents the robust prediction algorithm of the signal using the covariance information etc. In the predictor, the following information is used. (1) The observation matrices for the signal and the degraded signal. (2) The variance of the state for the degraded signal. (3) The auto-covariance information of the state for the degraded signal. (4) The cross-covariance information of the state for the signal with that for the degraded signal. (5) The variance of the observation noise. The estimation accuracy of the proposed robust RLS Wiener FIR predictor is superior to the existing RLS Wiener FIR predictor.
Search Articles
Keywords: Robust RLS Wiener FIR predictor, discrete-time stochastic systems, auto-covariance function, innovation approach, uncertain parameters
Pages: 86-101
DOI: 10.37394/23202.2020.19.13