Identification Methods for Two-Variable Difference Systems

Abstract

This paper proposes two identification methods for two-variable difference systems. The output and the input of the two-variable difference systems depend not only on time but also on spatial coordinates. A recursive least squares algorithm and a stochastic gradient algorithm are introduced to estimate the unknown parameters. Furthermore, in order to increase the convergence rate and to decrease the computational effort, a forgetting factor stochastic gradient algorithm is proposed. The simulation results indicate that the proposed methods are effective.