On the Use of the Laguerre Model for the Identification of Discrete-Time Systems

Abstract

A useful method for data compression in a parameter estimation scheme of linear time-invariant discrete systems is presented. The main feature of the method is that the given data sequences are initially transformed into information bearing sequences of considerably lower order through the discrete Laguerre expansion. By the use of the Laguerre difference equation model all subsequent computations are implemented in the Laguerre domain resulting thereby in a considerable reduction in computational effort. A simple rule to select the discount factor of Laguerre expansion is presented. The identification procedure is applied to simulated and real data.