On the use of L/sub 2/ gap metric for identifying Hammersteim models

Abstract

We consider the identification of nonlinear systems of the Hammerstein class. Hammerstein models are special in the sense that they can be transformed into linear models on which linear systems techniques could be applicable. We propose an iterative identification algorithm to obtain a model that minimizes the L/sub 2/ gap between the true and the identified model. An example is given in illustration.