On the Hermite Series Approach to Nonparametric Identification of Hammerstein Systems
| Authors: | Krzyzak Adam, Concordia University, Montreal, Canada
Sasiadek J.Z., Carleton University, Ottawa, Canada
Kegl Balazs, Universite de Montreal, Canada |
|---|
| Topic: | 1.1 Modelling, Identification & Signal Processing |
|---|
| Session: | Nonlinear System Identification II |
|---|
| Keywords: | Hammerstein system identification, Hermite series, convergence |
|---|
Abstract
Nonlinear dynamic systems of Hammerstein type are identified frominput and output measurements. Identification algorithms for amemoryless nonlinear part and for a linear dynamic part areproposed. Convergence and rates of convergence of the algorithmsare investigated. The class of nonlinearities considered in thepaper is very large and cannot be parameterized thereforenonparametric approach is used. The performance of identificationalgorithms is studied in simulation experiments.
