Wiener-Hammerstein Model Identification-Recursive Algorithms.

Abstract

Recursive algorithms for parameter estimation of Wiener-Hammerstein (W-H) models are developed. These algorithms are derived on the basis of minimizing cost functions of the output errors, the equation errors, and the prediction errors. Convergence conditions of the developed algorithms are established. Numerical results, to test the algorithms and illustrate their estimation accuracy, are given.