Wiener system identification using polynomial non linear state space model

This paper deals with identification of Wiener nonlinear systems. Such systems, consist of a linear dynamic block followed by a static non-linear subsystem. In this work, Polynomial Non Linear State Space(PNLSS) models are used to describe them. An output error identification method is performed, based on Levenberg-Marquardt algorithm; the parameters sensitivity functions are developed as a multivariable state space model. The method efficiency is investigated on numerical simulations in absence of noise and with noisy data for different signal to noise ratios.