Identification of Time-Varying Hammerstein-Wiener Systems

This paper focuses on identifying time-varying Hammerstein-Wiener systems. A modified algorithm based on extended Kalman filter is proposed. The algorithm is derived from the comparison between the first and the second order extended Kalman filter algorithms. The modification is based on the first order algorithm by adding a multiplier to its recursive step. It greatly enhances the convergence of the algorithm. The convergence performance of the algorithm is also studied and the conditions for the boundedness of the parameter estimation error are obtained. Simulation experiments of the proposed method and the traditional methods are carried out. Simulation results show the superior performance of the proposed method.