Hammerstein Model Identification Using Radial Basis Functions Neural Networks

A new method for the identification of the nonlinear Hammerstein Model consisting a static nonlinearity in cascade with a linear dynamic part, is introduced. The static nonlinearity is modeled by radial basis function neural networks (RBFNN) and the linear part is modeled by an autoregressive moving average (ARMA) model. A recursive algorithm is developed to update the weights of the RBFNN and the parameters of the ARMA model.