Decomposition based recursive least squares parameter estimation for Hammerstein nonlinear controlled autoregressive systems

A decomposition recursive least squares algorithm is proposed for the identification of a Hammerstein nonlinear controlled autoregressive system which is a memoryless nonlinear block followed by a linear ARX subsystem (H-CAR system for short). Using the decomposition based hierarchical identification principle, this paper decomposes the H-CAR system into several subsystems and then identifies each subsystem and finally separates the parameters of the original system from obtained parameter estimates. The proposed algorithm requires less computation burden compared with the recursive least squares algorithm. A simulation example is provided.