Weighted Parameter Estimation for Hammerstein Nonlinear ARX Systems

This paper proposes parameter estimation algorithms for Hammerstein nonlinear ARX systems. By making full use of the current and previous input-output data of the system, a weighted multi-innovation stochastic gradient algorithm is presented to improve the convergence rate of identification. The innovation term in the traditional identification algorithms can be treated as a particle in the particle-filtering technique, and the weight of each innovation then can be computed according to their importance. The simulation results indicate that the algorithm can improve the accuracy of parameter estimation.