Separable gradient estimation algorithm for Hammerstein systems based on decompositions

This paper studies parameter estimation problem of Hammerstein systems by using the gradient search principle. The Hammerstein system is a bilinear-parameter system which is linear about two parameter vectors, respectively. A separable gradient algorithm is developed for estimating the two parameter vectors based on the hierarchical identification principle. The algorithm is simple in principle and easy to implement online. The simulation results test the effectiveness of the proposed algorithm.