Maximum likelihood forgetting stochastic gradient estimation algorithm for Hammerstein CARARMA systems

This paper considers the identification problem of Hammerstein CARARMA systems, and derives a maximum likelihood stochastic gradient algorithm (ML-SG) by using the maximum likelihood principle and the negative gradient search. Furthermore, a forgetting factor is introduced to improve the convergence rate of the ML-SG algorithm. The simulation results indicate that the proposed algorithm are effective.