Multi-innovation parameter estimation for Hammerstein MIMO output-error systems based on the key-term separation∗

This paper uses the key-term separation principle and develops a multi-innovation stochastic gradient algorithm for Hammerstein MIMO output error systems. The basic idea is to decompose a Hammerstein MIMO system into two subsystems and to identify the parameters of each subsystem interactively. The parameter estimation accuracy can be improved with the innovation length increasing. The simulation example verifies the effectiveness of the proposed algorithm.