Recursive identification for Wiener-Hammerstein system

Abstract

The paper concerns identification of the Wiener-Hammerstein system consisting of a linear subsystem in cascade with a static nonlinearity f(·) followed by another linear subsystem with internal and output noises. Recursive estimates are given for coefficients of both linear subsystems and for the value f(y) at any fixed y. All estimates are proved to converge to the true values with probability one. A numerical example is given justifying the theoretical analysis.