Recursive identification for Wiener-Hammerstein systems using instrumental variable

Abstract

An identification method is discussed that deals with the Wiener-Hammerstein systems of general nonlinearity. By introducing a suitable instrumental variable a new algorithm is presented to recursively estimate the linear subsystems using stochastic approximation algorithm. The kernel nonparametric method is used to estimate the nonlinear function. The consistent analysis of the method is given under mild condition. A simulation example is provided justifying the proposed method.