Finite model order accuracy analysis in Wiener system identification

To two class unknown parameter vectors from Wiener system which were combined in nonlinear function, we can use prediction error method to identify these two class unknown parameter vectors and furthermore determine the asymptotic covariance matrix expressions of the system object model and noise model between Wiener system. In generalized case we construct a optimization problem which include the input power density based on this paper´s asymptotic covariance matrix expressions. By solve this optimization problem with some constrained condition, the optimal input signal spectrum about Wiener system is obtained. Finally, the efficiency and possibility of the proposed strategy can be confirmed by the simulation example results.