Fast estimation of Wiener kernels of nonlinear systems in thefrequency domain

摘要

A method for identification of discrete nonlinear systems in termsof the Volterra-Wiener series is presented. It is shown that use of aspecial composite-frequency input signal as an approximation to Gaussiannoise provides the computational efficiency of this method especiallyfor high order kernels. Orthogonal functionals and consistent estimatesfor Wiener kernels in the frequency domain are derived for this class ofnoise input. The basis of the proposed computational procedure forpractical identification is the fast Fourier transform (FFT) algorithmwhich is used both for generation of actions and for analysis of systemreactions