Abstract:
An identification algorithm is proposed nonlinear dynamical systems of fractional order of the Hammerstein class. Designed algorithm allows you to get strongly consistent parameter estimates in the presence of observation noise in the form of fractional white noise. The results of numerical experiments showed high efficiency of the proposed identification algorithm in comparison with the least squares method (LS).
Keywords:fractional order difference, fractional white noise, least squares method (LS), the system of a class of Hammerstein, output error.