Identification of a time-varying object using customizable models

A method of identifying a time-varying object by setting the model parameters on the basis of previous measurements is presented. Measurements take place on background of interference in the form of Gaussian white noise. The paper describes an algorithm for constructing a linear filter based on the Kalman filter and presents the simulation scheme in MATLAB Simulink format. The results of identification using different variants of filter settings are shown. The presented identification technique is promising one for the construction of systems based on neural networks.