Forecasting the State of Infocommunication Systems Using Time Series with Defined Fractal Properties

Abstract

The article developed a method for predicting the profile of the normal functioning of infocommunication systems formed on the basis of historical values by measuring the fractal dimension of the current increments of the predicted process. Instantaneous amplitudes of increments are used to find optimal combinations of the predicted process by the criterion of the minimum distance between the experimental and predicted functions. The approach is based on a combination of an effective apparatus for analyzing the fractal dimension of time series and modeling.