METHOD OF FORECASTING THE CONDITION OF RADIO-ELECTRONIC SITUATION OF MULTIPLE SYSTEMS IN THE CONDITIONS OF UNCERTAINTY

Abstract

The task of forecasting future values of the time series based on its previous values is the basis for planning in the economy, trade, energy and technical fields. Forecasting of a radio-electronic situation in the conditions of shortage of radio-frequency resource is a very important component of modern high-tech military conflicts, the transport basis of which are multi-antenna radio-emitting means. For this purpose, an analysis of the known methods of forecasting the radio-electronic situation is carried out in that article. It is established that nowadays there are many models of time series prediction, namely: regression and autoregressive models, neural network models, exponential smoothing models, Markov-based models, classification models, etc. Based on the above analysis, it is found that the most appropriate for use in the prediction problems of the electronic environment of multi-antenna communication systems are time series prediction methods, which are based on autoregressive models. The article proposes a technique for predicting the condition of the radio-electronic environment, which allows to increase the noise immunity of communication systems in the conditions of deliberate interference and the unsteady nature of the predicted process, in order to ensure electromagnetic compatibility and increase the efficiency of use of radio frequency resource by complexes. To solve the scientific problem, we use the general scientific methods of analysis and synthesis of complex technical systems, the theory of noise immunity of radio engineering systems and methods of mathematical modeling. It is advisable to use this technique while assessing the electronic environment and identifying measures to enhance the security of communications systems. The calculations show that the use of this method allows to reduce the error of the forecast by an average of 20%. It is advisable to practically implement the proposed methodology while developing software for programmable radio stations.