Method of Forecasting the Duration of Data Transmission Routes in Mobile Radio Networks

Abstract

The method of forecasting the time of overloading of data transfer routes in the mobile radio link (MRL), which is constructed using the neural network of Elman based on the calculation of the potential of neurons in the network is presented. The essence of the method is to make a decision on finding new routes based on the predicted time of overloading the data transfer routes in the MRL. This method allows predicting the time of overloading of data transfer routes in the MRL due to the reduction of the computational complexity of the neural network and the application of the algorithm for training the neural network.