Abstract
In this paper, based on the analysis of practical use of telecommunication systems, the necessity of a broad and scientifically proven implementation of statistical methods of their analysis and monitoring on the basis of open flow information is determined.A promising approach to processing of implicit knowledge forms is developed on the basis of the technology of neural structures. The architecture of neural networks allows to implement them using the technology of a high degree of integration. An effectiveness of using neural networks and their analog models is proved to solve the approximation problems of continuous functions of several variables and forecast of the processes that take place in telecommunication networks over the time.The procedures for initial processing parameters of telecommunication network for use as input data to the neural network are proposed. The developed procedures allow a closer consider and analyze the dynamics of information flows circulating in networks and identify the characteristics of random sequences and implementation of neural networks allows to predict the network behavior depending on seasonality and trend.
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call