Development of neural network models for the analysis of infocommunication traffic

Abstract

Development of neural network models for the analysis of infocommunication trafficThis article discusses the problems of today's infocommunication networks, the basis of which are multiservice networks serving all types of traffic, presented as a set of IP packets. The characteristic features of this traffic are analyzed, each of which is oriented to a certain class of services. The knowledge gained as a result of ongoing traffic research is an essential factor for increasing the effectiveness of decisions made in various fields of the telecommunications industry. The need for knowledge of the nature of traffic circulating in the network and the laws of its behavior is revealed and substantiated. Without this, it is impossible to effectively manage networks, develop solutions for their development, ensure network security and maintain the required level of quality. Despite the large number of works about building multi-service networks, a number of issues require further study. Analysis of traffic studies of modern converged, multiservice networks showed the lack of knowledge about its nature and laws of behavior, given the high variability of its characteristics. Thus, it can be argued that the parameters of the studied traffic are statistical, probabilistic in nature, can vary randomly over time and, accordingly, based on the study, the author proposes a study using statistical analysis methods. To study traffic, you should use the tools of probability theory and mathematical statistics.

The Concept of Building a Network of Digital Twins to Increase the Efficiency of Complex Telecommunication Systems

“…Standard methods of teletraffic models investigate with queues [48,49]. In conclusion, it is appropriate to formulate the influence of a network of digital twins on the chain "Data-Information-Knowledge-Wisdom" [50].…”