Optimization of Network Routing Using the Markov Decision Process and Hamiltonian Cycle

Abstract

An infocommunication network is a set of interconnected nodes, the main purpose of which is to transmit information over a certain period of time. Due to the fact that modern networks can include many elements and connections between them for efficient operation, it is necessary to react quickly and process the transmitted data. One of the tools for this is data routing on the network. At the same time, given the nature of the infocommunication network, it can be represented as a queuing system. A random process is characteristic of a queuing system. The study of a random process occurring in a system, expressing it mathematically, is the subject of queuing theory. In such a system, there is always a flow of applications coming into the system from users. The system, in turn, processes applications and returns the result to the user, respectively, it can be concluded that there are both incoming and outgoing flows of applications. This publication is devoted to the optimization of network routing using the Markov decision process and the Hamiltonian cycle.