Research classification traffic flows based on machine learning to ensure quality of service in radio systems and mobile cellular network

Methods for the efficiency classifying traffic packets based on the technology distributed communication networks such as Software Defined Networking, Internet Protocol Multimedia Subsystem, and Machine Learning are analyzed in order to ensure a certain level quality of service (QoS) and quality of experience (QoE) and reliability of stream transmission packets with an increase in the volume of transmitted useful and service traffic in radio systems and mobile cellular networks.This paper examines the problem researching methods for classifying flows of useful and service traffic packets using machine learning.To create a feature matrix for real-time traffic classification in order to maintain QoS, QoE and the reliability of the transmission packet flows, a new approach is proposed, which is based on calculating the statistical characteristics of packet flows.