Machine Learning Based 5G Network Slicing Management and Classification
Zong-Xun Wu, Yun-Zhe You, Chien-Chang Liu, Li‐Der Chou
Abstract
Due to the rapid development of the Internet, network bandwidth and stability are becoming more and more important. With the increase in the number of users, how to make each user have a high Quality of Service (QoS) is an urgent problem to be solved. 5G slicing allows flexible management of each user's network usage, which in turn optimizes the overall network usage and reduces the consumption of network resources. The 5G slicing can flexibly manage each user's network usage to optimize overall network usage and reduce network resource consumption. In this paper, use machine learning to analyze the network traffic, and analyze a total of 141 different applications on the network, and conduct experiments on different machine learning models. Based on the above experimental results, propose an algorithm for 5G slice management. Based on the above traffic analysis results, we will dynamically configure and optimize the resources of each slice according to the current network traffic of each user.