Scholarlink

5G technology has ushered in explosive development since the release of the relevant standards, and various industries have seized the opportunity to carry out digital transformation, the railway industry has also grasped the cutting-edge technology of communication, proposed the construction of railway-specific 5G system (5G for Railway, 5G-R). In order to fully utilize the best performance of 5G-R IoT and reduce its construction and maintenance costs, it is of great significance to carry out the research on 5G-R IoT service network slicing. This paper designs a 5G-R core network slicing model based on SDN (Software Defined Network) and NFV (Network Functions Virtualization) technologies, optimizes the model based on DQN (Deep Q Network) algorithm, and proposes a slicing algorithm for IoT services on the 5G-R core network side. Simulation results show that the algorithm proposed in this paper has feasibility and the average resource utilization is 28.4% higher than the average resource utilization of the comparison algorithm.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Research on IoT Service Slicing Algorithm Based on Deep Q Network for Railway 5G Core Network Side

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Yifan Zhao,
  • Jinbai Zou,
  • Chuanbo Hu

摘要

5G technology has ushered in explosive development since the release of the relevant standards, and various industries have seized the opportunity to carry out digital transformation, the railway industry has also grasped the cutting-edge technology of communication, proposed the construction of railway-specific 5G system (5G for Railway, 5G-R). In order to fully utilize the best performance of 5G-R IoT and reduce its construction and maintenance costs, it is of great significance to carry out the research on 5G-R IoT service network slicing. This paper designs a 5G-R core network slicing model based on SDN (Software Defined Network) and NFV (Network Functions Virtualization) technologies, optimizes the model based on DQN (Deep Q Network) algorithm, and proposes a slicing algorithm for IoT services on the 5G-R core network side. Simulation results show that the algorithm proposed in this paper has feasibility and the average resource utilization is 28.4% higher than the average resource utilization of the comparison algorithm.