Scholarlink

Network slicing has emerged as a transformative technology for 5G networks, enabling the creation of tailored logical and virtualized networks to meet the diverse needs of modern applications. Among these, Ultra-Reliable Low-Latency Communications (URLLC) stands out as a critical use case, demanding high reliability and minimal latency. This work introduces a novel intelligent resource allocation model designed specifically for optimizing Virtual Network Functions (VNFs) within the 5G core network, leveraging deep reinforcement learning. By addressing stringent URLLC requirements, our approach ensures efficient resource management while maintaining service quality. A comparative analysis highlights its superiority over existing algorithms with an improved reliability to 94% and a reduced latency to 0.74 ms.

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

Deep Reinforcement Learning Based Resource Allocation in Ultra Reliable 5G Core Networks

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Amel Khamoum,
  • Nacer Hamani,
  • Hakim Amrouche

摘要

Network slicing has emerged as a transformative technology for 5G networks, enabling the creation of tailored logical and virtualized networks to meet the diverse needs of modern applications. Among these, Ultra-Reliable Low-Latency Communications (URLLC) stands out as a critical use case, demanding high reliability and minimal latency. This work introduces a novel intelligent resource allocation model designed specifically for optimizing Virtual Network Functions (VNFs) within the 5G core network, leveraging deep reinforcement learning. By addressing stringent URLLC requirements, our approach ensures efficient resource management while maintaining service quality. A comparative analysis highlights its superiority over existing algorithms with an improved reliability to 94% and a reduced latency to 0.74 ms.