Scholarlink

<p>The integration of AI-driven 5G network optimization is essential for enabling immersive VR/AR applications in mobile learning environments. These applications require low-latency, high-bandwidth connectivity to enhance student engagement and learning outcomes. However, traditional 5G network management struggles with static resource allocation, dynamic network adaptation, and poor Quality of Experience (QoE) in VR/AR scenarios. To address these challenges, we propose an Adaptive Reinforcement Learning-Based Network Slicing (ARL-NS) framework for QoE-aware 5G optimization. The ARL-NS dynamically allocates 5G network slices in real-time using reinforcement learning agents that adjust resources based on continuous QoE feedback from VR/AR applications. Applied in a mobile VR/AR classroom setting, the ARL-NS framework improves latency (18.7 ms), throughput (96.8 Mbps), QoE score (0.91), and resource efficiency (89.6%), thereby ensuring a seamless and high-quality immersive learning experience. Simulation results show significant improvements in latency, user QoE, and resource utilization, supporting the scalable deployment of VR/AR educational applications over 5G networks.</p>

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

AI-driven 5G network optimization for immersive educational VR/AR applications in mobile learning environments

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年11月17日

查看全文
  • Nyu Long,
  • Weiheng Wang

摘要

The integration of AI-driven 5G network optimization is essential for enabling immersive VR/AR applications in mobile learning environments. These applications require low-latency, high-bandwidth connectivity to enhance student engagement and learning outcomes. However, traditional 5G network management struggles with static resource allocation, dynamic network adaptation, and poor Quality of Experience (QoE) in VR/AR scenarios. To address these challenges, we propose an Adaptive Reinforcement Learning-Based Network Slicing (ARL-NS) framework for QoE-aware 5G optimization. The ARL-NS dynamically allocates 5G network slices in real-time using reinforcement learning agents that adjust resources based on continuous QoE feedback from VR/AR applications. Applied in a mobile VR/AR classroom setting, the ARL-NS framework improves latency (18.7 ms), throughput (96.8 Mbps), QoE score (0.91), and resource efficiency (89.6%), thereby ensuring a seamless and high-quality immersive learning experience. Simulation results show significant improvements in latency, user QoE, and resource utilization, supporting the scalable deployment of VR/AR educational applications over 5G networks.