Scholarlink

<p>The advent of 6G networks has significantly expanded the potential of intelligent education by enabling ultra-high data rates, ultra-low latency, and enhanced reliability, thereby supporting immersive and personalized learning experiences. This study investigates the optimization and Service Quality Assurance (SQA) of 6G network slicing tailored for intelligent educational environments. Network slicing facilitates the creation of customized virtual networks designed to meet diverse Quality of Service (QoS) requirements, including bandwidth, latency, packet loss, and reliability, for applications such as virtual classrooms, AI-driven tutoring, interactive simulations, and VR-based learning. To address dynamic traffic variations and resource allocation challenges, an Efficient Golden Jackal Optimizer-tuned Elman Neural Network (EGJO-ENN) is proposed for real-time traffic prediction and adaptive bandwidth allocation. Multivariate QoS parameters are preprocessed using data cleaning and Z-score normalization to ensure balanced feature contribution and stable model convergence. The EGJO component enhances parameter optimization, while the ENN captures temporal traffic dependencies, enabling intelligent slice customization based on application-specific demands. Experimental evaluation demonstrates superior performance compared to existing approaches, achieving 98.13% accuracy, 7.21% MAPE, and 0.22 RMSE. The proposed framework improves reliability, latency control, bandwidth utilization, and overall resource efficiency, thereby ensuring enhanced SQA across diverse educational services. The results confirm the suitability of the EGJO-ENN framework as a scalable and adaptive solution for future 6G-enabled smart learning ecosystems.</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

Optimization and service quality assurance of 6G network slicing technology in intelligent education

  • 期刊论文

  • 发表日期：

    2026年7月6日

查看全文
  • Haoli Yang,
  • Zhongxia Liu

摘要

The advent of 6G networks has significantly expanded the potential of intelligent education by enabling ultra-high data rates, ultra-low latency, and enhanced reliability, thereby supporting immersive and personalized learning experiences. This study investigates the optimization and Service Quality Assurance (SQA) of 6G network slicing tailored for intelligent educational environments. Network slicing facilitates the creation of customized virtual networks designed to meet diverse Quality of Service (QoS) requirements, including bandwidth, latency, packet loss, and reliability, for applications such as virtual classrooms, AI-driven tutoring, interactive simulations, and VR-based learning. To address dynamic traffic variations and resource allocation challenges, an Efficient Golden Jackal Optimizer-tuned Elman Neural Network (EGJO-ENN) is proposed for real-time traffic prediction and adaptive bandwidth allocation. Multivariate QoS parameters are preprocessed using data cleaning and Z-score normalization to ensure balanced feature contribution and stable model convergence. The EGJO component enhances parameter optimization, while the ENN captures temporal traffic dependencies, enabling intelligent slice customization based on application-specific demands. Experimental evaluation demonstrates superior performance compared to existing approaches, achieving 98.13% accuracy, 7.21% MAPE, and 0.22 RMSE. The proposed framework improves reliability, latency control, bandwidth utilization, and overall resource efficiency, thereby ensuring enhanced SQA across diverse educational services. The results confirm the suitability of the EGJO-ENN framework as a scalable and adaptive solution for future 6G-enabled smart learning ecosystems.