Scholarlink

<p>The emergence of wireless networks will empower the smart factory to achieve the next level of efficiency, connectivity, and flexibility while contributing to the development of a sustainable ecosystem. However, the coexistence of cellular networks and WLAN in factories requires an efficient access service to ensure seamless coverage and provide high reliability for mobile services, such as motion control, automated guided vehicle control, and so on. The achievement of a balance between technological advancements and environmental considerations is essential for the long-term sustainability of smart factories. In this paper, we propose an intelligent multi-criteria access selection algorithm, called MASPC, which is the integration of an improved analytical hierarchy process-entropy weight method and deep reinforcement learning, to tackle the problems regarding access failure, ping-pong effect, and low utilization rate of spectrum resources and different sustainability constraints. Simulations show that the proposed algorithm can accurately make network selection decisions with a comprehensive consideration of network condition, service characteristics, user preferences, and sustainability constraint, and also depict significant performance improvement in terms of minimizing unnecessary handover, radio link failure, and throughput compared to other conventional access services.</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

Energy-aware computing of access service for wireless edge via distributed deep learning

  • 期刊论文

  • 发表日期：

    2025年2月14日

查看全文
  • Xiaoyi Jiang,
  • Fanqin Zhou,
  • Qiyang Zhang,
  • Lei Wei,
  • Yang Yang,
  • Daohua Zhu,
  • Lei Feng,
  • Dayang Wang

摘要

The emergence of wireless networks will empower the smart factory to achieve the next level of efficiency, connectivity, and flexibility while contributing to the development of a sustainable ecosystem. However, the coexistence of cellular networks and WLAN in factories requires an efficient access service to ensure seamless coverage and provide high reliability for mobile services, such as motion control, automated guided vehicle control, and so on. The achievement of a balance between technological advancements and environmental considerations is essential for the long-term sustainability of smart factories. In this paper, we propose an intelligent multi-criteria access selection algorithm, called MASPC, which is the integration of an improved analytical hierarchy process-entropy weight method and deep reinforcement learning, to tackle the problems regarding access failure, ping-pong effect, and low utilization rate of spectrum resources and different sustainability constraints. Simulations show that the proposed algorithm can accurately make network selection decisions with a comprehensive consideration of network condition, service characteristics, user preferences, and sustainability constraint, and also depict significant performance improvement in terms of minimizing unnecessary handover, radio link failure, and throughput compared to other conventional access services.