Scholarlink

Low Earth Orbit (LEO) satellites provide wide coverage and low-latency communication services, serving as an important supplement to ground communication systems. Due to the limited service time of individual LEO satellites, efficient handover strategies between LEO satellites are crucial to maintain continuous and reliable communication. This paper considers service time, channel quality, communication capacity, and satellite network load balancing to construct an optimization model for satellite handover strategies. To address this challenge, an improved deep reinforcement learning algorithm is applied. Simulation experiments demonstrate that this algorithm effectively reduces the number of handovers and handover failures while enhancing the load balancing of the satellite network.

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

Satellite Network Handover Scheme Based on Improved Deep Reinforcement Learning in Smart Grid Systems

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Ben Wang,
  • Xiaojuan Zhu,
  • Jianxin Gao,
  • Qi Xu,
  • Yang Yang

摘要

Low Earth Orbit (LEO) satellites provide wide coverage and low-latency communication services, serving as an important supplement to ground communication systems. Due to the limited service time of individual LEO satellites, efficient handover strategies between LEO satellites are crucial to maintain continuous and reliable communication. This paper considers service time, channel quality, communication capacity, and satellite network load balancing to construct an optimization model for satellite handover strategies. To address this challenge, an improved deep reinforcement learning algorithm is applied. Simulation experiments demonstrate that this algorithm effectively reduces the number of handovers and handover failures while enhancing the load balancing of the satellite network.