LEO巨型卫星网络端到端路由机制研究OA
Research on end-to-end routing mechanism for LEO mega-constellation networks
低轨巨型星座构成的卫星通信网络需要解决地面终端之间卫星直连方式的端到端路由问题,为此,首先分析了低轨巨型星座的特点,提出一种基于天地融合网络的移动和切换管理方案,同时提出一种基于位置导向的智能巨型卫星网络路由算法LoISR,解决了天地融合网络端到端的全路径通信问题.所提算法对目标卫星网络进行仿真并采集基准路由数据,基于该数据训练得到机器学习模型,再对模型进行裁剪和压缩,部署到卫星节点,并通过相关实验验证了所提算法的可行性和效率.
Satellite communication networks composed of low earth orbit mega-constellation network needs to solve the end-to-end rout-ing problem between ground terminals through direct satellite connections.To this end,the characteristics of low-orbit mega-constellations are first analyzed,and a mobility and handover management mechanism based on space-ground integrated network is pro-posed.At the same time,a location-oriented intelligent giant satellite network routing algorithm LoISR is proposed to solve the end-to-end full-path communication problem in space-ground integrated network.The proposed algorithm collects benchmark routing data through simulation of the target satellite network,a machine learning model is trained based on this data,then trims and compresses the model and deploys it to satellite nodes,the feasibility and efficiency of the proposed algorithm are verified through relevant experiments.
李克军;王羽;赵思琳;陈星池;文少杰;王野;李清
鹏城实验室,广东 深圳 518055鹏城实验室,广东 深圳 518055上海卫星互联网研究院有限公司,上海 201210||上海市卫星互联网重点实验室,上海 201210||卫星互联网全国重点实验室,上海 201210鹏城实验室,广东 深圳 518055上海卫星互联网研究院有限公司,上海 201210||上海市卫星互联网重点实验室,上海 201210||卫星互联网全国重点实验室,上海 201210鹏城实验室,广东 深圳 518055鹏城实验室,广东 深圳 518055
信息技术与安全科学
卫星网络天地融合网络路由算法机器学习
satellite networkspace-ground integrated networkrouting algorithmmachine learning
《天地一体化信息网络》 2026 (2)
92-107,16
评论