基于NDN的多层卫星网络分布式动态路由方法
发布时间：2018-08-05 10:30
【摘要】：多层卫星网络的拓扑结构具有明显的时变特征,难以提供持续且稳定的端到端路径,导致高效路由问题面临着巨大的挑战.将卫星网络设计为封闭的体系结构,采用强管理或调度式控制策略会导致其无法自由、高效地实现与其它系统的协同组网.基于IP的网络体系结构在应用于多层卫星网络时又存在路由效率不高的问题.基于命名数据网络NDN(Named Data Networking)体系结构,提出一种多层卫星网络分布式路由方法——SNDN(Satellite Named Data Networking).根据链路切换的周期性和可预知性,以时变图为建模工具,设计一个能自适应稠密和稀疏场景的路由算法来动态地计算时间相关的最快路径并附加为源路由信息,充分利用命名数据网络数据驱动的通信模式、数据包级的安全特性和逐跳转发的特点,由节点在包转发的过程中动态地完成转发表(FIB,Forwarding Information Base)和请求状态表(PIT,Pending Interest Table)的构建.可为面向未来空、天、地一体化协同组网的多层卫星网络路由研究提供一种新的思路.通过仿真实验及对比分析,验证了SNDN在路由效率和多播性能方面的优势.
[Abstract]:The topology of multilayer satellite networks has obvious time-varying characteristics, and it is difficult to provide continuous and stable end-to-end paths, which leads to great challenges for efficient routing problems. The satellite network is designed as a closed system structure, and the strong management or scheduling control strategy will lead to its freedom and efficient implementation of cooperative networking with other systems. IP-based network architecture has the problem of low routing efficiency when it is applied to multi-layer satellite networks. Based on the NDN (Named Data Networking) architecture of named data network, a distributed routing method for multi-layer satellite network is proposed: SNDN (Satellite Named Data Networking). According to the periodicity and predictability of link switching, using time-varying graph as a modeling tool, a routing algorithm is designed to dynamically calculate the time-dependent fastest path and attach it as source routing information. Taking full advantage of the data driven communication mode of named data network, the security characteristics of data packet level and the characteristics of jump forward, the node dynamically completes the construction of forwarding Information Base) and request pending Interest Table) in the process of packet forwarding. It can provide a new way for the research of multi-layer satellite network routing in the future. The advantages of SNDN in routing efficiency and multicast performance are verified by simulation and comparative analysis.
【作者单位】： 武汉大学计算机学院;地球空间信息技术协同创新中心;武汉大学测绘学院;
【基金】：国家自然科学基金(No.61373040,No.61572370)
【分类号】：TN927.2
[Abstract]:The topology of multilayer satellite networks has obvious time-varying characteristics, and it is difficult to provide continuous and stable end-to-end paths, which leads to great challenges for efficient routing problems. The satellite network is designed as a closed system structure, and the strong management or scheduling control strategy will lead to its freedom and efficient implementation of cooperative networking with other systems. IP-based network architecture has the problem of low routing efficiency when it is applied to multi-layer satellite networks. Based on the NDN (Named Data Networking) architecture of named data network, a distributed routing method for multi-layer satellite network is proposed: SNDN (Satellite Named Data Networking). According to the periodicity and predictability of link switching, using time-varying graph as a modeling tool, a routing algorithm is designed to dynamically calculate the time-dependent fastest path and attach it as source routing information. Taking full advantage of the data driven communication mode of named data network, the security characteristics of data packet level and the characteristics of jump forward, the node dynamically completes the construction of forwarding Information Base) and request pending Interest Table) in the process of packet forwarding. It can provide a new way for the research of multi-layer satellite network routing in the future. The advantages of SNDN in routing efficiency and multicast performance are verified by simulation and comparative analysis.
【作者单位】： 武汉大学计算机学院;地球空间信息技术协同创新中心;武汉大学测绘学院;
【基金】：国家自然科学基金(No.61373040,No.61572370)
【分类号】：TN927.2
【相似文献】
相关期刊论文 前10条
1 申敏;魏玄;;分布式IMS架构与关键技术[J];数字通信;2009年06期
2 陶志勇;方宁;;无线传感器网络分布式成簇算法优化[J];计算机系统应用;2012年07期
3 郭篁;王X;于悦;;容迟网络中基于社会感知的协作机制研究[J];电信科学;2012年10期
4 魏R蜆，
本文编号：2165525
本文链接：https://www.wllwen.com/kejilunwen/xinxigongchenglunwen/2165525.html
