面向IPv6物联子网的轻量级树型转发模型

发布时间：2018-05-05 11:07

本文选题：物联网 + 转发树　；参考：《软件学报》2014年08期

【摘要】：在IPv6物联网中,RPL路由模型已得到广泛的认可.然而对于规模较大的多跳网络结构,RPL面临着部分转发节点路由容量较大的问题.而且物联子网中扁平化的地址结构使得这一问题更为突出.设计了支持IPv6地址自动分配的轻量级树型转发模型TFAD(tree forwarding model with address automatically distributed),将物联子网中的节点构造成一棵层次转发树,树节点的IPv6地址在子树范围内高度聚合.各节点只需存储与其子节点数相当的转发项,即可完成TFAD模型的数据转发.此外,设计了TFAD模型的备份父节点机制,当网络出现故障时能够以子树为单位进行网络拓扑重构,实现物联子网的快速路由恢复.实验验证了TFAD模型的高效路由存储性能以及快速的路由学习能力和故障后路由恢复能力.
[Abstract]:The IPv6 routing model has been widely accepted in the Internet of things. However, for the larger multi-hop network structure, RPL is faced with the problem of large routing capacity of some forwarding nodes. And the flat address structure in the WW subnet makes the problem more prominent. A lightweight tree forwarding model, TFAD(tree forwarding model with address automatically distributed, is designed to support the automatic allocation of IPv6 addresses. The nodes in the WH subnet are constructed into a hierarchical forwarding tree, and the IPv6 addresses of the tree nodes are highly aggregated in the sub-tree range. Each node can complete the data forwarding of the TFAD model only by storing the forwarding items corresponding to the number of its children. In addition, the mechanism of backup parent node of TFAD model is designed. When the network fails, the network topology can be reconstructed by subtree, and the fast route recovery of the WH subnet can be realized. The experimental results show that the TFAD model is efficient in routing storage, fast in routing learning and in post-fault routing recovery.
【作者单位】：北京师范大学信息科学与技术学院;北京师范大学教育学部;首都师范大学信息工程学院;
【基金】：国家自然科学基金(61272446,61373161) 国家科技重大专项(2012ZX03005001-001)
【分类号】：TP391.44;TN929.5

【相似文献】