节点重要程度感知的分布式DTN结构检测机制

摘要

As nodes show strong social attributes in the intermittent connection wireless network,network structure detection can effectively improve the network performance. This paper presented a distributed network structure detection mechanism based on node importance in intermittent connection wireless network. First,nodes estimated their own importance degree based on the number of neighbors and forwarded messages. and selected the community center node in a distributed way. Then, ac-cording to the number of common neighbors nodes determined their own community center node. Eventually learned the entire network structure. The simulation results show that the proposed detection algorithm has higher accuracy than HCDA about 45%. Furthermore,this method with strong adaptability is suitable for a variety of mixed network scenarios.%间断连接无线网络具有较强的社会属性,感知网络结构能够有效改善网络性能.提出一种节点重要程度感知的网络结构检测机制,节点根据转发消息数量和邻居数量估计自身的重要程度,并以分布式的方式选取社区中心节点,进而依据与中心节点的共同邻居数确定本社区的邻居节点,完成网络结构检测.仿真结果表明,本机制检测准确率相较于HCDA提高大约45％,且所提出的方法扩展性较强,适用于各种混杂网络场景.