基于定向扩散的车辆传感器网络连通性研究  

Research on connectivity for vehicular sensor networks based on directed diffusion

作　　者：郑明才[1] 赵小超[2] 赵晋琴[1] 李新国[1] 

机构地区：[1]湖南第一师范学院信息科学与工程系,长沙410205 [2]湖南第一师范学院小教大专部,长沙410002

出　　处：《计算机应用研究》2012年第1期266-269,共4页Application Research of Computers

基　　金：湖南省科技计划资助项目(2009GK3076);湖南省教育厅科学研究项目(08C015);湖南省"十一五"教育规划青年课题(XJK08QXJ001);湖南省教改项目(2011(462))

摘　　要：为改善车辆传感器网络的实时连通性问题,提出一种路边节点梯度值可滑动的定向扩散梯度场DDGF-SRG,并给出了梯度场的实时动态调整规则。通过给路边节点分配固定梯度值,将规模巨大车辆传感器网络划分成若干以路边节点为核心的局部区域,以使梯度场的建立和动态更新可在各局部区域内分布式进行,提高梯度场建立和更新的实时性。通过给车辆节点分配一个可变的数据汇聚梯度值,建立车辆节点到路边节点的定向扩散梯度;通过给路边节点分配一个可变的数据汇聚梯度值,建立路边节点到低梯度层次的定向扩散梯度,并将各局部区域梯度场链接成一个全局的定向扩散梯度场。理论分析和仿真结果表明,DDGF-SRG梯度场在进行动态调整和利用车辆传感器网络的延迟容忍运载能力后,车辆传感器网络的实时连通性得到了极大改善。Aims to the problem of connectivity in vehicular sensor networks,proposed a kind of directed diffusion gradient field with slidable roadside-node gradient（DDGF-SRG）,and gave the dynamically adjusting rules for DDGF-SRG.By giving an invariable gradient value to every roadside node,divided the large scale vehicular sensor network into several local sub reigns around the relevant roadside node,to achieve the purpose of distributed setting up the directed diffusion gradient field for every sub reign and enhancing the real time operation function.By giving a variable data-aggregating gradient value to every vehicular node,formed the directed diffusion gradient from vehicular nodes to the corresponding roadside node.By giving a variable data-aggregating gradient value to every roadside node,set up the directed diffusion gradient from roadside nodes to the corresponding lower-gradient-level sub reign,and linked all the local directed diffusion gradients into one global directed diffusion gradient field,namely the DDGF-SRG.Theoretical analysis and simulation results invalidate that the real time connectivity will be improved greatly with the help of dynamical adjusting and vehicular-nodes＇ carrying capability for data packets in the period of delay tolerant time.

关 键 词：车辆传感器网络 定向扩散 梯度场 动态调整 延迟容忍 连通性 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]