车载自组织网络数据通信协同机制的研究

发布时间：2018-11-09 12:22

【摘要】：车载自组织网络(Vehicular Ad Hoc Networks,VANETs),是指通过无线传输介质使车辆之间以及车辆与路旁单元(Roadside Unit,RSU)能够通信的移动自组织网络(Mobile Ad Hoc Networks,MANETs)。与传统的MANETs不同的是,VANETs具有高度的移动性,且各节点的运动受到道路方向、周围车辆以及交通法规的限制。VANETs作为智能交通系统(Intelligent Transport System,ITS)的重要组成部分,一方面可以使各车辆通过该网络下载数据,另一方面,交管中心也可以通过VANETs收集路况及各车辆的相关信息等。现有的有关VANETs下载机制的研究通常把各个车辆的数据下载视为独立的过程,着重于解决由车辆移动性带来的网络连通性、时延及吞吐量等问题,而忽略了各个车辆数据下载的一致性以及下载过程中的相互协作;而有关VANETs数据收集机制的研究主要采用先验式的广播方式,或者借鉴无线传感网中的数据收集机制,这两者要么增加了收集过程中的网络负载,要么无法很好的适应VANETs快速变化的拓扑结构。本文首先研究并设计了基于联盟博弈理论的多跳VANETs合作下载机制(CMV),该机制的目标为最小化多个车辆从RSU下载相同数据时的网络开销。在提出的联盟博弈模型中,各个车辆通过互不相交的联盟下载所需内容。在每个联盟中,需选择一个“种子”车辆从最近的RSU下载数据,并把数据分享给同一个联盟中的成员。每个车辆的收益用一个价值函数表示,该函数与其下载跳数、位置及速度有关。同时,CMV考虑到了各个联盟内部的合作成本,为了计算出最佳的联盟组合方式,本文设计了一种分布式的联盟形成算法。仿真结果表明,相较于非合作模式,本文提出的合作下载机制能有效降低整个过程中的网络资源消耗。本文同时设计了VANETs下的按需数据收集机制(ODDC),该机制旨在控制整个数据收集过程中的时间和带宽消耗并获得较精确的收集结果。ODDC通过反应式的收集方式,建立了一颗生成树。该机制分为三个阶段:广播查询、本地收集、数据聚合。同时,ODDC考虑并利用了车辆移动性、密度及线性的物理拓扑。仿真结果显示,与现有机制比较,ODDC在高速公路场景下取得较高数据收集精度的同时,能有效降低数据收集过程所需时间以及网络开销。
[Abstract]:Vehicle-borne ad hoc network (Vehicular Ad Hoc Networks,VANETs) is a mobile ad hoc network (Mobile Ad Hoc Networks,MANETs) that communicates between vehicles and road side units (Roadside Unit,RSU) through wireless transmission media. Unlike the traditional MANETs, VANETs has a high degree of mobility, and the movement of each node is restricted by the direction of the road, surrounding vehicles and traffic regulations. VANETs is an important part of the Intelligent Transportation system (Intelligent Transport System,ITS). On the one hand, each vehicle can download the data through the network, on the other hand, the traffic management center can collect the road condition and the relevant information of each vehicle through VANETs. The existing research on VANETs download mechanism usually regards the data download of each vehicle as an independent process, focusing on solving the network connectivity, delay and throughput caused by the mobility of the vehicle. The consistency of each vehicle data download and the mutual cooperation in the downloading process are ignored. The research on VANETs data collection mechanism mainly adopts a transcendental broadcasting method, or draws lessons from the wireless sensor network data collection mechanism, which either increases the network load in the process of collection. It can't adapt to the fast changing topology of VANETs. This paper first studies and designs a multi-hop VANETs cooperative downloading mechanism based on alliance game theory (CMV), which aims to minimize the network overhead when multiple vehicles download the same data from RSU. In the proposed alliance game model, each vehicle downloads the required content through a disjoint alliance. In each alliance, select a "seed" vehicle to download data from the nearest RSU and share the data with members of the same alliance. The revenue per vehicle is expressed as a value function related to the number of hops, locations, and speeds. At the same time, considering the cooperation cost of each alliance, CMV designs a distributed coalition formation algorithm in order to calculate the best alliance combination method. The simulation results show that compared with the non-cooperative mode, the cooperative download mechanism proposed in this paper can effectively reduce the network resource consumption in the whole process. This paper also designs the on-demand data collection mechanism (ODDC),) under VANETs. The mechanism is designed to control the time and bandwidth consumption in the whole process of data collection and to obtain more accurate collection results. A spanning tree was built. The mechanism is divided into three stages: broadcast query, local collection and data aggregation. At the same time, ODDC takes into account and utilizes the mobility, density and linear physical topology of the vehicle. The simulation results show that compared with the existing mechanism, ODDC can effectively reduce the time and network overhead of the data collection process while achieving higher data collection accuracy in the freeway scene.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U495;U463.67

【相似文献】