城市场景下车载自组网路由算法的研究

发布时间：2018-04-06 02:24

本文选题：车载自组织网络　切入点：城市场景　出处：《兰州理工大学》2017年硕士论文

【摘要】：车载自组织网络为解决交通的拥堵问题、提高车辆驾驶的安全性、安全预警等方面提供了新的思路。在车载网络中,路由协议是核心技术。与传统的ad hoc网络相比,其具有很多的独特性。这也使得车载自组网路由协议的研究具有一定的难度。其中,如何考虑车载自组织网络复杂的场景,是设计路由协议时亟需解决的一个问题。本文以城市环境为背景,对车载自组网路由协议和场景模型展开研究,完成了三点研究内容:1.传统的随机路点模型简单易用,但是不能反映较为真实的交通场景,对路由协议的性能评价缺乏一定的说服力。因此,本文以Vanet Mobisim交通流软件中的IDM_IM模型为基础,建立了能够反映真实车辆运动规律的场景模型。通过仿真,分析了该模型的可靠性,为后续路由协议的性能评价提供了网络载体。2.针对GPSR协议采用贪婪算法策略在城市场景中选择转发路径时易出现局部最优化的问题,本文提出一种改进的GPSR协议。该协议提出基于方向和密度辅助转发的路由转发策略。在新协议中,车辆可以获取路段的实时车流密度和其与目的节点的靠近程度,从而在多个路段中选择车流密度大且靠近目的节点的路段。仿真结果表明,改进后的GPSR协议在路由开销、路由延时、数据包转发成功率方面优于GPSR协议。3.针对城市场景中链路质量不稳定、路由转发易受车流密度影响的问题,本文提出了面向城市环境的角度矢量分簇选择协议(VAC-BNR,Vector-angle-cluster and bridge nodes-based routing)。该协议首先将道路划分十字路口区域和十字路口间的直线路段区域。在直线路段区域,依据车辆的移动方向,将车辆划分不同的簇群,再计算每个簇群内节点的效用值,然后依据效用值设置节点转发数据包的优先级;而在十字路口区域,计算车辆的稳定因子,其包含车辆与周围车辆的相对速度、距离信息。并选择最稳定的车辆作为十字路口区域的转发节点。与其他同类协议相比,该协议在低车流密度下的分组转发率明显提高,同时路由开销也得到了较好的控制。
[Abstract]:The vehicle-borne ad hoc network provides a new way to solve the traffic congestion, improve the safety of vehicle driving, and provide a new way of safety warning.In vehicle network, routing protocol is the core technology.Compared with the traditional ad hoc network, it has a lot of uniqueness.This also makes it difficult to study the routing protocol of vehicular ad hoc networks.Among them, how to consider the complex scenario of vehicular ad hoc network is an urgent problem to be solved in the design of routing protocol.In this paper, based on the background of urban environment, the routing protocol and scenario model of vehicular ad hoc network are studied, and three research contents: 1: 1 are completed.The traditional random routing model is simple and easy to use, but it can not reflect the real traffic scene, so it is not persuasive to evaluate the performance of routing protocol.Therefore, based on the IDM_IM model of Vanet Mobisim traffic flow software, a scene model which can reflect the real vehicle motion law is established.Through simulation, the reliability of the model is analyzed, which provides the network carrier. 2. 2 for the performance evaluation of subsequent routing protocols.In order to solve the problem of local optimization when GPSR adopts greedy algorithm strategy to select forwarding paths in urban scenarios, an improved GPSR protocol is proposed in this paper.This protocol proposes a routing forwarding strategy based on directional and density assisted forwarding.In the new protocol, the vehicle can obtain the real-time traffic density and its proximity to the destination node, and then select the section with high traffic density and close to the destination node.Simulation results show that the improved GPSR protocol is better than GPSR protocol in routing overhead, routing delay and packet forwarding success rate.Aiming at the problem of unstable link quality in urban scenarios and traffic density affecting routing and forwarding, this paper proposes an angle vector clustering protocol (VAC-BNRN) for Vector-angle-cluster and bridge nodes-based routing.The agreement first divides the road into the intersection area and the straight section area between the intersections.In the straight section area, according to the moving direction of the vehicle, the vehicle is divided into different clusters, then the utility value of the nodes in each cluster is calculated, and the priority of forwarding packets is set according to the utility value.The stability factor of the vehicle is calculated, which contains the relative velocity and distance information between the vehicle and the surrounding vehicle.The most stable vehicle is chosen as the forwarding node in the intersection area.Compared with other similar protocols, the packet forwarding rate under low traffic density is significantly improved, and the routing overhead is also well controlled.
【学位授予单位】：兰州理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN929.5;U495

【参考文献】