车载自组织网络GPSR路由算法的研究

发布时间：2018-03-24 22:11

本文选题：车载自组织网络　切入点：路由协议　出处：《吉林大学》2017年硕士论文

【摘要】：车载自组织网络是智能交通系统的重要组成部分,而路由协议作为车载自组织网络的关键技术,很大程度上决定了车载网络的通信性能。在车载自组织网络中,车辆节点快速的移动,使得车载网络的拓扑结构频繁变化,车辆节点之间的邻居关系很不稳定,因此车载自组织网络在进行通信时对路由协议提出更高的要求。在此基础上,本文对车载自组织网络中的GPSR路由算法进行了改进。首先,本文对车载自组织网络和几类具有代表性的AODV、DREAM、GPS R路由协议进行介绍,研究了随机移动模型、平滑移动模型、城市场景移动模型、曼哈顿移动模型,详细地介绍了能够比较真实反映实际交通场景的Vanet Mobi Si m交通仿真器,并生成了可以直接应用在NS2中的交通场景文件。对比较适合车载自组织网络的GPSR路由协议的贪婪转发算法和周边转发算法进行了详细的分析,并利用NS2将GPSR路由协议和AODV协议进行了仿真对比研究。之后,本文分析了车载自组织网络中GPSR路由算法存在的缺陷,针对路由算法在进行数据转发时存在的邻居关系不稳定和通信路径冗余现象,提出了改进措施。在路由算法进行贪婪转发时,加入了邻居节点的累计通信时长参数,算法在选择下一跳转发节点时,在给定的距离范围内,挑选累计通信时长最大的邻居节点,这样可以使选择的下一跳节点即有距离优势又有稳定性;进行周边转发时,考虑邻居节点之间的位置关系,通过计算邻居节点与目的节点之间的角度大小,选择与目的节点位置关系最合适的邻居节点作为下一跳转发节点,这样规划出的路由路径更加合理。通过TCL语言和C++程序语言实现了改进的GPSR路由算法,将改进的GPSR路由协议嵌入到NS2中,结合Vanet Mobi Sim交通仿真器对改进算法进行了仿真实验。通过实验结果分析,改进的GPSR路由算法相对于原算法在丢包率、端到端平均时延和吞吐量性能指标上均有所提高,更加适合车载自组织网络。最后,指出了本文改进的GPSR路由算法存在的不足和下一步的工作计划,并展望了未来的车载自组织网络路由协议的发展方向。
[Abstract]:The vehicular ad hoc network is an important part of the intelligent transportation system. As the key technology of the vehicle ad hoc network, the routing protocol determines the communication performance of the vehicle network to a great extent. The rapid movement of vehicle nodes causes frequent changes in the topology of vehicular networks, and the neighbor relationship between vehicle nodes is unstable. Therefore, the vehicle-borne ad hoc networks require higher routing protocols when communicating. In this paper, the GPSR routing algorithm in vehicular ad hoc networks is improved. Firstly, this paper introduces the vehicle Ad hoc networks and several representative AODV ad hoc GPSR R routing protocols, and studies the random moving model and smooth moving model. The Vanet Mobi Si m traffic simulator, which can truly reflect the actual traffic scene, is introduced in detail in the urban scene movement model and the Manhattan mobile model. The traffic scene file which can be directly applied in NS2 is generated. The greedy forwarding algorithm and peripheral forwarding algorithm which are more suitable for GPSR routing protocol in vehicle-borne ad hoc networks are analyzed in detail. The GPSR routing protocol and the AODV protocol are compared with each other by NS2. After that, the defects of the GPSR routing algorithm in the vehicle Ad Hoc Network are analyzed in this paper. In view of the instability of neighbor relationship and the redundancy of communication path in routing algorithm, an improved method is put forward, in which the cumulative communication time parameter of neighbor node is added to the greedy forwarding algorithm. When the next hop forwarding node is selected, the neighbor node with the largest cumulative communication time can be selected within a given range of distance, which can make the next hop node have the advantage of distance and stability. Considering the location relationship between neighbor nodes, by calculating the angle between neighbor node and destination node, the neighbor node which is the most suitable location relationship with destination node is selected as the next hop forwarding node. By using TCL language and C program language, the improved GPSR routing algorithm is implemented, and the improved GPSR routing protocol is embedded in NS2. Vanet Mobi Sim traffic simulator is used to simulate the improved algorithm. The experimental results show that the improved GPSR routing algorithm improves packet loss rate, end-to-end average delay and throughput performance compared with the original algorithm. Finally, the shortcomings of the improved GPSR routing algorithm and the next work plan are pointed out, and the future development direction of the routing protocol for the vehicular ad hoc network is prospected.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN929.5;U495

【相似文献】