车载自组织网络中基于多跳的消息分发算法研究

发布时间：2018-04-08 18:42

本文选题：车载自组织网络　切入点：多跳　出处：《南京邮电大学》2014年硕士论文

【摘要】：近几年来，我国车辆数量的快速增长给城市交通带来了很多问题，这使得车载自组织网络技术越来越受到人们的关注。通过车辆间自由的组网通信，车载自组织网络能够实现交通安全信息、道路状况信息和娱乐消息等的实时高效传输，解决行车安全、道路拥堵等问题。然而，由于车载自组织网络独特的网络特性以及各种各样的应用所需的性能指标不同，使得设计出适用于所有场景的消息分发算法变得十分困难。本文主要研究车载自组织网络中不同应用场景下基于多跳的消息分发算法，主要研究工作包括以下内容：首先，针对车载自组织网络中与交通安全、交通状况相关的应用场景，本文对基于邻居信息的广播机制进行改进，提出一种多跳广播抑制算法NIMBS，利用车辆的邻居信息同时结合车辆的位置、行驶速度和方向形成邻居节点的有序序列，并最终确定接收车辆的转发等待时间。仿真结果表明该算法相比于类似抑制算法具有更低的数据包丢失率和时延，，有效地解决了广播算法存在的可扩展性、数据传输冗余以及广播风暴等问题。其次，在道路处于自由连通状态的前提下，研究了基于车辆速度的多车道连通概率分析模型。该模型将实际的多车道场景转换成每车道以相同速度行驶的虚拟多车道场景，并根据前人的分析结果推导出确切的道路连通概率计算表达式。经过大量的仿真实验，结果表明基于车辆速度的多车道连通概率分析模型与实际情况的趋势保持一致。最后，针对单播等端到端的消息分发场景，本文在上述分析模型基础上提出一种基于连通度概率感知的多跳消息分发算法CPAM。宏观层次上，算法关注网络连通度，利用基于车辆速度的多车道的分析模型计算各路段的连通概率，选择连通度高的路段进行消息的分发过程。而微观层次上，算法利用基于邻居信息的多跳广播抑制算法进行各路段的数据包传输，快速地将数据包从路段入口转发到路段出口。仿真结果表明，CPAM在数据包交付率以及时延性能上均优于带缓存的GPSR协议。
[Abstract]:In recent years, the rapid growth of the number of vehicles in China has brought a lot of problems to the urban traffic, which makes the vehicle-borne ad hoc network technology more and more attention.Through the free network communication between vehicles, the vehicle-borne ad hoc network can realize the real-time and efficient transmission of traffic safety information, road condition information and entertainment information, and solve the problems of traffic safety, road congestion and so on.However, due to the unique network characteristics of vehicle-mounted ad hoc networks and the different performance indicators required by various applications, it is very difficult to design a message distribution algorithm suitable for all scenarios.In this paper, we mainly study the multi-hop message distribution algorithm in different application scenarios in vehicle-borne ad hoc networks. The main research work includes the following contents:First of all, this paper improves the broadcast mechanism based on neighbor information in view of the application scenarios related to traffic safety and traffic conditions in the vehicle-borne ad hoc network.A multi-hop broadcast suppression algorithm (NIMBSs) is proposed, in which the neighbor information of the vehicle is combined with the vehicle location, speed and direction to form an ordered sequence of neighbor nodes, and the forwarding waiting time of the receiving vehicle is finally determined.The simulation results show that the proposed algorithm has lower packet loss rate and delay than the similar suppression algorithm, and effectively solves the scalability, data transmission redundancy and broadcast storm problems of the broadcast algorithm.Secondly, the multi-lane connectivity probability analysis model based on vehicle speed is studied under the condition that the road is in a free connectivity state.The model converts the actual multi-lane scene into a virtual multi-lane scenario with the same speed of each lane, and deduces the exact expression of the road connectivity probability based on the previous analysis results.After a lot of simulation experiments, the results show that the multi-lane connectivity probability analysis model based on vehicle speed is consistent with the actual situation.Finally, for the end-to-end message distribution scenario of unicast, this paper proposes a multi-hop message distribution algorithm CPAM based on connectivity probability perception based on the above analysis model.At the macro level, the algorithm focuses on network connectivity, uses the multi-lane analysis model based on vehicle speed to calculate the connectivity probability of each section, and selects the section with high connectivity to distribute messages.At the micro level, the algorithm uses the multi-hop broadcast suppression algorithm based on neighbor information to transmit the data packets from the entrance to the exit of the section.The simulation results show that the GPSR protocol is superior to the GPSR protocol in packet delivery rate and delay performance.
【学位授予单位】：南京邮电大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：U495;U463.67

【参考文献】