ROAMER: Roadside Units as message routers in VANETs

Vehicular Ad hoc Networks, also known as VANETs, enable vehicles that are not necessarily within the same radio transmission range to communicate with each other. VANETs also allow vehicles to connect to Roadside Units (RSUs). The latter are connected to the Internet, forming a fixed infrastructure that offers them the capability of communicating with each other and with roaming vehicles. RSUs support cooperative and distributed applications in which vehicles and RSUs work together to coordinate actions and to share and process several types of information. RSUs have so far been used for different roles such as data disseminators, traffic directories, location servers, security managers, and service proxies. In this paper, we focus on routing; namely we exploit RSUs to route packets between any source and destination in the VANET. To our knowledge, this is the first attempt to use the RSU backbone to efficiently route packets to very far locations in VANETs by using geographic forwarding. We evaluate the RSU backbone routing performance via the ns2 simulation platform. We compare our scheme to existing solutions and prove the feasibility and efficiency of our scheme in terms of query delay, packet success delivery ratio, and total generated traffic.     

车联网中基于任务紧急性的联合卸载方案

在车联网中,任务卸载可以有效地解决车辆的存储资源和计算资源不足的问题,单个的 MEC(mobile edge computing)服 务器通常无法满足车辆密集区域的任务卸载需求。 针对上述不足,设计了一种多 MEC 服务器的联合卸载方案( joint offloading method based on task urgency, JOMTU)。 车辆递交任务卸载请求给本地 MEC 服务器时,后者在负载严重的情况下,会根据任务 的紧急性和服务器负载情况等因素,将任务发送给附近 MEC 服务器处理以满足任务的截止日期要求。 仿真实验结果表明,与 传统的方案相比,所提出的方案将系统的整体任务失败率降低 17%,并且优化了整个网络的服务器负载情况、增加了网络的可 靠性。     

An evolutionary computation approach for designing mobile ad hoc networks

This paper presents the topological design of ad hoc networks in terms of distances among static nodes and speeds of mobiles nodes. Due to the complexity of the problem and the number of parameters to be considered, a genetic algorithm combined with the simulation environment NS-2 is proposed to find the optimum solution. More specifically, NS-2 provides the fitness function guiding the genetic search. The proposed framework has been tested using a railway scenario in which several static and mobile nodes are interacting. Results show the feasibility of the proposed framework and illustrate the possibility of genetic approach for solving similar application scenarios.     

The selective use of redundancy for video streaming over Vehicular Ad Hoc Networks

Video streaming over Vehicular Ad Hoc Networks (VANETs) offers the opportunity to deploy many interesting services. These services, however, are strongly prone to packet loss due to the highly dynamic topology and shared wireless medium inherent in the VANETs. A possible strategy to enhance the delivery rate is to use redundancy for handling packet loss. This is a suitable technique for VANETs as it does not require any interaction between the source and receivers.In this work, we discuss novel approaches for the use of redundancy based on the particularities of video streaming over VANETs. A thorough study on the use of redundancy through Erasure Coding and Network Coding in both video unicast and video broadcast in VANETs is provided. We investigate each strategy, design novel solutions and compare their performance. We evaluated the proposed solutions from the perspective not only of cost as bandwidth utilization, but also the offered receiving rate of unique video content at the application layer. This perspective is fundamental to understanding how redundancy can be used without limiting the video quality that can be displayed to end users.Furthermore, we propose the selective use of redundancy solely on data that is more relevant to the video quality. This approach offers increases in overall video quality without leading to an excessive overhead nor to a substantial decrease in the receiving rate of unique video content.     

车载网络中基于区域覆盖的机会传递算法研究

The highly mobility of vehicles, intermit-tent communication between the vehicles and the requirements of real time applications are some of the main challenges of multi-hop message delivery in Vehicular Ad Hoc Networks(VANETs).There are also additional challenges, especially when the des-tination for the message delivery is a moving vehi-cle. In this paper, we propose a novel multi-hop message delivery method, called the Coverage Area-based oPportunistic message forwarding algorithm (CAP). The main idea of CAP is to cover the poten-tial area of a moving target vehicle as much as pos-sible with the reachable area of the required deliver-y messages, so that the message can be delivered successfully. Because the success ratio and overhead of the multi-hop message delivery are two important but incompatible parameters in CAP, two tunings are discussed in the algorithm in order to maintain the balance of the two parameters. The simulation results show that compared with other reference ap-proaches, CAP provides an efficient message deliv-ery with a higher success ratio and a shorter mes-sage delay.     

基于相对速度的802.11p车载网络自适应退避算法*

IEEE 802.11p协议是针对车载环境应用制定的物理层与MAC层标准.车载网络中节点的高移动性和拓扑结构的频繁变更导致网络丢包率增加.为解决车辆移动特性引发的信道访问不公平性问题及降低网络丢包率,提出一种基于相对速度的自适应退避算法RSBA,依据移动节点间相对速度差异优化退避机制.通过NCTUns模拟实验表明,RS...     

公交移动协助城市车载网络分层路由协议

城市车载网络作为未来智能交通的重要组成部分,为解决城市交通问题提供了有效的解决方案.现有大部分路由方案未能很好地解决车辆高速移动、复杂城市交通环境和不同车流密度三大问题对路由转发的影响,导致路由协议的性能在高效和可靠性方面不足.根据城市交通信号传播环境、红灯等停以及公交运行轨迹固定、周期循环等特性,提出公交移动协助的城市车载网络路由协议,将车车之间的复杂、不确定路由转换为普通车辆-簇头-公交三层节点之间的确定、协作方式,发挥公交周期移动优势,并用簇头策略优化普通车辆与公交之间的通信.设计根据城市信号传播衰减与车辆移动相关的单跳链路筛选与多跳延迟相关的概率转发机制,保证公交转发的高效、可靠,并设计能自适应车流密度的簇头触发策略,仿真实验结果表明,整体方案转发跳数达到一定数量时,能取得较好的数据传输成功率和较低的网络延迟.     

基于车辆自组织网络的交通态势检测方法

随着汽车保有量的迅速增加,城市道路交通拥堵变得尤为严重,精确地检测交通态势可以帮助缓解交通问题。为此,提出一种基于车辆自组织网络（vehicular Ad hoc networks,VANETs）的交通态势检测方法---TraSD-VANET（traffic situation detection method based on VANETs）。在该方法中,车辆自动聚簇,然后主动向簇头汇报当前自身的位置和速度信息;簇头根据收到的信息计算簇内的车辆密度和路面上的加权平均速度,之后基于模糊逻辑判断簇内的交通态势。仿真结果表明,在四种车辆场景下,TraSD-VANET检测准确程度比协作检测方法 CoTEC （cooperative traffic congestion detection）平均高16％。该方法在道路交通态势检测中有重要的应用价值。     

基于车流密度的VANETs路由协议

在节点高速移动的车载自组织网络(VANETs)中,道路交通状况极大地影响着网络中的数据传输性能。在贪婪周边无状态路由(GPSR)协议的基础上加以改进,提出了基于车流密度的VANETs由协议。考虑了车流密度以及节点运动速度、方向等影响因素,设计车流密度的计算方法,利用新的转发策略替代GPSR的贪婪转发策略,能够选择车流密度较好的路径进行数据转发,避免因车流密度分布不均匀而产生的局部最大现象。同时能够对由于节点高速运动引起的链路中断进行预测,提出有效的修复机制,从而建立链路稳健的VANETs路由。采用Matlab仿真平台进行仿真实验,与已有的GPSR,GPCR协议进行比较分析。仿真结果表明:提出的路由协议相比于其它2种路由协议在时延和分组转发率方面得到显著改善,性能优越,非常适合在城市场景中。     

自组织车联网中GPSR路由协议的研究进展

由于贪婪周边无状态路由(GPSR)对于拓扑结构频繁变化的自组织车联网(VANETs)具有最适性,为此,针对GPSR提出了许多改进协议。首先对VANETs网络层路由协议进行分类比较,然后分析和总结近年来基于位置路由协议的核心路由机制和优缺点,重点分析典型的基于位置的路由协议GPSR在城市场景中存在的问题。最后,提出了GPSR未来可能的研究策略和发展方向。