【作者】 郭伟杰；

【导师】 黄刘生；

【作者基本信息】 中国科学技术大学 ， 计算机软件与理论， 2014， 博士

【摘要】 车联网VANET (Vehicular Ad-hoc Network)作为新型无线通信技术和现代汽车工业技术发展的产物,近年来受到了来自世界范围内学术界和汽车厂商的广泛关注,并被认为是建设智能交通系统和提高道路交通安全的关键技术。在车联网面向道路交通安全的诸多应用中,移动车辆周期性广播的Beacons消息和危险事件触发的告警消息Warnings对数据传输的可靠性和实时性提出了更高的要求。而车联网节点流通性高、网络拓扑多变和无线信道质量多样性的特点,使得车联网中面向安全应用的消息传输问题面临了更多挑战。基于车联网中消息分发问题的现有研究成果,我们分别从MAC和路由两个层面设计适合面向安全应用的消息分发协议与算法,并研究了跨MAC层和路由层的综合数据转发策略。为了对车联网中消息传输协议的性能进行有效验证,我们还对车联网环境下的实验仿真工具进行了研究,设计和开发了车联网综合仿真平台DVR_Sim,本文的主要研究内容及贡献如下四个方面。一、车联网中节点的高速移动性和车流分布的不均衡性严重影响到MAC协议的性能。传统基于信道竞争机制的MAC协议在车流密集时容易导致消息碰撞,而非竞争机制的MAC协议则因网络结构快速变化产生较多的协调开销。因此,我们在论文的第二章基于动态TDMA机制研究自适应车流分布变化的MAC协议。利用路边单元RSU的通信与协调能力,我们提出了自适应车流分布变化的动态时隙调度周期概念。随后重新设计了协议的帧格式和RSU时隙分配算法,并对空余时隙回收机制和相邻RSU之间无缝切换机制进行了优化。最后的仿真实验表明,在车流密度较大的交通状况下,我们设计的协议在Beacons消息的传输延迟、平均丢包率两方面明显优于IEEE802.11p MAC的性能,同时Beacons消息分发的网络有效吞吐量提高了40%。二、现有的消息分发协议中过度强调告警消息的优先级将会造成Beacons肖息丧失广播的机会,事实上,上述两类消息在紧急情况对于道路安全都非常重要。基于此我们在本文的第三章从MAC层角度研究上述两类消息访问无线信道的优先级和公平性问题,并针对车辆协同防碰撞应用场景设计了风险感知的动态MAC协议R-MAC.我们采用TDMA和CSMA/CA混合的无线信道访问协调机制,根据实时交通道路信息建立风险动态评估模型。R-MAC根据交通风险优先为告警消息分配必需的信道访问时隙,同时预留Beacons消息访问信道的机会。最后,多种交通场景下的仿真实验结果显示,在保证上述两类消息实时传输的情况下,我们设计的方案相比IEEE802.11p MAC协议提高了42%的消息传输成功率,并将两者获取无线信道服务机会的公平性提高了32%。三、为了实现告警消息Warnings的长距离分发,现有工作大多借助基于位置信息辅助的多播路由协议,但是难以满足告警消息的实时性和可靠性要求。因此,第四章我们利用车联网中通信节点发射功率可调的优势,并综合分析无线信号强度衰减以及发生在PHY/MAC层的传输延迟和消息重传等因素,跨层地研究告警消息分发过程中的最优中继选择算法,并设计了延迟感知的可靠广播协议DR-BP。最后,仿真实验结果表明相对于传统的S1P(Slotted1-Persistence, SIP)广播协议,我们设计的协议将告警消息长距离分发的实时性和可靠性分别提高了53%和38%。四、当前车联网环境下大部分仿真工具将交通流模拟和网络协议仿真的过程分离进行,难以实现网络仿真对交通流模拟的逆向反馈,更没有考虑真实交通环境下驾驶员对车辆操控的影响。为了更好地验证面向交通道路安全应用的数据分发协议性能,我们在第五章对车联网环境下的实验仿真工具进行了系统研究,设计并开发了车联网综合仿真平台DVR_Sim,实现了交通流模拟、网络协议仿真和驾驶员模拟三者的同步与双向实时反馈。更多还原

【Abstract】 Recent advances in new wireless communication technologies and the emergency of computationally rich vehicles are making vehicular ad-hoc network (VANET) research receive much attention from academic communities and major car manufactures around the world, which is considered as a potential technology for intelligent transportation system and improving the road traffic safety. The key for various safety-related applications is the real-time and reliable delivery of safety-related messages among vehicles, which include periodical beacons and risk-triggered warning messages. However, the variety of network topology and wireless channel caused by high mobility of vehicles bring more challenges to the delivery of safety-related messages in VANET.Based on the current research of messages delivery in VANET, we design and optimize novel protocols for the transmission of safety-related messages in the MAC and routing view two levels, where the cross-layer design is considered. Moreover, in order to verify the performance of transmission protocols effectively, we also deeply study the simulators used in VANET and develop an integrated simulation platform named DVR_Sim. The main contributions of this research are as follows.1. The performance of MAC protocols is affected seriously by the high mobility and unbalance of vehicles on roads. The performance of random and contention-based methods deteriorates significantly with increased traffic load, because of a corresponding increase in message collisions. However, the contention-free MAC methods require more coordination to perform allocation, especially when the network configuration changes rapidly. Thus, we study the adaptive collision-free MAC protocol based on dynamic TDMA mechanism for inter-vehicular communication in the second chapter. The RSU is exploited to maintain a dynamic slots assignment cycle for moving vehicles under its coverage. Then we redefine the frame form and slots assignment algorithm. Besides, the redundant slots reclamation and seamless handover between adjacent RSUs are also optimized carefully. Finally, the simulation results verify that our proposed scheme outperforms the IEEE802.11p MAC protocol in term of packet loss and transmission delay. What’s more, the goodput of beacon messages is improved by40%compared with IEEE802.11p.2. Most of the existing research always assumes that the risk-triggered messages have a higher priority to be transmitted than that of beacons. However, both of them are very important for road traffic safety in emergency situations. Thus, we study the efficiency and fairness of delivering beacons and warning messages on the medium access, and then design the risk-aware dynamic MAC protocol for vehicular cooperative collision avoidance system in the third chapter. We adopt the combination mechanism of TDMA and CSMA/CA and each frame is divided into two parts: TDMA segment for transferring beacons, and CSMA segment for delivering warning messages. Then, we propose a stochastic model to predict the average total number of potential collisions in a platoon of vehicles, which determines the size of CSMA segment in the R-MAC protocol meticulously. After that, extensive simulations under different traffic scenes show that our protocol improves the packets delivery ratio by42%and the Jain’s fairness index of the channel access by32%compared with IEEE802.11p, respectively.3. Although many geocast routing schemes are employed to widely propagate the warning messages among vehicles, it’s still difficult to meet the requirement of real-time and reliability for risk-triggered messages delivery. Therefore, we study the delay-aware reliable broadcast scheme based on transmit power control technology for VANET in the fourth chapter. The wireless channel fading, transmission delay and retransmissions characters occurring in the PHY/MAC layer are all considered when we design the cross-layer optimal relay selection algorithm. Finally, simulation results show that our protocol increases38%in packets delivery ratio and reduce53%in average transmission delay compared with traditional SIP broadcast protocol.4. Since the simulations of traffic and network protocol are conducted separately and the driver’s factor is not considered in most VANET simulations, we deeply study the simulators and develop an integrated simulation platform named DVR_Sim in the fifth chapter. The DVR_Sim can simulate the traffic, network protocol and driver’s factor simultaneously, which makes the experiment configuration more close to the real traffic scene.更多还原