移动性感知的车联网内容传输机制研究
发布时间：2018-08-11 12:16
【摘要】：早在18世纪,亚当斯密就在《国富论》中指出:“一切改良中,以交通运输的改良最为有效。”随着汽车在人们日常生活中所扮演的角色越来越重要,以“行人、车辆、道路资源最优化利用”为目标的智能交通系统成为国家经济发展的重要载体,而车联网是保障智能交通系统有效运行的神经中枢。为了保证车联网信息交互的可靠性、有效性及稳定性,从而提升公路交通系统的事故避免率、道路畅通性及运行智能化的水平,研究适用于车联网的内容传输机制具有重要意义。本论文针对车联网通信链路动态特性和多种不同类型的车载应用需求,采用模型构建、机制设计、理论分析及系统仿真等手段,围绕移动性感知的车联网内容传输机制展开研究。论文的创新性工作主要包括如下几个方面:1)针对车辆移动相关性和无线信道衰落相关性,提出了车联网场景下链路持续时间的理论模型,解决了拓扑变化频繁网络条件下无线通信链路连通性能的有效数学刻画问题。基于二维离散马尔科夫链模型,将时空相关性分别建模为车辆间或车对地相对距离状态的转移,得到了通信链路持续时间的概率质量函数。理论分析和仿真验证表明所提出的模型能够在低复杂度条件下,有效地描述车辆移动相关性和阴影衰落相关性对链路持续时间的影响;2)针对无中心车联网的多跳内容传输,提出了基于退避计数器的无信标传输协议,解决了动态拓扑条件下传输机制设计需要综合考虑传输有效性和链路稳定性的难题。提出了联合考量单跳传输距离和链路持续时间的新型基于退避计数器的无信标传输策略,并针对无信标协议链路构建时延,建立了半马尔科夫过程模型,推导了道路车流密度和车辆移动速度等因素对传输路径构建时延的影响。仿真结果表明,协议能够实现无中心车联网环境中的高稳定性多跳传输,理论模型能够有效描述无信标协议传输链路构建时延性能;3)针对有中心车联网的多跳内容传输,提出了稀疏路边基础设施布设条件下的车对地内容上传算法,解决了进行多跳传输路径构建需求联合考虑网络动态拓扑和道路车流状态的难题。利用车辆中继提升了基础设施的覆盖性能,构建了车辆向基础设施传输的中继车辆和调制编码方式选择问题,分别针对单跳和多跳链路提出了目标内容完整传输的启发式近似算法。理论推导和仿真结果表明,算法能够提升传输链路稳定性,在降低方案实现复杂度的同时,缩短了整体内容传输时延;4)针对车联网基于多播的内容传输,提出了多样车载用户需求和动态网络拓扑条件下的资源调度问题,解决了分层编码视频在道路区域同时满足传输全覆盖和高吞吐量两方面要求的问题。针对分层编码视频信息,考虑了道路车流密度以及车载用户设备所能够播放视频的解析度不同,将基础设施向视频信息请求车辆多播调度问题分解为“请求用户全覆盖的基本层视频调度子问题”和“有限多播资源条件下高吞吐量性能的扩展层视频调度子问题”。然后分别针对上述子问题,提出了两个启发式近似算法,在低复杂度的条件下实现了全覆盖和高吞吐量的目标;5)针对车联网基于移动群体感知的内容观测调度,提出了移动车辆群体感知框架,解决了非确定车辆轨迹、可变观测时段和有限观测成本条件下的车辆移动感知调度问题。针对非公共交通工具未来行驶路径很难获知的特性,推导得到了基于概率路径预测模型的空间观测覆盖度的表达式,构建了可观测时段和有限成本条件下移动车辆群体感知问题,提出了选择区域观测低相关的参与车辆集合及最大化观测时段单位成本观测性能的车辆群体雇用方案。理论推导和仿真结果表明,方案提升了在非确定车辆轨迹及有限观测成本预算条件下的空间覆盖度性能,同时算法具有低复杂度和可扩展性。
[Abstract]:As early as the 18th century, Adam Smith pointed out in the Wealth of Nations that "the improvement of transportation is the most effective of all improvements." With the increasingly important role of automobiles in people's daily life, the intelligent transportation system with the goal of "optimizing the use of pedestrians, vehicles and road resources" has become an important carrier of national economic development. In order to ensure the reliability, validity and stability of the information exchange of the vehicle networking, and thus enhance the accident avoidance rate of the highway traffic system, road smoothness and the level of intelligent operation, it is of great significance to study the content transmission mechanism suitable for the vehicle networking. According to the dynamic characteristics of the communication links and various types of on-board application requirements of vehicle networking, this paper studies the content transmission mechanism of vehicle networking with mobility awareness by means of model building, mechanism design, theoretical analysis and system simulation. A theoretical model of link duration in vehicle networking scenarios is proposed to solve the problem of effective mathematical characterization of wireless communication link connectivity under topology-varying networks. Based on the two-dimensional discrete Markov chain model, the temporal-spatial correlation is modeled as inter-vehicle or vehicle-to-ground correlation respectively. The theoretical analysis and simulation results show that the proposed model can effectively describe the impact of vehicle mobility correlation and shadow fading correlation on link duration under low complexity conditions; 2) for multi-hop content transmission in the no-central vehicle network; A novel beacon-free transmission protocol based on backoff counter is proposed to solve the problem that transmission efficiency and link stability should be considered in the design of transmission mechanism under dynamic topology. A novel beacon-free transmission strategy based on backoff counter is proposed, which considers the single-hop transmission distance and link duration jointly. The simulation results show that the protocol can achieve high stability and multi-hop transmission in the network environment without central vehicle, and the theoretical model can effectively describe the transmission of beacon-free protocol. Link construction delay performance; 3) For multi-hop content transmission with central vehicle interconnection, a vehicle-to-ground content upload algorithm under sparse roadside infrastructure is proposed, which solves the difficult problem of multi-hop transmission path construction considering network dynamic topology and road traffic status. A heuristic approximation algorithm is proposed for single-hop and multi-hop links respectively. The theoretical derivation and simulation results show that the algorithm can improve the stability of the transmission link and reduce the complexity of the scheme implementation. At the same time, the overall content transmission delay is shortened; 4) For the content transmission based on Multicast in the vehicle network, the resource scheduling problem under the condition of multi-vehicle user demand and dynamic network topology is proposed, which solves the problem that the layered coding video meets the requirements of both transmission coverage and high throughput in the road area. Video information is decomposed into "basic layer video scheduling sub-problem requiring full user coverage" and "extended layer view with high throughput performance under limited multicast resources" by considering the different traffic density on the road and the different resolution of video that can be played by on-board user equipment. Secondly, two heuristic approximation algorithms are proposed to solve the above sub-problems, which achieve the goal of full coverage and high throughput under the condition of low complexity. Fifthly, a mobile vehicle group awareness framework is proposed to solve the uncertain vehicle trajectory for content observation scheduling based on mobile group awareness. The problem of vehicle perceptual scheduling with variable observation period and limited observation cost is studied. In view of the fact that it is difficult to know the future path of non-public transport vehicles, the expression of spatial observation coverage based on probabilistic path prediction model is derived, and the mobile vehicle population with observable period and limited observation cost is constructed. For the perception problem, a group hiring scheme is proposed to select the vehicles with low correlation in the observation area and maximize the unit cost performance in the observation period. Low complexity and scalability.
【学位授予单位】：北京交通大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：U495;TN929.5
本文编号：2176965
[Abstract]:As early as the 18th century, Adam Smith pointed out in the Wealth of Nations that "the improvement of transportation is the most effective of all improvements." With the increasingly important role of automobiles in people's daily life, the intelligent transportation system with the goal of "optimizing the use of pedestrians, vehicles and road resources" has become an important carrier of national economic development. In order to ensure the reliability, validity and stability of the information exchange of the vehicle networking, and thus enhance the accident avoidance rate of the highway traffic system, road smoothness and the level of intelligent operation, it is of great significance to study the content transmission mechanism suitable for the vehicle networking. According to the dynamic characteristics of the communication links and various types of on-board application requirements of vehicle networking, this paper studies the content transmission mechanism of vehicle networking with mobility awareness by means of model building, mechanism design, theoretical analysis and system simulation. A theoretical model of link duration in vehicle networking scenarios is proposed to solve the problem of effective mathematical characterization of wireless communication link connectivity under topology-varying networks. Based on the two-dimensional discrete Markov chain model, the temporal-spatial correlation is modeled as inter-vehicle or vehicle-to-ground correlation respectively. The theoretical analysis and simulation results show that the proposed model can effectively describe the impact of vehicle mobility correlation and shadow fading correlation on link duration under low complexity conditions; 2) for multi-hop content transmission in the no-central vehicle network; A novel beacon-free transmission protocol based on backoff counter is proposed to solve the problem that transmission efficiency and link stability should be considered in the design of transmission mechanism under dynamic topology. A novel beacon-free transmission strategy based on backoff counter is proposed, which considers the single-hop transmission distance and link duration jointly. The simulation results show that the protocol can achieve high stability and multi-hop transmission in the network environment without central vehicle, and the theoretical model can effectively describe the transmission of beacon-free protocol. Link construction delay performance; 3) For multi-hop content transmission with central vehicle interconnection, a vehicle-to-ground content upload algorithm under sparse roadside infrastructure is proposed, which solves the difficult problem of multi-hop transmission path construction considering network dynamic topology and road traffic status. A heuristic approximation algorithm is proposed for single-hop and multi-hop links respectively. The theoretical derivation and simulation results show that the algorithm can improve the stability of the transmission link and reduce the complexity of the scheme implementation. At the same time, the overall content transmission delay is shortened; 4) For the content transmission based on Multicast in the vehicle network, the resource scheduling problem under the condition of multi-vehicle user demand and dynamic network topology is proposed, which solves the problem that the layered coding video meets the requirements of both transmission coverage and high throughput in the road area. Video information is decomposed into "basic layer video scheduling sub-problem requiring full user coverage" and "extended layer view with high throughput performance under limited multicast resources" by considering the different traffic density on the road and the different resolution of video that can be played by on-board user equipment. Secondly, two heuristic approximation algorithms are proposed to solve the above sub-problems, which achieve the goal of full coverage and high throughput under the condition of low complexity. Fifthly, a mobile vehicle group awareness framework is proposed to solve the uncertain vehicle trajectory for content observation scheduling based on mobile group awareness. The problem of vehicle perceptual scheduling with variable observation period and limited observation cost is studied. In view of the fact that it is difficult to know the future path of non-public transport vehicles, the expression of spatial observation coverage based on probabilistic path prediction model is derived, and the mobile vehicle population with observable period and limited observation cost is constructed. For the perception problem, a group hiring scheme is proposed to select the vehicles with low correlation in the observation area and maximize the unit cost performance in the observation period. Low complexity and scalability.
【学位授予单位】：北京交通大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：U495;TN929.5
【参考文献】
相关期刊论文 前1条
1 孟昊;钟章队;艾渤;;Ad Hoc网络路由协议研究及其性能比较[J];信息与电子工程;2009年02期
相关博士学位论文 前3条
1 孙海峰;VANETs路由算法研究[D];电子科技大学;2015年
2 何睿斯;车载网络复杂场景下无线信道测量与建模研究[D];北京交通大学;2015年
3 李利民;车联网V2I通信媒体接入控制技术研究[D];哈尔滨工业大学;2014年
，本文编号：2176965
本文链接：https://www.wllwen.com/kejilunwen/daoluqiaoliang/2176965.html
