车联网中面向QoS保障的资源分配和自适应传输技术

发布时间：2018-03-18 18:39

本文选题：车联网　切入点：视频　出处：《浙江大学》2017年博士论文　论文类型：学位论文

【摘要】：作为未来智能交通系统的主要组成部分之一,车联网应该具有保障其网络中各种业务的服务质量的能力。然而,车联网中车辆的高速移动、多种优先级业务对有限信道资源的抢占以及路边基础设施的稀疏布置共同造成车辆和路边基础设施间通信的时断时续、车辆之间快速变化的拓扑结构、高动态变化的信道质量以及不断变化的可用信道数目。除了上述复杂的网络状况外,车联网为了满足不同类型业务的服务质量要求,需要为不同类型的业务提供不同的资源分配策略,而同一种业务的多个服务质量指标往往是相互矛盾的,例如为了改善视频的播放流畅度,往往要降低画面质量。本学位论文针对车联网中作为商娱类业务代表的视频业务和作为智能交通类业务代表的信标消息广播业务,研究其服务质量保障技术,设计其资源分配和自适应传输方案。针对采用认知信道且路边基础设施稀疏布置的车联网,研究面向服务质量保障的自适应视频流传输。进行方案设计与性能分析,提出一种视频流自适应传输方案,给出两跳链路端到端连接性的数学表达。以保障视频流畅播放前提下提高视频画面质量为目标,建立使效用最大的视频层数优化问题。该效用函数反映车辆运动和未来位置、车辆的缓存器中现有的数据量、未来播放需要的数据量、预期能够获得的数据量等因素对请求的视频层数的影响。提出一个面向质量的自适应视频流算法,并且通过仿真验证其有效性。针对可用信道数因较高优先级业务占用而动态变化的车联网,研究面向多车辆视频业务的资源分配和自适应传输,进行方案设计与性能分析。提出一种基于博弈竞价的资源分配框架,将车辆的运动和资源的动态变化等实际因素考虑进来。该框架将通常由路边单元(RoadSideUnit,RSU)集中进行的资源分配转化为各车辆根据自己资源需求的紧迫程度的自由竞价,而将车辆的出价表示为动态变化的信道数目和车辆数目、车辆缓存器中存有的视频数据量、车辆位置、数据传输速率、车辆之间的相互竞争和影响等因素的函数。定义环境状态和其状态转移矩阵,以帮助车辆合理地出价。定义视频播放流畅前提下获得较高视频质量的车辆效用表达式。建立具有贝尔曼方程形式的累加效用和最大问题,提出一种基于动态规划的低复杂度解法,用于计算在不同车辆状态和环境状态下的最优出价和请求传输的最优视频层数。针对车联网中智能交通类业务的代表信标消息广播业务,研究面向冲突避免和位置追踪的自适应信标消息广播间隔和资源分配。提出相邻RSU之间公平有序的资源分配协商机制,以避免相邻RSU的覆盖区域内相近的车辆被分到相同的时频资源。定义车辆的信标消息广播间隔(Beacon Interval,BI)要求的数学表达,体现出不同车辆的不同行驶状态对BI的需求,用于反映车辆行驶安全和位置追踪准确的要求。根据车辆的BI需求,定义反映其满足程度的效用函数,并且将单个RSU覆盖范围内的资源分配问题建模为一个累积效用和最大化问题。为了便于求解该0-1整型优化问题,提出将该问题转化为无向加权图中最大独立集问题的转化方法,并且提出低复杂度的求解算法。上述方案能够在避免相邻RSU覆盖范围内车辆、同一RSU覆盖范围内车辆的信标消息广播冲突的前提下,尽量满足车辆行驶安全和位置追踪准确的要求。
[Abstract]:As one of the main part of future intelligent transportation system, vehicle networking capabilities should have its security business in the network service quality. However, the high-speed mobile vehicle networking, to seize the business of multiple priority limited channel resources and roadside infrastructure facilities caused by sparse arrangement of vehicles and roadside infrastructure communication on the rapidly changing topology between the vehicle and the number of available channels, high dynamic channel quality and changing. In addition to the complex network situation, car networking in order to satisfy the different types of service quality requirements, the need to provide different resource allocation strategies for different types of business, and with a business a plurality of service quality indicators are often contradictory, for example, in order to improve the video playback fluency, tend to reduce the quality of the picture. A paper for the car networking as entertainment business on behalf of the video service as the beacon message broadcasting service and intelligent traffic business representatives, to study the quality of service technology, the design of resource allocation and adaptive transmission scheme. According to the cognitive channel and roadside infrastructure sparse layout of the car networking, the research of adaptive video service oriented to ensure the quality of transmission. The performance analysis and design scheme, this paper proposes a video streaming adaptive transmission scheme, the expression given two hop link end to end the mathematics. In order to ensure the smooth video playback under the premise of improving the video quality as the goal, the establishment of the video layers optimization problem of maximum utility. The utility function reflects the movement of the vehicle and the future position, the existing vehicle data buffer, the future need to play the amount of data, is expected to be able to obtain the amount of data and other factors Impact on the requested video layers. Put forward an adaptive algorithm for quality video stream, and the simulation results show its validity. According to the number of available channels by high priority business occupancy and the dynamic change of car networking, research on resource allocation for multi vehicle video services and adaptive transmission, analyze the design and performance of proposed scheme. A resource allocation framework based on bidding game, the actual factors of vehicle motion and dynamic changes of resources into account. The framework will usually by the roadside unit (RoadSideUnit, RSU) focus on resource allocation into the vehicle according to the urgency of their bidding free resource requirements, and the vehicle will offer said the number of vehicles and the number of channels for the dynamic change, the amount of video data, there are vehicles in buffer position of the vehicle, the data transmission rate, the mutual competition between vehicles The function factors and influence. The transfer matrix for definition of environmental state and the state, to help the vehicle reasonable bid. The definition of smooth video playback of vehicle utility expression get higher video quality. Establish the cumulative utility and the biggest problem with the form of Behrman equation, is proposed based on a dynamic programming solution for low complexity. The number of bids and compute the optimal optimal video transmission request in different vehicle condition and environment condition. According to the car networking intelligent transportation business on behalf of the beacon message broadcasting service, based on collision avoidance and location tracking adaptive beacon message broadcast interval and resource allocation. The adjacent RSU resource allocation between fair and orderly negotiation mechanism in order to avoid coverage, adjacent RSU in similar vehicles were assigned to the same time frequency resource. Radio beacon message definition between vehicles Septum (Beacon Interval, BI) expression reflects the requirements of mathematics, the different needs of different vehicle driving state of BI, is used to reflect the running safety of the vehicle and position tracking accuracy. According to the vehicle's BI requirements, the utility function satisfies the definition reflects the degree of coverage and a single RSU modeling problem of resource allocation within the scope of as a cumulative sum utility maximization problem. In order to solve the 0-1 integer optimization problem, the problem is transformed into a weighted undirected graph transformation method in the maximum independent set problem, and put forward the algorithm of low complexity. The scheme can avoid the adjacent RSU range of vehicles, the premise beacon message broadcast conflict the same RSU within the coverage area of the vehicle, the vehicle to meet the safety and accurate location tracking requirements.

【学位授予单位】：浙江大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：TN929.5;U495

【相似文献】