协同车辆安全系统自适应信道拥塞控制研究

发布时间：2018-05-10 22:38

本文选题：车联网 + 协同车辆安全系统　；参考：《大连理工大学》2016年博士论文

【摘要】：协同车辆安全系统是车联网最重要也是最具有挑战性的一种应用。车辆追踪性能是协同安全系统的基础,协同安全系统主要依赖IEEE 802.11p周期性广播车辆状态信息,实现车辆间移动轨迹的追踪。然而,车联网中,车辆移动环境影响网络性能,在高车辆密度情形下,IEEE 802.11p信道产生拥塞,通信性能下降,而通信性能下降反过来直接导致车辆移动轨迹追踪精度下降。车辆移动物理动态性和通信过程之间紧耦合的关系,要求必须建立自适应于车辆移动环境的信道拥塞控制策略,充分公平使用信道带宽资源,保证不同车辆密度情形下车辆准确追踪性能。本文以建立车辆移动环境与IEEE802.11p交互模型和信道访问控制协议为目标,分别从以下几个问题展开研究：(1)目前的信道拥塞控制策略是在信道发生拥塞之后或者车辆密度变化之后才对网络参数进行调整,这就需要一定的时间,从拥塞的信道状态恢复至正常的信道状态,无法满足车辆追踪实时性需求。为了保证动态车辆密度环境下车辆间实时追踪,基于模型预测控制方法,提出了一种自适应于车辆密度的功率主动控制策略。首先,建立了动态车辆密度与信息分发速率交互模型,基于该模型,将一个时域内的功率控制问题转化为一个非线性优化问题,并设计了一个分布式功率迭代算法;然后,在一个时域优化问题的基础上,针对车辆密度变化的时变性,提出了基于车辆密度估计和滚动优化的功率自适应控制算法,该算法主动引导IEEE 802.11p在不同车辆密度情形下达到期望的信息分发速率性能,并实现准确的车辆追踪。(2)车联网中最近邻居车辆具有较高的碰撞危险系数,应该赋予最近邻居车辆较高的数据包发送优先权。然而,传统的信道拥塞控制方法不能保证不同车辆密度情形下最近邻居车辆的追踪精度需求。为了充分利用信道资源,且维持最近邻居车辆追踪精度,提出了一个自适应于车辆追踪精度需求的功率反馈控制框架。在该框架下,首先,设计了一个描述车辆密度、数据包成功接收概率和车辆追踪精度之间交互关系的参考模型,该参考模型基于期望的车辆追踪精度需求,获取期望的数据包成功接收概率,作为功率自适应控制过程的一个参考点；然后,设计了一个功率控制器,该控制器基于期望的数据包成功接收概率和实际的数据包成功接收概率之问的偏差,产生实时的功率控制序列,维持动态车辆密度和外界干扰影响下可接受的车辆追踪精度。(3)目前公平信道资源分配策略基于最大最小化方法实现,该方法没有考虑资源分配的效率问题,不能保证每个车辆的追踪精度最优。为了保证每个节点公平的信道资源访问机会,且实现准确的车辆追踪,提出了描述车辆追踪精度与接收节点数据包成功接收速率之间交互关系的效用函数；基于车辆追踪精度效用函数,利用博弈论方法,设计了一个带有惩罚函数的竞争窗口调整博弈模型,给出了博弈模型纳什均衡点引导算法,该算法能够使得信道访问机会比较大的节点受到其它邻居节点的惩罚,从而引导所有节点达到相同的信道访问概率；为了实现帕累托最优,设计了一个分布式竞争窗口调整算法,该算法通过在每个节点间交互竞争窗口信息,引导每个节点到达帕累托纳什均衡点,实现节点间相同的信道访问机会且保证准确的车辆追踪。(4)目前信道拥塞控制策略假设数据包的丢失主要由于信道中多个节点并发发送数据包产生的碰撞造成。然而,由于隐藏终端的存在,并不是所有成功发送的数据包都能被周围邻居节点成功接收。目前的信道拥塞控制策略没有考虑隐藏终端的影响。针对上述背景,基于比例-积分-微分控制方法,提出了一个自适应于车辆密度与隐藏终端数量的动态反馈功率控制策略。首先,建立了动态车辆密度和隐藏终端影响下的信息分发速率模型,基于该模型,给出了能够预测不同车辆密度情形下期望信息分发速率状态的参考模型；然后,设计了反馈功率控制模型,该模型基于期望信息分发速率和车辆实际信息分发速率之间的偏差,产生功率控制序列,引导车辆在动态车辆密度和隐藏终端因素影响下达到期望的车辆追踪精度性能。
[Abstract]:Cooperative vehicle safety system is the most important and most challenging application of vehicle network. Vehicle tracking performance is the basis of cooperative security system. The cooperative security system mainly relies on the IEEE 802.11p periodic broadcast vehicle state information to realize the tracking of the vehicle movement trajectory. However, the vehicle network, vehicle mobile environment affects the network. In the case of high vehicle density, the IEEE 802.11p channel is congested and the communication performance decreases, and the communication performance decline in turn leads to the decrease of the tracking precision of the vehicle. The relationship between the dynamic physical dynamics of the vehicle and the tight coupling between the communication processes requires the establishment of the channel congestion control adaptive to the vehicle mobile environment. The strategy is to make full and fair use of channel bandwidth resources to ensure accurate tracking performance of vehicles with different vehicle density. This paper aims at establishing vehicle mobile environment and IEEE802.11p interaction model and channel access control protocol, and studies the following problems respectively: (1) the current channel congestion control strategy is in channel occurrence. The network parameters are adjusted after the congestion or the vehicle density changes, which requires a certain time to recover from the congestion channel state to the normal channel state, which can not meet the real-time requirement of vehicle tracking. In order to ensure the real time tracking of vehicles in the dynamic vehicle density environment, the model predictive control method is proposed. A power active control strategy that adapts to vehicle density. First, an interactive model of dynamic vehicle density and information distribution rate is established. Based on this model, a power control problem in time domain is transformed into a nonlinear optimization problem, and a distributed power iterative algorithm is designed. Then, a time domain optimization problem is proposed. On the basis of the time variability of vehicle density change, a power adaptive control algorithm based on vehicle density estimation and rolling optimization is proposed. The algorithm initiatively guides IEEE 802.11p to achieve the desired information distribution rate performance under different vehicle density, and realizes accurate vehicle tracking. (2) the nearest neighbor vehicle in the network of vehicles. There is a higher collision risk factor, which should give priority to the nearest neighbor vehicle with higher packet transmission. However, the traditional channel congestion control method can not guarantee the tracking precision demand of the nearest neighbor vehicle under different vehicle density. In order to make full use of the channel resources and maintain the nearest neighbor vehicle tracking precision, the paper proposes a new method. A power feedback control framework adapts to the demand for vehicle tracking accuracy. Under this framework, a reference model is designed to describe the interaction between vehicle density, the successful receiving probability of the packet and the tracking accuracy of the vehicle. The reference model is based on the desired tracking precision requirement of the vehicle tracking, and obtains the successful reception of expected data packets. Probability, as a reference point for the adaptive control process of power, and then a power controller is designed. The controller is based on the deviation of the expected probability of receiving the packet and the successful reception probability of the actual packet, generating a real-time power control sequence, which is acceptable under the influence of dynamic vehicle density and external interference. (3) the current fair channel resource allocation strategy is based on the maximum minimization method. This method does not consider the efficiency of resource allocation, and does not guarantee the optimal tracking accuracy for each vehicle. In order to ensure the fair channel resource access for each node and achieve accurate vehicle tracking, a description car is proposed. The utility function of the interaction between the tracking precision and the receiving rate of the receiving node. Based on the utility function of the vehicle tracking precision, a game model with a penalty function is designed by game theory, and the Nash equilibrium point guidance algorithm of the game model is given, which can make the channel visit. In order to achieve Pareto optimality, a distributed competition window adjustment algorithm is designed to guide all nodes to the same channel access probability. In order to achieve Pareto optimality, a distributed competition window adjustment algorithm is designed to guide each node to Pareto Nash by interacting with the competition window information between each node. To achieve the same channel access between nodes and ensure accurate vehicle tracking. (4) the current channel congestion control strategy assumes that the loss of packets is mainly due to collisions generated by multiple nodes in the channel. However, because of the existence of hidden terminals, not all packets sent successfully can be weekly The current channel congestion control is successfully received. The current channel congestion control strategy does not take into account the impact of the hidden terminal. Based on the above background, a dynamic feedback power control strategy is proposed based on the proportional integral differential control method, which is adaptive to the vehicle density and the number of hidden terminals. First, the dynamic vehicle density and hidden terminal shadow are established. Based on the model, a reference model that can predict the rate state of the expected information distribution in different vehicle density is given. Then, the feedback power control model is designed. The model is based on the deviation between the expected information distribution rate and the vehicle actual information distribution rate, and produces a power control sequence. The guidance vehicle achieves the desired vehicle tracking accuracy under the influence of dynamic vehicle density and hidden terminal factors.

【学位授予单位】：大连理工大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：U495;TN929.5

【相似文献】