基于车联网的交通与通信防拥机制设计与验证
发布时间：2019-06-13 13:31
【摘要】：近年来,由于我国经济水平的不断提高,汽车数量日益增多,城市道路承受着巨大的压力,运用车联网技术能够有效地缓解严重的交通拥堵问题,但是随着车联网的快速发展和网络规模的扩大,如何有效地利用现有的网络资源,通过动态的路由选择实时地控制网络拥塞以提高车联网的通信效率显得尤为重要。本文将针对于城市道路中的交通拥堵问题与车联网通信中的网络拥塞问题,结合车联网技术与蚁群算法,分别从交通流与信息流两个维度研究与设计相应的防拥机制:基于改进蚁群算法的交通防拥机制:本文利用车联网中的V2I通信技术,通过改进蚁群算法进行启发式动态路径规划。算法使用车辆蚂蚁采集实时交通数据以预测短期内交通状况,并综合复杂动态的交通因素对状态转移概率及信息素计算规则进行改进,不仅能够规避正常的交通拥堵,还客观考虑和处理了由交通事故,道路施工,天气条件造成的特殊拥堵问题。此外,算法通过周期性生成数据包蚂蚁为道路网络中每个起始点对探索多条备选最优路径,并依据不同的车辆类型使用优先调度和先来先服务结合策略进行路径分配,使车辆合理分流。基于SDN-V的通信防拥机制:针对车联网节点的高移动性,网络拓扑的高动态性,连接间断性等问题,本文将SDN技术应用至车联网,设计分层SDN-V网络架构,分离车辆节点中的控制与数据平面,使车辆只需进行高效的数据转发工作。设置集中式控制器掌控全局网络拓扑结构,并调用改进的蚁群算法进行动态路由选择,以达到保持网络负载平衡,控制网络拥塞,提高车联网通信效率的目的。本文利用交通流仿真器SUMO与网络仿真器NS3双向耦合进行仿真实验,实验结果分别验证了基于改进蚁群算法的交通防拥机制对于缓解交通拥堵问题的实用性与基于SDN-V的通信防拥机制在控制网络拥塞,提高网络性能方面的有效性。
[Abstract]:In recent years, due to the continuous improvement of the economic level of our country, the number of cars is increasing day by day, the urban roads are under great pressure, the use of vehicle networking technology can effectively alleviate the serious traffic congestion problem, but with the rapid development of vehicle networking and the expansion of network scale, how to effectively make use of the existing network resources. It is particularly important to control network congestion in real time through dynamic routing to improve the communication efficiency of vehicle network. In this paper, aiming at the traffic congestion problem in urban roads and the network congestion problem in vehicle network communication, combined with vehicle networking technology and ant colony algorithm, the corresponding anti-hugging mechanism is studied and designed from two dimensions of traffic flow and information flow: traffic congestion mechanism based on improved ant colony algorithm: this paper uses V2I communication technology in vehicle networking to carry out heuristic dynamic path planning through improved ant colony algorithm. The algorithm uses vehicle ants to collect real-time traffic data to predict traffic conditions in the short term, and improves the state transition probability and pheromone calculation rules by synthesizing complex and dynamic traffic factors, which can not only avoid normal traffic congestion, but also objectively consider and deal with the special congestion problems caused by traffic accidents, road construction and weather conditions. In addition, the algorithm periodically generates packet ants to explore multiple alternative optimal paths for each starting point in the road network, and uses priority scheduling and first come first service combination strategy according to different vehicle types to distribute vehicles reasonably. Communication anti-embrace mechanism based on SDN-V: in order to solve the problems of high mobility of vehicle network node, high dynamics of network topology, intermittent connection and so on, this paper applies SDN technology to vehicle network, designs hierarchical SDN-V network architecture, separates control and data plane in vehicle node, so that vehicles only need to carry out efficient data forwarding. The centralized controller is set up to control the global network topology, and the improved ant colony algorithm is called for dynamic routing in order to maintain the load balance of the network, control the network congestion and improve the communication efficiency of the vehicle network. In this paper, the two-way coupling of traffic flow simulator SUMO and network simulator NS3 is used to carry out simulation experiments. The experimental results verify the practicability of traffic congestion prevention mechanism based on improved ant colony algorithm and the effectiveness of communication anti-hugging mechanism based on SDN-V in controlling network congestion and improving network performance.
【学位授予单位】：华东师范大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U495
本文编号：2498561
[Abstract]:In recent years, due to the continuous improvement of the economic level of our country, the number of cars is increasing day by day, the urban roads are under great pressure, the use of vehicle networking technology can effectively alleviate the serious traffic congestion problem, but with the rapid development of vehicle networking and the expansion of network scale, how to effectively make use of the existing network resources. It is particularly important to control network congestion in real time through dynamic routing to improve the communication efficiency of vehicle network. In this paper, aiming at the traffic congestion problem in urban roads and the network congestion problem in vehicle network communication, combined with vehicle networking technology and ant colony algorithm, the corresponding anti-hugging mechanism is studied and designed from two dimensions of traffic flow and information flow: traffic congestion mechanism based on improved ant colony algorithm: this paper uses V2I communication technology in vehicle networking to carry out heuristic dynamic path planning through improved ant colony algorithm. The algorithm uses vehicle ants to collect real-time traffic data to predict traffic conditions in the short term, and improves the state transition probability and pheromone calculation rules by synthesizing complex and dynamic traffic factors, which can not only avoid normal traffic congestion, but also objectively consider and deal with the special congestion problems caused by traffic accidents, road construction and weather conditions. In addition, the algorithm periodically generates packet ants to explore multiple alternative optimal paths for each starting point in the road network, and uses priority scheduling and first come first service combination strategy according to different vehicle types to distribute vehicles reasonably. Communication anti-embrace mechanism based on SDN-V: in order to solve the problems of high mobility of vehicle network node, high dynamics of network topology, intermittent connection and so on, this paper applies SDN technology to vehicle network, designs hierarchical SDN-V network architecture, separates control and data plane in vehicle node, so that vehicles only need to carry out efficient data forwarding. The centralized controller is set up to control the global network topology, and the improved ant colony algorithm is called for dynamic routing in order to maintain the load balance of the network, control the network congestion and improve the communication efficiency of the vehicle network. In this paper, the two-way coupling of traffic flow simulator SUMO and network simulator NS3 is used to carry out simulation experiments. The experimental results verify the practicability of traffic congestion prevention mechanism based on improved ant colony algorithm and the effectiveness of communication anti-hugging mechanism based on SDN-V in controlling network congestion and improving network performance.
【学位授予单位】：华东师范大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U495
【参考文献】
相关期刊论文 前8条
1 柴燕涛;董德尊;张鹤颖;朱成阳;廖湘科;;基于SDN架构的高性能网络拥塞避免策略[J];计算机工程与科学;2016年01期
2 贾建斌;陈颖文;徐明;;基于预测的机会车载网络中继选择策略研究[J];软件学报;2015年07期
3 胡晓宇;;SDN网络转发机制研究和应用场景分析[J];电子技术;2015年05期
4 赵娜;袁家斌;徐晗;;智能交通系统综述[J];计算机科学;2014年11期
5 赵海;田婷婷;樊景超;刘鹏;;一种分布式车联网多跳信息广播协议[J];东北大学学报(自然科学版);2014年09期
6 杜衡吉;李勇;;蚁群算法中参数设置对其性能影响的研究[J];现代计算机(专业版);2012年13期
7 李丽君;刘鸿飞;杨祖元;葛利嘉;黄席樾;;车用自组网信息广播[J];软件学报;2010年07期
8 范辉,华臻,李晋江,原达;一类最短路径问题蚂蚁算法求解[J];计算机工程与应用;2004年33期
，本文编号：2498561
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2498561.html
