基于连通概率的车载通信路由优化技术研究

发布时间：2018-04-22 04:01

本文选题：车载自组织网络 + 连通概率　；参考：《南京邮电大学》2017年硕士论文

【摘要】：近几年来,车辆的急剧增多给交通带来了不少问题,这使得车联网技术越来越受到人们的关注。车辆通过车载终端设备进行自由组网通信,实现交通安全、娱乐等消息的高效传输,以解决交通安全和娱乐问题。然而,由于车载自组织网络与移动自组织网络不同的网络特性,各种各样的应用所需的指标也不同,这使得设计出适合所有场景的路由机制是很困难的,所以要针对不同的应用场景设计不同的路由机制。相比于高速公路,城市公路存在路况复杂、车辆分布不均匀等问题,降低了车载网络中数据包的传递成功率。本文研究的是城市场景下的基于连通概率的地理位置路由优化算法,研究内容包括以下几个方面:首先,本文针对城市应用场景,对路径的连通概率进行预测与建模,将路段分为元胞,分析元胞中车辆的分布,提出了连通概率上下界算法,最后根据独立同分布的思想,得出路径上的连通概率。仿真结果表明了连通概率与车辆密度、通信范围之间的关系。其次,本文在上述研究基础上改进了基于连通概率的地理位置路由协议。本文选择城市道路的十字路口作为主干路由,其选择的依据是端到端时延、跳数、误码率以及本文提出的连通概率。将上述指标的分析表达式组合成带约束条件的最优化问题,再利用遗传算法的选择、交叉、突变等方法得到最优解(主干路由)。随后,把主干路由加到源节点待传输的数据包头部等待实际的转发。最后,数据包在上述已选主干路由上进行实际的传输。在相邻十字路口之间传输时,为了削弱车辆快速移动所带来的影响,本文针对直行道路和十字路口两种情况提出了基于邻居节点优先级的算法。根据邻居节点的位置、速度、方向等因素给其分配优先级,选择优先级高的邻居节点作为下一跳转发节点。仿真和分析表明,基于主干路由的连通概率地理位置路由协议的开销增大了,但是端到端时延、跳数、误码率和分组到达率的性能得到提升。
[Abstract]:In recent years, the rapid increase of vehicles has brought a lot of traffic problems, which makes the vehicle networking technology more and more attention. The vehicle carries on the free network communication through the vehicle terminal equipment, realizes the traffic safety, the entertainment and so on information high efficiency transmission, in order to solve the traffic safety and the entertainment question. However, due to the different network characteristics between vehicle-mounted ad hoc networks and mobile ad hoc networks, various applications require different metrics, which makes it difficult to design routing mechanisms suitable for all scenarios. Therefore, different routing mechanisms should be designed for different application scenarios. Compared with expressway, urban highway has many problems, such as complex road condition and uneven distribution of vehicles, which reduces the success rate of packet transmission in vehicular network. In this paper, we study the geographical location routing optimization algorithm based on connectivity probability in urban scenarios. The research includes the following aspects: firstly, this paper predicts and models the connectivity probability of the path in the urban application scenario. The section is divided into cells, the distribution of vehicles in the cell is analyzed, and the connected probability upper and lower bound algorithm is proposed. Finally, according to the idea of independent same distribution, the connected probability on the path is obtained. Simulation results show the relationship between connectivity probability and vehicle density and communication range. Secondly, this paper improves the geographic location routing protocol based on connectivity probability based on the above research. In this paper, the intersections of urban roads are chosen as trunk roads, which are based on end-to-end delay, hops, bit error rate and connectivity probability proposed in this paper. The analytical expressions of the above indexes are combined into an optimization problem with constraints, and the optimal solution is obtained by using the methods of selection, crossover and mutation of genetic algorithm. Then, the trunk road is added to the packet header to be transmitted by the source node to wait for the actual forwarding. Finally, the packet is actually transmitted on the selected trunk road. In order to reduce the influence of the rapid movement of vehicles between adjacent intersections, an algorithm based on neighbor node priority is proposed in this paper. According to the location, speed and direction of neighbor node, the neighbor node with high priority is selected as the next hop forwarding node. Simulation and analysis show that the overhead of the connected probabilistic geographic location routing protocol based on trunk routing is increased, but the performance of end-to-end delay, hops, bit error rate and packet arrival rate is improved.
【学位授予单位】：南京邮电大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN929.5;U495

【相似文献】