车联自组织网络中退避算法的研究
发布时间：2018-09-10 07:04
【摘要】：随着交通的飞速发展以及人们生活水平的大幅提高,车辆几乎成为了每家每户的必备品。然而,在车辆为我们带来便利的同时,也给交通增添了不可忽视的压力。传统的交通资源十分有限,当面对如此飞速增长的车辆数目时,就不能及时高效工作,甚至造成系统瘫痪,由此车联网应运而生。车联网是一种特殊的无线自组织网络,是实现交通智慧化的基础之一,它主要懫取车辆与车辆间,以及车辆与路边设备间的通信模式,并承载安全控制数据与业务数据的传输,为驾驶者提供实时准确的车况路况信息以及交通安全服务。但车联网的特殊性也决定了它在通信过程中面临的一系列挑战,尤其是在媒体接入控制协议(MAC)中,车辆在抢占信道时采取的退避算法,严重影响着信息传输性能。因此,本文针对车联网中广播模式下,IEEE 802.11p MAC协议的退避算法进行研究,主要工作如下:本文分析了车联网MAC协议的扩展需求,并总结了自组织网络中常用的退避算法及其缺陷,由此提出了基于改进MARKOV模型的竞争窗口调整方法。在传统的MARKOV模型中引入空闲状态,使模型更加适用于车辆复杂的通信环境及场景,并利用车辆密度来衡量交通拥塞程度,结合平稳分布及泰勒公式,推导出车辆所维护的竞争窗口值及其周围车辆密度的关系式。最终,结合IEEE 802.11p MAC协议中传统的固定竞争窗口值,设计出改进的退避算法,实现窗口值的自适应动态调整,优化了 802.11p中固定窗口值的缺陷,使车辆在抢占信道时采取更加智能准确的退避方法。仿真结果表明,本方案有效地减小了广播信息的碰撞率,并维持了较小的时延。基于车辆区别于其他通信节点的独有特性,以及其多样的通信环境,本文针对车联网设计了具备更强自适应性的退避方案:基于分类预测机制的退避方法。该方案首先提出分类策略,以车辆的邻居节点数,速度和停止时间为准则,建立属性集,对车辆状态进行分类,去冗余化。并引入反馈机制,使属性集具备自适应调整性,从而保证不同场景下的分类准确度。然后,利用上述准则,求取属性集所对应的综合竞争窗口值,由此生成窗口参考表。最后,利用车辆的历史行驶状态,建立隐马尔可夫预测机制(HMM),实现对车辆下一时刻竞争窗口的预测,从而确定退避时间。该方案充分考虑了车辆独有的特点及其复杂且受限的通信环境,建立了具备自适应调整及预测机制的退避方法,并通过仿真证明,该方案能使信息实时高效地传输,优化了车联网广播模式下的性能。
[Abstract]:With the rapid development of traffic and the improvement of people's living standards, vehicles have become a must for every household. However, while the vehicle brings convenience to us, it also adds the pressure that can not be ignored to the traffic. The traditional transportation resources are very limited, when faced with such a rapid increase in the number of vehicles, it can not work in time and efficiently, or even lead to system paralysis, so the vehicle network came into being. Vehicle networking is a special wireless ad hoc network, which is one of the bases of realizing traffic intelligence. It mainly takes the communication mode between vehicle and roadside equipment, and carries the transmission of safety control data and service data. Provide drivers with real-time and accurate traffic information and traffic safety services. However, the particularity of vehicle networking also determines a series of challenges it faces in the communication process, especially in the media access control protocol (MAC), the Backoff algorithm adopted by vehicles in preempting the channel seriously affects the performance of information transmission. Therefore, this paper studies the Backoff algorithm of IEEE 802.11p MAC protocol in the broadcast mode of vehicle networking. The main work is as follows: this paper analyzes the extended requirements of MAC protocol, and summarizes the common Backoff algorithms and their defects in the Ad Hoc Network. Based on the improved MARKOV model, a competition window adjustment method is proposed. The idle state is introduced into the traditional MARKOV model, which makes the model more suitable for the complex communication environment and scene of the vehicle, and uses the vehicle density to measure the degree of traffic congestion, combined with the stationary distribution and Taylor formula. The relation between the competing window value maintained by the vehicle and the density of the surrounding vehicle is derived. Finally, combined with the traditional fixed contention window value in IEEE 802.11p MAC protocol, an improved Backoff algorithm is designed to realize the adaptive dynamic adjustment of window value, and the defect of fixed window value in 802.11p is optimized. Make the vehicle take more intelligent and accurate Backoff method when preempting the channel. Simulation results show that the scheme can effectively reduce the collision rate of broadcast information and maintain a small delay. Based on the unique characteristics of vehicle which is different from other communication nodes and its diverse communication environment, this paper designs a more adaptive Backoff scheme for vehicle networking: a Backoff method based on classification prediction mechanism. In this scheme, a classification strategy is proposed, which takes the number of neighbor nodes, speed and stopping time as criteria, establishes attribute set, classifies vehicle status and deredundancy. A feedback mechanism is introduced to make the attribute set adaptive to ensure the classification accuracy in different scenarios. Then, the comprehensive competing window value corresponding to the attribute set is obtained by using the above criteria, and the window reference table is generated. Finally, the hidden Markov prediction mechanism (HMM),) is established to predict the next competitive window of the vehicle by using the historical driving state of the vehicle, so as to determine the Backoff time. In this scheme, the unique characteristics of the vehicle and its complex and limited communication environment are fully considered, and a Backoff method with adaptive adjustment and prediction mechanism is established. The simulation results show that the scheme can transmit information in real time and efficiently. The performance of vehicle network broadcast mode is optimized.
【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN929.5;U495
[Abstract]:With the rapid development of traffic and the improvement of people's living standards, vehicles have become a must for every household. However, while the vehicle brings convenience to us, it also adds the pressure that can not be ignored to the traffic. The traditional transportation resources are very limited, when faced with such a rapid increase in the number of vehicles, it can not work in time and efficiently, or even lead to system paralysis, so the vehicle network came into being. Vehicle networking is a special wireless ad hoc network, which is one of the bases of realizing traffic intelligence. It mainly takes the communication mode between vehicle and roadside equipment, and carries the transmission of safety control data and service data. Provide drivers with real-time and accurate traffic information and traffic safety services. However, the particularity of vehicle networking also determines a series of challenges it faces in the communication process, especially in the media access control protocol (MAC), the Backoff algorithm adopted by vehicles in preempting the channel seriously affects the performance of information transmission. Therefore, this paper studies the Backoff algorithm of IEEE 802.11p MAC protocol in the broadcast mode of vehicle networking. The main work is as follows: this paper analyzes the extended requirements of MAC protocol, and summarizes the common Backoff algorithms and their defects in the Ad Hoc Network. Based on the improved MARKOV model, a competition window adjustment method is proposed. The idle state is introduced into the traditional MARKOV model, which makes the model more suitable for the complex communication environment and scene of the vehicle, and uses the vehicle density to measure the degree of traffic congestion, combined with the stationary distribution and Taylor formula. The relation between the competing window value maintained by the vehicle and the density of the surrounding vehicle is derived. Finally, combined with the traditional fixed contention window value in IEEE 802.11p MAC protocol, an improved Backoff algorithm is designed to realize the adaptive dynamic adjustment of window value, and the defect of fixed window value in 802.11p is optimized. Make the vehicle take more intelligent and accurate Backoff method when preempting the channel. Simulation results show that the scheme can effectively reduce the collision rate of broadcast information and maintain a small delay. Based on the unique characteristics of vehicle which is different from other communication nodes and its diverse communication environment, this paper designs a more adaptive Backoff scheme for vehicle networking: a Backoff method based on classification prediction mechanism. In this scheme, a classification strategy is proposed, which takes the number of neighbor nodes, speed and stopping time as criteria, establishes attribute set, classifies vehicle status and deredundancy. A feedback mechanism is introduced to make the attribute set adaptive to ensure the classification accuracy in different scenarios. Then, the comprehensive competing window value corresponding to the attribute set is obtained by using the above criteria, and the window reference table is generated. Finally, the hidden Markov prediction mechanism (HMM),) is established to predict the next competitive window of the vehicle by using the historical driving state of the vehicle, so as to determine the Backoff time. In this scheme, the unique characteristics of the vehicle and its complex and limited communication environment are fully considered, and a Backoff method with adaptive adjustment and prediction mechanism is established. The simulation results show that the scheme can transmit information in real time and efficiently. The performance of vehicle network broadcast mode is optimized.
【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN929.5;U495
【参考文献】
相关期刊论文 前5条
1 孙伟;张和生;潘成;杨军;白U，
本文编号：2233734
本文链接：https://www.wllwen.com/kejilunwen/daoluqiaoliang/2233734.html
