车联网中继选择算法研究

发布时间：2018-03-28 04:10

本文选题：车联网　切入点：中继选择　出处：《重庆邮电大学》2014年硕士论文

【摘要】：信息领域正发生着由互联网到物联网的新一轮技术革命。车联网作为战略性新兴产业中物联网和智能交通两大领域的重要交集已引起学术界和工业界的极大关注。车联网中，车辆的快速移动以及接入点(Access Point, AP)覆盖范围有限等因素导致部分车辆无法与AP直接进行通信，可通过采用中继车辆(Relay Vehicle, RV)支持源车辆(Source Vehicle, SV)与AP之间的数据转发。在存在多个候选RV的情况下，，如何综合考虑物理信道特性、链路消息碰撞接入时延、RV的负载状况等多因素，选择最佳RV以保障用户通信需求，并实现系统性能优化已成为车联网的重要研究课题。本课题针对存在多个SV和多个可用RV，且存在自私车辆的车联网场景，提出一种基于多目标优化的RV选择算法，通过综合考虑SV的业务需求、信道特性、RV的可用带宽及由于信道竞争导致消息碰撞等因素，分别建模SV和RV的效用函数，并基于各SV及RV性能最优建模多目标优化模型，最后采用理想点法进行SV及RV的最优匹配，以确定最佳RV选择方案。针对多个SV和多个RV合作实现RV优化选择的应用场景，提出一种基于博弈论的RV选择算法，通过综合考虑多种因素对算法性能的影响，建立SV及RV的合作博弈建模，使用二分图最优匹配方法（Kuhn－Munkras算法）对博弈模型进行求解，从而得出对应系统综合性能最优的最佳RV选择方案。针对车联网中两类典型业务，即时延敏感型业务和吞吐量敏感型业务，本文提出了一种基于簇的RV选择算法，分别就簇头选择和簇间切换机制开展研究，提出基于效用函数优化的簇头选择策略以及基于拟切换簇成员及目标簇的效用增益最优的簇切换策略。本文针对车联网具体网络场景及用户业务需求提出RV优化选择策略，可以作为深入研究车联网RV选择技术的参考，具有一定的创新性、理论价值和现实意义。
[Abstract]:The field of information is undergoing a new round of technological revolution from Internet of things to Internet of things. As an important intersection of Internet of things and intelligent transportation in the strategic emerging industries, car networking has attracted great attention from academia and industry. The rapid movement of vehicles and the limited coverage of access points (APs) make some vehicles unable to communicate directly with AP. The relaying vehicle Relay vehicle (RV) can be used to support data forwarding between the source vehicle Source vehicle (SVV) and AP. In the case of multiple candidate RVs, how to consider the physical channel characteristics, link message collision access delay, RV load and other factors, etc. Choosing the best RV to protect the user's communication requirements and realize the system performance optimization has become an important research topic of vehicle networking. In this paper, a multi-objective optimization based RV selection algorithm is proposed for vehicle networking scenarios with multiple SV and available RVs and selfish vehicles. By considering the business requirements of SV comprehensively, this paper proposes a new RV selection algorithm based on multi-objective optimization. Based on the available bandwidth of RV and channel competition, the utility function of SV and RV is modeled, and the multi-objective optimization model is built based on each SV and RV. Finally, the optimal matching of SV and RV is carried out by using the ideal point method to determine the optimal RV selection scheme. In this paper, a game theory based RV selection algorithm is proposed for the application of multiple SV and multiple RV cooperation to realize RV optimal selection. The cooperative game modeling of SV and RV is established by considering the influence of many factors on the performance of the algorithm. Kuhn-Munkras algorithm is used to solve the game model, and the optimal RV selection scheme with optimal comprehensive performance of the corresponding system is obtained. In this paper, a clust-based RV selection algorithm is proposed for two types of typical services, instant delay sensitive services and throughput sensitive services in vehicle networking. Cluster head selection and inter-cluster switching mechanism are studied respectively. A cluster head selection strategy based on utility function optimization and a cluster handover strategy based on optimal utility gain of quasi-switched cluster members and target clusters are proposed. This paper proposes a RV optimal selection strategy for the specific network scenarios and user business requirements of vehicle networking, which can be used as a reference for the in-depth study of RV selection technology in vehicle networking. It has certain innovation, theoretical value and practical significance.
【学位授予单位】：重庆邮电大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN929.5;TP391.44

【参考文献】