基于多场景和多目标的车载网络通信NSGA-II方法  

NSGA-Ⅱ Method for Vehicle Network Communication Based on Multiple Scenes and Objectives

作　　者：喻喜平[1] 范文林 YU Xi-ping;FAN Wen-lin(Wuhan Railway Vocational College of Technology,Wuhan 430012,China;School of Economics/Central China Normal University,Wuhan 430079,China)

机构地区：[1]武汉铁路职业技术学院,湖北武汉430012 [2]华中师范大学经济学院,湖北武汉430079

出　　处：《山东农业大学学报（自然科学版）》2019年第6期1076-1079,1103,共5页Journal of Shandong Agricultural University：Natural Science Edition

摘　　要：为了最大限度增加车载自组织网络(VANET)的服务质量,同时实现部署成本最小化,本文提出了一种多目标进化的NSGA-Ⅱ方法,用于确定车载通信网络的路边基础设施位置。该方法除了考虑每个路边单元(RSU)类型的有效通讯范围,还考虑了给定RSU类型可以同时处理的最大车辆数,以及这些车辆接受服务的时长和场景中使用的应用类型。同时,通过交通模式和通信应用(文本/语音/视频)的不同组合对各种场景进行建模。结果表明,所提多目标进化算法能够计算出精确折衷解,在相对超容量(RHV)指标方面,显著优于同类方法。In order to maximize the quality of service of vehicular ad hoc network(VANET)and minimize the deployment cost,a multiple-objective evolutionary NSGA-Ⅱmethod is proposed to determine the roadside infrastructure location of vehicular communication network.The proposed method considers not only the effective communication range of each roadside unit(RSU)type,but also the maximum number of vehicles that can be processed simultaneously for a given RSU type,as well as the service time of these vehicles and the application types used in the scenario.At the same time,various scenarios are modeled through different combinations of traffic modes and communication applications(text/voice/video).The experimental results show that the proposed multiple-objective evolutionary algorithm can calculate the exact compromise solution,and is significantly better than the similar methods in the relative hyper volume(RHV)index.

关 键 词：车载网络通信 NSGA-Ⅱ 

分 类 号：TN91[电子电信—通信与信息系统]