基于RSU辅助和自适应分簇的异构车载网络选择方法  

Heterogeneous Vehicular Network Selection Method Based on RSU-assisted and Adaptive Clustering

作　　者：聂雷 胡字升 鲍海洲 NIE Lei;HU Zisheng;BAO Haizhou(School of Computer Science and Technology,Wuhan University of Science and Technology,Wuhan 430065,Hubei,China;Hubei Province Key Laboratory of Intelligent Information Processing and Real-Time Industrial System,Wuhan 430065,Hubei,China)

机构地区：[1]武汉科技大学计算机科学与技术学院,湖北武汉430065 [2]智能信息处理与实时工业系统湖北省重点实验室,湖北武汉430065

出　　处：《计算机工程》2025年第3期162-171,共10页Computer Engineering

基　　金：国家自然科学基金(61802286);湖北省教育厅科学研究计划青年项目(Q20221108);武汉科技大学智能信息处理与实时工业系统湖北省重点实验室基金(ZNXX2022009)。

摘　　要：异构车载网络(HVN)环境中车载终端用户的服务质量(QoS)体验高度依赖于网络选择方法。针对单车辆选择接入网络容易导致优质网络阻塞和网络资源分配不均的问题,提出一种基于路侧单元(RSU)辅助和自适应分簇的异构车载网络选择方法,适用于融合5G/6G的城市HVN环境。首先,该方法借助RSU的计算存储能力评估候选网络的性能,分别利用层次分析法(AHP)和熵权法(EWM)计算候选网络的主观权重和客观权重,在利用简单加权法计算其综合权重后,根据车辆的业务类型采用基于直觉模糊集的灰色关联度分析(IFS-GRA)法对候选网络排序;然后,在网络负载较高时对车辆进行自适应分簇,利用分层结构有效降低网络的拥塞概率;最后,车辆从RSU处获取满足其业务需求的最优接入网络。实验结果表明,该方法相较CHSO-GRA和MANSA方法分别减少了30.72%和9.57%的网络切换次数,相较DUVC和CHSO-GRA方法分别增加了8.01%和39.36%的吞吐量,提高了网络资源的利用率,也实现了网络的负载均衡。The Quality of Service(QoS)experience of vehicular end-users in a Heterogeneous Vehicular Network(HVN)environment is highly dependent on the network selection method.Selecting the access network from individual vehicles can easily lead to the blocking of the quality network and the uneven allocation of network resources.To address this problem,a network selection method based on Road Side Unit(RSU)-assisted and adaptive clustering,called RANS,is proposed and applied to urban HVN environments integrating 5G/6G.First,the method evaluates the performance of candidate networks using the computation and storage capacity of RSU and calculates the subjective and objective weights of candidate networks based on the Analytic Hierarchy Process(AHP)and Entropy Weighting Method(EWM),respectively.After calculating the comprehensive weights using a simple weighting method,the candidate networks are ranked according to service type using the Intuitionistic Fuzzy Set-based Grey Relational Analysis(IFS-GRA)method.The adaptive clustering of vehicles is performed when the network load is high,and a hierarchical structure is utilized to effectively reduce the congestion probability of the network.Finally,each vehicle accesses the optimal network from the RSUs that satisfy the service requirements.The experimental results reveal that the proposed RANS method reduces the number of network handovers by 30.72%and 9.57%compared to the CHSO-GRA and MANSA methods,respectively,thereby increasing the throughput by 8.01% and 39.36% compared to the DUVC and CHSO-GRA methods,respectively,achieving better load balancing while improving network resource utilization.

关 键 词：路侧单元辅助 自适应分簇 网络选择 负载均衡 5G/6G网络 异构车载网络 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]