摘要
针对VANET网络DSDV路由协议依赖于更新消息的周期性传播导致其网络开销增长过大的问题,提出了一种基于深度学习的VANET网络DSDV路由协议GD-DSDV.GD-DSDV路由协议的主要思想是对车辆节点及节点间的链路质量进行评价,并利用机器学习中的梯度下降法对评价指标进行训练,最终得到优化的数据传输路由,从而达到减小网络开销的目的.文中描述了GD-DSDV路由协议的实现过程并从分组平均递交率、路由开销和平均时延等方面进行分析比较.分析结果表明GD-DSDV协议具有比DSDV协议更加优良的性质,可以有效减小路由开销,对现有VANET网络的动态变化具有更强的适应能力.
DSDV routing protocol of VANET network relies on the periodic spread of update information,as a result,its network spending was growing tremendously.In order to deal with this problem,GD-DSDV,a kind of DSDV routing protocol of VANET which was based on deep learning,was proposed.Its main idea was to evaluate nodes as well as links between these nodes of a vehicle,and gradient descent in machine learning(ML)was used to instruct evaluation indicators so as to obtain an optimized routing for data transmission,thereby reducing network spending.The implementation of GD-DSDV was described,several aspects were compared and analyzed,such as average packet delivery rate,routing cost and average delay.Through the above analysis,it shows that GD-DSDV performs betterthan DSDV,which can effectively reduce routing spending and adapt dynamic changes of present VANET network.
作者
赵毅峰
符琦
孙庞博
王飞
Zhao Yifeng;Fu Qi;Sun Pangbo;Wang Fei(School of Computer Science and Engineering,Hunan University of Science and Technology,Xiangtan 411201,China)
出处
《湖南科技大学学报（自然科学版）》
CAS
北大核心
2020年第1期83-89,共7页
Journal of Hunan University of Science And Technology：Natural Science Edition
基金
湖南省教育厅科技重点项目资助(19A174)。
