Research on intelligent prediction method of internet of vehicles security performance based on GConvGRU-GAT
LI Tiancheng
XUE Peiyou
XU Lingwei
Abstract：With the continuous development of 5G technology and the Internet of Things,the Internet of Vehicles business is emerging in an endless stream,and its application scenarios are becoming more and more complex and changeable.In the communication process of Internet of Vehicles,the security perform-ance is affected by the complex and ever-changing environment of the vehicle networking,making it diffi-cult to make real-time and accurate predictions.Therefore,this paper proposes an intelligent prediction method for the security performance of the Internet of Vehicles based on GConvGRU-GAT in this paper.Firstly,under the N-Nakagami channel,a model of the secure communication system of the Internet of Vehicles was established,and the security performance was analyzed by detecting the signal-to-noise ratio of the communication link,and the influence of different influencing factors on the security performance was effectively analyzed,and the communication dataset was constructed.In order to better predict the communication system in real time,this paper integrates Graph Attention Network(GAT),Graph Conv-olutional Network(GCN)and Gated Recurrent Unit(GRU)to design a mobile security performance pre-diction network model based on GConvGRU-GAT by integrating Graph Attention Network(GAT),Graph Convolutional Network(GCN)and Gated Recurrent Unit(GRU).Experimental results show that compared with other algorithms,the GConvGRU-GAT algorithm has better prediction effect,and the per-formance is improved by 84.6%compared with the GCN model.
Keywords：internet of vehiclessecure communicationssecurity performance predictiongraph attention neural network
Publication Date：2025-08-01
Online Publishing Date：2026-08-28(First online date of this platform, not the publication date of the document)
Pages：10( 507-515,526 )