Research On Cooperative Downloading For Internet Of Vehicle Based On Trust

With development of Internet of Vehicle(IOV),application demands of IOV have been transformed into rich and colorful network services.However,expansion of application scope and increase of vehicle number have led to a sharp increase in IOV services demands,and because IOV uses wireless communication for file transmission,the communication of the vehicle is easily interfered by attackers.How to distribute content efficiently and securely has become an important research topic in the field of IOV.In order to solve problems of low security and insufficient downloading capacity in IOV,this thesis proposes a trust-based IOV cooperative downloading allocation scheme,where trust includes credibility and reliability.Main contents are as follows:1)Credibility calculation method based on vehicle interaction.To address the problem that existing IOV trust computing methods yield inaccurate trust values due to ignoring the characteristics of cooperative downloading,this thesis proposes a credibility computing method combined with characteristics of cooperative downloading vehicles.A trust management model based on vehicle interaction is constructed.The cooperative downloading protocol is designed and the security of the protocol is analyzed.The credibility is calculated by evaluation and interactive history of cooperative downloading,combined with characteristics of multiple participating vehicles in cooperative downloading.Simulation verifies the superiority of the proposed method in terms of identifying malicious vehicles.2)Reliability calculation method based on predicted speed.To address the problem that current trust computing methods of IOV ignore the impact of downloading capacity by vehicle speed prediction physical properties,which results in low cooperative downloading capacity,a reliability computing method based on predicted speed is proposed.A traffic flow speed prediction method considering the similarity of time influence coefficient is proposed to realize traffic flow speed prediction,and the adjacency matrix is constructed by the similarity of time influence coefficient and distance.The feature matrix with spatial features is obtained by graph convolution,and the prediction speed value is obtained by SMasked-Transformer(SMT)model based on improved masked model.Reliability of cooperative downloading vehicle is calculated based on the predicted speed.Simulation verifies the advantages of the proposed traffic flow speed prediction method,and the advantages of incorporating predicted speed-based reliability into trust calculation.3)Cooperative downloading allocation scheme based on trust.To address the problem of insufficient downloading capacity and lack of security of current cooperative downloading allocation schemes,a trust-based cooperative downloading allocation method is proposed.The Lagrange duality method is adopted,downloading quantity is taken as optimization objective function,and constraint conditions are constructed from aspects of maximum transmission rate and maximum forwarding rate,and the optimal solution of the allocation scheme is obtained by iterative Lagrange duality method.Simulation verifies the advantages of the proposed scheme in terms of downloading capacity.