Research On Security Mechanism Of Multiple Attacks For Vehicular Named Data Networking

The Internet of Vehicles(Io V)is one of the key application of intelligent transportation systems,aimed at improving traffic efficiency,road safety,and travel comfort.Named Data Networking(NDN)is a content-centric networking architecture that has been applied to Io V to form Vehicular Named Data Networking(VNDN),which can effectively address the limitations of traditional Io V based on TCP/IP communication.However,VNDN also faces new security issues,including risks associated with the network itself,such as broadcast storms and link disconnections,and various types of attacks such as interest flooding,on-off attacks,and black hole attacks,which seriously affect the security and data sharing efficiency of the network.To address these issues,this paper proposes a multi-attack security mechanism for VNDN,which can improve network performance,monitor network operation status,achieve multi-attack detection,and effectively mitigate the impact of attacks by combining trust evaluation methods.The main research contents of this paper are as follows:(1)A directed routing and forwarding strategy based on content retrieval has is designed.First,the base-level roadside units collect the content names held by the vehicle nodes within their communication range to establish a path index,then the upper-level roadside units aggregate the index information to construct a global hierarchical path index.When a vehicle node requests content,it first queries the path index from the roadside unit and forwards the data according to the index query result.This forwarding strategy effectively solves the broadcast storm problem and improves network link stability.(2)A lightweight attack detection method based on traffic features is proposed.This method,based on the directed routing strategy,can effectively reduce the false positive rate caused by broadcast storms.First,the roadside unit dynamically monitors the traffic status of vehicle nodes within its communication range in the form of a sliding window,extracts relevant traffic features,and then constructs an attack detection framework based on the random forest model.Finally,the particle swarm optimization algorithm is used to optimize parameters to achieve accurate detection of multiple attacks.(3)A multi-attack trust mechanism for VNDN is proposed.Firstly,the attack behavior in the network is analyzed and the attack behavior features are extracted.Then,a trust evaluation scheme is designed based on features such as PIT occupancy rate,interest satisfaction rate,PIT entry existence time,and data packet loss rate,and the node trust is updated using the regularized incomplete beta function.Finally,the attack behavior of the node is identified based on the trust value,and the impact of the attack is mitigated by dropping packets with a certain probability.The ndn SIM platform is used to construct a simulation environment for VNDN,and the proposed methods are evaluated for performance.The experimental results show that the security mechanism proposed in this paper can improve network performance,accurately detect multiple network attack behaviors,and effectively mitigate the impact of attacks to ensure network security.