Research On Security Detection Mechanism For Collusion Attacks In Internet Of Vehicles

As a typical application of Internet of things technology in the transportation industry,the Internet of vehicles(Io Vs)provides users with intelligent and comfortable transportation services and experiences.With the in-depth development and application of the Io Vs,the security risks facing by the Internet of vehicles are increasing,especially collusion attacks are highly destructive,highly concealed,which are very harmful to the Internet of vehicles.At present,there are many challenges such as high detection difficulty,low recognition efficiency and defense difficulty in collusion attacks of the Io Vs.Therefore,aiming at collusion attacks in the Io Vs,combined with the dynamic characteristics of the Internet of vehicles,this paper studies the detection and suppression methods of collusion attacks in the Io Vs.The specific research contents are as follows:(1)Aiming at collusion attacks,this paper proposes a security detection mechanism for collusion attacks in the Io Vs,which is composed of collusion attack detection module and collusion attack inhibition module.The mechanism not only effectively detects collusion vehicle nodes,but also inhibits collusion attacks in the network.(2)In the collusion attack detection module of the mechanism,a collusion attack detection method of reputation is proposed for the reputation collusion attack.This method uses the reputation mechanism to identify a single untrusted vehicle node and applies a multi-criteria detection method availably detect collusion vehicle nodes.For the message collusion attack,a collusion attack detection method of message is proposed.This method uses the time validity and location validity of safety event messages to judge the authenticity of road safety events.Then,by observing the vehicle nodes that send false messages,the two-dimensional frequency detection method is applied to detect collusive vehicle nodes.(3)In the collusion attack inhibition module of the mechanism,a collusion attack inhibition model based on evolutionary game theory is proposed.The model analyzes the characteristics of non-cooperative behavior of collusion vehicle nodes,designs the game revenue matrix,and uses the relationship between the game revenue and the reputation of vehicle nodes to suppress the malicious behavior of collusion vehicle nodes,so as to reduce the harm of collusion attacks to the Internet of vehicles.In this paper,the ONE simulator is used to construct a connected vehicle traffic environment,and the proposed mechanism is embedded into the simulator for performance evaluation.The experimental results show that the security detection mechanism for collusion attacks proposed in this paper can effectively detect collusion vehicle nodes and inhibit the malicious behavior of collusion vehicle nodes.Furthermore,the mechanism guarantees not only the security and reliability of communication but also the operation efficiency of the Internet of vehicles.The mechanism provides technical support for the basic security application of Internet of vehicles.