Research On Vehicle CAN Bus Information Security And Detection Technology

With the development of the IoT and the progress of science and technology,traditional cars are gradually becoming intelligent and networked.Cars are no longer simple means of transportation,but become a huge mobile communication carrier.Vehicle-to-vehicle,vehicleto-road and other components constitute a huge information system,which makes vehicles become a new target of network security attacks.In recent years,car attacks have occurred frequently,and the problem of automobile information security has erupted in an all-round way.Therefore,it is urgent to study the problem of automobile information security.CAN bus is the most widely used communication network in automobile.At the beginning of the design,CAN bus did not fully consider its security.With the rapid development of Internet of vehicles,CAN bus has exposed serious information security problems.Attackers can attack CAN bus by means of replay,traversal,flood,fuzzy test and other means,which seriously threatens the safety of people and vehicles and pedestrians.This paper focuses on the information security and anomaly detection of CAN bus.(1)Based on the knowledge of the physical architecture,message structure and communication mechanism of CAN bus,this paper studied the security threats faced by IVI,on-board t-box,App,Tsp cloud platform and physical interface OBD,as well as the security vulnerabilities caused by the vulnerability of CAN bus itself,and proposed an information security solution for CAN bus(2)Studied the working mechanism of OBD diagnostic interface of networked vehicles and the connection mode of CAN bus,and deeply studied replay,flood,traversal,fuzzy test,injection and other attack modes.Simulate the attacker attacks on cars.By replaying the attack to turn the car's engine speed dial;By flooding attack,make the car dashboard dysfunction;By traversing the attack,controlling the door,the amount of electricity,etc.(3)In this paper,intrusion detection technology and anomaly detection technology are studied,and proposed an anomaly detection algorithm based on information entropy for CAN bus of networked vehicles,and established an anomaly detection model for CAN bus of networked vehicles.The feasibility of this method to detect CAN bus anomaly is verified by the experimental analysis of anomaly detection in automobile.