Font Size: a A A

Research On CAN Bus Attack And Defense Detection Technology Based On Intelligent Network Of Car

Posted on:2018-07-04Degree:MasterType:Thesis
Country:ChinaCandidate:H YangFull Text:PDF
GTID:2322330536957345Subject:Information and Communication Engineering
Abstract/Summary:PDF Full Text Request
With the development of the Internet of Things industry,the breakthrough technology proposed in this filed,the intelligent made cars begin to enter people's lives.The vehicles,mobile devices,infrastructures,and the owners can easily exchange data and intercommunicate with each other through the cloud Services.Internet of Things technology greatly improves people's driving experience,triggering a lifestyle change,which is the future direction of development of the car.At present,as more and more electronic controlled equipment applied to the automobile,the the different electronic modules need to be cooperated with each other.The cooperation will make the process of vehicle run smoothly and the form a complicated communication network between the smart cars.With the increasing scale of the network and the growing number of electronic devices,the types and quantity of the software running on the car are increasing.The robustness and reliability of code execution and network information security issues follow.The hacker can attack the vehicle software through the inside and outside communication path of the vehicle,and send abnormal message to the CAN bus,to affect the control system of the vehicle.These potential risks are not just stealing information,money,they truly pose a threat to our lives.Through the abnormal detection,security and other technologies to protect the car CAN bus information security is of great significance.In this paper,according to the current CAN network used in intelligent networked vehicles,the security problems such as reproducible,tampering can be analyzed and dissected in depth,and the security vulnerabilities in CAN bus are analyzed.Designed the CAN bus for the attack method,describes how to reverse the bus message information,in order to achieve the purpose of controlling the car.Then,based on the analysis of attack means and bus message structure,an anomaly detection model for CAN bus is proposed.The main results are as follows:1)On the basis of researching and analyzing the CAN communication protocol of vehicle,the paper uses the security vulnerability existing in CAN network to design effective attack method.The use of reverse technology to analyze CAN packet data packets,crack the car message command information,in order to achieve the purpose of controlling the car.Car CAN network is different from the traditional computer network,the packet does not have the computer network IP packet such as the source address and destination address.Based on the characteristics of CAN packet structure,this paper proposes an exception detection model framework,which detects anomalies from the packet ID and message data bits respectively,and can detect the attack on CAN network.2)An anomaly detection system based on feature and information entropy was proposed for CAN message identification ID bit.By detecting the probability distributions of different message IDs in the CAN bus,the information entropy of the CAN bus was calculated and the information entropy of the normal CAN bus was taken as the threshold value of the anomaly detection.At the same time the normal bus CAN ID as a white list,identify abnormal bus CAN ID characteristics.The results of the simulation show that the anomaly detection method based on information entropy and feature combination can effectively monitor flooding attacks,replay attacks,and a small number of high priority packet attacks.3)An anomaly detection system based on support vector machine(SVM)was proposed for the data bits of the on board bus message.According to the characteristics of the data bits,the data bits of the on-board bus messages were divided into eight characteristics.Then we distinguished the normal data packets from the abnormal data packets according to the support vector machine(SVM)detection method.The results of the show that the anomaly detection system based on support vector machine has a good detection effect on the tampering attack of bus message data.As a new research field,this paper has made some preliminary explorations on CAN bus vulnerability excavation,intrusion attack,anomaly detection,security protection and so on.The result is an important reference for the further research and application of CAN bus safety protection.
Keywords/Search Tags:Intelligent network of car, Internet of things, CAN bus information, Entropy, Support Vector Machines, Anomaly detection
PDF Full Text Request
Related items