An Experimental Study Towards Driver Identification For Intelligent And Connected Vehicles

Although there were no concerns about the hacking of vehicles in the past,with the continuous development of information technology and network technology,the latter are being used to replace mechanical systems with electronically-controlled systems,vehicles are now able to communicate with other devices through wired or wireless interfaces such as GPS,Radio,Bluetooth,Wi Fi or even 4G,modern vehicles are developing into Intelligent and Connected Vehicles(ICVs),various security threats are gradually filling up the interiors of many vehicles.In this paper,we present a survey on security threats and protection mechanisms in ICVs.Moreover,this study divides the threat into three different levels and proposes a novel threat assessment model to establish efficient measures for ICVs.What’s more,we introduce the DREAD model to score threats.Our goal is to establish and improve the security evaluation system,refine the risk rating and escort the healthy development of ICVs.Electronic Controls Units(ECU),monitoring and controlling the different subsystems of a car,are interconnected through Controller Area Network(CAN)and compose the global internal network of the ICVs.Actually,the CAN bus also has many shortcomings,specifically,the broadcast transmission on the CAN bus,as well as the unencrypted authentication strategy make the CAN bus vulnerable to various attacks.It is noticed that the final step of almost all attacks in available works,must be at the CAN bus.We can say that the CAN bus is the last line of defense for ICVs.We present in this paper a comprehensive study and use a low-cost experimental device to perform a comprehensive reverse engineering of the automotive CAN bus of ICVs(i.e.,BYD Qin and Luxgen U5).We also designed and completed experiments for replay attacks and spoofing attacks,and the reverse engineering can provide theoretical support for driver identification.All drivers have habits behind the wheel.When a driver is driving,his driving habits will be reflected by vehicle data.While we do not know of attempts by automotive manufacturers or car hackers to violate privacy,a key question is: could they(or their collection and later accidental leaks of data)violate the driver’s privacy? Using ICV data can not only restore the current driving state,but also realize the identification of the driver.Different from previous works about driver identification,we don’t use any manufacturer’s CAN protocol to parse vehicle’s data and don’t use any external sensor data.We directly collect the raw data of the CAN bus,and use machine learning algorithms to classify the data.At first,ten drivers drive the same car(i.e.,Luxgen U5)to carry out a series of driving exercises on the prescribed path,and we use the automotive diagnostic tool to get all real-time CAN bus data from the On-Board Diagnostic(OBD)port.Then,we use Feature scaling and Principal Component Analysis(PCA)algorithm to preprocess the data.Finally,we use K-Nearest Neighbor(KNN)algorithm and Naive Bayes algorithm combined with voting mechanism to successfully identify the driver’s identity.The experimental results show that the recognition rate of ten drivers is 100%.