详情

原文传递 Secure and Private Sensing for Driver Authentication and Transportation Safety

题名：	Secure and Private Sensing for Driver Authentication and Transportation Safety
作者：	Voris, J.; Artan, N. S.; Li, W.
关键词：	Driver authentication##Transportation safety##Vehicular security##Intelligent transportation systems##Smart cities##Machine learning##
摘要：	Recent technology trends have allowed affordable and efficient collection of driver data. This has enabled a variety of potential applications, including more accurate pricing determinations for insurance and finer grained traffic planning for improved public safety. Although this technological growth provides for a wealth of new opportunities, given the safety implications of driving, there are many security and privacy issues that must be considered for their deployment. For instance, some applications require access to a vehicle's engine via a debug interface, known as On-Board Diagnostics (OBD-II), which may provide a vector for attack. Other systems may involve GPS tracking, which can potentially violate a driver's privacy. Our research seeks to find solutions to these shortcomings by using local sensing and monitoring to support the development of new driver devices and applications, such as driver authentication, while preserving vehicular security and privacy. We propose a novel approach to data collection for commercial driving applications and vehicle safety that puts users in control of how their information is used. By collecting local driving data in a manner that is decoupled from critical car components and Internet connections, our system can support transportation applications, such as driver authentication, without sacrificing vehicle security or driver privacy. The legitimate driver of a vehicle traditionally gains authorization to access their vehicle via tokens such as ignition keys, some modern versions of which feature RFID tags. However, this token-based approach is not capable of detecting all instances of vehicle misuse. Technology trends have allowed for affordable and efficient collection of various sensor data in real time from the vehicle, its surroundings, and devices carried by the driver, such as smartphones. This report describes the result of our research effort investigating the use of this sensory data to actively identify and authenticate the driver of a vehicle by determining characteristics which uniquely categorize individuals’ driving behavior. Our approach is capable of continuously authenticating a driver throughout a driving session, as opposed to alternative approaches which are either performed offline or as a session starts. This means our modeling approach can be used to detect mid-session driving attacks, such as carjacking, which are beyond the scope of alternative driver authentication solutions. A simulated driving environment was used to collect sensory data of driver habits including steering wheel position and pedal pressure. These features are classified using a Support Vector Machine (SVM) learning algorithm. Our results show that our approach is capable of using various aspects of how a vehicle is operated to successfully identify a driver under 2.5 minutes with a 95% confidence interval and with at most one false positive per driving day.
总页数：	47
报告类型：	科技报告