A Critical Roadmap to Driver Authentication via CAN Bus: Dataset Review, Introduction of the Kidmose CANid Dataset (KCID), and Proof of Concept

TL;DR

The paper addresses the vulnerability of modern vehicles to CAN-bus-based unauthorized use and theft by evaluating driver-authentication approaches and the limitations of existing datasets. It introduces the Kidmose CANid Dataset (KCID), a raw CAN-bus dataset from 16 drivers across four vehicles, enriched with demographic information and both daily-driving and fixed-route data, to enable robust, generalizable authentication research. A driver-authentication anti-theft framework is proposed and demonstrated via a proof-of-concept prototype on a Raspberry Pi, including live-road trials in an unaltered vehicle to show practical feasibility. Beyond authentication, KCID supports applications in insurance profiling, safety assessment, mechanical anomaly detection, and impaired-driving detection, highlighting KCID’s potential to advance automotive behavioral biometrics research and real-world security deployments.

Abstract

Modern vehicles remain vulnerable to unauthorized use and theft despite traditional security measures including immobilizers and keyless entry systems. Criminals exploit vulnerabilities in Controller Area Network (CAN) bus systems to bypass authentication mechanisms, while social media trends have expanded auto theft to include recreational joyriding by underage drivers. Driver authentication via CAN bus data offers a promising additional layer of defense-in-depth protection, but existing open-access driver fingerprinting datasets suffer from critical limitations including reliance on decoded diagnostic data rather than raw CAN traffic, artificial fixed-route experimental designs, insufficient sampling rates, and lack of demographic information. This paper provides a comprehensive review of existing open-access driver fingerprinting datasets, analyzing their strengths and limitations to guide practitioners in dataset selection. We introduce the Kidmose CANid Dataset (KCID), which addresses these fundamental shortcomings by providing raw CAN bus data from 16 drivers across four vehicles, including essential demographic information and both daily driving and controlled fixed-route data. Beyond dataset contributions, we present a driver authentication anti-theft framework and implement a proof-of-concept prototype on a single-board computer. Through live road trials with an unaltered passenger vehicle, we demonstrate the practical feasibility of CAN bus-based driver authentication anti-theft systems. Finally, we explore diverse applications of KCID beyond driver authentication, including driver profiling for insurance and safety assessments, mechanical anomaly detection, young driver monitoring, and impaired driving detection. This work provides researchers with both the data and methodological foundation necessary to develop robust, deployable driver authentication systems...

Figures (6)

• Figure 1: Our two data collection strategies.
• Figure 2: The on-board diagnostics II (OBD-II) ports of two vehicles used in this study. We leveraged the OBD-II ports to collect data for our driver authentication dataset. A data cable (connected to either a laptop or a standalone data collection device) would be plugged into the port to facilitate data collection.
• Figure 3: Fixed route from Grand Island, FL, to Bethany Lutheran Church in Leesburg, FL. © OpenStreetMap contributors. Available under the Open Data Commons Open Database License (ODbL). See https://www.openstreetmap.org/copyright.
• Figure 4: Real-time driver monitoring by State Farm car insurance, showing (a) overall safety scores across multiple categories and (b) detailed driving events from a single trip.
• Figure 5: Workflow for the proposed driver authentication anti-theft system. The system continuously monitors driving behavior, providing visual feedback through LED indicators and implementing graduated intervention stages for unauthorized access attempts.