CAN-Trace Attack: Exploit CAN Messages to Uncover Driving Trajectories

2025-07-13

AI Summary

Existing trajectory privacy protection mechanisms rely heavily on GPS, rendering them vulnerable to signal outages and offering insufficient protection. This paper introduces CAN-Trace—the first privacy attack framework for trajectory reconstruction leveraging in-vehicle Controller Area Network (CAN) bus messages. It constructs a weighted road graph using CAN signals such as vehicle speed and throttle position, then applies graph-matching algorithms against real-world road networks to infer driving trajectories without GPS. A robust evaluation metric is proposed to accommodate sensor data loss and matching inaccuracies. Experiments demonstrate attack success rates of 90.59% in urban and 99.41% in suburban environments—substantially outperforming GPS-dependent approaches. These results expose a critical, previously underestimated privacy threat: long-term driving trajectory inference via ubiquitous CAN bus telemetry.

Technology Category

• Planning, Routing, and Scheduling: Activity and Plan Recognition
• Computer Vision: Motion & Tracking
• Application Domains: Mobility, Driving & Flight

Application Category

• Security and Privacy: Large-scale security measurements
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

Driving trajectory data remains vulnerable to privacy breaches despite existing mitigation measures. Traditional methods for detecting driving trajectories typically rely on map-matching the path using Global Positioning System (GPS) data, which is susceptible to GPS data outage. This paper introduces CAN-Trace, a novel privacy attack mechanism that leverages Controller Area Network (CAN) messages to uncover driving trajectories, posing a significant risk to drivers' long-term privacy. A new trajectory reconstruction algorithm is proposed to transform the CAN messages, specifically vehicle speed and accelerator pedal position, into weighted graphs accommodating various driving statuses. CAN-Trace identifies driving trajectories using graph-matching algorithms applied to the created graphs in comparison to road networks. We also design a new metric to evaluate matched candidates, which allows for potential data gaps and matching inaccuracies. Empirical validation under various real-world conditions, encompassing different vehicles and driving regions, demonstrates the efficacy of CAN-Trace: it achieves an attack success rate of up to 90.59% in the urban region, and 99.41% in the suburban region.

Problem

• Exploits CAN messages to reconstruct driving trajectories
• Addresses GPS outage vulnerability in trajectory detection
• Proposes graph-matching algorithm for privacy attack on driving data

Innovation

• Uses CAN messages for trajectory reconstruction
• Applies graph-matching to road networks
• Introduces metric for evaluating matched candidates