VIN2VICTIM: From Vehicle Identification to Targeted Tracking via Data...
summary
VIN2VICTIM converts a single VIN into targeted physical tracking through four stages: owner deanonymization, trajectory reidentification, home/work inference and departure prediction. Using T‑Drive, GeoLife and Porto data, two spatio-temporal points reidentify 96–99% of taxi traces and ~82% of personal traces; home clusters appear for 92–100% of users. The pipeline survives 60‑minute sampling and geo‑indistinguishability at ε=0.1 leaves 74–78% unicity. Practical mitigations include rotating vehicle identifiers, suppressing location after parking, and on-vehicle edge filtering.
AI-written summary · about
chapters (8)
- 0:00VIN2VICTIM research
- 2:21Vehicles vs phones in tracking
- 6:47Four-stage attack pipeline
- 9:50Live attack demo
- 12:04Attack effectiveness on datasets
- 13:54Current defenses and limits
- 16:07Root causes and solutions
- 17:33Future of vehicle data protection
related talks
48mNonconsensual Tracking and Victim Safety: A Case Study in Abusability
08mEvil Uses of Data
20mWIP: Blurred Lines -- A GDPR-Compliant Framework for Anonymising Automotive...
01h 11mPRIVACY IN AUTOMATED AND CONNECTED VEHICLES
06mModelling Memory for Individual Re-identification in Decentralised Mobile Contact Tracing ...
40mSecuring Tesla and Mobileye from Split-Second Phantom Attacks
01h 06mSelf-Driving and Connected Cars: Fooling Sensors and Tracking Drivers
25mPrivacy Concerns in the Connected Car Ecosystem