Crooks Are Using CAN Injection Attacks To Steal Cars
“In most cars on the road today, these internal messages aren’t protected: the receivers simply trust them,” [Ken Tindell, CTO of Canis Automotive Labs] detailed in a technical write-up this week. The discovery followed an investigation by Ian Tabor, a cybersecurity researcher and automotive engineering consultant working for EDAG Engineering Group. It was driven by the theft of Tabor’s RAV4. Leading up to the crime, Tabor noticed the front bumper and arch rim had been pulled off by someone, and the headlight wiring plug removed. The surrounding area was scuffed with screwdriver markings, which, together with the fact the damage was on the kerbside, seemed to rule out damage caused by a passing vehicle. More vandalism was later done to the car: gashes in the paint work, molding clips removed, and malfunctioning headlamps. A few days later, the Toyota was stolen.
Refusing to take the pilfering lying down, Tabor used his experience to try to figure out how the thieves had done the job. The MyT app from Toyota — which among other things allows you to inspect the data logs of your vehicle — helped out. It provided evidence that Electronic Control Units (ECUs) in the RAV4 had detected malfunctions, logged as Diagnostic Trouble Codes (DTCs), before the theft. According to Tindell, “Ian’s car dropped a lot of DTCs.” Various systems had seemingly failed or suffered faults, including the front cameras and the hybrid engine control system. With some further analysis it became clear the ECUs probably hadn’t failed, but communication between them had been lost or disrupted. The common factor was the CAN bus.
Read more of this story at Slashdot.