As vehicles become more and more autonomous, they rely on computer systems to provide location services and communicate with other vehicles that may be on the road.
This dependence on systems that track a vehicle’s position contributes to the collection of personally identifiable position data. That, in turn, could lead to the automatic and routine surveillance of users’ everyday travels — data that would certainly be useful to various enterprises, such as insurance companies.
Tracking Location Services
Researchers at Hanyang University in Ansan, South Korea, may have come up with a way to confuse these tracking systems so they cannot establish a personally identifiable path for a vehicle. In their paper, “Preserving Location Privacy of Connected Vehicles With Highly Accurate Location Updates,” researchers explained how they took advantage of the two different communications systems that a vehicle can use: the location-based system (LBS) and the vehicle-to-vehicle (V2V) communication capability.
The LBS shouldn’t be trusted from a security standpoint. Operating much like a traditional tracker, an LBS usually uses an LTE internet network and can generate a time-series analysis of location samples. This gives the path information of the driver, which can establish a location history.
There is also a beacon for vehicle-to-vehicle communication within these automobiles. This technology shouts out local, safety-related messages to any vehicles that might be close in proximity.
Obfuscating V2V Communication Data
Researchers used both of these radio channels to implement their security method, called mutually obfuscating paths (MOPs). Once two vehicles are close enough to send beacon messages to each other, MOPs have the two vehicles stop independently sending location data to the LBS. Instead, each one sends the data from both of the vehicles.
This ends up generating multiple paths: the real one and the alternative taken by the other vehicle. However, the LBS doesn’t know which is which and discards the data.
Motherboard reported that the researchers evaluated the scheme via simulations. The best results were found in high-traffic situations, since this scenario offers the most V2V communication for location services. With multiple vehicles present, there were many alternate paths that could be sent to the LBS by the user vehicle to confuse any potential tracking systems.
Ultimately, this approach may pave the way for privacy to coexist with location mapping as part of the autonomous vehicles of the future.