Trackers like Apple's FindMy devices have seen an incredible boom throughout the last few years and are used for a variety of use-cases such as tracking lost luggage, stolen bikes, or simply finding one's lost keys. But they may also be used by stalkers to track their victims. Our goal was to analyze the ecosystem and propose potential defenses against stalking.

The trackers themselves are incredibly simple and generally consist of a low end microcontroller with a BLE (Bluetooth Low Energy) module. At regular intervals, these trackers send out a beacon that uniquely identifies them. Nearby devices such as phones or laptops record these beacons and send them, together with their location, to a central server. The owner of the tracker can then query the central server for recorded locations of their beacon. Trackers piggy back on the internet connection and (GPS) location of the nearby device as, due to battery concerns, they lack internet connectivity or location services.

This reliance on external devices to provide location information and internet connectivity may be used by a victim to defend themselves. By creating a relay network that listens for beacons and relaying them to various locations, we can create a network that disturbs the location information. For example, I may forward the beacon from a tracker that I recorded near my office at EPFL to a colleague at ETH where they send out the beacon (impersonating the tracker) using an arbitrary BLE device. Nearby devices then pick up the relayed beacon and report it to the central server. If the relayed beacon is picked up by more devices than the original beacon then the wrong location is reported.

In our PrivacyShield paper we explore different relaying strategies and measure their effectiveness. We also introduce a relaying network with open source prototype implementations.

This work was published at Usenix SEC'26 and Florian Hofhammer did the major work on the design and implementation of the anti-stalking framework.