BLE-Doubt: Smartphone-Based Detection of Malicious Bluetooth Trackers

Abstract

Stalkers can hide Bluetooth Low-Energy (BLE) trackers, like the Apple AirTag and Tile Finder, in their targets’ clothing or vehicles to surveil their locations. Existing countermeasures to detect BLE-based stalking are promising but have several shortcomings: they only work against Apple products, they are slow to detect trackers, and there is no publicly available characterization of how well they work. We present an open-source, general method for detecting maliciously deployed BLE trackers. Our algorithm detects malicious devices in just a few minutes, whereas previous algorithms take hours or days. We show in a small but novel validation study that our algorithm performs with high precision and recall for most extant trackers, although AirTags pose additional challenges. Along with our algorithm and validation, we provide an open-source Android application capable of real-time detection of these devices. We also characterize the behavior of the AirTag and discuss the risk factors which make it particularly hard to detect. We conclude with a discussion for future work to make tracking devices safer for the public.