Every Byte Matters: Traffic Analysis of BluetoothWearable Devices
summary
Dataset of popular smartwatches, fitness and health trackers collects Bluetooth Classic and LE traces while manually performing labeled actions; features based on packet sizes and timings feed machine learning that achieves high precision and recall at (a) recognizing the exact device, (b) identifying human actions like heart‑rate checks and exercising, (c) detecting which app is being opened on Wear OS, and (d) spotting fine‑grained events such as recording an insulin injection. Standard defenses (padding, delays, dummy traffic) prove costly and largely ineffective.
AI-written summary · about
related talks
24mAttacking Bluetooth LE Design and Implementation in Mobile + Wearables Ecosystems
30mUnfit Story of Fitness Trackers: Hacking BLE Fitness Trackers
30mUnfit Story of Fitness Trackers: Hacking BLE Fitness Trackers
51mBluetooth Hacking: Tools And Techniques
43mWhen encryption is not enough: Attacking Wearable
54mSecuring Your In-Ear-Fitness Coach: Challenges in Hardening Next Generation Wearables
02h 34mFIRSTCON21 - June 9 | Stream 1 (12:00-14:45 UTC)
48mCreating a Bluetooth Toolbox