How your phone, laptop, or watch can be tracked by their Bluetooth transmissions

Unique fingerprints lurk in radio signals more often than not, it seems


Over the past few years, mobile devices have become increasingly chatty over the Bluetooth Low Energy (BLE) protocol and this turns out to be a somewhat significant privacy risk.

Seven boffins at University of California San Diego – Hadi Givehchian, Nishant Bhaskar, Eliana Rodriguez Herrera, Héctor Rodrigo López Soto, Christian Dameff, Dinesh Bharadia, and Aaron Schulman – tested the BLE implementations on several popular phones, PCs, and gadgets, and found they can be tracked through their physical signaling characteristics albeit with intermittent success.

That means the devices may emit a unique fingerprint, meaning it's possible to look out for those fingerprints in multiple locations to figure out where those devices have been and when. This could be used to track people; you'll have to use your imagination to determine who would or could usefully exploit this. That said, at least two members of the team believe it's worth product makers addressing this privacy weakness.

The academics describe their findings in a paper [PDF], "Evaluating Physical-Layer BLE Location Tracking Attacks on Mobile Devices," which is scheduled to be presented at the IEEE Symposium on Security and Privacy in 2022.

BLE message transmissions have become more common in phones, laptops, watches, and the like thanks to operating system support for services like Apple’s Continuity protocol, for moving work across devices, and Find My, for locating lost devices. More recently, the US-based researchers explain, software for tracking COVID-19 has used mobile devices as BLE beacons, broadcasting signals in the service of public health.

Applications utilizing BLE commonly try to conceal identifying data by doing things like re-encrypting the MAC address of the transmitting device, they explain. But this sort of MAC address randomization can't conceal baked-in hardware characteristics that may be usable to uniquely identify the transmitting machine.

The boffins looked at at handful of popular mobile devices – ​​iPhone 10 (iOS), Thinkpad X1 Carbon (Windows), MacBook Pro 2016 (macOS), Apple Watch 4 (watchOS), Google Pixel 5 (Android), and Bose QuietComfort 35 wireless headphones – and found they could often successfully fingerprint the physical BLE chip layer.

In other words, they measured variations in the radio-frequency characteristics of BLE transmissions in a way that allowed them to distinguish BLE devices from one another, making identified devices theoretically trackable.

Radio-frequency fingerprinting has been a subject of academic research for years, on systems like RFID [PDF], Bluetooth [PDF], and WiFi [PDF].

Device fingerprinting

The UC San Diego group claims that no one has previously evaluated how practical a fingerprinting attack on BLE might be in the real world and that no one has previously proposed a BLE fingerprinting tool that can measure the physical-layer imperfections exposed by such systems' transmissions.

The BLE chipsets in the sample devices share a common architectural pattern: They include Wi-Fi circuitry, to reduce power consumption and to save space. As a result, both BLE and Wi-Fi in these devices rely on the same 2.4 GHz in-phase/quadrature (I/Q) receiver frontend.

"A consequence of this hardware design choice is that BLE transmissions contain the same hardware imperfections as Wi-Fi," the academics explain in their paper.

"The imperfections are introduced by the shared I/Q frontend of the chipset. They result in two measurable metrics in BLE and WiFi transmissions: Carrier Frequency Offset (CFO) and I/Q imperfections, specifically: I/Q offset and I/Q imbalance."

Given that prior research has shown these metrics can uniquely fingerprint WiFi devices, the boffins set out to show that the same could be done for now ubiquitous BLE signals.

They encountered a number of challenges that made identification more difficult. First, distinguishing devices with the same chipset, like Apple's iPhone, is harder than distinguishing devices with different chipsets. Second, device temperature variations complicate matters, potentially requiring reassessment so that an idle device can be linked to a device running an app.

And third, devices transmit with different power levels, which affects the range at which they can be detected – iPhones evidently broadcast their COVID beacons at a higher power level than Android devices.

Other potential issues, like the difference between using an expensive software-defined radio for signal scanning and an inexpensive hobbyist model, turned out to be something that could be compensated for through calibration.

Real-world testing

The group collected two datasets of BLE beacons. The first came from scanning for signals at six coffee shops, a university library, and a food court, each for about an hour. They gathered packets from 162 devices during that period and found about 40 per cent were uniquely identifiable.

The second data set came from setting up a software-defined radio at the exit of a room where it was exposed to hundreds of devices daily. The researchers recorded Apple and Google COVID–19 Exposure Notification BLE beacons transmitted by bypassers' devices during 10 hours periods on two separate days, a week apart.

They saw 647 unique MAC addresses across the two 20 hours of data collection and were able to uniquely identify 47.1 per cent of those; 15 per cent had imperfections that overlapped with only one other device.

The boffins also tried an experiment where they tracked 17 different targets as they moved about. The average false negative rate came to 3.21 per cent while the average false positive rate came to 3.5 per cent, which means their system identified a device it was mostly accurate.

In an email to The Register, UC San Diego doctoral students Hadi Givehchian, Nishant Bhaskar, both primary authors of the paper, said they expected Apple's AirTag and Samsung SmartTag Plus would be trackable using the same technique.

"The BLE chipsets in locator beacons are likely to have the same manufacturing variations that we observed in other BLE-only wireless devices we tested," they said.

Shut it down

Two possible defenses are suggested in the paper: adding a random time-varying extra frequency offset to the crystal oscillator, which BLE can apparently handle, to make signal measurements less predictable; and running a background process that keeps changing the computation as the MAC address gets randomized, which would make the battery run down faster.

Turning devices off should limit tracking, but efforts to disable tracking may not work as expected.

"To the best of our knowledge, powering down a personal [device] entirely will stop it from beaconing," said Givehchian and Bhaskar. "However, we found that simply disabling Bluetooth on some phones will not stop the beacons. For example, on some Apple devices disabling Bluetooth in the Control Center (the menu accessed by swiping down from the top of the screen) may not stop it from beaconing."

Ultimately, the researchers conclude that tracking people via BLE can be done, and some people are more vulnerable than others, depending on conditions and the commonness or uniqueness of the device targeted.

"Based on our results we do believe this attack is feasible and practical, so device vendors should consider mitigations," said Givehchian and Bhaskar. "Many devices in use today have unique fingerprints, and the hardware needed for the attack costs less than $200.

"However, we also observed the attack is not guaranteed to be successful in all situations, the target [device] may be misidentified in a large crowd, and its fingerprint will change when the device heats up or cools down." ®

Similar topics


Other stories you might like

  • India reveals home-grown server that won't worry the leading edge

    And a National Blockchain Strategy that calls for gov to host BaaS

    India's government has revealed a home-grown server design that is unlikely to threaten the pacesetters of high tech, but (it hopes) will attract domestic buyers and manufacturers and help to kickstart the nation's hardware industry.

    The "Rudra" design is a two-socket server that can run Intel's Cascade Lake Xeons. The machines are offered in 1U or 2U form factors, each at half-width. A pair of GPUs can be equipped, as can DDR4 RAM.

    Cascade Lake emerged in 2019 and has since been superseded by the Ice Lake architecture launched in April 2021. Indian authorities know Rudra is off the pace, and said a new design capable of supporting four GPUs is already in the works with a reveal planned for June 2022.

    Continue reading
  • Prisons transcribe private phone calls with inmates using speech-to-text AI

    Plus: A drug designed by machine learning algorithms to treat liver disease reaches human clinical trials and more

    In brief Prisons around the US are installing AI speech-to-text models to automatically transcribe conversations with inmates during their phone calls.

    A series of contracts and emails from eight different states revealed how Verus, an AI application developed by LEO Technologies and based on a speech-to-text system offered by Amazon, was used to eavesdrop on prisoners’ phone calls.

    In a sales pitch, LEO’s CEO James Sexton told officials working for a jail in Cook County, Illinois, that one of its customers in Calhoun County, Alabama, uses the software to protect prisons from getting sued, according to an investigation by the Thomson Reuters Foundation.

    Continue reading
  • Battlefield 2042: Please don't be the death knell of the franchise, please don't be the death knell of the franchise

    Another terrible launch, but DICE is already working on improvements

    The RPG Greetings, traveller, and welcome back to The Register Plays Games, our monthly gaming column. Since the last edition on New World, we hit level cap and the "endgame". Around this time, item duping exploits became rife and every attempt Amazon Games made to fix it just broke something else. The post-level 60 "watermark" system for gear drops is also infuriating and tedious, but not something we were able to address in the column. So bear these things in mind if you were ever tempted. On that note, it's time to look at another newly released shit show – Battlefield 2042.

    I wanted to love Battlefield 2042, I really did. After the bum note of the first-person shooter (FPS) franchise's return to Second World War theatres with Battlefield V (2018), I stupidly assumed the next entry from EA-owned Swedish developer DICE would be a return to form. I was wrong.

    The multiplayer military FPS market is dominated by two forces: Activision's Call of Duty (COD) series and EA's Battlefield. Fans of each franchise are loyal to the point of zealotry with little crossover between player bases. Here's where I stand: COD jumped the shark with Modern Warfare 2 in 2009. It's flip-flopped from WW2 to present-day combat and back again, tried sci-fi, and even the Battle Royale trend with the free-to-play Call of Duty: Warzone (2020), which has been thoroughly ruined by hackers and developer inaction.

    Continue reading

Biting the hand that feeds IT © 1998–2021