Researchers at Germany’s Karlsruhe Institute of Technology sat 197 volunteers in front of a bog-standard WiFi setup and worked out who each one was with almost 100% accuracy. There was no camera involved. Nobody needed a phone in their pocket. All it took was the radio traffic your router is already scattering around the room.
That final detail is what separates this from the decade’s worth of “WiFi can see you” research that came before it. The earlier work all required purpose-built hardware. This does not.
A camera in every respect except the lens
“By observing the propagation of radio waves, we can create an image of the surroundings and of persons who are present,” said Thorsten Strufe, a professor at KASTEL, KIT’s Institute of Information Security and Dependability.
“This works similar to a normal camera, the difference being that in our case, radio waves instead of light waves are used for the recognition,” Strufe said.
Since the setup interprets waves travelling through a room rather than pinging a device you own, there is nothing you need to be carrying. “Thus, it does not matter whether you carry a WiFi device on you or not.”
Powering down your own handset is no escape route either. “It’s sufficient that other WiFi devices in your surroundings are active.”
Worth a second pass, that line. What protects you here is how other people’s gadgets behave, not how yours does.
The unencrypted stream that makes it work
Earlier efforts to see people through walls relied either on LIDAR sensors, which gauge distance by emitting light and interpreting what bounces back, or on channel state information, the fine-grained record of how a radio signal distorts as it ricochets off walls, furniture and human bodies. Either route demands specialist kit or considerably more measurement effort.
KIT’s approach dispenses with all of that. By the researchers’ account, an off-the-shelf WiFi device does the job.
What makes it possible is beamforming feedback information, better known as BFI. Any device joined to a WLAN sends regular reports back to the router so the wireless link can be tuned. Crucially, those reports travel unencrypted, so they are readable by anybody within range.
Gather enough of them and the system produces images of a person from several angles at once, and that multi-angle view is what allows an identity to be pinned down. After the machine learning model has been trained on a given individual, re-identifying them is a matter of seconds.
Across the 197 participants, identities were inferred with almost 100% accuracy no matter the viewing perspective or how the person happened to walk. Changing your gait will not help.
Every router is a camera you will never spot
“This technology turns every router into a potential means for surveillance,” said Julian Todt of KASTEL. “If you regularly pass by a cafe that operates a WiFi network, you could be identified there without noticing it and be recognized later — for example by public authorities or companies.”
To their credit, the team declines to hype up the danger as it stands today. Felix Morsbach points out that a spy agency or a criminal who wants eyes on you already has simpler options, such as compromising CCTV installations that are already in place or internet-connected video doorbells.
What worries them is the direction of travel. “However, the omnipresent wireless networks might become a nearly comprehensive surveillance infrastructure with one concerning property: they are invisible and raise no suspicion.”
And there is the real story. A surveillance camera is a tangible thing you can notice, photograph and lodge a complaint about. A router tucked behind an espresso machine gives away nothing about the purposes its signals serve.
Why the repair has to land in the standard itself
“The technology is powerful, but at the same time entails risks to our fundamental rights, especially to privacy,” Strufe said.
Authoritarian governments are the team’s chief concern, since WiFi-based identification could be aimed at protesters or other groups with none of the visible equipment that usually tips people off that they are being watched. Nothing to smash, no lens to tape over.
Their case rests on ubiquity: wireless networks are already installed in homes, offices, restaurants and public spaces, so the protections have to be built in before running this at scale becomes cheap and straightforward. Accordingly, they want safeguards baked into the upcoming IEEE 802.11bf WiFi standard.
Funding for the research came via the Helmholtz “Engineering Secure Systems” topic, and the findings were presented at the ACM Conference on Computer and Communications Security in Taipei.
For a single concrete takeaway: the weak point is BFI, and current WiFi leaves it unencrypted by design. That is not an oversight, nor an implementation flaw you could patch on your own router. Changing it falls to a standards body, and 802.11bf is the venue where it will either happen or be missed.