Ordinary WiFi Can Now Identify You Without a Camera

Researchers in Germany have shown that an ordinary Wi-Fi network can pick a specific person out of a crowd with near-perfect accuracy, no camera involved. In a test with 197 people, the system recognized individuals in seconds, from different angles, no matter how they walked.

The work comes from the Karlsruhe Institute of Technology and was written up by ScienceDaily. The idea is simple to describe and unsettling to think about. Thorsten Strufe, a professor at KASTEL, KIT's Institute of Information Security and Dependability, put it plainly. "By observing the propagation of radio waves, we can create an image of the surroundings and of persons who are present," says Strufe. He compared it to a normal camera, except radio waves do the job that light usually does.

When phones, laptops, and other gadgets talk to a router, they routinely send back small unencrypted status messages called beamforming feedback information, which help the network aim its signal. Your body changes how those radio waves bounce around a room. The team's system reads those distortions and builds radio "images" of a person from several viewpoints, then a trained machine-learning model matches those patterns to an identity. Once the model is trained, the researchers say, identification takes only a few seconds. In their tests it hit roughly 99.5% accuracy across the 197 participants, holding up regardless of viewing angle or walking style.

Keep one distinction clear. This is not a system that already knows your name the moment you walk past a café. It re-identifies a person, meaning it can tell that the same body seen at one spot is the same body seen at another. Turning that into a legal name requires a separate step and a separate database. 

The researchers are blunt about the downside. "The technology is powerful, but at the same time entails risks to our fundamental rights, especially to privacy," emphasizes Strufe. The team is especially worried about authoritarian governments using it to monitor protesters without the obvious infrastructure of cameras, and they are pushing for privacy safeguards to be built into the upcoming IEEE 802.11bf Wi-Fi sensing standard.