Ordinary WiFi can now identify you with near-perfect accuracy
Researchers have discovered that ordinary WiFi networks can be used to identify and track people without cameras or special equipment, using signals already being transmitted by nearby devices.
## WiFi Signals Used for Surveillance WiFi signals could potentially be used to identify people and map their surroundings without relying on cameras or requiring the person being observed to carry a connected device. This is according to a recent study by researchers at the Karlsruher Institut für Technologie (KIT), who have shown that unencrypted signals routinely exchanged between WiFi devices and routers can be used to create radio-based images of people and recognize them within seconds. The technique works by analyzing radio waves moving through a space, allowing a person to be identified without them having a phone, smartwatch, or other WiFi enabled device with them. Even turning off their own device would not necessarily prevent the system from working, as it is sufficient that other WiFi devices in their surroundings are active. ## Ordinary WiFi Routers Could Become Surveillance Tools The researchers warn that the findings reveal a potentially serious privacy risk because WiFi networks are already widespread in homes, offices, restaurants, and public spaces. They say that this technology turns every router into a potential means for surveillance, and that people could be identified without noticing it and recognized later by public authorities or companies. The possibility of WiFi based monitoring is especially notable because it is invisible and raises no suspicion. Unlike a visible security camera, a wireless network normally gives people no obvious indication that its radio signals could potentially be used to recognize who is nearby. ## No Special Surveillance Hardware Is Needed The new technique does not require specialized equipment or complex measurements. Instead, it uses standard WiFi devices and takes advantage of normal communications produced by legitimate users connected to a WLAN. Connected devices routinely send information back to the router to help optimize wireless communication, and these signals, known as beamforming feedback information (BFI), are transmitted without encryption. By analyzing this information, the system can generate images of people from multiple viewpoints. Those images can then be used to determine a person's identity. Once the machine learning model has already been trained to recognize individuals, the identification process takes only a few seconds. ## Researchers Achieved Almost 100% Identification Accuracy The team tested the technique in a study involving 197 participants. Their system was able to infer people's identities with almost 100% accuracy, regardless of the viewing perspective or the way a person walked. The researchers are particularly concerned about how such technology could be used in authoritarian countries, where it could potentially be applied to monitor protesters or other groups without the obvious surveillance infrastructure associated with conventional cameras. The researchers are calling for protective measures and privacy safeguards to be incorporated into the forthcoming IEEE 802.11bf WiFi standard to prevent these capabilities from becoming easier to exploit at scale.
