Your WiFi Can Identify and Track You Without Cameras or Phones

A team of researchers at Germany's Karlsruhe Institute of Technology (KIT) have shown how ordinary WiFi signals can be used to identify individuals without the need for cameras or smartphones. In tests with 197 people, the system achieved nearly 100% accuracy in recognizing individuals regardless of their angle or walking style.

The technique relies on analyzing how WiFi radio waves interact with people and objects in a room. Similar to how cameras capture images using visible light, this system works by studying radio wave changes. Unlike cameras, no special devices like smartphones are required. Even if an individual switches off their own phone, they may still be tracked as long as other WiFi devices nearby continue transmitting information.

WiFi devices frequently exchange data with routers to optimize wireless connections. This data, known as Beamforming Feedback Information (BFI), tells the router about radio signal paths in the area. The researchers discovered that BFI can also reveal how people influence these radio waves. Since BFI is transmitted without encryption, someone within WiFi range could potentially collect it without needing specialized surveillance equipment.

A machine-learning algorithm can then convert these signal changes into patterns that resemble images from various perspectives. Once the system learns what a person looks like through WiFi signals, identifying them takes only a few seconds.

The researchers warn about the privacy risks posed by this technology. Given that WiFi networks are ubiquitous in homes, offices, restaurants, and other public spaces, someone could potentially be identified without realizing they were being monitored. The team urges for stronger privacy protections to be incorporated into future WiFi standards to prevent large-scale surveillance.