WiFi Sees All: Your Router is Now a Spy

WiFi signals are being explored as a novel method for surveillance, capable of identifying individuals and mapping environments without cameras or requiring users to carry connected devices. Professor Thorsten Strufe of KASTEL at KIT explains that observing radio wave propagation allows for recognition, similar to how a camera uses light waves. This technique works even if a person has no WiFi-enabled devices on them or if their own devices are turned off, as surrounding active WiFi devices suffice.

The widespread presence of WiFi networks in homes, offices, and public spaces presents a potential privacy risk, as any router could become a surveillance tool. Julian Todt from KASTEL warns that individuals could be identified discreetly and later recognized. Unlike visible CCTV, this WiFi-based monitoring is invisible and raises no suspicion.

This new approach bypasses the need for specialized hardware like LIDAR sensors, relying instead on standard WiFi devices. It analyzes beamforming feedback information (BFI), which is routinely transmitted unencrypted by connected devices to optimize communication. By examining this data, a machine learning model can generate images for identification.

The system has achieved nearly 100% identification accuracy in studies. Researchers are concerned about its potential misuse in authoritarian regimes for monitoring, and they advocate for built-in privacy protections in the upcoming IEEE 802.11bf WiFi standard. The project received funding under the Helmholtz "Engineering Secure Systems" topic and was presented at the ACM Conference on Computer and Communications Security (CCS) in Taipei.