Be careful when monitoring radio networks

If you pass by a café that operates a Wi-Fi network, you can be identified, even if you don’t have a smartphone with you. Researchers from the Karlsruhe Institute of Technology (KIT) have discovered that it is possible to identify people using Wi-Fi signals alone. They point out that this poses a significant risk to privacy.

To deduce a person’s identity, it is no longer necessary for them to carry a smartphone or tablet. All it takes is Wi-Fi devices communicating with each other in the person’s environment. This creates an image comparable to taking a photo with a camera, but based on radio waves. The research team calls for adequate confidentiality guarantees.

“By observing the propagation of radio waves, we can create a picture of the environment and the people there,” explains Professor Thorsten Strufe from the Institute for Information Security and Reliability at KASTEL-KIT. “It works in the same way as a normal camera, the difference being that in our case radio waves instead of light waves are transformed into an image,” explains the cybersecurity expert.

“So it doesn’t matter whether you have a Wi-Fi device on you or not.” Even turning off a device doesn’t help: “It’s enough that other Wi-Fi devices nearby are active.”

Wi-Fi routers as “silent observers”

“This technology turns every router into a potential means of surveillance,” warns Julian Todt of KASTEL. “If you regularly pass by a café that operates a Wi-Fi network, you could be identified there without realizing it and be recognized later, for example by public authorities or businesses.”

Strufe points out that there are indeed simpler methods for secret services or cybercriminals to observe people, for example by accessing CCTV cameras or video doorbells. “However, ubiquitous wireless networks could become an almost complete surveillance infrastructure.” In fact, Wi-Fi networks exist in almost every home, office, restaurant, and public space today.

No special equipment required

Unlike attacks using LIDAR sensors or previous Wi-Fi-based methods, which use channel state information (CSI), that is, measured data that indicates how a radio signal changes when it hits walls, furniture or people, attackers do not need special hardware. This method only requires standard Wi-Fi devices. It works by exploiting legitimate users whose devices are connected to the Wi-Fi network.

These regularly send feedback signals within the network, also called beamforming feedback information (BFI), to the router, in unencrypted form so that they are readable by third parties. This creates images from different viewing angles that can be used to identify the respective people. Once the underlying machine learning model is trained, identification takes just seconds.

A success rate of almost 100%

In their study of 197 participants, the research team was able to infer people’s identities with almost 100% accuracy, regardless of perspective or their walking style.

“Technology is powerful, but at the same time it poses risks to our fundamental rights, especially when it comes to privacy,” emphasizes Strufe. Researchers warn that this is particularly critical in authoritarian states where technology could be used to observe protesters. Therefore, they urgently call for protection measures and privacy guarantees in the upcoming Wi-Fi standard IEEE 802.11bf.

The researchers will present their findings at the ACM Conference on Computer and Communications Security (ACM CCS 2025), held October 13-17 in Taipei, Taiwan.