A new technology developed by researchers in Germany suggests that people can be identified even if they are not carrying a smartphone or any WiFi-enabled device. Simply moving through an area with an active wireless network—such as a café, office, or public space—may be enough. The findings raise serious privacy concerns and have sparked calls for stronger safeguards.
A research team from the Karlsruhe Institute of Technology (KIT) has demonstrated that individuals can be recognized by passively capturing WiFi signals already present in their surroundings, without their knowledge and without the need for specialized equipment.
Radio Waves as an “Invisible Camera”
According to the researchers, radio waves emitted by WiFi devices interact with the human body as they propagate through space. These interactions create distinctive patterns that can be captured and analyzed using machine-learning techniques.
“By observing how radio waves spread through an environment, we can create an image of both the space and the people within it,” explains Professor Thorsten Strufe from KASTEL, KIT’s Institute for Information Security and Dependability. “It works much like a conventional camera—except that it uses radio waves instead of light.”
Crucially, the method does not depend on whether a person is carrying a WiFi-enabled device. Even turning off your phone offers no protection if other WiFi devices nearby remain active.
Every Router a Potential Observer
The researchers warn that this technology could effectively turn every WiFi router into a potential surveillance tool. “If you regularly pass by a café with an active WiFi network, you could be identified there without realizing it—and recognized again later,” says researcher Julian Todt.
While intelligence agencies or criminals already have simpler ways to monitor people—such as CCTV systems or smart doorbells—the team points out that wireless networks pose a unique risk: they are ubiquitous, invisible, and rarely perceived as a threat.
No Specialized Equipment, High Accuracy
Unlike earlier WiFi-based sensing techniques or systems relying on specialized sensors such as LiDAR, this approach can be implemented using standard WiFi hardware. It exploits routine feedback signals exchanged between devices and routers—signals that are transmitted without encryption and can be intercepted by anyone within range.
By collecting this data, researchers can generate radio-based “images” of people from multiple angles, enabling individual identification. Once trained, the machine-learning model can recognize individuals within seconds.
In a study involving 197 participants, the system identified individuals with near-perfect accuracy, regardless of viewing angle or walking style.
A Growing Privacy Concern
The researchers stress that while the technology is powerful, it also poses significant risks to fundamental rights—particularly privacy. These concerns are especially acute in authoritarian regimes, where such tools could be used to monitor protesters or track individuals without consent.
For this reason, the team is calling for urgent privacy protections and safeguards to be incorporated into the upcoming WiFi standard IEEE 802.11bf, before the technology becomes widely deployed.
“The technology itself is impressive,” Strufe concludes, “but without clear rules and protections, it could easily be turned against the very society it was meant to serve.”