
Italian scientists have proved that a regular Wi-Fi signal can be used to track people who don’t have any devices with them.
WhoFi technology generates a unique biometric identifier based on how a person’s body interacts with surrounding Wi-Fi signals. Approach, described in preprint, uses signal distortion caused by the human body to re-identify people as they move through spaces covered by different Wi-Fi networks.
Researchers Danilo Avola, Daniele Pannone, Dario Montagnini, and Emad Emam claim that the developed system can identify people with an accuracy of 95.5%. The technology is based on a method known as channel state information (CSI), which involves capturing changes in Wi-Fi signals as they pass through the physical environment. The results were achieved using Wi-Fi signals from two TP-Link N750 routers.
«The basic idea is that as a Wi-Fi signal propagates through an environment, its waveform changes due to the presence and physical characteristics of objects and people along the way. These changes, captured in the form of CSI, contain a lot of biometric information», — the article says.
According to the team, changes in the phase and amplitude of the signals are specific enough to serve as a kind of digital human fingerprint, especially when analyzed by a deep neural network. The researchers used the currently popular AI transformer architecture.
Wi-Fi signals are a serious alternative to traditional surveillance tools such as cameras. They can operate regardless of lighting conditions, pass through walls, and avoid image interception, making data acquisition and transmission more secure. At the same time, the technology raises concerns about passive surveillance. Experiments with them known at least since 2019. A similar system was introduced in 2020 but had a maximum accuracy of 75%.
Sources: PC World, Interesting Engineering
Spelling error report
The following text will be sent to our editors: