Researchers at the Karlsruhe Institute of Technology (KIT) have identified a privacy risk in modern WiFi systems that could allow passive monitoring of people through everyday wireless networks, without requiring them to carry any device.
The study focuses on beamforming feedback data generated by WiFi systems to improve signal quality between routers and connected devices. This data, which is exchanged routinely in modern wireless standards, is not consistently encrypted and can be captured from the surrounding environment.
According to the researchers, these signal patterns reflect subtle changes in how radio waves interact with human movement and physical space. When processed using machine learning techniques, the captured data can be used to distinguish individuals based on their unique movement-related signal signatures.
Unlike traditional surveillance methods, this approach does not depend on cameras, microphones, or direct device compromise. Instead, it relies entirely on passive observation of existing wireless traffic, meaning that ordinary WiFi routers deployed in homes or public spaces could unintentionally contribute to data collection.
In experimental settings involving a large group of participants, the system was able to identify individuals with very high accuracy after training on observed signal patterns. The results suggest that environmental radio signals contain more personal information than previously assumed.
The researchers caution that as WiFi infrastructure becomes more advanced and widely deployed, it may also increase opportunities for unintended sensing of human activity. They highlight the need for stronger protections around wireless feedback data to reduce the risk of privacy exposure in everyday environments.