Researchers Develop System to Track People Using Wi-Fi Signal Distortions

Researchers from La Sapienza University in Rome have developed a system called ‘WhoFi’ that can track individuals based on the way their bodies distort Wi-Fi signals. The system works by analyzing subtle changes in Wi-Fi signal propagation caused by the presence of human bodies. These alterations, captured as Channel State Information (CSI), provide biometric data unique to each person. The technique, trained with a deep neural network, has shown up to 95.5% accuracy in identifying individuals, even without the need for them to carry a phone. This approach takes advantage of how the human body interacts with electromagnetic transmissions, resulting in distinctive signal distortions. A previous method, EyeFi, achieved around 75% accuracy, but WhoFi has improved on this significantly. The results suggest that Wi-Fi signals could serve as a robust, privacy-preserving biometric modality, marking a significant step in signal-based re-identification systems. While this technology has the potential to enhance security and tracking, it raises concerns about privacy implications, especially in public and private spaces where Wi-Fi is ubiquitous.