WhoFi: Revolutionary Surveillance Technology Tracks Individuals via Wi-Fi Signal Disruptions
July 24, 2025 — In a groundbreaking development at the intersection of wireless communication and biometric surveillance, researchers at La Sapienza University in Rome have introduced "WhoFi," a novel technology capable of identifying and tracking individuals based on how their bodies alter Wi-Fi signals. This innovation promises to reshape surveillance capabilities but also reignites urgent debates over privacy and ethics.
How WhoFi Works
WhoFi leverages a technique known as Channel State Information (CSI), which records how Wi-Fi signals fluctuate as they encounter people and objects in the environment. Every individual’s unique physical characteristics and movements cause distinct disruptions or perturbations in these signals. By capturing these subtle variations using standard Wi-Fi networks, WhoFi constructs detailed biometric "fingerprints" without relying on traditional visual or physical data.
The technology uses a deep neural network trained to interpret these disruptions and convert them into compact, unique signature vectors for each person. This AI model can consistently recognize individuals with an impressive accuracy rate of 95.5%, even when environmental conditions change or people move within different surroundings.
Unlike conventional biometric systems such as fingerprint scanners or facial recognition cameras, WhoFi does not require direct physical contact or visual input. It can operate seamlessly across broad areas covered by Wi-Fi networks, including through walls, in darkness, and in obstructed spaces—limitations that often challenge visual surveillance methods.
Technical Details and Development
The WhoFi system involves an encoding framework that transforms raw Wi-Fi signal data into meaningful latent representations, which are then processed into normalized signature vectors. The research team comprises Danilo Avola, Daniele Pannone, Dario Montagnini, and Emad Emam, who detailed their findings in a 2025 preprint published on arXiv. This work builds upon an earlier 2020 project called "EyeFi," enhancing accuracy and robustness.
Potential Applications and Advantages
The ability to identify individuals based solely on their Wi-Fi signal impact opens new avenues for security monitoring and access control in environments where cameras and direct biometric scans are impractical or pose privacy challenges. For instance, WhoFi could be employed in critical infrastructure, restricted zones, or smart buildings to detect unauthorized presence discreetly and reliably.
Additionally, because Wi-Fi waves penetrate many obstacles and function in adverse conditions—such as poor lighting, fog, smoke, or around corners—WhoFi offers a stealthy and resilient alternative to conventional cameras and sensors.
Privacy Concerns and Ethical Considerations
Despite its technical promise, WhoFi’s capabilities have raised significant privacy alarms. Tracking individuals covertly through Wi-Fi without their consent could reveal sensitive information about their location, habits, and routines. The researchers acknowledge these risks and emphasize that WhoFi does not gather personal identity data in the traditional sense; it creates anonymous biometric signatures rather than capturing names or images.
As the team notes, “By leveraging non-visual biometric features embedded in Wi-Fi CSI, this study offers a privacy-preserving and robust approach for Wi-Fi-based re-identification, and it lays the foundation for future work in wireless biometric sensing.” Nonetheless, the tension between public safety benefits and individual privacy protections remains unresolved.
Current Status and Future Outlook
At present, WhoFi remains an academic prototype without commercial deployment or government adoption plans. However, the compelling advantages of this technology imply it could see real-world applications soon, particularly as societies grapple with balancing surveillance effectiveness and privacy rights.
Researchers, policymakers, and privacy advocates alike are watching developments around WhoFi and similar wireless biometric systems closely, mindful that the proliferation of such tools could fundamentally change how surveillance is conducted in public and private spaces.
This article was written by Paul Arnold, edited by Gaby Clark, and reviewed for accuracy by Andrew Zinin. It is based on the original research paper: Avola et al., "WhoFi: Deep Person Re-Identification via Wi-Fi Channel Signal Encoding," arXiv, 2025, DOI: 10.48550/arxiv.2507.12869.
For more updates on emerging technologies and their societal impacts, subscribe to Science X Newsletter.
