Wi-Fi signals could be used to uniquely identify individuals WhoFi complements biometrics prompting privacy fears

Date:
Mon, 28 Jul 2025 17:04:00 +0000

Description:
WhoFi tech can identify people by how they disrupt Wi-Fi, raising major concerns over privacy and surveillance potential.

FULL STORY ======================================================================WhoFi uses WiFi signal distortions to fingerprint individuals without visual data Deep neural network maps signal changes to identify people with nearperfect accuracy Academic research opens new privacy debates around biometric
tracking via WiFi signals

Researchers at La Sapienza University in Rome have created WhoFi, a system which claims to be able to identify individuals by analyzing WiFi signals.

The system tracks people by interpreting how their presence disrupts WiFi patterns, offering a potential alternative to conventional biometric methods.

The technology works by examining Channel State Information, or CSI, which measures changes in WiFi signals caused by people and objects - and a deep neural network then interprets these disturbances as individual fingerprints. No cameras or physical contact required

The researchers claim the system delivers 95.5% accuracy in identifying
people even under different environmental conditions.

The team behind WhoFi includes Danilo Avola, Daniele Pannone, Dario
Montagnini and Emad Emam, who previously proposed a system called EyeFi in 2020. The new system is more accurate and capable of reidentifying people via nonvisual biometric signatures embedded in CSI.

WhoFi does not rely on cameras or physical contact. It needs only an existing WiFi network to sense human presence and movement.

The technology can operate in darkness, through walls, and even around obstacles, making it a discreet option compared to video surveillance
systems.

The researchers stress that WhoFi does not collect personal data or reveal identities in the conventional sense, noting, By leveraging nonvisual biometric features embedded in WiFi CSI, this study offers a
privacypreserving and robust approach for WiFibased ReID, and it lays the foundation for future work in wireless biometric sensing.

Still, its clear that the ability to track individuals without their
knowledge is a potential privacy nightmare.

Breaches of routine privacy can reveal patterns of daily behavior, such as regular locations or movements, potentially exposing sensitive personal habits.

So far, WhoFi remains an academic project with no known plans for commercial or government deployment. Yet the advantages in surveillance capability are clear. It can bypass poor lighting and crowded environments and is less conspicuous than cameras or visual scanners.

A number of similar Wi-Fi-based detection technologies have surfaced in various forms over the years.

Gamgee developed a fall detection system that could alert others if someone fell or if an intruder entered the home.

Comcasts Xfinity service introduced Wi-Fi Motion, which turns everyday devices like smart fridges, printers, or TVs into motion sensors.

Other researchers have gone further, using Wi-Fi signals to "see" through walls. A UC Santa Barbara team created a system that outlines objects and
even reads letters through barriers.

A similar study from Carnegie Mellon University demonstrated how standard Wi-Fi routers can detect a persons location and body position inside a room.

You can read more about the research behind WhoFi in this paper published on the arXiv preprint server.

Via Tech Xplore You may also like Gamgee's fall detection system could also
be used for intruder detection Wi-Fi routers can help detect motion in your home, but are also sparking privacy worries Researchers used Wi-Fi signals to see through walls. Privacy nightmare waiting to happen?



======================================================================
Link to news story: https://www.techradar.com/pro/wi-fi-signals-could-be-used-to-uniquely-identify -individuals-whofi-complements-biometrics-prompting-privacy-fears


--- Mystic BBS v1.12 A47 (Linux/64)
* Origin: tqwNet Technology News (1337:1/100)