Skip to content
Submit incident
AII-1002-05851
Well documentedThree or more independent publishers report the core facts.Supporting sources are pending verification.

WhoFi System Enables Biometric Identification Through Wi-Fi Signal Analysis

Researchers at La Sapienza University developed WhoFi, a deep learning system that identifies and tracks individuals by analyzing how their bodies distort Wi-Fi signals, achieving 95.5% accuracy. The technology raises concerns about covert biometric surveillance since it can track people through walls without their knowledge or consent, using ubiquitous Wi-Fi infrastructure.

23 July 2025 Β· La Sapienza University of Rome Β· Privacy and surveillance
23 July 2025
AII-1002-05851
Quick facts
Category
Privacy and surveillance
Country
Italy
Date
23 July 2025
Drafted by
Raidu Incident Detection

What happened

Researchers at La Sapienza University in Rome developed WhoFi, a system using Channel State Information (CSI) from Wi-Fi router signals to identify and track individuals. The system analyzes how a person's body alters Wi-Fi radio signals in unique, recognizable patterns, converting these patterns into vector signatures via deep learning.

In testing on 14 participants across various clothing scenarios, WhoFi achieved 95.5% accuracy in identifying individuals. The system operates through walls and is unaffected by lighting conditions, clothing, or line of sight, relying instead on how radio waves interact with internal body structures.

Reported impact

The technology raises concerns about invisible, unwanted or illegal surveillance. Anyone transmitting Wi-Fi signals could unknowingly share biometric information about nearby individuals, even those not carrying devices.

Potential applications range from benign retail tracking (sending customers promotional messages upon return) to tracking individuals' physical locations and movements across time and space, including in private residences if data were obtained.

Index analysis

WhoFi demonstrates significant capability for covert biometric identification and tracking without visual data capture. The 95.5% accuracy on basic Wi-Fi hardware (TP-Link N750 routers) suggests the technique could be widely deployed using existing infrastructure.

Classification

Category
Privacy and surveillance
Industry
Digital security
Country
Italy
Harm status
Potential harm
Intent
Unintentional
Timing
Unknown
Harm types
Human or fundamental rights
Harmed
General public
AI systems
WhoFi

Sources and evidence

Sources not yet published

This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.

How sources are verifiedSuggest a source for this incident