WiFi signals can quietly surveil you, says study, PgII

Key Highlights:

• A new study reveals that WiFi beamforming feedback information (BFI) can identify individuals with high accuracy.
• Researchers at the Karlsruhe Institute of Technology achieved 99.5% accuracy in identifying 160 individuals based on their walking patterns.
• The study used a WiFi setup with two access points and four listening perspectives operating in the 6 GHz band.
• Unlike CSI, BFI is readily available on off-the-shelf hardware, posing a more serious privacy risk.

Detailed Insights:

• Beamforming, a feature in modern WiFi standards, helps routers efficiently direct signals, but phones and laptops broadcast unencrypted reports (BFI) that can be intercepted.
• The study demonstrated that BFI alone is strongly identifying and can recognize individuals even when they alter their walking styles, such as carrying a backpack or walking briskly.
• WiFi-based tracking can create an "inverse panopticon," where individuals are silently profiled without their knowledge, unlike visible surveillance methods like CCTV cameras.
• Identifying individuals through their gait allows linking other WiFi-based tasks, such as recognizing activities or estimating occupancy, to those identities over time.
• Current mitigation strategies, like adding noise to training fields, are not mature and primarily target CSI instead of BFI, leaving a gap in addressing the privacy risks associated with BFI.

Scientific/Technical Concepts Involved:

• Beamforming: A WiFi technology that focuses radio signals towards a specific device to improve signal strength and efficiency.
• Channel State Information (CSI): Information about the communication channel's properties, often requiring specialized hardware to obtain.
• Beamforming Feedback Information (BFI): Unencrypted reports broadcast by devices that describe how they perceive the wireless channel.
• Neural Network: A computing system inspired by the biological neural networks that constitute animal brains.