Tools for Ground-Truth-Free Passive Client Density Mapping in MAC-Randomized Outdoor WiFi Networks

Area of Science:

• Computer Science
• Network Engineering
• Data Privacy

Background:

• Data privacy regulations restrict WiFi network operators' ability to track client activity.
• Randomized MAC addresses in client devices complicate traditional methods of counting and mapping users via Probe Requests.

Purpose of the Study:

• To develop a method for mapping WiFi client densities in outdoor networks despite MAC randomization.
• To extend the statistical proportionality between client counts and Probe Requests to spatial mapping.
• To eliminate the need for ground truth measurements (e.g., cameras) for calibration.

Main Methods:

• Proposed a toolkit of nine distinct tools.
• Utilized raw, randomized MAC Probe Request counts as input.
• Developed a method to transform Probe Request data into a calibrated client density map.

Main Results:

• Successfully extended the client count proportionality to spatial mapping.
• Created a density map of client distribution in outdoor WiFi networks.
• Enabled client mapping without reliance on external ground truth data.

Conclusions:

• The proposed toolkit effectively addresses the challenge of localizing randomized MAC clients.
• This approach provides a viable solution for WiFi network operators to understand client distribution for commercial and security purposes.
• The method offers a privacy-preserving way to gain insights into network usage patterns.