Non-Intrusive Privacy-Preserving Approach for Presence Monitoring Based on WiFi Probe Requests.

Area of Science:

• Computer Science
• Data Science
• Network Security

Background:

• Monitoring human presence and movement is vital for public safety, urban planning, and event management.
• Existing methods face challenges due to privacy regulations and randomization in device communication.

Purpose of the Study:

• To propose a novel, non-intrusive, privacy-preserving method for detecting people's presence and movement patterns.
• To address the challenge of randomized network management messages in WiFi-enabled devices.

Main Methods:

• Developed a de-randomization technique using clustering and matching of network management messages and radio channel characteristics.
• Calibrated the method with a public dataset, validated in rural and indoor environments, and tested in a crowded urban setting.

Main Results:

• Achieved over 96% device detection accuracy in rural and indoor environments when validated individually.
• Demonstrated over 70% accuracy for grouped devices in rural and 80% in indoor settings.
• Confirmed accuracy, scalability, and robustness in a crowded urban environment.

Conclusions:

• The proposed de-randomization method offers a viable, low-cost solution for analyzing population movement patterns.
• Identified computational complexity and parameter tuning as areas for future optimization and automation.