Indoor Crowd 3D Localization in Big Buildings from Wi-Fi Access Anonymous Data.

Area of Science:

• Computer Science
• Geostatistics
• Network Engineering

Background:

• Indoor crowd localization and counting in large public buildings face challenges with infrastructure, signal processing, and privacy.
• Conventional methods using cameras or sensors require extensive calibration, noise reduction, and complex data processing, often with privacy concerns.
• There is a growing need for privacy-preserving crowd monitoring techniques.

Purpose of the Study:

• To develop a privacy-preserving technique for estimating the localization of people in large public buildings.
• To utilize existing, already-deployed Wi-Fi infrastructure for crowd estimation.
• To leverage anonymous data for crowd analysis.

Main Methods:

• The study applies geostatistical techniques to access data from Wi-Fi Access Points (APs).
• It specifically uses the time series of the number of accesses per AP.
• No personal data or device identification is required, ensuring anonymity.

Main Results:

• Geostatistical methods generate a 3D spatial distribution representation of people within the building.
• The technique relies on the interaction between mobile devices and APs.
• Encouraging results were obtained from data collected at Wroclaw University of Science and Technology.

Conclusions:

• The proposed method offers a viable, privacy-preserving approach for indoor crowd localization and counting.
• It effectively utilizes existing Wi-Fi infrastructure, reducing deployment costs.
• The geostatistical approach provides high-quality spatial distribution insights.