A Wi-Fi-Based Passive Indoor Positioning System via Entropy-Enhanced Deployment of Wi-Fi Sniffers

Area of Science:

• Computer Science
• Electrical Engineering
• Signal Processing

Background:

• Accurate indoor positioning is essential for various applications.
• Existing systems often require active user participation or specialized hardware.
• Passive systems offer a less intrusive alternative for device tracking.

Purpose of the Study:

• To develop a passive indoor positioning system (IPS) utilizing Wi-Fi signals.
• To optimize the deployment of Wi-Fi sniffers for enhanced positioning accuracy.
• To implement and evaluate a real-time passive IPS.

Main Methods:

• A modified Genetic Algorithm (GA) with an entropy-enhanced objective function was used for optimal sniffer deployment.
• Received Signal Strength Indicators (RSSIs) were collected as Wi-Fi fingerprints.
• A weighted k-nearest neighbourhood (WKNN) algorithm was employed for positioning.
• A surveying robot automated RSSI data collection across Wi-Fi bands (2.4 GHz and 5 GHz).

Main Results:

• Optimized sniffer placement using GA improved positioning accuracy.
• An offline positioning accuracy of 2.2 meters Root Mean Squared Error (RMSE) was achieved with 20 Wi-Fi sniffers.
• A proof-of-concept online passive IPS demonstrated real-time DUT detection and positioning.

Conclusions:

• The proposed passive IPS is effective for indoor device localization.
• Optimized sniffer deployment significantly enhances positioning accuracy.
• The system enables real-time tracking without requiring user collaboration or additional device interfaces.