Automatic Detection of Missing Access Points in Indoor Positioning System ^†

Area of Science:

• Computer Science
• Electrical Engineering
• Ubiquitous Computing

Background:

• Accurate indoor localization is crucial for various applications, but Wi-Fi infrastructure degradation can significantly impact system performance.
• Existing Wi-Fi fingerprinting localization methods often suffer from reduced accuracy when Access Points (APs) become unavailable or change.
• Manual recalibration of localization models is time-consuming and impractical for dynamic environments.

Purpose of the Study:

• To develop an automated system for detecting missing Access Points (APs) in Wi-Fi-based indoor localization.
• To enhance the robustness and accuracy of Wi-Fi localization systems by enabling automatic model updates.
• To provide a universal solution applicable to any fingerprinting-based Wi-Fi localization model.

Main Methods:

• Implementation of a Wi-Fi localization system comprising a localization model and an AP detection module.
• Utilizing received signal strength (RSS) data from multiple mobile terminals to identify relevant APs and trigger model updates.
• Employing the Random Forest algorithm for the core localization model, with automatic detection of missing APs.

Main Results:

• Reduced mean horizontal error by 5.5 meters and floor prediction classification error by 0.26 during Wi-Fi infrastructure malfunctions.
• Demonstrated the system's capability to accurately detect missing and present APs across various occupancy scenarios and AP failure rates.
• Validated the universal applicability of the automatic AP detection module for fingerprinting-based localization.

Conclusions:

• The proposed automated AP detection and model rebuilding mechanism significantly enhances the reliability of Wi-Fi indoor localization.
• The system effectively mitigates accuracy loss caused by dynamic changes or failures in Wi-Fi infrastructure.
• This approach offers a practical and scalable solution for maintaining high-performance indoor localization systems.