Improving Indoor WiFi Localization by Using Machine Learning Techniques.

Area of Science:

• Computer Science
• Engineering
• Geomatics

Background:

• Global Positioning System (GPS) is effective outdoors but limited for indoor environments.
• Accurate indoor positioning is crucial for emerging applications and services.
• Existing indoor positioning methods often face challenges with accuracy and robustness.

Purpose of the Study:

• To present a novel architecture for high-accuracy indoor positioning.
• To leverage machine learning techniques for improved indoor location estimation.
• To provide insights into optimal machine learning methods for indoor positioning tasks.

Main Methods:

• Implementation of a novel architecture for indoor positioning.
• Application of machine learning techniques for error reduction.
• Utilizing a divide-and-conquer strategy to enhance positioning precision.

Main Results:

• Achieved a mean absolute error (MAE) of approximately 1 meter for latitude and longitude.
• Demonstrated a precise and practical solution for indoor positioning challenges.
• Identified effective machine learning techniques for indoor positioning.

Conclusions:

• The proposed novel architecture offers a significant advancement in indoor positioning accuracy.
• Machine learning and divide-and-conquer strategies are effective for overcoming indoor positioning limitations.
• The findings provide a valuable resource for developing future indoor positioning systems.