Fingerprint building and positioning based on wireless sensor networks for underground.

Area of Science:

• Wireless Sensor Networks
• Localization Algorithms
• Machine Learning for Positioning

Background:

• Accurate underground positioning is challenging due to signal obstruction and environmental changes.
• Existing wireless sensor network (WSN) positioning methods struggle with dynamic environments and fingerprint drift.
• Reliable fingerprint construction and maintenance are crucial for robust WSN localization.

Purpose of the Study:

• To propose an adaptive fingerprint construction and positioning algorithm for underground WSNs.
• To enhance the reliability and accuracy of WSN-based positioning in changing environments.
• To minimize positioning errors by effectively managing fingerprint data updates.

Main Methods:

• Dividing the underground area into sub-areas using the neighborhood principle.
• Implementing a reliability mechanism with calibration nodes to validate fingerprint availability.
• Utilizing a neighborhood mapping fingerprint model trained by backpropagation neural networks for fingerprint construction.
• Employing an adaptive network-based fuzzy inference system for real-time positioning.

Main Results:

• The proposed algorithm achieved an average positioning error of 3.03 meters.
• Performance was validated with a seven-day interval between training and testing datasets.
• The algorithm demonstrated adaptability to environmental changes, ensuring consistent positioning accuracy.

Conclusions:

• The developed algorithm effectively addresses the challenges of dynamic environments in WSN positioning.
• The combination of neighborhood mapping and adaptive fuzzy inference systems yields robust and accurate localization.
• This approach offers a promising solution for reliable underground positioning using wireless sensor networks.