Intelligent Wireless Sensor Network Sensor Selection and Clustering for Tracking Unmanned Aerial Vehicles

Area of Science:

• Engineering
• Computer Science
• Signal Processing

Background:

• Effective Wireless Sensor Network (WSN) management is crucial for applications like Unmanned Aerial Vehicle (UAV) tracking.
• Traditional sensor clustering methods often create large, redundant clusters, increasing network costs and potentially degrading performance.
• Low-cost, bearing-only sensors present unique challenges in WSN management for target tracking.

Purpose of the Study:

• To develop an optimized sensor clustering technique for WSNs, specifically for UAV tracking using bearing-only sensors.
• To reduce network communication and computation costs associated with sensor data collection.
• To maintain high tracking performance while minimizing the number of active sensors.

Main Methods:

• Combination of a predictive posterior distribution methodology with a simplified objective function for sensor selection.
• Development of an optimization algorithm to identify and form smaller, efficient sensor clusters prior to data collection.
• Evaluation of the proposed method against traditional techniques for UAV tracking.

Main Results:

• The novel objective function successfully identifies optimal sensor clusters.
• The developed optimization algorithm reduces selected sensor cluster sizes by an average of 50%.
• The reduced cluster sizes maintain comparable tracking performance to traditional methods using more sensors.

Conclusions:

• The proposed method offers a significant improvement in WSN management for UAV tracking.
• Smaller sensor clusters lead to reduced network costs without compromising tracking accuracy.
• This approach is effective for low-cost, bearing-only sensor networks.