Optimized Radio Frequency Footprint Identification Based on UAV Telemetry Radios.

Area of Science:

• Electrical Engineering
• Computer Science
• Aerospace Engineering

Background:

• Unmanned Aerial Vehicles (UAVs) proliferation necessitates robust detection and identification for airspace security.
• Telemetry radios are critical for Beyond Visual Line of Sight (BVLOS) UAV operations, offering a unique signal source for identification.

Purpose of the Study:

• To develop and evaluate a novel UAV identification system leveraging transient signals from telemetry radios.
• To optimize signal detection and improve the accuracy of Radio Frequency (RF) fingerprint identification for UAVs.

Main Methods:

• An optimized EC-α algorithm for transient signal start-point detection.
• A Convolutional Neural Network (CNN) for feature extraction from raw I/Q data.
• Clustering using Self-Organizing Map (SOM) and a proposed Clustering Signals Joint Identification (CSJI) algorithm.

Main Results:

• The proposed UAV RF identification system achieved an average accuracy of 92.3% at 30 dB SNR in indoor tests.
• Demonstrated superior performance compared to Support Vector Machine (SVM) and K-Nearest Neighbors (KNN) algorithms.
• Outdoor experiments confirmed the practical feasibility of the developed identification approach.

Conclusions:

• The novel UAV identification framework based on telemetry radio transient signals is effective and accurate.
• The optimized EC-α, CNN, SOM, and CSJI algorithms contribute to enhanced RF fingerprint identification.
• This approach offers a promising solution for UAV detection and security in critical airspace.