Drone Classification and Localization Using Micro-Doppler Signature with Low-Frequency Signal

Abstract

Security issues, such as unsafe operation and terrorist activity have become more critical due to the rising popularization of drones in recent years. To address these issues, drone classification and localization techniques become more desirable. In this paper, we propose a new approach for drone classification and localization using micro-Doppler signature. After obtaining the micro-Doppler signature generated from drone propeller rotation, dimension reduction is carried out for feature extraction. Next, extracted features are fed into four different classifiers. Finally, classification and localization are accomplished by discriminating features corresponding to different types of drones and different locations of interest. We carry out the practical experiment with radio frequency signal at low-frequency band. The experimental results show that the proposed approach can effectively classify and localize drones with good robustness.