Abstract-The problem of unmanned aerial vehicles classification using continuous wave radar is considered in this paper. Classification features are extracted from micro-Doppler signature. Before the classification, the micro-Doppler signature is filtered and aligned to compensate the Doppler shift caused by the target's body motion. Eigenpairs extracted from the correlation matrix of the signature are used as informative features for classification. The proposed approach is verified on real radar measurements collected with 9.5 GHz radar. Planes, quadrocopter, helicopters and stationary rotors as well as birds are considered for classification. Moreover, a possibility of distinguishing different number of rotors is considered. The obtained results show the effectiveness of the proposed approach. It provides capability of correct classification with a probability of around 95%.

Abstract-The radar micro-Doppler signature of a target is determined by parts of the target moving or rotating in addition to the main body motion. The relative motion of parts is characteristic for different classes of targets, e.g. the flapping motion of a bird's wings vs. the spinning of propeller blades. In the present study, the micro-Doppler signature is exploited to discriminate birds and small unmanned aerial vehicles (UAVs). Emphasis is on micro-Doppler features that can be extracted from spectrograms and cepstrograms, enabling the human eye or indeed automatic classification algorithms to make a quick distinction between man-made objects and bio-life. In addition, in case of man-made objects, it is desired to further characterize the type of mini-UAV to aid the threat assessment. Also this characterization is done on the basis of micro-Doppler features.

No abstract

No abstract

No abstract