RDIwS: An Efficient Beamforming-Based Method for UAV Detection and Classification

The detection and classification of unmanned air vehicles (UAVs) are disturbing challenges within contemporary radar systems, wherein the physical characteristics of UAVs, including their size and radar cross section (RCS), exert a substantial influence on radar’s detection capabilities. Smaller UAVs, characterized by reduced RCS values, often escape radar detection. In response to these challenges, this study introduces an efficient radar signal processing technique based on beamforming, termed range–Doppler integration while steering (RDIwS). RDIwS significantly enhances the signal-to-noise ratio (SNR) associated with UAVs, resulting in an increased detection probability and classification accuracy for these UAVs. Importantly, the RDIwS approach demonstrates superior performance to traditional multiple-input multiple-output (MIMO) methods and established beamforming-based techniques, showcasing its potential to significantly advance UAV detection and classification across various operational contexts. For four targets located at different angles and distances scenario, and at −30 dB SNR and false alarm probability of
$10^{-{5}}$
, the RDIwS beamforming-based method achieved a detection probability of 75% compared to 5% for steering-only beamforming, and no detection for MIMO radar case.