Accurate Detection and Classification of Low RCS Targets using their Micro Doppler Signatures

Abstract

Many countries are currently trying to perfect their drone attack detection tech nology for self-defence and early drone detection/classification to initiate the appro priate counter-drone systems for protecting the people and civil/military infrastruc tures. To combat all these threats, we need a technology/system that can handle the small sizes of such targets for their detection with range information, accurate micro Doppler (m-D) imaging, and classification, which are the main research contributions of this thesis. First, a C-band (5.3 GHz) continuous wave (CW) radar is designed and developed to detect the m-D signatures of low-RCS targets such as RC planes, two/three-blade propeller systems, mini-helicopters, quadcopters, and bionic birds. These targets are operated before the radar (in stationary position), and m-D signatures for their x individual and simultaneous operations are collected. The m-D signatures of low-RCS targets are generated using short-time Fourier transform (STFT). But STFT has one limitation: fixed time and frequency resolutions or fixed window size. So,there is a trade-off between time and frequency resolution.