Automatic ground target classification using forward scattering radar

Abstract

Experimental study is undertaken of the feasibility of forward scattering radar (FSR) and its application to automatic ground target classification. The radar itself, fundamental theoretical analysis, target recognition algorithm and the target's classification subsystem are introduced. For target recognition, the effect of shadow inverse synthetic aperture radar is used. The radar experimental set-up and experimentation results are discussed. For classification, a system is proposed, which extracts features from the radar measurements by using Fourier transform and principle component analysis and uses a nearest neighbour classifier. Speed estimation in FSR is also introduced. By analysing 850 experimentally obtained car signatures, the performance of the system is evaluated and the effectiveness of the system is confirmed. The limitations of the work and its future are also discussed.