Target recognition using a set of scale, rotation and translation invariant features extracted directly from the radar echoes

In this work we propose a method for classifying targets extracting a feature vector directly from the received signal without passing through the target radar image. The method assumes that the target is a rigid body and possesses at least four dominant scatterers. Besides these assumptions, no restrictions are imposed on the kind of relative motion between radar and target. The classification is performed in two main steps: feature extraction and template matching. Simulation results show application examples and performance of the proposed approach.