Application of adaptive chirplet representation for ISAR feature extraction from targets with rotating parts

Abstract

The problem of feature extraction from inverse synthetic aperture radar (ISAR) data collected from targets with rotating parts is addressed. In traditional ISAR imaging, rigid-body motion is usually assumed. When non-rigid-body motions are present, it is not possible to obtain a focused image of both the target and the rotating part. To solve this problem, the radar signal is first parameterised using the adaptive chirplet signal representation. The signal from the body and that from the rotating part are then separated in the parameter space. Point-scatterer simulation results show that better geometrical features of the body and better micro-Doppler features of the rotating part can be extracted after the separation. The algorithm is also demonstrated using the measurement data from an in-flight aircraft and a walking person.