A COMPARATIVE STUDY ON ISAR IMAGING ALGORITHMS FOR RADAR TARGET IDENTIFICATION

Abstract

Inverse synthetic aperture radar (ISAR) images represent the two-dimensional (2-D) spatial distribution of the radar cross- section (RCS) of an object and, thus, they can be applied to the problem of target identiflcation. The traditional approach to ISAR imaging is the range-Doppler algorithm based on the 2-D Fourier transform. However, the 2-D Fourier transform often results in poor resolution ISAR images, especially when the measured frequency bandwidth and angular region are limited. Instead of the Fourier transform, high resolution spectral estimation techniques can be adopted to improve the resolution of ISAR images. These are the autoregressive (AR) model, multiple signal classiflcation (MUSIC), and matrix enhancement and matrix pencil MUSIC (MEMP-MUSIC). In this study, the ISAR images from these high-resolution spectral estimators, as well as the FFT approach, are identifled using a recently developed identiflcation algorithm based on the polar mapping of ISAR images. In addition, each ISAR imaging algorithm is analyzed and compared in the framework of radar target identiflcation. The results show that the dynamic range as well as the resolution of the ISAR images plays an important role in the identiflcation performance. Moreover, the optimum size of the subarray (i.e., covariance matrix) for MUSIC and MEMP-MUSIC in terms of target identiflcation is experimentally derived.