LAUSR

Cross-range scaling is an important procedure for inverse synthetic aperture radar (ISAR) imaging since the cross-range resolution is always unknown due to the noncooperative targets. In fact, the essential problem of scaling is to estimate the rotation velocity of the target from echoes. Considering that most of the scaling methods have limitations on the requirement of some prior knowledge of the target’s motion, difficulties in handling some complicated-structure targets, and low-signal-to-noise ratio (SNR) conditions, we propose a novel scaling approach based on the multidelay discrete polynomial-phase transform (DPT) combined with keystone transform. Different from the second-order-DPT-based method, the proposed method separates the different components on the 2-D spectrum of time and delay time, which avoids the interference of cross terms and enhances the anti-noise ability. The experiments on linear frequency-modulated (LFM) signals, simulated ISAR echoes, and real-measured data verify the effectiveness and robustness of the proposed method.