Abstract
Long synthetic aperture radar (SAR) offers the advantage of achieving higher resolution by utilizing longer synthetic aperture times, which makes it a promising technology for ocean observation in the future. However, compared to SAR systems with shorter synthetic aperture times, it suffers more severely from issues such as image defocusing, blurring and artifacts during the observation of maritime targets, due to motion errors. To improve the quality of SAR imaging against motion errors in long synthetic aperture time scenarios, this paper proposes a combined motion compensation (MOCO) method based on subaperture processing. The method first divides the full aperture data into several subapertures. Within each subaperture, the platform is assumed to move at approximately constant velocity. The major imaging step is then combined with two motion compensation operations, which are performed individually within each subaperture. Then, the processed subaperture data are stitched together, and finally, the residual errors are compensated by the third MOCO, resulting in the final image. By simulating maritime observation targets with point targets, simulation results demonstrate that the proposed MOCO algorithm effectively reduce the influence of motion errors, suppress the sidelobe interference to the imaging, and improve the focusing accuracy. Compared with other classical MOCO algorithms, the ISLR_r and ISLR_a metrics show improvements of 0.2662 and 0.8170 dB, respectively. Further verification of the proposed method is conducted by processing the imaging results of measured sea surface data. The proposed algorithm produces clearer wave textures and achieves better imaging performance on targets such as ships in the sea. This result validates the effectiveness and superiority of the proposed method. The proposed method effectively addresses the need for high-precision motion error compensation in high-resolution imaging within long synthetic aperture time system.

This publication has 30 references indexed in Scilit: