Li Caipin, He Mingyi
School of Electronics and Information, Northwestern Polytechnical University, N 1, Dongxiang Rd, Xi'an 710129, China.
China Academy of Space Technology (Xi'an), Weiqu Street, Xi'an 710100, China.
Sensors (Basel). 2017 May 6;17(5):1058. doi: 10.3390/s17051058.
Geosynchronous Orbit Synthetic Aperture Radar (GEO SAR) has recently received increasing attention due to its ability of performing staring observations of ground targets. However, GEO SAR staring observation has an ultra-long integration time that conventional frequency domain algorithms cannot handle because of the inaccurately assumed slant range model and existing azimuth aliasing. To overcome this problem, this paper proposes an improved chirp-scaling algorithm that uses a fifth-order slant range model where considering the impact of the "stop and go" assumption to overcome the inaccuracy of the conventional slant model and a two-step processing method to remove azimuth aliasing. Furthermore, the expression of two-dimensional spectrum is deduced based on a series of reversion methods, leading to an improved chirp-scaling algorithm including a high-order-phase coupling function compensation, range and azimuth compression. The important innovations of this algorithm are implementation of a fifth-order order slant range model and removal of azimuth aliasing for GEO SAR staring observations. A simulation of an L-band GEO SAR with 1800 s integration time is used to demonstrate the validity and accuracy of this algorithm.
地球同步轨道合成孔径雷达（GEO SAR）因其对地面目标进行凝视观测的能力，近年来受到越来越多的关注。然而，GEO SAR凝视观测具有超长的积分时间，由于斜距模型假设不准确以及存在方位向混叠，传统频域算法无法处理。为克服这一问题，本文提出一种改进的Chirp-scaling算法，该算法采用五阶斜距模型，其中考虑了“走走停停”假设的影响以克服传统斜距模型的不准确性，并采用两步处理方法消除方位向混叠。此外，基于一系列反演方法推导了二维频谱的表达式，得到了一种改进的Chirp-scaling算法，包括高阶相位耦合函数补偿、距离向和方位向压缩。该算法的重要创新之处在于实现了五阶斜距模型以及消除了GEO SAR凝视观测中的方位向混叠。利用一个积分时间为1800 s的L波段GEO SAR进行仿真，验证了该算法的有效性和准确性。
