直角坐标多级后投影聚束SAR成像算法

在局部极坐标系下，快速多级后向投影算法(FFBPA)可以以较低的采样率对子孔径成像，但在不同局部极坐标系之间需大量的2维图像域插值实现图像融合。相比极坐标系，图像融合在直角坐标系下更容易实现。但在直角坐标系下进行子孔径成像的奈奎斯特采样率较高，这将影响直角坐标系下的成像效率。该文针对此问题提出一种谱压缩技术，通过对距离时域和距离频域两次补偿，大幅压缩了直角坐标系下子孔径成像的方位谱宽度。该文算法具有堪比原始后向投影算法(BPA)的成像质量和优于FFBPA的计算效率，且能够应用于非线性轨道SAR。最后通过星载0.1 m仿真实验和机载0.2 m实测实验验证了算法的有效性。

Cartesian Coordinates Factorized Back-projection Algorithm for Spotlight SAR

The Fast Factorized Back-Projection Algorithm (FFBPA) can reconstruct images in low sampling rate in Local Polar Coordinates (LPC). However, massive 2 dimensional image interpolations are required in image fusion from different LPCs. Image fusion is much easier in Cartesian Coordinates (CC), whereas, the Nyquist sampling rate of images in CC is higher, resulting in decline in the efficiency. To solve this problem, a spectrum compressing method is proposed. By compressing in range-time domain and range-frequency domain, the azimuth spectrum is greatly compressed. The image quality of the proposed method is similar to that of Back-Projection Algorithm (BPA) and is superior to that FFBPA. This method can also be used in SAR of nonlinear track. In the end, the validity of this method is proved by spaceborne SAR simulation data of 0.1 m resolution and airborne SAR real data of 0.2 m resolution.