基于尺度变换原理的SAR波数域成像算法

Wavenumber Domain Algorithm Based on the Principle of Chirp Scaling for SAR Imaging

距离徙动算法(RMA)作为一种合成孔径雷达(SAR)频域成像算法,理论上能够达到最优性能。然而,该算法采用逐像素点卷积运算实现Stolt映射,其计算效率无法满足SAR大数据量处理需求。据此,该文提出基于尺度变换原理(PCS)的RMA成像算法。首先,将SAR回波数据沿距离向进行划分,利用子带参考距离处2阶距离方位耦合项与高阶项对子带信号进行补偿;然后,转化非线性Stolt映射为线性形式;最后,利用PCS原理实现Stolt插值,以实现高效率的数据重采样。所提PCS-RMA算法仅利用快速傅里叶变换和复矢量相乘操作即可实现改进型Stolt映射,兼具良好的聚焦性能与较高的计算效率。基于多组仿真数据与X波段1.2GHz带宽的机载SAR实测数据处理结果,验证了所提算法的有效性,同时该算法可进一步应用于弹载/星载/无人机载SAR数据的快速成像处理。

As a frequency-domain algorithm for Synthetic Aperture Radar(SAR) imaging, the Range Migration Algorithm(RMA) can theoretically achieve optimal performance. However, because its Stolt mapping is performed using pixel-by-pixel convolution, the computational efficiency of RMA is inadequate for massive SAR data processing requirements. In this paper, we propose a modified RMA based on the Principle of Chirp Scaling(PCS). First, SAR echo data is divided along the range direction, and the subswath signal is compensated by the second-order range-azimuth coupling term and high-order terms at the reference distance.Then, the nonlinear Stolt mapping is modified to become linear. Finally, Stolt interpolation is realized using PCS to efficiently resample the processed data. Demonstrating both well-focused performance and high computational efficiency, the proposed PCS-RMA employs only fast Fourier transforms and complex vector multiplication operations to achieve modified Stolt mapping. The processing results of several simulation data and X-band-measured airborne SAR data with a pulse bandwidth of 1.2 GHz verify the effectiveness of the proposed algorithm. The proposed algorithm can also be employed for the rapid processing of missile-borne,spaceborne, and drone-borne SAR data.