基于压缩感知理论的合成孔径激光雷达成像算法

SAL Imaging Algorithm Based on Compressed Sensing Theory

合成孔径激光雷达是一种主动式有源激光成像雷达,它具有远高于微波合成孔径雷达的成像分辨率,能够实现对远距离目标的精细成像。由于激光信号的带宽极大,普通的硬件设备难以满足奈奎斯特采样定理的要求。提出一种基于压缩感知理论的合成孔径激光雷达成像算法,该方法利用光外差方法探测回波信号,在此基础上通过对外差信号进行随机采样提取信号中的有效信息,使用正交匹配追踪算法实现对目标高分辨距离像图像的重构,最后结合频率变标算法得到目标的高分辨二维图像。仿真结果表明运用新算法对合成孔径激光雷达的回波信号进行采样,能使用远低于奈奎斯特定理所规定的采样率完成信号采样,并有效压低信号旁瓣,实现对目标的超高分辨成像。

Synthetic aperture LADAR （SAL） is an active laser imaging system. It has much higher imaging resolution than the microwave synthetic aperture radar, and can realize the refined imaging for a far distance target. Due to the extra large bandwidth of laser signal, the hardware of SAL is difficult to satisfy the requirement for Nyquist sampling rate. A new signal sampling method based on the compressed sensing （CS） theory is proposed to reduce the sampling rate of echoes in this paper. In the method, the echoes are randomly sampled by using an optical heterodyne process, and then the high- resolution range profile is recovered by using the orthogonal matching pursuit algorithm （OMP）. The 2-D image is recovered by using frequency scaling algorithm. Simulations demonstrate that high-quality images can be reconstructed even though the sampling rate is much lower than the Nyquist sampling rate.