在ISAR成像中,距离和方位分辨率分别受发射信号带宽和成像积累角的限制。基于压缩感知(CS)理论,该文提出了一种2维联合超分辨ISAR成像算法。首先建立ISAR观测信号模型并构造2维超分辨字典,然后利用ISAR图像的稀疏先验信息将2维联合超分辨成像建模为最小l1范数的优化问题,最后提出一种快速算法求解该优化问题。该方法进行距离维和方位维的联合处理,有效利用了回波数据的2维耦合信息;通过共轭梯度(CG)运算,快速傅里叶变换(FFT),Hadamard乘积等操作,有效提高了算法的实现效率。仿真和实测实验验证了该算法的有效性。
For ISAR imaging, the range and cross-range resolutions are constrained by the bandwidth of transmitted signal and Coherent Processing Interval (CPI). In this paper, a novel algorithm of Two-Dimension(2D) joint super-resolution ISAR imaging is addressed based on Compressive Sensing (CS) theory. The ISAR observation signal model is established, where the 2D super-resolution dictionary is formed. By exploiting the sparse prior information of ISAR image, 2D super-resolution imaging is mathematically converted into the l1 norm optimization. The super-resolution ISAR imaging can be realized with accuracy via fast optimization algorithm. In the proposed algorithm, the 2D coupling information of the echo can be effectively utilized through the joint processing of range and azimuth dimension. Besides, the efficiency of the proposed algorithm is improved by using the Conjugate Gradient (CG) algorithm, Fast Fourier Transform (FFT) and Hadamard multiplication operations. Simulation and real-data experiments verify the effectiveness of the proposed algorithm.