基于稀疏贝叶斯正则化的阵列SAR高分辨三维成像算法

阵列合成孔径雷达(Linear Array Synthetic Aperture Radar, LASAR) 3维成像技术是一种具有重要潜在应用价值的雷达成像新体制，但受线阵天线及平台尺寸限制，传统匹配滤波成像算法难以实现LASAR高分辨3维成像。该文利用LASAR回波信号及观测目标的先验分布特性，提出了一种基于快速稀疏贝叶斯正则化重构的LASAR高分辨3维成像算法。该算法先结合贝叶斯估计准则及最大似然估计原理，构造LASAR目标重构的稀疏贝叶斯最小化代价函数；再利用迭代正则化方法求解联合范数最优化问题实现LASAR稀疏目标高分辨3维成像。另外，针对稀疏贝叶斯正则化成像运算量大的问题，结合位置预测快速成像思路，利用阈值分割算法对稀疏粗成像进行强目标提取，进而提升算法运算效率。仿真数据和实测数据验证了该文算法的有效性。 

LASAR High-resolution 3D Imaging Algorithm Based on Sparse Bayesian Regularization

Linear Array Synthetic Aperture Radar (LASAR) is a novel and promising radar imaging technique. It is difficult to achieve high-resolution LASAR three-dimensional (3D) imaging using the traditional imaging methods based on match filter, because of limitations by the sizes of the linear array antenna and the platform. In this paper, by exploiting the prior distribution of the LASAR echoes and the observed scene, an LASAR high-resolution 3D algorithm based on sparse Bayesian regularization is proposed. The algorithm first combines the Bayesian principle and maximum likelihood estimation theory, and then a sparse Bayesian minimum cost function is constructed for LASAR target recovery. Second, using an iterative regularization reconstruction method, high-resolution imaging of LASAR sparse targets is achieved by solving a joint-norms optimization problem. In addition, for the problem of a large amount of sparse Bayesian regularization imaging, combined with the position prediction fast imaging idea, the threshold segmentation algorithm is used to extract the strong target of sparse coarse imaging, and then the algorithm operation efficiency is improved. Simulation and experiment results are presented to confirm the effectiveness of the algorithm.