Abstract: Through-the-Wall Radar（TWR） imaging enables precise position targets localization and high-resolution imaging. However, interior wall position uncertainties bring obvious mess to it under multipath effect. To solve the above problem, on the basis of reasonable models the target multipath echo of through-the-wall radar, the quasi-Newton method and group-sparse based imaging reconstruction algorithm is proposed for TWR imaging. The proposed imaging algorithm can achieve precise target localization and ghost-free based on the interior wall localization reestablished precisely. The feasibility and validity of the proposed imaging method is verified by the numerical simulation results.

Key words: Through-the-Wall Radar（TWR）, group-sparse reconstruction, quasi-Newton method, imaging algorithm

摘要： 多径效应环境中边墙位置的不确定性给穿墙雷达系统的目标定位和高分辨率成像的要求带来很大的挑战。针对上述问题，在对穿墙雷达多径回波信号进行合理建模的基础上，提出一种基于拟牛顿法和块稀疏重建的成像算法。所提成像算法在精确重建边墙位置的基础上具有成像目标定位准确，无虚假目标的优点，数值仿真实验结果验证了所提成像算法的有效性和正确性。

关键词: 穿墙雷达, 块稀疏重建, 拟牛顿法, 成像算法