利用线性Bregman迭代的ISAR高分辨成像

High Resolution ISAR Imaging by Linearized Bregman Iteration

为获得ISAR方位向高分辨成像性能,基于压缩感知理论,提出了一种具有方位向高分辨能力的算法。该算法基于线性Bregman迭代(LBI)实现ISAR方位向高分辨。首先将重构精度高的LBI推广到复数域;然后进行了算法的理论分析,与正交匹配追踪算法对比分析了LBI算法的有效性以及稀疏度和采样率对重构精度的影响,仿真结果表明推广到复数域后的LBI能够有效实现复数稀疏信号的重构,验证了算法的优越性;最后将算法运用于ISAR方位向成像。仿真结果验证了LBI算法可明显提高ISAR方位向分辨率。

In order to achieve high resolution azimuth imaging performance in inverse synthetic aperture radar（ISAR）,a new imaging algorithm based on compressive sensing（CS）theory is proposed.High resolu-tion in azimuth is implemented via linearized Bregman iteration（LBI）.Firstly,LBI with high reconstruction precision is extended to the complex field,and then the algorithm is analyzed theoretically,the effectiveness of LBI and the influence yielded by sparsity and sampling rate is compared with orthogonal matching pursuit （OMP）algorithm.Results show that the extended LBI can realize effective reconstruction and verify the su-periority of LBI.Finally,the algorithm is applied to ISAR azimuth imaging.Simulation results indicate that the presented algorithm can improve ISAR azimuth resolution.