参数迭代最小化稀疏信号重构ISAR成像算法  

ISAR Imaging Algorithm for Parameter Iterative Minimization Sparse Signal Recovery

作　　者：冯俊杰[1,2] 张弓 FENG Junjie;ZHANG Gong(College of Electronics and Information Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,China;School of Electrical Engineering,Liupanshui Normal University,Liupanshui,Guizhou 553004,China)

机构地区：[1]南京航空航天大学电子信息工程学院,南京211106 [2]六盘水师范学院电气工程学院,贵州六盘水553004

出　　处：《计算机工程》2018年第10期228-234,共7页Computer Engineering

基　　金：国家自然科学基金(61471191);航空科学基金(20152052026);贵州省科学技术基金(黔科合LH字[2014]7471号);贵州省重点学科项目(ZDXK201535);六盘水师范学院微波信息技术创新团队项目(LPSSYKJTD201402)

摘　　要：为实现稳健逆合成孔径雷达(ISAR)成像,提出基于参数迭代最小化贝叶斯稀疏信号重构的ISAR成像算法。建立ISAR稀疏成像信号模型,通过推导目标参数稀疏贝叶斯模型的联合概率密度函数,将ISAR成像转化为贝叶斯准则下的稀疏约束最大后验概率估计。对目标散射系数和噪声功率交替迭代优化求解,从而实现目标重构。实验结果表明,与SL0算法、OMP算法和BP算法相比,该算法的参数能够自适应调整,具有更好的成像效果。In order to obtain the robustness Inverse Synthetic Aperture Radar(ISAR)image,an iterative minimization Bayesian learning sparse signal recovery algorithm is proposed.Firstly,ISAR imaging is established,and the imaging problem is converted to sparse constraint Maximum A Posteriori(MAP)estimation under Bayesian criterion.Then,the target reconstruction and the noise power can be obtained simultaneously by exchanging iterative solution,and the target can be recovered.Compared with the conventional sparse recovery algorithms,the proposed algorithm can adjust parameters automatically.Real data experimental results show that the paramters of the proposed algorithm can be adjusted autonomically and have better imaging effect comapred with SL0 algorithm,OMP algorithm and BP algorithm.

关 键 词：逆合成孔径雷达 迭代最小化 稀疏信号重构 成像 最大后验概率 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]