基于约束条件的广义正交匹配追踪CS雷达成像算法  

Generalized Orthogonal Matching Pursuit CS Radar Imaging based on Constraints

作　　者：夏朝禹 高瑜翔[1,2] 谢建峰 楚春阳 XIA Chaoyu;GAO Yuxiang;XIE Jianfeng;CHU Chunyang(College of Communication Engineering,Chengdu University of Information Technology,Chengadu 610225,China;Meteorological Information and Signal Processing Key Laboratory of Sichuan HigterEducation Institutes,Chengdu 610225,China)

机构地区：[1]成都信息工程大学通信工程学院,四川成都610225 [2]气象信息与信号处理四川省高校重点实验室,四川成都610225

出　　处：《成都信息工程大学学报》2020年第4期400-405,共6页Journal of Chengdu University of Information Technology

基　　金：四川省数育厅高校创新团队项目(15TD0022)。

摘　　要：强高斯噪声破坏了成像区域的稀疏性,造成传统压缩感知(CS)雷达B-scan像中出现若干虚假目标。针对以上问题提出一种基于约束条件的广义正交匹配追踪(C-gOMP)改进算法,可以显著提高CS雷达在强高斯杂波背景下的成像性能。首先,该算法将回波数据进行贪婪迭代;然后,使用代价函数对迭代后的系数施加更深层次的约束以保证整个函数的收敛性,即在重构过程中针对高斯分量进行抑制。仿真结果表明,在相同实验条件下,C-gOMP获得的距离向分辨率为传统匹配滤波法的2倍。在SNR为O dB时,成像成功率比gOMP高出20%,得到的二维B-scan像的M,sg.系数约为gOMP的2倍。Gaussian noise will destroy the sparsity of the imaging region,which causes the traditional compressed sensing(CS)radar b-scan inage to produce some fake targets.Aiming at the above problems,an improved generalized orthogo-nal matching pursuit(C-gOMP)algorithm based on constraints condition is proposed in this paper,which can signifi-cantly improve the imaging performance of CS radar under the background of strong Gaussian clutter.Firstly,,this arith-metic carry out greedy iteration of echo data,then,the cost function is used to impose a deeper constraint on the iterated coefficients to ensure the convergence of the whole function.Namely,Gauss components are restrained during signal re-construction.The simulation results show that under the same experimental conditions,the range resolution of the C-gOMP algorithm is twice than that of the traditional matched filtering method.When SNR=0 dB,the positioning success rate is up to 20%higher than that of the gOMP algorithm,and the image coefficients of Mau.obtained by C-gOMP are a-bout twice than that of gOMP algorithm.

关 键 词：雷达成像 压缩感知 噪声抑制 广义正交匹配追踪 

分 类 号：TN951[电子电信—信号与信息处理]