旋转分数阶Fourier域构建SAR方位向成像算法的研究  

Research on Rotating Fractional Fourier Transform Domain for Azimuth-imaging Algorithm of Synthetic Aperture Radar

作　　者：王振力 王群[1] 陈先意 马如坡 刘晓迁 WANG Zhenli;WANG Qun;CHEN Xianyi;MA Rupo;LIU Xiaoqian(Computer Information and Network Security Department,Jiangsu Police Institute,Nanjing 210031,China;School of Computer and Software,Nanjing University of Information Science and Technology,Nanjing 210044,China)

机构地区：[1]江苏警官学院计算机信息与网络安全系,南京210031 [2]南京信息工程大学计算机与软件学院,南京210044

出　　处：《遥感信息》2020年第6期1-5,共5页Remote Sensing Information

基　　金：国家科技攻关计划项目(2018YFB1003205);国家自然科学基金项目(61502242);江苏省教育科学“十三五”规划课题项目(D/2020/01/22);江苏省高等学校自然科学研究面上项目(19KJB510022);江苏警官学院高层次引进人才科研项目(JSPI19GKZL405)。

摘　　要：距离多普勒算法在许多模式SAR尤其是正侧视SAR成像处理中被广为使用,但是其较低精度的SAR图像质量越来越不能满足实际应用需求。为解决传统方法方位向成像质量低的问题,提出旋转分数阶Fourier变换域构建SAR方位向上的高分辨率成像算法。文章详细推导SAR方位向信号运用分数阶Fourier变换时最优阶数的计算表达式,理论分析表明,该阶数具有唯一性且取决于方位向调频率、方位向抽样序列长度和方位向抽样频率;以获得的最优阶数为基础,完成旋转分数阶Fourier变换域和方位向高分辨率成像算法的构建。机载SAR模拟数据和星载SAR实测数据成像结果表明该算法的有效性。Range Doppler(RD)algorithm is widely used in synthetic aperture radar(SAR)imaging such as in the typical side-looking radar.However,the poor images produced by this algorithm fail to satisfy the needs of practical applications.In order to resolve the problem of low azimuth-imaging quality of this traditional method,a high performance SAR azimuth-imaging algorithm based on rotating fractional Fourier transform domain is proposed.In this paper,the expression of the optimal order for SAR azimuth signals using fractional Fourier transform is detailed derived.The theoretical analysis shows that the order is unique and depends on the modulation frequency,sampling length and sampling frequency on the azimuth direction.The rotating fractional Fourier transform domain and azimuthal high performance imaging algorithm are constructed based on the proposed optimal order.The effectiveness of the proposed algorithm is demonstrated by the imaging results of the simulation data of airborne SAR and the measured data of spaceborne SAR.

关 键 词：FOURIER变换 分数阶FOURIER变换 合成孔径雷达 距离多普勒算法 图像质量评估 

分 类 号：TP72[自动化与计算机技术—检测技术与自动化装置] TN95[自动化与计算机技术—控制科学与工程]