全极化雷达的多任务压缩感知目标识别方法  

Full-polarization radar target recognition of multitask compressive sensing

作　　者：翟庆林[1] 刘盛启 胡杰民[1] 占荣辉[1] 

机构地区：[1]国防科技大学电子科学与工程学院 [2]中国人民解放军31011部队

出　　处：《国防科技大学学报》2017年第3期144-150,共7页Journal of National University of Defense Technology

基　　金：国家自然科学基金资助项目(61471370;61401479)

摘　　要：为有效利用全极化雷达高分辨距离像(High Resolution Range Profile,HRRP)的丰富特征信息和全极化样本中各单极化HRRP均对应于相同目标姿态的特性,提出一种基于多任务压缩感知的全极化雷达目标识别方法。该方法约束在不同极化字典中选择来自相同角域的字典原子对相应极化方式下的HRRP进行表示,可以有效利用不同极化HRRP之间的相关信息提高目标识别性能。基于电磁散射数据对所提出的方法进行了测试,实验结果证明了方法的有效性。To efficiently utilize the information which can be extracted for target recognition and the character that different polarization channels characterize the same structure signature of a target using different polarization modes to boost recognition performance, a method for full-polarization HRRP recognition based on multitask compressive sensing was proposed. Each singlepolarization HRRP was represented by the atoms adaptively selected from its associated dictionary, and the atoms derived from different dictionaries corresponded to the same index set. Compared with the conventional methods, the proposed method has the significant advantage of exploiting the correlation among single-polarization HRRPs to enhance recognition performance. Experiments were carried out on simulated data, and the results demonstrate the efficiency of the proposed method.

关 键 词：雷达目标识别 全极化高分辨距离像 多任务压缩感知 

分 类 号：TN919[电子电信—通信与信息系统]