随机跳频压缩感知雷达动态高分辨距离成像  

Dynamic HRRP generation for random frequency-hopping compressive sensing radar

作　　者：游鹏[1] 刘振[1] 泰玉亮 王宏强[1] 黎湘[1] 

机构地区：[1]国防科技大学电子科学与工程学院,湖南长沙410073

出　　处：《中南大学学报（自然科学版）》2014年第4期1119-1127,共9页Journal of Central South University:Science and Technology

基　　金：国家杰出青年科学基金资助项目(61025006);湖南省杰出青年科学基金资助项目(11JJ1010);国家自然科学基金面上项目(61271442;61171133)

摘　　要：针对实际中难以事先确定随机跳频压缩感知雷达所需脉冲数的问题,设计一种高分辨距离像动态合成策略。该策略的核心是序贯发射和接收跳频脉冲,同时用接收数据序贯更新对目标距离像的估计,直至满足设定的停止条件为止。为实现快速序贯更新,提出一种高效的复数域同伦算法。该算法实现快速递归稀疏恢复,避免每次加入一个新的观测时都要求解一个新的优化问题。此外,利用距离像的相关性特征,给出一种合理的停止准则。研究结果表明:采用所提出的动态合成策略,能够利用雷达回波信号自适应地确定最优脉冲数,在非合作运动目标成像方面具有显著优势。Considering that the required pulse number was difficult to determine in advance for random frequency-hopping （RFH） compressive sensing radar, a dynamic synthesis strategy for high resolution range profile （HRRP） generation was proposed, whose core was to sequentially transmit and receive pulses, and to update the HRRP with the sequential received pulses until a proper stopping rule was satisfied. To efficiently implement the sequential update, a fast complex-valued homotopy algorithm was developed. This algorithm performs as an efficient recursive procedure of sparse recovery, thus avoids solving a new optimization problem from scratch. Furthermore, the proper stopping rule was given according to the correlation characteristics of HRRP. The results show that the optimal pulse number required in each CPI can be sought to adapt to the echo signal, and the established dynamic approach is more suitable for uncooperative moving targets.

关 键 词：高分辨距离像 随机跳频 压缩感知 复数值同伦 递归稀疏恢复 

分 类 号：TN95[电子电信—信号与信息处理]