基于ADMM的低仰角目标二维DOA估计算法  

Two-dimensional DOA Estimation for Low-angle Target Based on ADMM

作　　者：马健钧 魏少鹏 马晖 刘宏伟[1] MA Jianjun;WEI Shaopeng;MA Hui;LIU Hongwei(National Laboratory of Radar Signal Processing,Xidian University,Xi’an 710071,China)

机构地区：[1]西安电子科技大学雷达信号处理国家重点实验室,西安710071

出　　处：《电子与信息学报》2022年第8期2859-2866,共8页Journal of Electronics & Information Technology

基　　金：国家自然科学基金青年基金(61901344);高等学校学科创新引智计划(B18039);博士后创新人才支持计划(BX20180239)。

摘　　要：针对面阵米波(VHF)雷达低仰角目标2维DOA估计问题,该文提出一种基于交替乘子法(ADMM)的快速2维DOA估计算法。该方法首先利用均匀面阵条件下方位、俯仰角无耦合的特性,将2维角度估计问题转化为两个1维角度估计问题,通过方位、俯仰维波束合成实现对目标信息提取;其次根据信号模型建立信号空域超完备表达式,利用ADMM方法完成对方位、俯仰角估计。该方法避免了2维联合估计复杂计算量,复杂度大大降低,且运算过程无需特征分解,进一步提高了运算效率。仿真结果表明了该算法的优越性。For the two-dimensional DOA estimation problem of low elevation target of Very High Frequency(VHF)array radar,a fast two-dimensional algorithm based on Alternating Direction Method of Multipliers(ADMM)is proposed.Firstly,the two-dimensional DOA estimation problem is transformed into two onedimensional DOA problems by using the uncoupled characteristics of azimuth and elevation under uniformed planar array,and the target information is extracted by azimuth and elevation dimensional digital beamforming,and then based on signal mode,the over-complete expression in the signal space domain is established.Finally,the ADMM algorithm is used to estimate azimuth and elevation.ADMM algorithm avoids the complicated calculation of two-dimension joint estimation,reduces greatly the complexity,and the algorithm process does not need the eigenvalue decomposition,which improves further the operation efficiency.Simulation results show the superiority of the algorithm.

关 键 词：2维DOA估计 低仰角目标 米波面阵雷达 ADMM算法 

分 类 号：TN958[电子电信—信号与信息处理]