基于流形分离与频谱细化技术的低复杂度DOA估计方法  

Low-complexity DOA Estimation Method Based on Manifold Separation and Spectrum Refinement Technique

作　　者：姜科良 黄治江 解楠[1] JIANG Keliang;HUANG Zhijiang;XIE Nan(Institute of Electronic Engineering,China Academy of Engineering Physics,Mianyang 621900,China)

机构地区：[1]中国工程物理研究院电子工程研究所,四川绵阳621900

出　　处：《现代信息科技》2024年第18期10-15,共6页Modern Information Technology

摘　　要：基于阵列流形分离(MST)的共形阵列空间谱估计技术,采用快速傅里叶变换(FFT)来降低谱搜索的计算复杂度,但是在该方法中,空间谱搜索的俯仰角和方位角均为0~360°,存在计算搜索冗余问题。针对共形阵列空间谱估计问题,提出基于联合阵列流形分离技术和频谱细化技术的空间谱估计技术,避免计算搜索冗余问题,相比阵列流形分离技术计算复杂度显著降低。采用一个32阵元球面共形阵列进行计算机仿真,验证了所提出方法的估计性能和计算复杂度。Conformal array spatial spectrum estimation technique based on MST uses FFT to reduce the computational complexity of spectral search,but in this method,elevation angle and azimuth angle of the spatial spectrum search are both 0°~360°,and there is a problem of computational search redundancy.To address the problem of conformal array spatial spectrum estimation,spatial spectrum estimation technique based on joint MST and spectrum refinement technique is proposed to avoid computational search redundancy problem,reducing the computational complexity significantly compared to MST.A 32-element spherical conformal array is used for computer simulation to verify the estimation performance and computational complexity of the proposed method.

关 键 词：DOA估计 MUSIC算法 频谱细化技术 CHIRP-Z变换 低复杂度 

分 类 号：TN911.7[电子电信—通信与信息系统]