基于空时频联合分析的方位估计方法  

Research on DOA estimation algorithm using spatial time-frequency analysis

作　　者：宋海岩[1] 佟宁宁[1] 秦进平[1] SONG Haiyan;TONG Ningning;QIN Jinping(College of Electrical and Information Engineering,Heilongjiang Institute of Technology,Harbin 150050,China)

机构地区：[1]黑龙江工程学院电气与信息工程学院,黑龙江哈尔滨150050

出　　处：《黑龙江工程学院学报》2018年第3期42-46,共5页Journal of Heilongjiang Institute of Technology

基　　金：黑龙江省普通本科高等学校青年创新人才培养计划(UNPYSCT-2015101);哈尔滨市科技创新人才研究专项资金项目(青年后备人才项目)(2016RQQXJ117);黑龙江省博士后资助项目(LBH-Z15045)

摘　　要：随着时频分析理论和算法的迅猛发展,时频分析技术已经在方位估计(Direction Of Arrival,DOA)领域引起广大专家学者的广泛关注。首先讨论将时频分析理论应用于方位估计进而获得高空间分辨能力的方法。其次,为了综合利用所有相关的空时频分布点,提出联合对角化方法。该方法采用最小方差无畸变处理器(Minimum Variance Distortionless Response processor,MVDR),克服传统多重信号分类处理器(Multiple Signal Classification,MUSIC)需要估计信号子空间和噪声子空间的不足。数值仿真结果验证所提出的基于空时频联合分析的方位估计方法的有效性,同时表明该方法在低信噪比以及相干源条件下的优异性能。With the rapid development of the theory and algorithms for time-frequency analysis,it has already inspired some notable investigation in the context of Direction Of Arrival（DOA）estimation.This paper shows how time-frequency analysis can be applied to the DOA domain and achieve a higher spatial resolution.In order to combine all the relevant STFD points,the joint diagonalization technique is proposed.At the same time,the work adopts Minimum Variance Distortionless Response processor（MVDR）,rather than MUSIC,to avoid the estimation of the signal subspace and noise subspace.Numerical simulations illustrate the effectiveness of the DOA estimation algorithm based on the STFDs structure and also show that under some challenging scenarios such as low SNR and coherent arrivals,our proposed algorithms can distinguish closely spaced sources compared with conventional methods.

关 键 词：方位估计 空时频分布 联合对角化 最小方差无畸变处理器 

分 类 号：TN911.7[电子电信—通信与信息系统] TB566[电子电信—信息与通信工程]