基于时频子空间分解的宽带线性调频信号DOA估计  

DOA Estimation of Wideband Chirp Signals Based on Time-Frequency Subspace Decomposition

作　　者：黄克骥[1] 田达[1] 陈天麒[1] 

机构地区：[1]电子科技大学电子工程学院,成都610054

出　　处：《电子与信息学报》2004年第3期344-349,共6页Journal of Electronics & Information Technology

摘　　要：针对具有时变方向向量的宽带线性调频信号,该文建立了基于短时Wigner-Ville分布(WVD)的空间时频分布矩阵,通过对各个空间时频矩阵的特征分解获得对应的信号子空间和噪声子空间,给出了基于时频子空间投影实现多个时频点综合估计信号DOA的算法。利用空间时频分布的前后向平滑解决了具有相同时频特性信号的均匀线阵DOA估计问题。算法不需要聚汇和插值等复杂的矩阵变换,精度较高,计算简便.仿真实验显示该算法性能显著优越于基于矩阵插值的宽带调频信号DOA估计算法.In allusion to wideband chirp signals with time-varying direction vector, this paper constructs STFD(Spatial Time-Frequency Distribution) by short WVD(Wigner-Ville Distribution). By eigenvalue decomposition to get signal subspace and noise subspace, a algorithm of DOA estimation based on time-frequency subspace projection is proposed . Through spatial smoothing, the problem of DOA estimation of signals with the same time-frequency signature is solved. This method need not focusing and interpolation, has high precision and low computation. Simulation shows the algorithm has better performance than the method of DOA estimation of wideband signals based on matrix interpolation proposed recently.

关 键 词：宽带线性调频信号 空间时频分布矩阵 信号子空间 噪声子空间 DOA估计 

分 类 号：TN911.23[电子电信—通信与信息系统]