摘要
针对阵列信号波达方向(DOA)和频率联合估计计算量大、参数配对较困难等问题,提出了一种基于子空间辨识方法的DOA和频率联合估计算法.该算法构造了一个特殊的状态空间模型,并通过选取辅助矩阵来抑制噪声.由子空间辨识方法得到广义可观测矩阵的估计值,再利用总体最小二乘(TLS)方法得到系统矩阵的估计值,由系统矩阵得到DOA和频率的估计值.该方法具有参数自动配对和计算量小的特点.计算机仿真验证了该算法的有效性.
In order to reduce the computational complexity and pair the parameters more easily,an algorithm for joint DOA and frequency estimation based on subspace identification is presented.Firstly,we construct a special state-space model and select an auxiliary matrix to restrain the noise.Secondly,we use subspace identification method to estimate the extended observable matrix,and get the estimation of system matrices using total least square(TLS) method.Finally,we get the estimation of DOA and frequency from the system matrices.The computational complexity of the algorithm is comparatively small,and the parameters estimated could be paired automatically.Simulation results are presented to demonstrate the effectiveness of the algorithm.
出处
《北京理工大学学报》
EI
CAS
CSCD
北大核心
2012年第1期77-81,共5页
Transactions of Beijing Institute of Technology
基金
国家自然科学基金资助项目(51075175)
关键词
子空间辨识
波达方向估计
频率估计
联合估计
总体最小二乘
subspace identification
DOA estimation
frequency estimation
joint estimation
total least square(TLS)