摘要
本文首先介绍了宽带阵列信号的模型,然后给出了一种新的基于信号子空间的宽带信号DOA估计算法。新算法直接分解接收数据协方差矩阵得到聚焦矩阵,通过聚焦变换将不同频率点的信号子空间聚焦到参考频率点,聚焦后得到单一频率点的数据协方差矩阵,再应用窄带MUSIC算法进行DOA估计。由于是直接分解接收数据协方差矩阵得到聚焦矩阵,因此新算法无需预估计波达方向,可以避免角度预估计对测向性能的影响,而且减少了计算量,提高了宽带信号DOA估计的实时性。最后在16元均匀线阵的情况下,选用宽带高斯信号源进行仿真,将新算法与RSS算法和TLS算法相比较,验证了新算法的优良性能。
This paper introduces the model of wideband array signal processing, then it gives a new method for wideband DOA estimation, which is based-on signal subspace. This new method directly decomposes the covariance matrix of the received data to get the focus matrix. The covariance matrix of single freque signal subspace of different frequencies to the focusing frequency, and then the MUSIC ncy is obtained by focusing method is used to estimate the direction of arrival. This new method does not need preliminary angle estimation, so it can avoid the impact of preliminary angle estimation on performance of DOA ; it also greatly reduces the amount of computation and the consumed time for DOA estimation to favor the real-time application of wideband DOA estimation. Computer simulation was carried out using uniform linear array with 16 sensors and wideband Gauss signal source, and simulation results verify the good performance of this new method though comparing it with RSS method and TLS method.
出处
《电子测量与仪器学报》
CSCD
2007年第6期54-57,共4页
Journal of Electronic Measurement and Instrumentation
关键词
RSS算法
TLS算法
DOA
聚焦矩阵
BSS算法
rotational signal subspace method, total least squares approach, direction-of-arrival, focus matrix, method which based-on signal subspace.
