色噪声背景下基于声矢量阵列孔径扩展的相干目标波达方向估计  

Coherent Target Direction-of-Arrival Estimation Based on Aperture Expansion of Acoustic Vector Array Under Colored Noise

作　　者：刘兆霆[1,2] 阮谢永[2] 刘中[1] 

机构地区：[1]绍兴文理学院数理信息学院,浙江绍兴312000 [2]南京理工大学电子工程与光电技术学院,江苏南京210094

出　　处：《兵工学报》2011年第3期257-263,共7页Acta Armamentarii

摘　　要：在目标波达方向(DOA)估计中,更大的阵列孔径往往能够获得更高估计分辨率和估计精度。然而,在色噪声背景或相干目标情况下,子空间类DOA估计算法性能急剧下降。针对这个问题,采用2个分置且相互垂直的均匀线性矢量阵列,通过阵列之间的协方差矩阵前向/后向平滑进行解相干预处理,并消除有色噪声的干扰;然后利用传播因子方法(PM)得到目标的方位角和仰角估计,无需奇异值分解和角度搜索;算法同时结合去模糊处理使得当阵元间隔大于半波长并在一定范围内增加时,其估计性能得到显著改善,且无需参考估计值的计算,节省了运算量。最后论文给出了仿真,验证了算法的有效性。An array with larger aperture generally offers enhanced resolution and direction-of-arrival（DOA） accuracy.However,in the presence of spatially colored noise or coherent sources,the performances of the subspace-based estimators degrade critically.A computationally efficient coherent sources direction finding method in colored noise using two uniform linear vector sensor sub-arrays far separated but orthogonally oriented is proposed.It is based on the forward-backward average covariance matrix of the observations between two sub-arrays,the matrix is used to eliminate the effect of the spatially colored noise and de-correlate the sources coherency.Then,a so-called propagator method（PM） is employed to obtain the azimuth-elevation angle estimates without eigen-decomposition and 2D iterative searching.The proposed method combined with the disambiguation processing can improve the resolution and DOA accuracy by increasing the sensor spacing in a certain range without the computational expensive derivation of coarse reference estimates.Monte-Carlo simulations are presented to verify its effectiveness.

关 键 词：声学 声矢量阵列 有色噪声 相干目标 波达方向 

分 类 号：TN911.7[电子电信—通信与信息系统]