中文摘要：

单矢量水听器高分辨定向技术研究对于远距离目标测向有重要意义．根据单矢量水听器自身具有流型的特点，将矩阵滤波器用于定向前的预处理，可以有效抑制不感兴趣方向的噪声．将滤波后的数据协方差进行特征值分解，得到新的噪声子空间，并修改传统MUSIC算法方位谱表达式，得到了适用于噪声条件下单矢量水听器的高分辨定向算法．仿真结果表明，在足够高的信噪比条件下改进算法方位估计结果可以达到渐近无偏估计．当信噪比较低时，改进算法有效提高了分辨概率并减小了波束宽度，能在噪声环境下实现高分辨方位估计．试验数据处理结果验证了算法的有效性．

英文摘要：

The high-resolution direction of arrival （DOA） estimation algorithm, based on a single vector hydrophone is a very important mechanism for detecting the direction of a target in long distance range. The spatial filter was used for pretreatment to restrain the noise from uninterested directions. The method is based on the premises that a single vector hydrophone has its own array type. The research study explores the idea that a new noise subspaee can be obtained after decomposing the covariance matrix of the filtered data and the improved algorithm used for the high-resolution DOA estimation of a single vector hydrophone with noise can also be established after amending the azimuth estimation fmTnula of traditional multiple signal classification （MUSIC） algorithm. The simulation indicate that the new algorithm achieves the asymptotically unbiased estimation under the condition of high signal to noise ra- tio （SNR） , as it raises the resolving probability effectively and narrows the beam width with low SNR. The experi- mental findings indicate that the new algorithm can be used to get high-resolution direction-of-arrival with noise, as well as validate the correctness of the new algorithm.