基于圆阵声源定位的目标上下界判定方法

Method of Determining the Upper and Lower Bounds of Target Based on Sound Source Localization with Circular Array

通过圆形阵列和波束形成可以实现对声源目标的定位，在近距离的噪声声源定位中，声源的方位角和俯仰角范围分别为0～360°和０～90°，但通过波束形成计算方位角时，只能给出半个空间０～180°内的目标方位角，当目标出现在180°～360°时，采用原有的公式和阵列顺序，计算出的功率谱是杂乱的，无法直接给出，需要通过对阵列数据进行旋转并二次计算才能得到方位角，显著增加运算时间。通过研究，提出了互相关测量法，初步判断声源所在方位，再通过波束形成算法计算出方位角，可以减小运算量，约降低1／3的计算时间。

Positioning of the sound source target can be achieved through a circular array and beamforming. In close-noise sound source localization, sound source azimuth and elevation angle range are of 0 ～ 360 and 0 ～90 degrees. But through beamform- ing calculated azimuth can only give half of the space within 0 ～ 180 degrees target,when the target occurs in 180 ～360 degrees it cannot be obtained directly. Using the original formula and the array sequence to calculate power spectrum is confusion, rotating the array data and the second calculation is needed to obtain the azimuth ,which greatly increasing the computational time. Through re- search, cross-correlation measurement method was provided, the initial position of the sound source was judged, and then through beamforming algorithm to calculate the azimuth, reduced the amount of computation by approximately one-third of the computation time. Abstract：Positioning of the sound source target can be achieved through a circular array and beamforming. In close-noise sound source localization, sound source azimuth and elevation angle range are of 0 ~ 360 and 0 ~ 90 degrees. But through beamform- ing calculated azimuth can only give half of the space within 0 - 180 degrees target,when the target occurs in 180 ~ 360 degrees it cannot be obtained directly. Using the original formula and the array sequence to calculate power spectrum is confusion, rotating the array data and the second calculation is needed to obtain the azimuth ,which greatly increasing the computational time. Through re- search, cross-correlation measurement method was provided, the initial position of the sound source was judged, and then through beamforming algorithm to calculate the azimuth, reduced the amount of computation by approximately one-third of the computation time.