基于波束形成的列车噪声源识别麦克风平面阵列性能仿真研究

麦克风阵列的几何结构形成了麦克风阵列工作时性能的基本限制,是影响噪声源识别系统性能的关键因素之一.对十字阵(37阵元)、六角阵(37阵元)、矩形阵(36阵元)在均匀加权时方向性、角度分辨率等性能进行仿真定量对比研究：十字阵具有最好的方向性为71.6 dB,六角阵具有最好的空间对称性且能抑制栅瓣;阵元数目变化时主瓣宽度的对比,表明阵元数大于100时增加阵元对改善角度分辨率作用不大;主瓣宽度与频率关系,显示随频率增加3种阵列输出信号畸变程度基本相同.

The simulation of planar microphone array performance on train noise source identification based on beamforming

The geometry structure of microphone array shape is important for their performance, and plays the key factors impacting for external noise source identification. The simulation of the array directivity, angle resolution and other performance of cross-array and hexagonal-array (37 elements), rectangular-array (36 elements) were performed. The results indicate that cross-array has the best directivity is 71.6 dB, hexagonal-array has the best space symmetry and also is able to suppress grating lobe. The comparison of main lobe width with array element number variation shows that increasing element number to improve angular resolution is not helpful when the number is greater than 100.Through analyzing the relationship between main lobe width and frequency concludes that the output signal distortion degree of three kinds of array is practically the same.