Research On Identification Method For Vehicle Noise Source By Using Microphone Array

Vehicle noise is urgent to be control as it threatens the environment and endangers the safety of passengers.Noise source identification is the research on the spatial distribution of the noise,which provides an important basis for the following vehicle noise control.Based on the microphone array technology,this thesis studies the sound source identification method and validates it with the test.Firstly,the mathematic model of the uniform linear array and the uniform planar array is established based on the theory of array statistical model.Comparing the performance of uniform linear array and planar array,the simulation results show that the planar array has stronger space recognition ability,which determines that the planar array is used as the microphone array model in this paper.Secondly,comparing the simulation results of the two commonly used spatial spectrum estimation algorithms show that the Multiple Signal Classification(MUSIC)algorithm has a higher resolution at the low signal-to-noise ratio condition.But the MUSIC algorithm cannot distinguish two coherent or close signal which just using the characteristic value decomposition of the covariance matrix.Then a modified MUSIC algorithm is contrasted with traditional MUSIC algorithm,the simulation results show that the modified algorithm can distinguish the signal location with close distance,and the recognition accuracy is improved.Finally,the low-frequency volume source is used as the sound source model to verify that the modified MUSIC algorithm has good accuracy and stability.The sound location of static vehicle at four conditions is tested by the modified MUSIC algorithm and planar microphone array,the result shows the vehicle main noise place by the three-dimensional spatial spectrum and projection contour map.This paper aims to study on vehicle noise source identification method,the conclusion can be applied to more field of sound source localization,which has reference significance to the research of microphone array and high-resolution sound source identification method.