Investigation Of Sources Localization Technique Optimization Based On Microphone Arrays

As an important tool for signal processing and source localization,microphone arrays have important value in engineering applications.As the application scenario is widened,the performance requirements for the source localization of the microphone array are also increasing.This article has carried out a series of work on the theme of performance optimization of microphone array sound source localization.Firstly,beamforming,DAMAS and MUSIC,which are widely used in sound source localization methods,are introduced,and each method is numerically verified.In order to improve the spatial resolution of beamforming,suppress spatial aliasing and side lobe intensity under the condition of limiting the number of microphones,a technique for synthesizing arrays of mobile microphones has been developed.In order to avoid the Doppler effect caused by the movement of the microphone,an approximate frequency correction formula based on Taylor series expansion is adopted.The numerical test concludes that a single mobile microphone synthesis array can achieve effective sound source localization,and the method is robust under different SNR conditions;the moving microphone synthesis array can significantly suppress the beam under the condition that the number of microphones is constant.The formed side lobes and spatial aliasing,and the algorithm has the characteristics of flexible construction of the synthetic array,so that the suppression effect can be more diverse,which brings more convenience to the experiment.Under the framework of compressed sensing,the sound sourcelocalization model and two solving algorithms of OMP and BP are introduced.The comparison between OMP and BP algorithm is carried out,and the conclusion that the BP algorithm has higher precision and the OMP algorithm is more efficient is obtained.The numerical simulation of the compressed sound source localization was carried out.The numerical experiments were carried out on the results of the sound source not on the grid point and the number of sound sources.The beamforming compression grid method is added to the entire algorithm flow as a pre-processing process of compressed sensing,so that the compression sensing positioning efficiency is significantly improved.