Research On Sound Source Location System Based On MEMS Digital Microphone Array

The microphone array is used to sample the sound field time and space in a certain geometry, and is widely used in video conferencing, speech enhancement,vehicle detection, velocity estimation and orientation, hearing aids and machine audiometry. The development of micro, flexible and efficient sound source positioning system is an important trend in the development of microphone array technology. The spatial power spectrum of the signal can estimate the direction of the direction of the signal. However, when the aperture of the array decreases, the width of the main powerof the spatial power spectrum increases, resulting in a decrease in the angular resolution of the array space. In order to obtain high positioning accuracy in the micro-micrometer array, it is necessary to improve the spatial spectrum for the broadband and coherent characteristics of the sound source. In this paper, based on the MEMS microphone array of sound source positioning system, is a small, low power consumption, high integration and real-time general-purpose sound field positioning module.In this paper, we first analyze the influence of channel error (phase error) and array aperture on the positioning performance of DOA estimation, and point out the reason why the spatial resolution of microarray spatial spectrum is low. Then, the high resolution MVDR algorithm and the super - resolution spectrum estimation MUSIC algorithm based on the minimum variance - no - distortion criterion are studied deeply,and the two algorithms are used in the spatial spectrum estimation of the wide - band far - field speech signal. Finally, the MEMS microphone array system is built, and the experimental results of the two algorithms are compared and analyzed. The experimental results show that the spatial resolution of the microarray has been greatly improved.