Research And Implementation Of Adaptive Real-time Positioning System Based On Microphone Array

With the development of artificial intelligence,sound source localization has become a hot topic in research fields such as intelligent robots,speech signal processing and so on.The high real-time and accuracy of positioning make it play a vital role in many aspects.The increase in the number of microphones can theoretically improve the positioning accuracy,but it could poses a huge challenge to real-time.In addition,for sound sources at different distances,the wrong choice of near-field model or far-field model in the calculation also brings large errors to the positioning.In view of real-time nature of multi-microphone array positioning,this article chooses a fiveelement cross microphone array.It is proposed to use FPGA as the processing platform to give full playto it advantages of parallel processing,and write the IP core of FPGA with Verilog to realize the data calculation to achieve high real-time performance.The model of the sound field are studied.For the positioning of the sound source of the far-field sources,the MUSIC algorithm is often used to determine the azimuth and pitch angle of the target sound source.However,due to the complexity of external factors,it is ofen the case that the peak value is not obvious or there are multiple peaks.To solve this problem,the time-domain generalization technique in the signal processing process is proposed,and this technique is used to eliminate the influence of the amplitude value.The effectiveness of this method was verifield through MATLAB modeling and simulation.Secondly,for the blind adaptation of the sound field model,this study proposes a fusion localization method adaptive to far-field and near-field sources.This method determines the model of the sound source by comparing the variance of the calculation results of the two models in the farfield and near-field.According to the corresponding model,secondary sound source localization is performed,which incresses the accuracy of localization under the same complexity.Finally,the sound source localization software and hardware system model is built,and the audio acquisition module is self-made.Base on the ZYNQ-7000 hardware platform,the underlying IP core is written in Verilog language,and the upper layer control program is written in C language.Experimental results show that with this improved method,the positioning system error is reduced by more than three times,and it overcomes the situation that the sound source localization must be determined by field source.