Research On Array Localization System Based On Digital Acoustic Sensor

With the progress of science and technology and the demand of market,acoustic array localization technology has become a hot spot in the field of array signal research.Acoustic array localization system uses acoustic sensor array to collect sound signal data,and then uses sound source localization algorithm to process the collected sound signal to obtain the position of sound signal.At present,the traditional acoustic array localization system has the problems of large volume,high power consumption and complex structure;sound source localization algorithm has the problem of long operation time and cannot meet the real-time requirements.For the above problems,in this thesis,an acoustic array localization system with small volume,low power consumption and strong expansibility is designed by combining digital acoustic sensors with ZYNQ platform;the sound source localization algorithm is optimized,the computation time is shortened and the real-time performance is improved.The work of this thesis is as follows:(1)The near-field model and far-field model of the received signal of the acoustic sensor array are analyzed,and the structure of the acoustic sensor array is theoretically deduced and analyzed.Since the actual collected signal data contain noise,in order to obtain higher quality signal data,signal preprocessing is studied.(2)The data acquisition and processing system based on the combination of digital acoustic sensor array and processor ZYNQ 7020 is designed,which completes multi-channel data synchronous acquisition,demodulation,buffering and network port transmission.The human-computer interaction interface of acoustic array localization system is designed by MATLAB platform,which realizes the functions of real-time receiving data,interface localization display and saving localization results.(3)In terms of sound source localization algorithm,the conventional beamforming algorithm and MUSIC high-resolution spectrum estimation algorithm are analyzed,and the sound source localization algorithms based on broadband signals are further deduced as ISM algorithm and CSM algorithm.Because of the time-consuming operation of the conventional beamforming algorithm,the regional shrinkage algorithm is combined with the conventional beamforming algorithm to shorten the operation time of the algorithm and improve the real-time performance.(4)Combined with the actual application scenarios,the experimental platform of acoustic array localization system is built,and the algorithms are compared under different distances and different frequencies.The results show that the scheme of this system is effective and can achieve sound source localization well;at the same time,the optimized conventional beamforming algorithm takes less time than the conventional beamforming algorithm.It meets the real-time and accuracy requirements of acoustic array localization system.