The Research And Realization Of The Sound Source Traceability Of The Microphone Array In Multiple Scenes

With the development of Internet and communication technology,the problem of sound source tracing has become a hot topic in acoustic research.However,due to the complexity and instability of sound field,it is also a difficulty in acoustic research.Sound source tracing has a wide range of applications and needs in industrial,military and civil fields.For example,it is applied to troubleshooting,monitoring,video conference,whistle capturing and other aspects.Therefore,this study not only has a very important theoretical significance,but also has a wide range of practical application value.The sound source tracing mainly uses microphone array for research.According to different kinds of sound scenes in different environments,this paper selects targeted sound acquisition equipment,carries out algorithm analysis,and studies the sound source tracing problem.The application of this paper is the acoustic part of the project supported by the open fund of the Key Laboratory of the Ministry of education of the virtual geographical environment,and the research scene from the anechoic room to the student workshop,to the outdoor small area non-stationary sound field,and finally to the outdoor large area near stable sound field,the sound environment is more and more complex,and the processing methods are also diverse.In view of different application situations,a variety of processing schemes are proposed.The main works are summarized as follows:(1)Based on the generalized cross-correlation method,the source tracing in anechoic room is studied.The generalized cross-correlation method in time delay estimation algorithm is studied in detail,and the sound source tracing experiment based on microphone array is carried out in the anechoic room.Finally,the sound source location results are obtained.(2)Research on indoor sound source tracing based on inverse distance weight interpolation algorithm.In the laboratory,the distributed wireless intelligent acoustic sensor array is used to collect data.The data is uploaded to the server for background call by building a network communication link platform.Based on the collected sound pressure level data,the inverse distance weight space interpolation algorithm is used to realize the location of static single sound source,the location of multiple sound sources and the tracking of dynamic sound source.In addition,the spatial-temporal distribution map of indoor sound field is obtained.In this paper,the inverse distance weight interpolation algorithm is introduced into the research of sound source tracing for the first time.The experimental results show that the algorithm is feasible.(3)The research of outdoor sound source tracing based on Kriging interpolation algorithm.In the noisy outdoor environment,the built-in sensor of smart phone is used to collect the information of acoustic environment.Compare and test the microphone performance of smart phones,then install “sound and scene information collection” APP in the qualified mobile phones.The collected information includes recording audio,recording time and GPS information.After preprocessing the recorded audio,the above three kinds of information are used to calculate the sound pressure level based on the position.Finally,Kriging interpolation is used to get the position information of the sound source.In this paper,Kriging space interpolation algorithm is applied to the source tracking of sound field for the first time.The experimental results show that good results are achieved and the tracking of dynamic target sound source is realized.(4)Research on the target sound source tracing in outdoor large area sound field.On the basis of the above content(3),the problem of outdoor large area and multi scene noise source is studied.The playground,community,square comprehensive and shopping mall are selected for the experiment.The experimental results show that the location of the target sound source and the distribution of the sound field in different periods are consistent with the actual scene,which verifies the feasibility of this experimental method.