Research On Signal Processing Of Sound Location And Recognition System

Passive acoustic detection technology has attracted wide attention due to its ability to solve many problems that traditional detection technology cannot handle.Among them,passive acoustic source detection technology,with its strong concealment,portable detection equipment,and precise measurement advantages,has been widely applied in various fields.It is the application result of the joint development of sensor and array design,signal pre-processing,wireless communication,and high-precision data acquisition technologies.In this thesis,we focus on the design of a real-time sound source identification and localization system,and conduct research from three aspects:theoretical research,software and hardware design,and practical environmental testing of the system.(1)First,the overall flow and framework of the passive acoustic detection system were designed,which consists of hardware framework and software framework.In terms of hardware,appropriate acoustic sensors were selected and the array size was determined based on the actual environment.The IAC-IMX6UL-KIT was chosen as the processor for the system,and ADS7828 was selected as the synchronous sampling chip.In terms of software,the overall flow of modules such as the acquisition module,recognition module,and localization module were designed.Based on the established framework,further research can be conducted on preprocessing,sound source recognition,and sound source localization theory and implementation.(2)Explored the characteristics of sound signals and the impact of environmental factors on sound propagation,and established corresponding models for sound signal propagation.Conducted preprocessing of the signals using bandpass filters and the Least Mean Square(LMS)adaptive algorithm.Applied the Mel Frequency Cepstrum Coefficient(MFCC)for feature extraction of the target sound source signals and employed the Back Propagation(BP)neural network for classification and recognition.Designed a quaternion triangular array model and derived relevant localization formulas.Utilized the Time Difference of Arrival(TDOA)algorithm based on arrival time differences for localization,and employed the generalized cross-correlation algorithm to calculate the corresponding time delays.Finally,implemented the corresponding functional modules and algorithms in the hardware environment by writing code based on the simulation results.(3)The functionality of each module of the designed sound detection system was tested in the actual environment.The test results show that the sound source real-time localization and detection system designed in this thesis can achieve the expected functions,and the recognition and positioning accuracy can meet the requirements.In addition,this thesis also analyzed the errors in the test results.