基于声源定位的头部姿态检测系统研究与实现

Research and Implementation of Snoring Head Posture Detecting System Based on Sound Source Localization Technology

研究了声源定位技术在鼾声检测上的应用,主要研究鼾声的声学特性、声源定位技术原理,并利用声源定位技术对人体打鼾时头部姿态进行识别.采用广义互相关函数法对多路声音采集系统采集到的鼾声信号进行延迟估计计算,根据得到的相对时延信息进行声音定位,结合定位信息和朝向信息推断出当前受试者打鼾时头部姿态.采用数据采集卡进行高速信号采集,使用虚拟仪器Labview进行信号处理和算法实现,实验中受试者以不同头部姿态在该系统监测下模拟打鼾,通过该系统检测头部姿态,综合判断正确率达到80%.

Sound source localization technology in the application of snoring detecting was studied,mainly including the aspects of acoustic characteristics of snoring as well as sound source localization technology principle and was used to detect the head posture of snoring human. The generalized cross-correlation （ GCC） function algorithm was used to calculate relative time-delaying estimation of snoring signals from mulita-channel sound acquisition system and sound position was obtained based on the relative delay information. Face orientation and head location information were combined to deduce the current head posture. The head posture detecting system contained high speed signal acquisition which was compiled by using DAQ. Besides,algorithm and signal processing was compiled by Labview. In the experiment, the subject simulated snoring under the detecting of system and the accuracy of the results is up to 80% .