基于正四面体阵列的声源定位算法及误差分析  

Sound Source Location Algorithm and Error Analysis Based on Tetrahedral Array

作　　者：徐勤奇[1] 杨鹏[1,2] 

机构地区：[1]河北工业大学控制科学与控制工程学院,天津300130 [2]河北省控制工程技术研究中心,天津300130

出　　处：《计算机仿真》2013年第7期296-299,共4页Computer Simulation

基　　金：河北省自然科学基金项目(F2010000137)

摘　　要：研究麦克风阵列的声源定位问题,由于在语音通信中,声源和传播环境复杂,造成定位精度不高。为解决上述问题,结合声阵列在机器人系统中的应用,提出用正四面体阵列来对目标声源进行定位,利用时间差的定位方法,推导了方向角、仰角、距离的计算公式和由于时间误差引起定位误差的公式,并对结果进行了仿真分析,结果表明,正四面体阵列与平面阵列相比取得了较好的效果,不但提高了精度还降低了声阵列的盲区,尤其是在定距方面,克服了平面阵列随着距离的增大而定距误差变的很大的缺点,实现远距离定位。Study the sound source localization of the microphone array. In voice communications, the sound source and the propagation environment aer complicated, resulting in low positioning accuracy. In order to solve the above problems, associated with the application of acoustic array in robot system, the tetrahedral array was proposed to locate sound source target. Using the time difference positioning method, the paper deduced the location formulas and the fallibility formulas, and then the simulation shows that the method has a good performance compared with plane array, not only improves location accuracy but also decreases the blind space. Especially in the positioning distance, it overcomes the disadvantage that the error of planar array becomes biger as the distance increases.

关 键 词：听觉 声源定位 正四面体阵列 误差分析 

分 类 号：TP912.3[自动化与计算机技术]