基于广义互相关时延估计(TDOA)算法的声源定位跟踪系统设计与实现
Design and Implementation of Sound Source Localization and Tracking System Based on Time Difference of Arrival(TDOA)Cross-correlation Algorithm
徐雪慧 1梅振龙2
作者信息
- 1. 武汉职业技术学院 电子信息工程学院,湖北 武汉 430074
- 2. 国家脉冲强磁场科学中心,湖北 武汉 430073
- 折叠
摘要
为实现在二维平面内对声源进行实时定位和动态跟踪,运用时延估计(TDOA)互相关算法设计一种原声监听头阵列的CC-TDOA声源定位跟踪系统,实现对多个原声监听头进行同步采样,再进行信号放大及运算处理,运用LCD屏实时显示目标声源的距离和方位角度,同时运用二维云台控制激光笔对准声源,并持续动态跟踪声源.模拟仿真及真实环境实验测试表明,采用基于到达时间差的互相关定位算法,计算量小,精度较高,测试角度误差小于 2º,距离误差小于 1.2%,可以满足多种智能应用场合中声源实时定位与跟踪的要求.
Abstract
To achieve real-time localization and dynamic tracking of sound sources in a two-dimensional plane,a CC-TDOA sound source localization and tracking system is designed and implemented based on the Time Difference of Arrival(TDOA)cross-correlation algorithm.The system synchronizes the sampling of multiple sound sensors,amplifies and processes the signals,and uses an LCD screen to display the distance and azimuth angle of the target sound source in real time.Additionally,a two-dimensional gimbal is used to control a laser pointer to align with the sound source and continuously track it.Simulation and real-world experimental tests demonstrate that the TDOA-based cross-correlation localization algorithm has low computational complexity and high accuracy,with an angle error of less than 2 degrees and a distance error of less than 1.2%.This system can meet the requirements of real-time localization and tracking of sound sources in various intelligent applications.
关键词
声源定位/互相关/TDOA定位算法/跟踪系统设计Key words
sound source localization/cross-correlation/Time Difference of Arrival(TDOA)algorithm/tracking system design引用本文复制引用
出版年
2024
NETL
NSTL
万方数据