中文摘要：

设计了一个基于TelosB节点的无线传感器网络声源定位实验系统,该系统由控制节点和信息采集节点以及基站构成.重点解决了由于信息采集节点的休眠唤醒延时导致的采样异常,以及串口和无线传输速率限制造成的数据拥塞.在定位算法部分,采用了经典的基于广义互相关函数的时延估计定位算法.为了消除节点器件失配、相关噪声等非理想特性造成的干扰,增加了声音信号预滤波处理和广义互相关函数滤波等改进措施.通过实验表明,采用上述方法实现的定位系统能够在室内环境实现声源的定位,对于一个面积为9m2的定位系统,超过70%的定位结果绝对误差小于30cm.

英文摘要：

A wireless sensor network acoustic source localization system based on TelosB is designed in this article.This system consists of a control node,information collection nodes and a base station.Abnormal sampling lead by wake-up delay of the information collection nodes was resolved as well as the data block of the serial port.In localization algorithm,the time delay estimation localization algorithm based on the classical generalized cross correlation function was adopted.At the same time,in order to eliminate the effects brought by the non-ideal characteristics such as device mismatch and synchronization related noise among different nodes,we pre-filtered the signal and the generalized cross correlation function.It is demonstrated in experiments that the sound localization can realize acoustic source localization in indoor environment.For a positioning area of 9m2,absolute error of over70%localization results are less than 30 cm.