分布式声源定位算法实现

Distributed Source Localization Algorithm

为了提高声源目标在三维定位中的精度,建立声源定位三维仿真模型,通过测量声源到各传感器之间的到达时间差,计算出声源到达各传感器的距离差.以泰勒级数展开算法为核心,估计出声源的三维坐标,通过Matlab对均方根误差、声源坐标的估计进行仿真,通过VC++与Matlab进行混合编程实现声源定位仿真系统,实现传感器坐标的串口的实时采集.仿真结果表明所提出的算法具有较高的定位精度,估计位置与声源实际位置偏差不超过3 m.

In order to improve the accuracy of positioning the sound source in three-dimention, a 3- dimensional simulation model of sound localization was established. The differences, of between the distances of the sound source to sensors were the differences between the obtained times that the sound source takes to travel to sensors. The three-dimentional coordinates of the sound source were estimated by using the Taylor series expansion algorithm. The estimation was simulated based on root mean square error by Matlab. The simulation system of sound source localization was realized by VG＋＋ and Matlab programming. The serial real-time acquisition of the sensor coordinates was realized. The results show that the proposed algorithm is of higher accuracy ——the deviation of the estimated location and the actual location is less than 3 meters.