基于坐标系变换的三维声源定位算法

Three-dimensional acoustic localization algorithm based on coordinate conversion

采用基于声达时间差(TDOA)的声源定位技术,给出了一种基于坐标系变换的三维声源定位算法,该算法弥补了以往算法繁琐、计算量大、定位不准确的缺点.将一个人为设定的坐标系转化为特定的坐标系,使得在原坐标系下处于任意位置的4个麦克风的坐标转化在新坐标系下具有一定自由度的特定3点的坐标和另一任意点的坐标,并在此新坐标系下快速计算出目标声源的坐标,从而确定声源的实际位置.将实际测量的声源位置与算法的计算结果进行比较,结果表明了该算法的有效性和实用性.

The existing acoustic localization algorithms based on time difference of arrival （TDOA） are computationally quite elaborate and are inaccurate to locate sound source. This work presented a new three-dimensional acoustic localization algorithm based on coordinate conversion. The algorithm calculates the sound source＇s coordinate in a new coordinate system converted from the original coordinate system in which four microphones have random coordinates. In the new coordinate system, three microphones have fixed freedom, while the last one＇s coordinate is random. After that, the actual position of the sound source can be quickly figured out. Comparison between the calculated results according to the algorithm with the results measured directly proves that the algorithm is efficient and practical.