牛顿迭代算法在声场定位中的应用

在声场定位系统中，发声系统和拾音系统之间往往没有直接联系，为了能精确判断出声源所处的位置，设计了一种由4个拾音器和1个单片机控制的声场定位系统.声源发出特定频率的声音信号，系统通过3个拾音器接收到声音信号，经过处理计算得到时间延时差.利用牛顿迭代算法，将时间延时差代入方程进行迭代计算，即能精确计算出声源具体位置，第4个拾音器用于校准所得结果.采用该算法只需最多4次迭代运算就能计算出声源位置坐标.利用上位机软件模拟测试算法，并在实物系统中测试，验证了牛顿迭代算法在声场定位中的可行性和精确性.该声场定位系统位置坐标可在液晶屏上显示，可实时描绘出声源移动轨迹.

Sound field positioning using Newton iterative algorithm

A novel measuring system with four pickups and a micro-control unit was presented to localize a certain acoustic source as the sound sources do not link directly to acoustic pickups in sound field posi- tioning systems. Three of the four pickups were used to capture the acoustic signals emitted by a source with a certain frequency, and the time differences of arrival between any two of three arrival signals were processed using Newton iterative algorithm to ascertain the source. The result was calibrated using the rest pickup. At most four iterations were needed to determine where the source was. The feasibility and accuracy of the algorithm were verified respectively in simulation environment and practical systems while the coordinates and movements of the source were displayed in real time.