摘要
在使用分布式麦克风阵列对目标声源进行定位时,为使定位误差达到最小,目前常采用最小二乘估计算法。但该算法会导致距目标声源越远的节点对定位结果的影响反而会越大,如果优先选取距目标声源较近的节点进行运算,又会造成未参与运算的节点定向信息的浪费。针对此问题,利用非线性规划理论,建立分布式声定位的非线性规划模型。仿真结果表明:采用非线性规划模型后,距目标声源较远的节点对定位精度的影响比最小二乘法小;当监测区域内可用节点数目较少时,此方法能有效提高定位精度。
While using distributed microphone array to locate target sound source,least-square estimation algorithm is often used in order to make the locating errors smallest.However,the algorithm can lead to the more farther away the node from target sound source,the larger effect on locating results.If the priority to select the node closer to the target sound source to calculate,but will inevitably lead to waste localization information of some node,which doesn't participate in operation.Aiming at this problem,nonlinear programming model for distributed acoustic location is founded.Simulation results show that the effect of remote node farther away from the target sound source on localization precision is smaller than the least squares method;and this method can effectively improve localization precision when a small number of nodes in the monitoring area is available.
出处
《传感器与微系统》
CSCD
北大核心
2013年第2期46-49,共4页
Transducer and Microsystem Technologies
关键词
分布式
目标声源
定位精度
非线性规划模型
distributed
target sound source
localization precision
nonlinear programming model