| 基于线性校正总体最小二乘准则的三维说话人定位算法 |
| |
| 引用本文： | 张奕,殷福亮,陈喆.基于线性校正总体最小二乘准则的三维说话人定位算法[J].通信学报,2009,30(12):106-112. |
| |
| 作者姓名： | 张奕 殷福亮 陈喆 |
| |
| 作者单位： | 大连理工大学电子与信息工程学院,辽宁,大连,116023 |
| |
| 基金项目： | 国家自然科学基金资助项目，高等学校博士学科点专项科研基金资助项目(200801410015)Foundation Items:The National Natural Science Foundation of China(NSFC)，The Specialized Research Fund for the Doctoral Program of Higher Education of China |
| |
| 摘 要： | 针对到达时间差(TDOA)度量误差和麦克风阵列误差,提出了一种基于线性校正总体最小二乘准则的三维说话人定位算法.该定位算法通过对球面误差准则进行总体最小二乘优化来控制阵列误差的影响,并通过对优化过程引入方向参数的二次约束来控制算法的方差.仿真实验表明,本算法可以同时兼顾定位的方差性能和偏差性能,给出稳健的定位结果.
|
| 关 键 词： | 三维说话人 定位麦克风阵列 到达时间差 总体最小二乘 |
Linear correction total least-squares approach to 3D acoustic source localization |
| |
| ZHANG Yi,YIN Fu-liang,CHEN Zhe.Linear correction total least-squares approach to 3D acoustic source localization[J].Journal on Communications,2009,30(12):106-112. |
| |
| Authors: | ZHANG Yi YIN Fu-liang CHEN Zhe |
| |
| Abstract: | An total least-squares based 3D acoustic source localization algorithm(LCTLS) by taking both sensors position errors and TDOA errors into account was proposed.The proposed algorithm optimized the localization errors by total least-squares criterion for reducing the bias of localization which was caused by sensors position errors,and maintained a stable variance of localization by taking a quadratic constraint of acoustic position vector.Simulation results show that the LCTLS can balance the performance of bias and variance of location,and obtain robust localization results. |
| |
| Keywords: | 3D sound localization microphone array TDOA total least squares |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《通信学报》浏览原始摘要信息 |
|
点击此处可从《通信学报》下载全文 |
