基于改进时延估计的声源定位算法

Acoustic Source Localization Algorithm Based on Improved Time Delay Estimation

针对强噪声和强混响条件下,室内声源定位算法收敛速度慢和定位精度低等问题,提出一种基于改进时延估计的声源定位方法.该方法建立在单源多元混响模型下,首先用四元十字型麦克风阵列估计时延;然后在广义互相关时延估计算法的基础上,引入二次相关法以削弱噪声干扰,同时采用LMS(最小均方)自适应滤波算法弥补广义互相关方法的不足,提高混响环境下的时延估计精度;最后,通过远场近似几何方法定位声源.实验结果表明,与相位变换加权广义互相关函数(GCC-PHAT)算法相比,该方法具有较好的抗噪能力与抗混响能力,能获得更准确的定位结果.

Aiming at the problems of slow convergence rate and low positioning accuracy of the indoor acoustic source localization algorithms under the condition of strong noise and strong reverberation,we proposed an acoustic source localization method based on improved time delay estimation.The method was based on a single source multiple reverberation model.Firstly,the quaternary cross microphone array was used to estimate the time delay.Secondly,on the basis of the generalized crosscorrelation time delay estimation algorithm,quadratic correlation method was introduced to weaken the noise interference.At the same time,the least mean square（LMS）adaptive filtering algorithm was used to compensate for the deficiencies of the generalized cross-correlation method and improve the time delay estimation precision in the reverberation environment.Finally,the acoustic source was located by the far-field approximation geometric method.Experimental results show that,compared with the generalized cross-correlation-phase transform（GCC-PHAT）algorithm,the proposed method has better anti-noise and anti-reverberation abilities,and achieves more accurate positioning results.