结合Sigmoid函数的自适应时延估计算法研究  

Study of Adaptive Time Delay Estimation Algorithm Combining Sigmoid Functions

作　　者：李典航 王申营 程骏 彭云锋 邹腾跃 LI Dianhang;WANG Shenying;CHENG Jun;PENG Yunfeng;ZOU Tengyue(College of Mechanical and Electrical Engineering,Fujian Agricultureand Forestry University,Fuzhou 350100,China)

机构地区：[1]福建农林大学机电工程学院,福建福州350100

出　　处：《自动化仪表》2025年第4期31-35,41,共6页Process Automation Instrumentation

摘　　要：当面对具有混响和强噪声的复杂环境时,传统的广义二次互相关时延估计算法往往存在精度不足的问题。为此,提出了一种改进的自适应时延估计算法。该算法的核心思想是利用前置自适应滤波器对声源信号进行降噪处理并获取信噪比参数,将信噪比参数与Sigmoid函数相结合并引入相位变换(PHAT)加权函数的加权因子中,使得时延估计算法具备根据噪声强度自适应调整加权的能力。以人声声源信号为例,利用Matlab进行不同信噪比环境下的性能仿真试验,并在五元十字麦克风阵列下完成声源定位的仿真。试验结果表明,在不同低信噪比下,改进算法的声源定位距离偏差比优于传统算法。因此,在复杂环境中,改进算法具有更高的定位精度和鲁棒性。该研究为复杂环境下的声源定位提供了新的解决方案,对于提升智能会议系统、远程监控系统、声源探测仪表等领域的声源定位精度具有重要意义。When facing complex environments with reverberation and strong noise,traditional generalized quadratic intercorrelation time delay estimation algorithms often suffer from insufficient accuracy.For this reason,an improved adaptive time delay estimation algorithm is proposed.The core idea of the algorithm is to use the front adaptive filter to reduce the noise of the sound source signal and obtain the signal-to-noise ratio parameter,combine the signal-to-noise ratio parameter with the Sigmoid function and introduce it into the weighting factor of the phase transform(PHAT)weighting function,so that the time delay estimation algorithm has the ability to adaptively adjust the weighting according to the noise intensity.Taking the human voice source signal as an example,Matlab is used to carry out the performance simulation test under different signal-to-noise ratio environments,and the simulation of sound source localization is completed under the five-element cross microphone array.The experimental results show that the sound source localization distance deviation ratio of the improved algorithm is better than that of the traditional algorithm under different low signal-to-noise ratios.Therefore,the improved algorithm has higher localization accuracy and robustness in complex environments.This study provides a new solution for sound source localization in complex environments,which is of great significance for improving the accuracy of sound source localization in intelligent conference systems,remote monitoring systems,sound source detection instruments and other fields.

关 键 词：声源定位 自适应 时延估计 信噪比 SIGMOID函数 加权因子 相位变换 

分 类 号：TH841[机械工程—仪器科学与技术]