摘要
应用多普勒效应原理,实现了对固定频率、匀速运动的单声源的速率测定。利用MATLAB拟合了固定传感器接收到的该声源运动时的声音信号,用Hilbert变换得到了该信号的瞬时频率,并以离散时间域的多普勒效应为目标函数,采用非线性最小二乘拟合得到其速率。算法采用小波分析及经验模态分解(EMD)方法分别对原始声音信号和瞬时频率进行去噪。实验结果表明,该算法精度较高,是一种可行的方法。
An algorithm is proposed for velocity determination of single moving sound source with uniform velocity and frequency. A synthesized sound signal is formed to simulate the moving sound source recorded by a fixed recorder, on which Hilbert transform is applied to obtain the instantaneous frequency of the sound signal. Utilizing the Doppler effect as a model to fit sampled instantaneous frequency by nonlinear least square, the velocity of a mobile can be obtained. Wavelet and empirical mode decomposition (EMD) are used to reduce the noise in the noisy sound signal and instantaneous frequency respectively. Experimental results show that this algorithm is accurate and feasible.
出处
《重庆邮电大学学报(自然科学版)》
2007年第6期689-693,共5页
Journal of Chongqing University of Posts and Telecommunications(Natural Science Edition)
基金
重庆市自然科学基金项目(2006BB2374)
关键词
HILBERT变换
瞬时频率
经验模态分解
最小二乘
Hilbert transform
instantaneous frequency
empirical mode decomposilion (EMD)
least square