摘要
针对基于时延估计的机电设备故障声定位中的低信噪比和脉冲噪声情况,用α稳定分布建模噪声,改进了非整数自适应时延估计方法。共变相关法对观测序列进行时延估计粗测,将得到的估计值作为非整数自适应时延估计器的初值;将共变相关法中共变序列作为时延估计器的输入信号,在最小平均p范数准则下迭代得到非整数时延估计值。共变序列保留了原始序列间的时延信息,削弱了不相关的噪声。计算机仿真对比实验验证了改进的方法在脉冲环境和低信噪比条件下有更好的性能。
Fractional adaptive time delay estimation(TDE)method,modeling noise with_stable distribution,is improved to adapt low signal to noise ratio(SNR)and impulse noise in equipment fault acoustic localization based on TDE.Covariate correlation method gets integer part of TDE,it is used as initial value of adaptive delay estimator.Covariant sequence is used as input of time delay estimator,finally,fractional TDE value is obtained by iteration under minimum average p norm criterion.Covariant sequence preserves time delay information between original sequences,weakens irrelevant noise.Computer simulations and comparative experiments verify that improved method has better performance under impulse environment and low SNR.
作者
李均浩
刘文红
LI Junhao;LIU Wenhong(School of Electronic Engineering,Shanghai Dianji University,Shanghai 201306,China;School of Electronics and Information,Shanghai Dianji University,Shanghai 201306,China)
出处
《应用声学》
CSCD
北大核心
2019年第2期253-260,共8页
Journal of Applied Acoustics
基金
国家自然科学基金项目(61172108)
上海电机学院横向科研合作项目(17B56)
关键词
设备声故障定位
_稳定分布
共变
互相关法
非整数时延估计
Equipment acoustic fault location
stable distribution
Covariation
Cross correlation method
Fractional time delay estimation
