A Blind Source Separation Method for Convolved Mixtures by Non-Stationary Vibration Signals
A Blind Source Separation Method for Convolved Mixtures by Non-Stationary Vibration Signals
复制标题
DOI：
10.21307/ijssis-2017-575
复制
发表时间：
2013-01
影响因子：
1.2
通讯作者：
Ye Hongxian;L. Wenchang;Huang Xiaoping
中科院分区：
文献类型：
--
作者：
Ye Hongxian;L. Wenchang;Huang Xiaoping
提出了一种简单可行的分离非平稳振动信号卷积混合信号的BSS方法。该方法通过两个迭代过程来实现：第一步是估计滤波器的系数。采用独立性准则，通过简化滤波器系数，采用反向传播方法得到未知滤波器。下一个步骤是估计源振动信号。利用前一步骤得到的滤波器获得耦合振动信号，并通过解耦过程获得估计源。仿真和实验结果表明，该方法是有效的。这种改进的方法可以用于分离具有非平稳机械振动的卷积混合物。
Abstract A simple and feasible of BSS method to separate the convolution mixture of non-stationary vibration signals is present in this paper. The method is carried out by means of two iterative procedures: The first procedure is to estimate the coefficients of the filters. The independence criterion is used and the unknown filters are obtained by a back propagation procedure by means of simplifying the coefficients of filters. The next procedure is to estimate the source vibration signals. The coupled vibration signals are obtained by means of the filters gotten from the former procedure and the estimation sources are obtained through decoupled procedure. Simulation and experiment results show that the method is effective. This improved method can be used to separate convolutive mixtures with non-stationary mechanical vibrations.