摘要
基于信息融合的思想,简介了独立分量分析方法;以Matlab为辅助工具,应用独立分量分析方法中比较成熟的快速算法FastICA,给出了语音信号分离的独立分量分析方法的具体途径,并对其分离效果进行了分析;然后,应用该方法对轴承的故障噪声特征信号成功地实现了提取。这两个实例的验证结果表明,采用独立分量分析方法对噪声源信号分离是有效的,应用前景是广阔的。
Based on the MATLAB tool, the well-known algorithm named FastICA for ICA was used to give the approach for noise sound source signals separation, and some analyses were discribed. The examples of sound signals separation and beating fault signals extraction demonstrate that the ICA method is effective for noise signals separation.
出处
《信息技术》
2010年第2期91-93,共3页
Information Technology
关键词
独立分量分析
信号分离
噪声
语音信号
轴承故障信号
independent component analysis
signals separation
noise
sound signals
beating fault signals
