匹配滤波和ICA消除触觉传感器工频噪声

Power frequency noise reduction technique of tactile sensor signal using matched filter and ICA

针对触觉传感器信号易受机器人动力系统的电磁干扰而带来工频噪声,设计一种采用匹配滤波和独立分量分析消除触觉传感器工频噪声的新方法。首先通过匹配滤波判断触觉传感器信号中是否含有工频噪声,其次运用独立分量分析法消除相应的工频干扰。实验结果表明:匹配滤波可以有效判断传感器信号是否受到工频干扰,独立分量分析法不仅能够有效地消除传感器信号中的工频噪声,且不会对其他频率成分造成影响,具有良好的工频滤波效果。

A new method of removing power frequency noise in tactile sensor signal was designed by matched filtering and independent component analysis(ICA),for tactile sensor signal was susceptible to electromagnetic interference of the robot power system.First,the matched filter was used to determine whether tactile sensor signal contain frequency noise,and then the frequency noise was eliminated by the independent component analysis method.The results show that matched filter can effectively judge whether the sensor signal is affected by frequency interference,and independent component analysis method can remove the frequency noise effectively and has no effect on the other frequency components.