诱发电位(ＥＰ)信号在检测神经系统状态时有重要意义．但ＥＰ信号总是淹没在自发脑电波(ＥＥＧ)信号中,因此,为利用ＥＰ信号诊断神经系统的损伤和病变,需要从二者的混合信号中提取出ＥＰ信号．独立分量分析(ＩＣＡ)是一种新近发展起来的信号分离方法．本文把ＩＣＡ方法应用于ＥＰ信号的提取,并与传统的滤波方法进行了比较．计算机模拟表明,采用ＩＣＡ方法进行噪声分离的结果明显优于信号滤波方法． Evoked Potentials (EP) signals are important in detecting the status of the nervous system. However, EP signals are always submerged in EEG signals. Therefore, in order to diagnose nervous system lesions and lesions using EP signals, EP signals need to be extracted from the mixed signals of the two signals. Independent Component Analysis (ICA) is a newly developed method of signal separation. In this paper, ICA method is applied to the extraction of EP signal and compared with the traditional filtering method. Computer simulation shows that the ICA method for noise separation results significantly better than the signal filtering method.