基于事件相关电位的脑－机接口系统微弱信号分析方法研究

脑－机接口(BCI,brain computer interface)技术作为一种新型的人机交互方式，近年来受到越来越多的关注。基于听觉刺激模式诱发的事件相关电位P300建立了脑－机接口系统，主要研究微弱P300信号的单导联小样本数据的特征提取和模式识别方法。首先经过预处理和小波变换对脑电信号中的P300成分进行特征提取，建立特征向量，然后使用支持向量机的方法对特征提取结果进行模式分类。对7名受试者进行事件相关电位P300诱发实验，结果表明，系统的分类正确率可达到85％以上，且支持向量机的训练时间都在0.05 s以内。该方法可为脑－机接口的进一步应用奠定基础。

Research on Weak Signals Analysis of Brain-Computer Interface Based on the Event-Related Potentials

As a new human-machine interaction way,brain-computer interface(BCI) has already become a hot spot in recent years.Based on event-related potentials P300 stimulated by auditory,a BCI system is built to study feature extraction and pattern recognition of weak P300 signals.The signals are derived from single lead and small sample dataset.Firstly,simple pre-processed and wavelet transform are applied to obtain feature vector.Then support vector machine method is presented to realize pattern recognition.The experiments are taken among seven subjects to collect P300 signals,and the results prove that the classification accuracy is above 85%.The training time of support vector machine is in 0.05 s.Therefore,the proposed methods can expand the application of BCI technology.