探讨通道对视觉诱发电位单次提取精度的影响

采用“模拟自然阅读”诱发电位作为脑-机接口通信载体，结合小波变换和误差逆传播（ErrorBackPropagation，HIJBP）神经网络对视觉诱发脑电信号（VEP）进行分类以产生脑机接口控制信号。将被试者4个通道记录到的脑电信号分别作为特征，信号时程都为300ms，时段分别取100ms~400ms、200ms-500ms和300ms~600ms。试验结果表明四个单通道都达到较好的分类效果，有利于产生脑机接口控制信号。

Effect to Channel in Single-Trial Visual Evoked Potential Estimation

Adopting the＂Imitating-natural-reading＂elicited potentials as carriers between brain and computer.The classification of visual evoked potential（VEP） with wavelet transform and BP neural network was used to obtain brain-computer interface （BCI） control signal. Taking the signals from the four EEG recordings as features and setting the time length of signal in a trial as 300ms, the classification rates were tested on average with three time intervals,namely, 100ms-400ms, 200ms-500ms, and 300ms-600ms. The test results showed that four single-channel up to a good classification results.It is useful to produce BCI control signal.