近年來，以視覺誘發電位(visual evoked potential, VEP)為基礎之大腦人機界面(Brain-Computer Interface, BCI)已被廣泛使用，利用不同頻率或相位進行編碼，形成各種控制指令。透過非侵入式腦波訊號(Electroencephalogram, EEG)的擷取及辨識，可使其成為與外界溝通的橋樑，而不需要透過肌肉活動來控制。VEP-BCI具有高傳輸率與少訓練時間等優點，但是腦波訊號屬於非線性(nonlinear )、非穩態(non-stationary)且易受雜訊干擾的隨機程序，不論是60Hz的電訊號，或是量測時其它的生理訊號，如:眼動訊號(Electrooculography, EOG)、肌電訊號(Electromyography, EMG)，皆會嚴重干擾量測結果。 本文提出較傳統EEG處理方式不同的濾波方法，利用訊號空間投影法(Signal Space Projection, SSP) 設計空間濾波器，僅允許大腦皮質視覺區(Visual Cortex) 的訊號通過，相對而言，可濾除雜訊及不必要的生理訊號，並提高SNR，對於VEP及SSVEP皆有顯著效果。The brain computer interface (BCI) based on the visual evoked potential (VEP) has been widely used in many applications in recent years. By tagging of flickers with different frequencies and phases, user’s gazed target can be recognized by analyzing the frequencies or phases of evoked VEPs. Though VEP-based BCI has the advantages of high information transfer rate (ITR) and less training time, the extraction of steady state visual evoked potential (SSVEP) is sometimes not complete due to its characteristics of nonlinearity, non-stationary and noise susceptibility. Either 60Hz noise or electrophysiological signals, such as Electrooculography (EOG) and Electromyography (EMG) would disturb Electroencephalogram (EEG) seriously. Accordingly, this study adopts signal space projection (SSP) to design a spatial filter, passing signals only from visual cortex. The concept of spatial filter is different from the traditional frequency filtering process. In contrast, the method could filter out noise and additional electrophysiological signals, enhancing signal-to-noise ratio (SNR), and it has noticeable effects on VEP and SSVEP.
