一种基于双特征的联合脑-机接口系统设计

A system Design of a Hybrid BCI Based on the Dual characteristics

与传统基于单一脑电信号的脑-机接口相比,基于多种特征信号的联合脑-机接口能有效提高脑-机接口性能。在基于稳态视觉诱发电位和P300诱发电位的联合使用的可行性基础上,提出了新的刺激编码方式,构建了一种基于两种特征的联合脑-机接口系统。通过设计3×3字符刺激矩阵,矩阵中纵列按各自设定频率闪烁诱发稳态视觉诱发电位,横行随机出现蓝色框诱发P300。实验表明,当受试者注视并关注目标字符,两种特征脑电信号能够被同时诱发,且对脑电信号中两种特征进行识别能够检测出受试者选取的字符。与传统基于P300的字符脑-机接口相比,刺激诱发时间减少了一半,从根本上提高了脑-机接口的速度。在以后工作中,系统可以扩展到更大矩阵(如6×6),构建更为实用的联合脑-机接口系统。

Hybrid brain- computer interface （BCI） based on the multi- feature of EEG signals can effectively improve the system performance comparing to the traditional BCI based on the single EEG feature. In consideration to the combination of SSVEP and P300, a new encoding - stimulus model is proposed and then a hybrid BCI paradigm model based on the combination of SSVEP and P300 EEG is devised. In a 3 by 3 character stimulus matrix, columns of characters are flicked at a set frequency to induce SSVEP and a random blue frame is shown on rows to induce P300. Experiments show that when subjects stare a target character, SSVEP and P300 can be induced simultaneous- ly. Therefore, the identification of P300 and SSVEP can detect the target character which subjects selected. Com- pared with the traditional P300 BCI paradigm for character - input, the proposed system is able to reduce running time and fundamentally improve the system processing speed. As a future work, an enhancement with more characters in a matrix （ such as 6 by 6） can be extended so to achieve a more practical BCI system.