Study On Related Technology In Brain Computer Interface And Online Systems Implementation

As a new communication system, Brain-computer interface can provide a way of communication and environment control for those patients with serious dyskinesia who have a normal mind, improving their living quality. After years of research, the domestic and foreign research on BCI has made significant progress, but having some distance from the practical application. In this paper, we studied some issues in BCI systems, and implemented an online BCI system.As to solve the commonly used ocular artifacts (OA) removal methods existing the problem of lossing EEG. This paper proposed two new OA removal methods, that are wSOBI and dCCA. The effectiveness of two new methods is tested by simulation data and real EEG. According to the results of two evaluation indexes, the OA removal effects of two new methods are significantly better than wICA, wCCA and SOBI. Moreover, the efficiency of two new methods both meet online systems requirement. Because the dCCA method only adopts two-channel EEG, and its processing speed is faster, so it has more advantages to apply to BCI online systems.To solve the problem, that is non-balanced bilinguals using the foreign language need to overcome the mother-tongue interference. This paper adopts a new paradigm based on the time interval factor to extract the Chinese-English code-switching ERP. Firstly, we extract five features of the ERP after preprocessing, among which the power spectrum, the wavelet sub-band energy and wavelet entropy are three new features proposed by this paper. According to the feature analysis and ANOVA, three new features are more effective than incubation and peak, and the N2effect under the condition of long interval is more notable than that of short interval, which supports the prevailing inhibitory control theory.Aiming at the problem that patients with serious dyskinesia suffer from low living quality, based on Alpha waves and BCI2000platform, we implemented a real-time control system of virtual small light. Through some signal processing steps special to this system, the valid control commands are gained. And the feasibility and reliability of the system is verified by online control experiments and three evaluation indexes. It suggests that this design which is an BCI online system easily to operate. Because subjects can effectively control the small light by opening and closing eyes without complex training, so it is a natural human-computer interaction form. All in all, this design with a good expansion has laid a foundation for the practical application of BCI technology.