Procedia Engineering · 2011 · 23 citations · 3 references
EngineeringBiomedical EngineeringElectrical EngineeringBiomedical Signal ProcessingElectrophysiologyNeural EngineeringBrain-computer InterfacesElectroencephalogram Signal ProcessingElectroencephalographyEvoked PotentialsNeural Signal ProcessingVisual Evoked PotentialsAccuracyBrain-computer Interface Signal AcquisitionBrain-computer Interface Signal AnalysisBrain-computer Interface Signal ProcessingClassification Algorithms For Brain-computer InterfacesTraining TimeVisual Evoked Potential AnalysisWavelet Packet Transform
Brain-Computer Interface (BCI) has already become a hot spot in recent years. The Steady State Visual Evoked Potential (SSVEP) is a very suitable input signal of BCI system because of its high information transfer rate and short training time. A novel method was presented in this paper to analyze the feature of SSVEP. A four-direction motion control system of ball moving on the computer screen was set up to test the performance of the proposed method. The system utilized wavelet packet technology to find the normalized powers of special sub-wavebands as feature vector for the linear classifier and then converted them to the control commands. The experimental results have proved comparing with conventional FFT method, the wavelet packet technology method can obtain the feature vector faster and the control accuracy can reach more than 83 percent.
3