Search

A novel method is proposed here to determine whether a time series is deterministic even in the presence of noise. The method is the extension of an existing method based on smoothness analysis of the signal in state space with surrogate data testing. While classical measures fail to detect determinism when the time series is corrupted by noise, the proposed method can clearly distinguish between pure stochastic and originally deterministic but noisy time series. A set of measures is defined here named partial smoothness indexes corresponding to principal components of the time series in state space. It is shown that when the time series is not pure stochastic, at least one of the indexes reflects determinism. The method is first successfully tested through simulation on a chaotic Lorenz time series contaminated with noise and then applied on EEG signals. Testing results on both our experimental recorded EEG signals and a benchmark EEG database verifies this hypothesis that EEG signals are deterministic in nature while contain some stochastic components as well.

Artifact removal is an essential part in electroencephalogram (EEG) recording and the raw EEG signals require preprocessing before feature extraction. In this work, we implemented three filtering methods and demonstrated their effects on the performance of different classifiers. Bandpass digital filtering, median filtering and facet method are three preprocessing approaches investigated in this paper. We used data set lib from the BCI competition 2003 for training and testing phase. Our accuracy varied between 80% and 96%. In our work, we demonstrated that the problems of choosing the classifier and preprocessing methods are not independent of each other. Two of our approaches could achieve the 96% accuracy i.e. 31 of 32 characters were predicted correctly. These two approaches have different classifier and different preprocessing method. It means that the performance of each classifier can be enhanced with a specific preprocessing method. In our approach, we used only three electrodes of 64 applied electrodes. Therefore it can noticeably reduce the time and cost of EEG measurement.

In this paper, the performance of five classifiers in P300 speller paradigm are compared. Theses classifiers are Linear Support Vector Machine (LSVM), Gaussian Support Vector Machine (GSVM), Neural Network (NN), Fisher Linear Discriminant (FLD), and Kernel Fisher Discriminant (KFD). In classification of P300 waves, there has been a trend to use SVM classifiers. Although they have shown a good performance, in this paper, it is shown that the FLD classifiers outperform the SVM classifiers. FLD classifier uses only ten channels of the recorded electroencephalogram (EEG) signals. This makes them a very good candidate for real-time applications. In addition, FLD approach does not need any optimization similar to other methods. In addition, in this paper, it is shown that the efficiency of using Principal Component Analysis (PCA) for feature reduction results in decreasing the time for the classification and increasing the accuracy.

This paper describes the implementation of a human computer interface based on eye tracking. Current commercially available systems exist, but have limited use due mainly to their large cost. The system described in this paper was designed to be a low cost and unobtrusive. The technique was video-oculography assisted by corneal reflections. An off-the shelf CCD webcam was used to capture images. The images were analyzed in software to extract key features of the eye. The users gaze point was then calculated based on the relative position of these features. The system is capable of calculating eye-gaze in real-time to provide a responsive interaction. A throughput of eight gaze points per second was achieved. The accuracy of the fixations based on the calculated eye-gazes were within 1 cm of the on-screen gaze location. By developing a low-cost system, this technology is made accessible to a wider range of applications.