Search

Most EEG–based motor imagery classification research focuses on the feature extraction phase of machine learning, neglecting the crucial part for accurate classification which is the classification. In contrast, this paper concentrates on the classifier development where it thoroughly studies the performance analysis of k-Nearest Neighbour (k-NN) classifier on EEG data. In the literature, the Euclidean distance metric is routinely applied for EEG data classification. However, no thorough study has been conducted to evaluate the effect of other distance metrics to the classification accuracy. Therefore, this paper studies the effectiveness of five distance metrics of k-NN: Manhattan, Euclidean, Minkowski, Chebychev and Hamming. The experiment shows that the distance computations that provides the highest classification accuracy is the Minkowski distance with 70.08%. Hence, this demonstrates the significant effect of distance metrics to the k-NN accuracy where the Minknowski distance gives higher accuracy compared to the Euclidean. Our result also shows that the accuracy of k-NN is comparable to Support Vector Machine (SVM) with lower complexity for EEG classification.

Multiple-mode stub-loaded resonator with quad-channel diplexer and tri-band bandpass filter are presented and analyzed theoretically in this paper. The multi-mode stub-loaded resonator employs the basic structure of a triplemode resonator. Herein, the triple-mode resonator is modified by introducing a mid-coupled line between the odd-mode resonances to produce a quad-mode resonator. The proposed resonator is applied in the quad-channel diplexer design, composing of two independent dual-band bandpass filters. In turn, the triple-band bandpass filter based on a parallel quad-mode and a dual-mode stub-loaded resonator is developed. To validate the performance of the proposed resonators, two experimental examples, including a quad-channel diplexer and a triple-band bandpass filter are fabricated and measured. Each of the designed circuits occupies small area, i.e. about 0.32Agx0.22Ag and 0.]8Agx0.20Az respectively. Good agreements between simulated and measured results are achieved. [ABSTRACT FROM AUTHOR]