P01Se01EEG data were recorded from participant 1(session 1). For further details, please see the associated README file.P01Session01.zipP01Se02EEG data were recorded from participant 1(session 2). For further details, please see the associated README file.P01Session02.zipP01Se03EEG data were recorded from participant 1(session 3). For further details, please see the associated README file.P01Session03.zipP01Se04EEG data were recorded from participant 1(session 4). For further details, please see the associated README file.P01Session04.zipP02Se01EEG data were recorded from participant 2(session 1). For further details, please see the associated README file.P02Session01.zipP03Se01EEG data were recorded from participant 3(session 1). For further details, please see the associated README file.P03Session01.zipP04Se01EEG data were recorded from participant 4(session 1). For further details, please see the associated README file.P05Se01EEG data were recorded from participant 5(session 1). For further details, please see the associated README file.P06Se01EEG data were recorded from participant 6(session 1). For further details, please see the associated README file.P07Se01EEG data were recorded from participant 7(session 1). For further details, please see the associated README file.P08Se01EEG data were recorded from participant 8(session 1). For further details, please see the associated README file.P09Se01EEG data were recorded from participant 9(session 1). For further details, please see the associated README file.P10Se01EEG data were recorded from participant 10(session 1). For further details, please see the associated README file.P11Se01EEG data were recorded from participant 11(session 1). For further details, please see the associated README file.P12Se01EEG data were recorded from participant 12(session 1). For further details, please see the associated README file.P02Se02EEG data were recorded from participant 2(session 2). For further details, please see the associated README file.P03Se02EEG data were recorded from participant 3(session 2). For further details, please see the associated README file.P04Se02EEG data were recorded from participant 4(session 2). For further details, please see the associated README file.P05Se02EEG data were recorded from participant 5(session 2). For further details, please see the associated README file.P06Se02EEG data were recorded from participant 6(session 2). For further details, please see the associated README file.P07Se02EEG data were recorded from participant 7(session 2). For further details, please see the associated README file.P08Se02EEG data were recorded from participant 8(session 2). For further details, please see the associated README file.P09Se02EEG data were recorded from participant 9(session 2). For further details, please see the associated README file.P10Se02EEG data were recorded from participant 10(session 2). For further details, please see the associated README file.P11Se02EEG data were recorded from participant 11(session 2). For further details, please see the associated README file.P12Se02EEG data were recorded from participant 12(session 2). For further details, please see the associated README file.P02Se03EEG data were recorded from participant 2(session 3). For further details, please see the associated README file.P03Se03EEG data were recorded from participant 3(session 3). For further details, please see the associated README file.P04Se03EEG data were recorded from participant 4(session 3). For further details, please see the associated README file.P05Se03EEG data were recorded from participant 5(session 3). For further details, please see the associated README file.P06Se03EEG data were recorded from participant 6(session 3). For further details, please see the associated README file.P07Se03EEG data were recorded from participant 7(session 3). For further details, please see the associated README file.P08Se03EEG data were recorded from participant 8(session 3). For further details, please see the associated README file.P09Se03EEG data were recorded from participant 9(session 3). For further details, please see the associated README file.P10Se03EEG data were recorded from participant 10(session 3). For further details, please see the associated README file.P11Se03EEG data were recorded from participant 11(session 3). For further details, please see the associated README file.P12Se03EEG data were recorded from participant 12(session 3). For further details, please see the associated README file.P02Se04EEG data were recorded from participant 2(session 4). For further details, please see the associated README file.P03Se04EEG data were recorded from participant 3(session 4). For further details, please see the associated README file.P04Se04EEG data were recorded from participant 4(session 4). For further details, please see the associated README file.P05Se04EEG data were recorded from participant 5(session 4). For further details, please see the associated README file.P06Se04EEG data were recorded from participant 6(session 4). For further details, please see the associated README file.P07Se04EEG data were recorded from participant 7(session 4). For further details, please see the associated README file.P08Se04EEG data were recorded from participant 8(session 4). For further details, please see the associated README file.P09Se04EEG data were recorded from participant 9(session 4). For further details, please see the associated README file.P10Se04EEG data were recorded from participant 10(session 4). For further details, please see the associated README file.P11Se04EEG data were recorded from participant 11(session 4). For further details, please see the associated README file.P12Se04EEG data were recorded from participant 12(session 4). For further details, please see the associated README file.

Brain-computer interface (BCI) allows collaboration between humans and machines. It translates the electrical activity of the brain to understandable commands to operate a machine or a device. In this study, we propose a method to improve the accuracy of a 3-class BCI using electroencephalographic (EEG) signals. This BCI discriminates rest against imaginary grasps and elbow movements of the same limb. This classification task is challenging because imaginary movements within the same limb have close spatial representations on the motor cortex area. The proposed method extracts time-domain features and classifies them using a support vector machine (SVM) with a radial basis kernel function (RBF). An average accuracy of 74.2% was obtained when using the proposed method on a dataset collected, prior to this study, from 12 healthy individuals. This accuracy was higher than that obtained when other widely used methods, such as common spatial patterns (CSP), filter bank CSP (FBCSP), and band power methods, were used on the same dataset. These results are encouraging and the proposed method could potentially be used in future applications including BCI-driven robotic devices, such as a portable exoskeleton for the arm, to assist individuals with impaired upper extremity functions in performing daily tasks.