Table 1.

Comparison of our proposed method with recent electroencephalography (EEG)-based work on command generation, accuracy, and window size.

Reference	Brain area	Activity	Brain–computer interface (BCI) type	Modality	Application	Commands	Accuracy (%)	Window size
Kim et al. (2014)	Complete brain	Eye movement	Active	EEG + Eye tracker	Quadcopter control	8	91.67	5 s
Bai et al. (2015)	Complete brain	Motor imagery and P300	Active + reactive	EEG	Opening, closing, selection of files in Internet Explorer	9 (can achieve 50)		4 s window for motor imagery and 600 μs for P300
Hortal et al. (2015)	Motor and parietal	Mental imagination	Active	EEG + EOG	Robotic arm control for pick and place task	6	Task 1:71.13 and Task 2:61.51	0.5 s to synchronize output to brain–machine interface
Naseer and Hong (2015a)	Prefrontal and motor cortex	Mental arithmetic, mental counting and motor imagery	Active	Functional near-infrared spectroscopy	Decoding answers to four-choice questions	4	73.3	2–7 s
Ma et al. (2015)	Parietal and occipital	P300 and eyeblink	Reactive + active	EEG + EOG	Mobile robot control	9	87.3 for average of 5 trials	~1.6 s
Combaz and Van Hulle (2015)	Whole brain	P300 and steady-state visually evoked potentials (SSVEP)	Reactive	EEG	Applications to locked-in patients option selection	12	Maximum achieved >95	200 μs before stimulation to 800 μs after stimulation for experiment 1
Ramli et al. (2015)	Motor and occipital	Eye gaze	Reactive	EEG + EOG	Application to BCI applications (wheelchair control)	6	97.88	0.5 s
Yin et al. (2015)	Parietal and occipital cortex	P300 and SSVEP	Reactive	EEG	Speller paradigm with applications to BCI systems control	Up to 64 commands	95.18
The proposed method	Frontal	Mental task + eye movement	Active	NIRS + EEG	Applications to quadcopter control	8	76.5% for NIRS and 86% for EEG	1 s for EEG and 2 s for NIRS