Paper Info
Title
A BCI Speller with 120 Commands Encoded by Hybrid P300 and SSVEP Features
Abstract
Implementing higher speed and larger command sets for brain-computer interfaces (BCIs) has always been the pursuit of researchers, which is helpful to realize the technological applications. The hybrid BCIs jointly induce different electroencephalogram (EEG) signals and could improve system performance effectively. This study designed an online BCI with 120 commands and high-speed by hybrid P300 and steady-state visual evoked potential (SSVEP) features. A time-frequency-phase encoding strategy was used to encode 120 commands in a short time, this strategy used time-locked P300s and frequency and phase-locked SSVEPs with a wide frequency band. The step-wise linear discriminant analysis (SWLDA) and ensemble task-related component analysis (eTRCA) were severally used to decode P300s and SSVEPs. As a result, online average spelling ac-curacy across six subjects was 83.89%. Average and highest information transfer rate (ITR) for this system was 151.53 bits/min and 175.09 bits/min, respectively. Meanwhile, the shortest time for out-putting one command was only 1.45 s. These results demonstrate the feasibility and effectiveness of this high-speed BCI with 120 commands, furthermore, this study used a wider frequency band of SSVEPs to encode 120 commands, which is helpful to extend larger command sets and achieve higher system performance.
Year
DOI
Venue
2023
10.1007/978-981-19-8222-4_19
Human Brain and Artificial Intelligence
Keywords
DocType
ISSN
Brain-computer interface (BCI), Hybrid BCI, P300, Steady-state visual evoked potential (SSVEP), High-speed, Large command sets
Conference
1865-0929
Citations 
PageRank 
References 
0
0.34
0
Authors
7
Name
Order
Citations
PageRank
Xiao Xiaolin100.68
Wen Shengfu200.34
Han Jin300.34
Yang Man400.34
Yin Erwei500.34
Minpeng Xu62717.17
Dong Ming710551.47