LAUSR

Objective: The aim of this work was to build an asynchronous brain-computer interface (BCI) system based on steady-state visual evoked potentials (SSVEPs) that outputs continuous, stable and smooth control commands in the up, down, left and right directions. Real-time feedback is presented on the computer screen to enhance collaborative participation in the human-computer interaction. Methods: Four stimulus sources flickering at different frequencies were sequentially fixed around the computer monitor to allow for subjects to complete control tasks by gazing at different stimulus sources. The subjects autonomously switched between the idle and working states by controlling the alpha amplitude. A sliding window voting discrimination (SWVD) strategy was incorporated into the canonical correlation analysis (CCA) algorithm for asynchronous classification. Results: Experiments were performed with 18 subjects using both the synchronous and asynchronous paradigms. The average accuracy was 95.42±3.35% with a data length of 3 s for synchronous operation; in addition, most subjects were able to successfully control a target to move precisely and smoothly with a 1 s sliding window during asynchronous operation. Conclusion: The proposed design scheme is feasible for our online asynchronous BCI system. Significance: By applying the SWVD strategy and optimizing the experimental paradigm, the classical CCA algorithm was successfully applied for continuous control in an asynchronous BCI system. With the developed system, obvious improvements in the information transmission rate (ITR) and sensitivity were achieved, which will be beneficial for the development of practical BCI systems.