Asynchronous steady-state visual evoked potential (SSVEP) BCIs hold promise for real-world control, but phase shifts between ongoing EEG and fixed-phase templates often hurt performance. The authors observed and validated a phase-sliding oscillation (PSO) phenomenon: with a fixed-phase template, phase-sliding trajectories from sliding-window sequences with different initial phases oscillate stably at the stimulus frequency. PSO-AC rebuilds EEG into sequences with varying initial phases, converts them into stable time-frequency representations through wavelet synchrosqueezing transform to extract phase-decoupled features, and uses a probabilistic temporal model to separate control from non-control states.
In offline experiments (n=22), PSO-AC reached a mean control/non-control accuracy of 94.2% with a 2-s data length, outperforming a state-of-the-art baseline and maintaining its edge in 9-class classification under different time delays, the authors report. In online robotic-arm experiments, the method produced more stable command triggering and higher control efficiency, cutting the command cost per successful trial from 4.30 to 1.14. The authors note that asynchronous SSVEP-based BCIs show great potential for real-world control. The paper’s DOI is 10.1142/s0129065727500225.