Multi-Objective Optimization-Based High-Pass Spatial Filtering for SSVEP-Based Brain–Computer Interfaces

Many spatial filtering methods have been proposed to enhance the target identification performance for steady-state visual evoked potential (SSVEP)-based brain–computer interface (BCI). The existing approaches tend learn filter parameters of a certain using only training data from same stimulus, and they rarely consider information other stimuli or volume conduction problem during process. In this article, we propose novel multi-objective optimization-based high-pass method improve SSVEP detection accuracy robustness. filters are derived via maximizing correlation between signal individual template whilst minimizing targets template. optimization will also be subject constraint that sum elements is zero. evaluation study on two self-collected datasets (including 12 four frequencies, respectively) shows outperformed compared such as canonical analysis (CCA), multiset CCA (MsetCCA), squared correlations (SSCOR), task-related component (TRCA). was verified public 40-class benchmark dataset recorded 35 subjects. experimental results demonstrated effectiveness approach enhancing performance.