Performance investigation of MVMD-MSI algorithm in frequency recognition for SSVEP-based brain-computer interface and...

Area of Science:

• Neuroscience
• Robotics
• Signal Processing

Background:

• Brain-computer interfaces (BCIs) are crucial for robotic control.
• Steady-state visual evoked potential (SSVEP) based BCIs face challenges with signal artifacts and reliability.
• Existing methods struggle with nonlinear and non-stationary electroencephalogram (EEG) signals.

Purpose of the Study:

• To enhance the performance and reliability of SSVEP-based BCIs for robotic control systems.
• To develop a novel algorithm for improved artifact reduction in EEG data.
• To validate the proposed algorithm's effectiveness in real-world robotic applications.

Main Methods:

• Proposed the Multivariate Variational Mode Decomposition-Multivariate Synchronization Index (MVMD-MSI) algorithm.
• MVMD-MSI decomposes nonlinear, non-stationary EEG signals into intrinsic mode functions (IMFs).
• Evaluated the algorithm using a 6-degrees-of-freedom robot in offline and online experiments.

Main Results:

• The MVMD-MSI algorithm demonstrated significant improvements in SSVEP decoding performance.
• Achieved high accuracy (98.31%) at 1.8 seconds in online experiments for robotic arm control.
• Showed robust performance with fewer EEG channels and shorter data lengths.

Conclusions:

• The MVMD-MSI algorithm represents a significant advancement for SSVEP analysis in robotic control.
• The method enhances decoding performance and artifact reduction in BCI systems.
• Confirmed feasibility and effectiveness for real-time SSVEP BCI-based robotic applications.