A Novel Event-Related Desynchronization/Synchronization with Gamma Peak EEG model for Motor State Identification

Electroencephalography (EEG) is one of the imperative physiological signals for therapeutic and rehabilitative robotic control. The EEG features in time-domain, frequency-domain, and time-frequency have been extensively studied, especially differentiating between right and left movements of hands and legs. However, there are limitations in addressing real-world practical implementation due to multiple channels and computationally intensive and complex algorithms. This research proposes a novel model that can provide a high-level accuracy using only a single channel with straightforward signal processing algorithms using Event-Related Desynchronisa-tion (ERD) and Event-Related Synchronisation (ERS) signals. We demonstrate to distinguish between the arm's 'Rest' and 'Active' states by using a single-channel EEG signal. It is a practical model that can easily extend to any embedded device for a portable real-world application.