Development of IoT Based Brain Computer Interface for Motor Function Detection Using EEG Signal Processing

In motor function rehabilitation, the manual control of exoskeleton system is replaced with the closed-loop brain-computer interface (BCI). Powered ankle foot orthosis was used for gait rehabilitation and locomotion assistance. The neurological disorder patients need external controlled orthosis for help. Electromyogram (EMG) and electroencephalogram (EEG) were used for the development of these devices. The EEG signal was acquired and the Movement-related cortical potentials (MRCP) is utilized for actuator control of exoskeletons. To detect the motor function and to perform gait rehabilitation, this paper presents a hybrid method which combines the brain computer interface with IoT for motor function detection and exoskeleton control. The proposed system preprocesses the EEG signal, extracts the motor functions and controls the actuators. In recent existing methods the Empirical Mode Decomposition (EMD) algorithm is used to detect the MRCPs from scalp EEG. In the proposed method ensemble-EMD (EEMD), which is the noise-assisted data processing technique is investigated. EEG from frontal, central, and parietal lobes namely the C3, C4 channels were acquired. Particle swarm Optimization and fine, weighted, cubic, cosine, and medium KNN techniques was used for preprocessing and detection of motor functions. The control function was given to the ankle foot orthosis for locomotion assistance and gait rehabilitation. For implementation MATLAB, Python and eclipse were used