Improved lightweight DL algorithm for biometric identification from EEG signal

Biometrics is a growing field, which permits identification of individuals by means of unique physical features. Electroencephalography (EEG)-based biometrics utilizes the small intra-personal differences and large inter-personal differences between individuals’ brainwave patterns. However, traditional EEG-based subject identification techniques frequently require a lot of electrodes, making it cumbersome and impractical for real-world applications. In this research, we suggest a method for subject identification using lightweight convolutional neural networks (CNN) while minimizing the number of electrodes. We propose a approach for subject identification in EEG that aims to minimize the number of electrodes while leveraging the power of CNNs. To achieve this, we divide the conductive electrodes of the cortex into (64, 32, 16) distinct groups. By exploiting the automatic feature extraction capabilities of CNNs, we process the EEG data from each electrode group individually. Remarkably, Electrodes (16) achieved an accuracy rate of 97.72%, 32 odd electrodes achieved an accuracy rate of 98.16%, while 32 even electrodes achieved an accuracy rate of 99.3%, and electrodes (64) achieved an accuracy rate of 95.47%. These results clearly demonstrate the robustness and efficacy of our method in accurately identifying individuals based on their EEG patterns. By decreasing the number of electrodes and capitalizing on the distinctive patterns captured by the electrode groups, our method provides a practical and efficient solution for subject identification in EEG.