Securing digital identities has become one of the defining challenges of our era. Passwords get phished, fingerprints get lifted, and facial recognition systems have been fooled by photographs. Against this backdrop, authenticating a person via the electrical activity of their own brain is arguably one of the few modalities an attacker simply cannot replicate without a living, cooperating subject. This paper reports on the design and evaluation of a web-based brainwave authentication system that leverages simulated EEG signals and machine learning to verify user identity. Because consumer-grade EEG hardware remains expensive and logistically demanding, we sidestep that barrier by training on publicly available EEG datasets alongside physiologically grounded synthetic signals. Frequency-band features drawn from alpha, beta, theta, and gamma components are fed into Support Vector Machine (SVM) and Random Forest classifiers, achieving authentication accuracies of 94.6% and 96.2% respectively. The system is wrapped in a Django web application—complete with registration, signal submission, and a results dashboard—making the pipeline immediately usable for academic evaluation and extensible toward real hardware integration.
Catherine reshma GJ (Mon,) studied this question.