Secret-Key Agreement by Asynchronous EEG over Authenticated Public Channels

Area of Science:

• Neuroscience
• Cybersecurity
• Signal Processing

Background:

• Electroencephalogram (EEG) signals offer a unique biometric modality for secure communication.
• Existing biometric key agreement systems face challenges with signal noise and data security.

Purpose of the Study:

• To develop and evaluate a novel system for sequential secret key agreement using EEG signals.
• To assess the system's performance based on key agreement rate, extraction rate, leakage rate, and entropy.

Main Methods:

• Utilized 6 performance metrics from asynchronously recorded EEG signals via an EMOTIV EPOC+ headset.
• Conducted experiments with 76 participants performing a single mental task.
• Optimized and rigorously evaluated the proposed secret key agreement system.

Main Results:

• Achieved a 100% key agreement rate and a 9% key extraction rate.
• Demonstrated a low leakage rate of 0.0003 and high mean block entropy (0.9994).
• Generated keys successfully passed the NIST randomness test.

Conclusions:

• The proposed EEG-based system provides a secure and effective method for secret key agreement.
• System performance remained robust even when eavesdroppers had access to public channel data.
• This approach offers a promising direction for developing next-generation secure communication technologies.