Concurring of Neural Machines for Robust Session Key Generation and Validation in Telecare Health System During...

Area of Science:

• Computer Science
• Artificial Intelligence
• Cybersecurity

Background:

• Electronic health (e-health) is crucial for secure patient-physician communication, especially during the COVID-19 pandemic.
• Telecare systems provided essential remote services for non-invasive patients during the crisis.
• Data security and privacy are paramount in neural cryptographic engineering for health systems.

Purpose of the Study:

• To propose a technique for agreeing session keys using dual artificial neural networks within the Telecare Health COVID-19 domain.
• To enhance data security and privacy through neural cryptographic engineering and Tree Parity Machine (TPM) synchronization.
• To validate the robustness of generated session keys against data attacks.

Main Methods:

• Generation of session keys using dual artificial neural networks (Tree Parity Machine - TPM).
• Partial sharing of intermediate keys between patient and doctor for neural synchronization.
• Key validation across different lengths (40, 60, 160, 256 bits) and testing for randomization.

Main Results:

• Dual neural TPM networks demonstrated a high magnitude of co-existence in Telecare Health Systems during COVID-19.
• The proposed technique showed high protection against data attacks on public networks.
• Partial session key transmission effectively prevented intruders from guessing patterns, with high randomization observed.

Conclusions:

• The dual neural network approach offers a secure and robust method for session key generation in Telecare Health.
• This technique significantly enhances data protection and privacy for remote health monitoring.
• The method is effective in mitigating security risks associated with public network communications in e-health.