Secure and privacy improved cloud user authentication in biometric multimodal multi fusion using blockchain-based...

Area of Science:

• Computer Science
• Cybersecurity
• Artificial Intelligence

Background:

• Increasing cyber threats necessitate robust user authentication in cloud systems.
• Existing methods struggle to balance authentication accuracy with user privacy.
• Need for decentralized, tamper-proof, and transparent authentication solutions.

Purpose of the Study:

• To develop an innovative user authentication solution for cloud-based systems.
• To enhance security and privacy in user authentication processes.
• To address the trade-off between accuracy and privacy in current authentication methods.

Main Methods:

• Integration of multimodal biometrics for accurate user identification.
• Application of deep learning, specifically Instance-based learning-based DetectNet (IL-DN), for swift and precise authentication.
• Implementation of privacy-preserving techniques for biometric data encoding.
• Utilization of blockchain technology, smart contracts, and an enhanced Proof of Work (PoW) mechanism for a decentralized and secure system.

Main Results:

• The proposed system demonstrates high authentication accuracy using multimodal biometrics via IL-DN.
• Effective privacy preservation of biometric data through advanced encoding techniques.
• Establishment of a decentralized, tamper-proof, and transparent authentication framework via blockchain.
• Comprehensive evaluation of performance metrics including accuracy, privacy, security, and resource utilization.

Conclusions:

• The research offers a comprehensive solution for secure and privacy-enhanced user authentication in cloud environments.
• The integrated approach effectively addresses the limitations of current authentication systems.
• This work significantly contributes to filling the research gap in secure and private cloud authentication.