Blockchain consensus algorithm for supply chain information security sharing based on convolutional neural networks.

Area of Science:

• Computer Science
• Information Security
• Supply Chain Management

Background:

• Traditional supply chains suffer from data silos and information asymmetry, hindering secure data sharing.
• Existing methods lack real-time data verification and efficient processing, impacting operational efficiency.
• Blockchain and Convolutional Neural Networks (CNN) offer potential solutions for secure and efficient data management.

Purpose of the Study:

• To address data silos and information asymmetry in supply chain security sharing.
• To develop a secure and efficient data sharing mechanism using CNN and blockchain.
• To improve the overall operational efficiency, transparency, and sharing efficiency of supply chains.

Main Methods:

• Integration of Convolutional Neural Networks (CNN) for data analysis and privacy preservation.
• Application of blockchain technology for immutable and transparent data sharing.
• Introduction of a federated learning (FL) mechanism with an improved consensus algorithm, PoDaS (Proof of Data Sharing).
• PoDaS algorithm utilizes computational consumption from FL as proof of workload, combining advantages of Proof of Work (PoW) and Proof of Stake (PoS).

Main Results:

• The PoDaS algorithm demonstrated superior performance in block generation time compared to PoW and PoS.
• PoDaS achieved a model accuracy of 96.00%, significantly outperforming traditional PoW and PoS algorithms.
• The FL process, verified and recorded via blockchain consensus, ensured security and reliability.

Conclusions:

• The proposed CNN and blockchain-based approach effectively solves data silos and information asymmetry in supply chain security.
• The PoDaS consensus algorithm enhances federated learning efficiency and model accuracy for secure data sharing.
• The system promotes real-time data access, verification, and improved operational efficiency in supply chains.