Decentralized distribution for secure GAN using IPFS with the Hyperledger blockchain

Centralized storage and security problems make training and machine learning models vulnerable to manipulation and stealing. To provide safe transfer of GAN models and data, our proposed method, IH-EAGAN, offers the design a highly effective, efficient, and security-level solution. The proposed solution is based on IPFS (InterPlanetary File System) decentralization storage for GAN’s encoder model and decoder model being utilized in conjunction with hyperledger blockchain technologies, which allows participants to manage their transactions, utilizes smart contracts and has a ledger. The Hyperledger Fabric is a private and permissioned blockchain system. For the purpose of ensuring that this data is unchangeable and that any interaction with the ledger through the Chaincode will be logged by fabric for accountability and transparency, hyperledger is supported by a secret chaos key, which is known only by two legitimate parties (sender side and receiver side) for execution encryption peer to peer for verification and authentication. To construct a robust, integrated system that protects privacy against security breaches.