PPBFL: A Privacy Protected Blockchain-based Federated Learning Model
Yang Li, Chunhe Xia, Wanshuang Lin, Tianbo Wang
TL;DR
PPBFL tackles privacy and incentive challenges in federated learning by integrating blockchain with IPFS storage, a novel PoTW consensus to reward training effort, and dual adaptive local/global differential privacy to protect both local data and the global model. It introduces a ring-signature-based mixing CID mechanism to conceal participant identities and a reverse differential privacy scheme to counterbalance privacy noise in the global model, preserving utility under composition. The approach is validated on MNIST and Fashion-MNIST, showing improved security guarantees and competitive accuracy across IID and Non-IID settings, with greater robustness when global DP is applied. This work provides a practical, scalable framework for privacy-aware, incentive-aligned, blockchain-enabled federated learning suitable for distributed, trustless environments.
Abstract
With the rapid development of machine learning and a growing concern for data privacy, federated learning has become a focal point of attention. However, attacks on model parameters and a lack of incentive mechanisms hinder the effectiveness of federated learning. Therefore, we propose A Privacy Protected Blockchain-based Federated Learning Model (PPBFL) to enhance the security of federated learning and encourage active participation of nodes in model training. Blockchain technology ensures the integrity of model parameters stored in the InterPlanetary File System (IPFS), providing protection against tampering. Within the blockchain, we introduce a Proof of Training Work (PoTW) consensus algorithm tailored for federated learning, aiming to incentive training nodes. This algorithm rewards nodes with greater computational power, promoting increased participation and effort in the federated learning process. A novel adaptive differential privacy algorithm is simultaneously applied to local and global models. This safeguards the privacy of local data at training clients, preventing malicious nodes from launching inference attacks. Additionally, it enhances the security of the global model, preventing potential security degradation resulting from the combination of numerous local models. The possibility of security degradation is derived from the composition theorem. By introducing reverse noise in the global model, a zero-bias estimate of differential privacy noise between local and global models is achieved. Furthermore, we propose a new mix transactions mechanism utilizing ring signature technology to better protect the identity privacy of local training clients. Security analysis and experimental results demonstrate that PPBFL, compared to baseline methods, not only exhibits superior model performance but also achieves higher security.