DP2Guard: A Lightweight and Byzantine-Robust Privacy-Preserving Federated Learning Scheme for Industrial IoT

2025-07-21

AI Summary

To address the dual challenges of high encryption overhead and weak resilience against adaptive model poisoning attacks in privacy-preserving federated learning (PPFL) for industrial IoT, this paper proposes a lightweight and robust collaborative defense framework. Methodologically, it replaces conventional encryption with lightweight gradient masking; integrates singular value decomposition and cosine similarity for feature-aware anomaly detection; designs a dynamic trust-score-based adaptive aggregation mechanism; and leverages blockchain for immutable aggregation result recording and auditable traceability. The key contribution lies in the first holistic integration of gradient masking, trustworthy aggregation, and on-chain auditing—ensuring data privacy while significantly enhancing robustness against four advanced poisoning attack types. Experiments on two public datasets demonstrate over 30% reduction in communication and computational overhead, alongside superior training efficiency compared to state-of-the-art PPFL approaches.

Technology Category

• Machine Learning: Privacy
• Computer Vision: Adversarial Attacks & Robustness
• Natural Language Processing: Safety and Robustness

Application Category

• Security and Privacy: Privacy-enhancing technologies
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing

Abstract

Privacy-Preserving Federated Learning (PPFL) has emerged as a secure distributed Machine Learning (ML) paradigm that aggregates locally trained gradients without exposing raw data. To defend against model poisoning threats, several robustness-enhanced PPFL schemes have been proposed by integrating anomaly detection. Nevertheless, they still face two major challenges: (1) the reliance on heavyweight encryption techniques results in substantial communication and computation overhead; and (2) single-strategy defense mechanisms often fail to provide sufficient robustness against adaptive adversaries. To overcome these challenges, we propose DP2Guard, a lightweight PPFL framework that enhances both privacy and robustness. DP2Guard leverages a lightweight gradient masking mechanism to replace costly cryptographic operations while ensuring the privacy of local gradients. A hybrid defense strategy is proposed, which extracts gradient features using singular value decomposition and cosine similarity, and applies a clustering algorithm to effectively identify malicious gradients. Additionally, DP2Guard adopts a trust score-based adaptive aggregation scheme that adjusts client weights according to historical behavior, while blockchain records aggregated results and trust scores to ensure tamper-proof and auditable training. Extensive experiments conducted on two public datasets demonstrate that DP2Guard effectively defends against four advanced poisoning attacks while ensuring privacy with reduced communication and computation costs.

Problem

• Lightweight privacy-preserving federated learning for Industrial IoT
• Robust defense against adaptive poisoning attacks
• Reduced communication and computation costs

Innovation

• Lightweight gradient masking replaces encryption
• Hybrid defense with SVD and cosine similarity
• Trust score-based adaptive aggregation scheme