Poisoning Attacks on Federated Learning-based Wireless Traffic Prediction

2024-04-22

2024 IFIP Networking Conference (IFIP Networking)

AI Summary

Federated learning (FL)-based wireless traffic prediction (WTP) systems are vulnerable to low-knowledge fake traffic injection (FTI) attacks, and existing defenses lack sufficient robustness—particularly for regression tasks. Method: This paper proposes the first lightweight FTI attack tailored to regression settings and introduces GLID, a global-local inconsistency detection framework. GLID employs quantile-based statistical analysis for dimension-wise anomalous parameter cleansing and integrates model parameter consistency verification with privacy-preserving mechanisms. Contribution/Results: Experiments on real-world wireless traffic datasets demonstrate that the proposed FTI attack achieves significantly higher success rates than baseline methods. GLID reduces prediction error by 37.2% compared to unprotected FL, outperforms state-of-the-art defenses in robustness, and simultaneously ensures computational efficiency, prediction accuracy, and privacy preservation.

Technology Category

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

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Security and Privacy: Large-scale security measurements

Abstract

Federated Learning (FL) offers a distributed framework to train a global control model across multiple base stations without compromising the privacy of their local network data. This makes it ideal for applications like wireless traffic prediction (WTP), which plays a crucial role in optimizing network resources, enabling proactive traffic flow management, and enhancing the reliability of downstream communication aided applications, such as IoT devices, autonomous vehicles, and industrial automation systems. Despite its promise, the security aspects of FL-based distributed wireless systems, particularly in regression-based WTP problems, remain inadequately investigated. In this paper, we introduce a novel fake traffic injection (FTI) attack, designed to undermine the FL-based WTP system by injecting fabricated traffic distributions with minimal knowledge. We further propose a defense mechanism, termed global-local inconsistency detection (GLID), which strategically removes abnormal model parameters that deviate beyond a specific percentile range estimated through statistical methods in each dimension. Extensive experimental evaluations, performed on real-world wireless traffic datasets, demonstrate that both our attack and defense strategies significantly outperform existing baselines.

Problem

• Federated Learning
• Wireless Traffic Prediction
• False Traffic Injection Attack

Innovation

• Fake Traffic Injection (FTI)
• Global-Local Inconsistency Detection (GLID)
• Federated Learning Defense