Self-Adaptive and Robust Federated Spectrum Sensing without Benign Majority for Cellular Networks

2025-07-16

AI Summary

To address spectrum scarcity in cellular networks, existing federated learning-based spectrum sensing (FLSS) approaches face two critical challenges: difficulty in training on unlabeled data and insufficient robustness against data poisoning attacks. This paper proposes an adaptive robust FLSS framework that jointly integrates semi-supervised learning with energy detection to enable efficient model training in fully unlabeled scenarios. It further introduces a novel “vaccine-inspired” defense mechanism—the first of its kind—that relaxes the conventional assumption of benign-majority clients and achieves Byzantine resilience even under non-benign-majority conditions. Experimental results demonstrate that the framework attains near-perfect (≈100%) spectrum sensing accuracy on both real-world and synthetic unlabeled datasets. Moreover, it maintains strong robustness against high fractions of malicious clients—e.g., up to 50%—significantly enhancing the security and practicality of distributed spectrum sensing.

Technology Category

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

Application Category

• 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
• Security and Privacy: Large-scale security measurements

Abstract

Advancements in wireless and mobile technologies, including 5G advanced and the envisioned 6G, are driving exponential growth in wireless devices. However, this rapid expansion exacerbates spectrum scarcity, posing a critical challenge. Dynamic spectrum allocation (DSA)--which relies on sensing and dynamically sharing spectrum--has emerged as an essential solution to address this issue. While machine learning (ML) models hold significant potential for improving spectrum sensing, their adoption in centralized ML-based DSA systems is limited by privacy concerns, bandwidth constraints, and regulatory challenges. To overcome these limitations, distributed ML-based approaches such as Federated Learning (FL) offer promising alternatives. This work addresses two key challenges in FL-based spectrum sensing (FLSS). First, the scarcity of labeled data for training FL models in practical spectrum sensing scenarios is tackled with a semi-supervised FL approach, combined with energy detection, enabling model training on unlabeled datasets. Second, we examine the security vulnerabilities of FLSS, focusing on the impact of data poisoning attacks. Our analysis highlights the shortcomings of existing majority-based defenses in countering such attacks. To address these vulnerabilities, we propose a novel defense mechanism inspired by vaccination, which effectively mitigates data poisoning attacks without relying on majority-based assumptions. Extensive experiments on both synthetic and real-world datasets validate our solutions, demonstrating that FLSS can achieve near-perfect accuracy on unlabeled datasets and maintain Byzantine robustness against both targeted and untargeted data poisoning attacks, even when a significant proportion of participants are malicious.

Problem

• Addressing spectrum scarcity in 5G/6G via dynamic allocation
• Overcoming labeled data scarcity in federated spectrum sensing
• Mitigating data poisoning attacks in FL-based sensing systems

Innovation

• Semi-supervised FL with energy detection
• Vaccination-inspired defense mechanism
• Robust against data poisoning attacks