Adapt under Attack and Domain Shift: Unified Adversarial Meta-Learning and Domain Adaptation for Robust Automatic Modulation Classification

2025-11-02

AI Summary

Deep learning-based automatic modulation classification (AMC) suffers from dual vulnerabilities: susceptibility to adversarial attacks and sensitivity to domain shifts, hindering real-world deployment. To address this, we propose the first unified framework integrating meta-learning and domain adaptation. In the offline phase, meta-adversarial training enhances model generalization against unseen attacks; in the online phase, feature-level domain alignment enables low-label-cost cross-domain adaptation. Our approach jointly optimizes for both adversarial robustness and distributional robustness—overcoming the limitations of conventional single-objective optimization. Extensive experiments demonstrate that the framework significantly improves classification accuracy under multiple concurrent threats (e.g., diverse adversarial perturbations and domain mismatches), while substantially enhancing model reliability and generalizability in dynamic wireless environments.

Technology Category

• Machine Learning: Adversarial Learning & Robustness
• Computer Vision: Adversarial Attacks & Robustness
• Multiagent Systems: Adversarial Agents

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Vertical and domain-specific search

Abstract

Deep learning has emerged as a leading approach for Automatic Modulation Classification (AMC), demonstrating superior performance over traditional methods. However, vulnerability to adversarial attacks and susceptibility to data distribution shifts hinder their practical deployment in real-world, dynamic environments. To address these threats, we propose a novel, unified framework that integrates meta-learning with domain adaptation, making AMC systems resistant to both adversarial attacks and environmental changes. Our framework utilizes a two-phase strategy. First, in an offline phase, we employ a meta-learning approach to train the model on clean and adversarially perturbed samples from a single source domain. This method enables the model to generalize its defense, making it resistant to a combination of previously unseen attacks. Subsequently, in the online phase, we apply domain adaptation to align the model's features with a new target domain, allowing it to adapt without requiring substantial labeled data. As a result, our framework achieves a significant improvement in modulation classification accuracy against these combined threats, offering a critical solution to the deployment and operational challenges of modern AMC systems.

Problem

• Enhancing AMC robustness against adversarial attacks and domain shifts
• Unifying meta-learning and domain adaptation for dynamic environments
• Improving modulation classification accuracy under unseen threats

Innovation

• Meta-learning trains model on adversarial samples
• Domain adaptation aligns features with new domains
• Unified framework resists attacks and distribution shifts