TASER: Task-Aware Spectral Energy Refine for Backdoor Suppression in UAV Swarms Decentralized Federated Learning

As backdoor attacks in UAV-based decentralized federated learning (DFL) grow increasingly stealthy and sophisticated, existing defenses have likewise escalated in complexity. Yet these defenses, which rely heavily on outlier detection, remain vulnerable to carefully crafted backdoors. In UAV-DFL, the lack of global coordination and limited resources further render outlier-based defenses impractical. Against this backdrop, gradient spectral analysis offers a promising alternative. While prior work primarily leverages low-frequency coefficients for pairwise comparisons, it neglects to analyze the intrinsic spectral characteristics of backdoor gradients. Through empirical analysis of existing stealthy attacks, we reveal a key insight: the more effort attackers invest in mimicking benign behaviors, the more distinct the spectral concentration becomes. Motivated by this, we propose Task-Aware Spectral Energy Refine (TASER) -- a decentralized defense framework. To our knowledge, this is the first efficient backdoor defense that utilizes spectral concentration instead of complex outlier detection, enabling mitigation of stealthy attacks by structurally disrupting the backdoor task. To suppress the backdoor task, TASER preserves main-task-relevant frequency coefficients and discards others. We provide theoretical guarantees and demonstrate through experiments that TASER remains effective against stealthy backdoor attacks that bypass outlier-based defenses, achieving attack success rate below 20% and accuracy loss under 5%.

Key Contributions

Empirical discovery that stealthy backdoor attacks, by mimicking benign gradients, inadvertently produce more concentrated spectral energy in mid-frequency bands — enabling frequency-domain detection
TASER defense framework that scores frequency components using approximate Fisher information and inter-batch consistency, preserving main-task-relevant coefficients while discarding backdoor-carrying ones
Decentralized, communication-efficient design requiring no global coordinator, achieving attack success rate below 20% with under 5% accuracy loss against stealthy attacks that bypass outlier-based defenses

🛡️ Threat Analysis

Model Poisoning

The paper's entire contribution is a defense against backdoor/trojan attacks in federated learning, where malicious UAV nodes inject trigger-based hidden behavior into the collaborative model. TASER is a backdoor suppression framework that targets the spectral signature of stealthy backdoor gradients.