Prepublicación2026

Submitted to Journal of Network and Computer Applications

DecoyTrace: Toxic Decoys for Active Defense in Decentralized Federated Learning

DecoyTrace usa ciberengaño para defender DFL sin servidor. Un nodo señuelo móvil envía modelos cebo a sus vecinos, detecta respuestas envenenadas, aísla el origen y recupera el entrenamiento, con 60 configuraciones en NEBULA.

Resumen

Decentralized Federated Learning (DFL) eliminates the central aggregation server, reducing the single point of observation that traditional defenses against attacks rely on. As a result, peer-to-peer networks become exposed to malicious updates containing backdoors or semantic poisoning, since such updates can remain close to benign ones in the parameter space while behaving very differently. This may evade defenses based on passive parameter inspection. However, existing deception-based defenses have mainly been designed for centralized FL and do not jointly address local observation, poisoning propagation, source attribution, and containment in strictly serverless DFL. To address these limitations, this paper presents DecoyTrace, a proactive cyber deception-based defense for strictly serverless DFL environments. DecoyTrace deploys a mobile DecoyNode that generates decoy challenges using chaotic maps, disseminates a dual model (clean vs. decoy) based on neighbor trust, and evaluates them using three-state semantic metrics. Upon confirmation, a distributed protocol isolates the source and performs a model reset or recovery to preserve training progress. Evaluated across sixty configurations on the NEBULA platform (five datasets, three topologies, and four attack/defense scenarios), DecoyTrace systematically restores lost utility. The F1-score remains within 0.03 of the baseline on MNIST/FashionMNIST (mitigating drops of up to 0.37), matches or exceeds the baseline on EMNIST and CIFAR-100, and remains between 0.05 and 0.10 below the baseline on CIFAR-10, the most visually complex convolutional scenario evaluated. Furthermore, containment reduces CPU and network usage by up to two-thirds. These results demonstrate the feasibility of unifying deception, identification, and containment in DFL, while also identifying its limitations in complex tasks and multi-attractor threat models.

Autores

Pedro Beltrán López, Enrique Tomás Martínez Beltrán, Pantaleone Nespoli, Manuel Gil Pérez, Alberto Huertas Celdrán

Palabras clave

  • Aprendizaje federado descentralizado
  • Cyber Deception
  • Active Defense
  • Poisoning Attacks
  • Backdoor Attacks
  • Mitigación de ataques

Publicaciones relacionadas

Publicaciones relacionadas por tema, método o aplicación.

Artículo de revista2026

Computer Networks

RepuNet: A Reputation System for Mitigating Malicious Clients in DFL

Isaac Marroqui Penalva, Enrique Tomás Martínez Beltrán, Manuel Gil Pérez, Alberto Huertas Celdrán

RepuNet estudia cómo reducir la influencia de clientes maliciosos en DFL. Combina señales del modelo y de la comunicación en una reputación que ajusta el peso de cada par durante la agregación, con evaluación en NEBULA.

Artículo de revista2026

Information Fusion

Decentralized Federated Learning with Multimodal Prototypes for Heterogeneous Data

Enrique Tomás Martínez Beltrán, Gérôme Bovet, Gregorio Martínez Pérez, Alberto Huertas Celdrán

Modalis aborda clientes con modalidades diferentes o ausentes en aprendizaje descentralizado. Intercambia prototipos compactos de clase y alinea representaciones locales para colaborar con datos heterogéneos.

Investigación relacionada