FedJam: Multimodal Federated Learning Framework for Jamming Detection

TL;DR

FedJam is a novel multimodal federated learning framework that enhances wireless jamming detection by combining spectrograms and network KPIs, ensuring privacy, efficiency, and robustness in diverse real-world scenarios.

Contribution

The paper introduces FedJam, the first multimodal federated learning framework for on-device jamming detection that fuses spectrograms and KPIs with a lightweight architecture.

Findings

FedJam outperforms unimodal baselines by up to 15% in accuracy.

Requires 60% fewer communication rounds to converge.

Maintains low resource utilization and robustness under heterogeneous data distributions.

Abstract

Jamming attacks pose a critical threat to wireless networks, yet existing detection methods remain largely unimodal, centralized and resource-intensive, limiting their performance, scalability, and deployment feasibility, respectively. To address these limitations, we present FedJam, a multimodal Federated Learning (FL) framework for on-device jamming detection and classification. FedJam locally fuses spectrograms and cross-layer network Key Performance Indicators (KPIs) using a lightweight dual-encoder architecture with an integrated fusion module and multimodal projection head, that enables privacy-preserving training and inference without transmitting raw data. We prototype and deploy FedJam on a wireless experimental testbed and evaluate it using the first, over-the-air multimodal dataset comprising synchronized samples across benign and three distinct jamming attack types. FedJam outperforms state-of-the-art unimodal baselines by up to 15% in accuracy, while requiring 60% fewer communication rounds to converge, and maintains low resource utilization. Its advantage is especially pronounced in realistic scenarios, where it remains extremely robust under heterogeneous data distributions across devices.