Skip to main content Skip to main navigation


Cross-Silo Horizontal Federated Learning Methods in Network Traffic Analysis

Sogo Pierre Sanon; Rekha Reddy; Christoph Lipps; Hans Dieter Schotten
In: European Wireless 2023. European Wireless (EW-2023), October 4, Rome, Italy, IEEE, 10/2023.


Federated Learning (FL) is a Machine Learning (ML) technique allowing multiple parties to collaboratively train a modelwithout sharing their raw data with each other. This approach is particularly useful in scenarios where data privacy, datasovereignty, and data protection are a concern and ensures organizations comply with data protection laws like the GeneralData Protection Regulation (GDPR). In network traffic analysis, FL is applicable in scenarios where decentralized data sourcesare used for training, for example in network traffic prediction, which is a critical task for resource allocation and networkoptimization. Therefore, in this work, the performance of different aggregation methods in FL for network traffic predictionis examined. Aggregation methods such as weighted federated averaging, secure aggregations, and robust aggregations areconsidered for a comparative study. The performance of these methods is evaluated on the GEANT project’s public dataset.This work provides insights into the selection of appropriate FL methods in network traffic analysis and sheds light on thetrade-offs between accuracy, and robustness in the presence of malicious clients. The results indicate that the combination of medianaggregation with Homomorphic Encryption (HE) is a good choice as it provides feasibility for performance, security, and robustness.