A Simple yet Accurate Autoadaptive Model of Network Traffic for Detection of Attacks on Low Latency Services
Résumé
This paper addresses the problem of detection of attacks in computer networks. More precisely, we consider attacks on emerging low-latency services, which typically require a specific traffic management system. We present a simple yet very efficient hybrid method that takes advantage of both autoencoders and transformer models. The original method is compared with the current stateof-the-art on a large real-life dataset of network traffic to show the relevance of the proposed approach, especially for low falsepositive rates. A quick ablation analysis shows that the efficiency of the method relies on the combined use of the two approaches jointly in our hybrid model.
Origine | Fichiers éditeurs autorisés sur une archive ouverte |
---|---|
licence |