On the Collaborative Inference of DDoS: An Information-theoretic Distributed Approach

摘要: Literature contributions have shown that information theoretic techniques can effectively detect various types of Distributed Denial Service (DDoS) attacks. However, such are often centralized with a limited measurement vantage point and suffer from the issue single failure. Furthermore, flourishing distributed cloudbased environments, ought to adapt settings for scalability performance reasons. In this paper, we address problem collaborative DDoS detection using information-theoretic techniques. To end, propose an entropy-based mechanism supports agreement identify suitable tuning network parameters inference in real-time. Empirical evaluations real attacks demonstrate proposed approach is indeed capable cooperatively inferring while achieving resiliency scalability.