No abstract

Existing Distributed Denial-of-Service (DDoS) attacks detections in software defined networks (SDNs) typically only carry out detection in a single domain. In reality, abnormal traffic usually affects multiple network domains. Thus, the cross-domain attacks detection has been proposed to improve the detection performance. However when participating in the detection each SDNs domain needs to provide a large number of real traffic data where private information may be leaked out. Existing multiparty privacy protection schemes often achieve privacy guarantees by sacrificing accuracy or increasing the time cost. It is a challenging task to have both high accuracy and reasonable time consumption.In this paper, we propose Predis, a privacy-preserving cross-domain attacks detection scheme for SDNs. Predis combines perturbation encryption and data encryption to protect privacy, and uses a computationally simple and efficient algorithm k-Nearest Neighbor (kNN) as its detection algorithm. We also improve the kNN to achieve better efficiency. Through theoretical analysis and extensive simulations, we demonstrate that Predis is capable to achieve efficient and accurate attacks detection, while keeping sensitive information of each domain secure.

No abstract

No abstract