DDoS Attack Detection in Software Defined Network using Ensemble K-means++ and Random Forest
Diash Firdaus, Rendy Munadi, Yudha Purwanto
Abstract
SDN (Software Defined Network) is the future of networking and has attracted great interest as a new paradigm in networking. SDN has centralized control by separating control plane and data plane, it will be very vulnerable to DDoS attacks. To improve security, it requires high detection accuracy and efficiency. To detect DDoS attacks on SDN we propose DDoS detection using Machine Learning with Ensemble Algorithm. At the experimental stage, we used InSDN as a dataset. This study consists of two methodologies. The first step is the clustering and classification method, the clustering and classification method has two stages, the first stage is feature selection and normalization, and the second stage is Ensemble Algorithm clustering and classification. The second step is the detection validation method in SDN using the Mininet emulator. We use Ensemble Algorithm K-means++ and Random Forest to obtain High detection accuracy and efficiency.