Implementation and Analysis of an Improved PCA technique for DDoS Detection
Sonam Salaria, Sakshi Arora, Nishita Goyal, Pooja Goyal, Shifaly Sharma · 2020
With ever-increasing numbers of IoT devices becoming an essential part of our everyday life, the risk of compromised data security has grown many folds. SDN is state-of-the-art technology in the networking domain. Combined with IoT, it provides centralized control to the controller enabling better management of the traffic and prevention of certain DDoS attacks in layer 3 and layer 7 of the OSI model. But the centralized management of the network makes the controller vulnerable to a new type of DDoS attack which targets the controller itself, thereby bringing the entire network down. This paper presents an improved Principal Component Analysis (PCA) technique with weighted principal components to counter this attack. The entire network is partitioned into different subnets and the improved PCA technique is applied to each part independently. The proposed method helped counter the DDoS attacks targeted at the controller or the switch with 95.24% accuracy in contrast with 92.3% of the previous partitioned PCA scheme.