Detection and mitigation of botnet based DDoS attacks using catboost machine learning algorithm in SDN environment
R Sanjeetha, A. Gokul Raj, Kolli Saivenu, Mumtaz Irteqa Ahmed, B. Sunal Sathvik, Anita Kanavalli · International Journal of Advanced Technology and Engineering Exploration · 2021
Software-Defined Networks is an emerging new technology in the field of networks that had a lot of impact in fields such as cloud computing and data centers.In traditional networks, the data plane, which is responsible for the forwarding of data, and the control plane, which decides the path of the packet are tightly coupled.But in software-defined networks, the control plane and data plane are separated.A special device called the controller acts as the control plane.The switches act as the data plane which forwards the packets based on the flow rules defined by the controller. *Author for correspondenceThe controller is connected to all the switches and the controller communicates with the switches using a secured protocol called OpenFlow.Such a system encourages modularity, freedom to choose the software and the hardware, and is quite robust when deployed as a network.The introduction of SDN brought in certain advantages that were absent in traditional networks, but such networks are still susceptible to DDoS attacks which can disrupt all the services in the network.This requires the SDN to have an efficient, quick and accurate detection and mitigation mechanism for such attacks.The use of XGBoost is proposed for the purpose of DDoS attack detection and has been observed to outperform traditional algorithms such as Support Vector Machine (SVM) and random forest in terms of accuracy and speed [1].