A New Approach to Network Traffic Efficiency and DDOS Attack Detection on Software-defined Networks
Abdülkadir Çakır, Enes Açıkgözoğlu · Research Square · 2022
Abstract Many devices have been connected to each other and a wide platform has been formed with the development of internet technologies. The continuous expansion of this platform has revealed requirements such as single-point management, accessibility, bandwidth management, and efficient use of the network. Considering that software-defined networks are systematically managed with software, it is predicted that they will meet the specified network requirements more easily. In this study, a model that detects both efficient use of the network and Distributed Denial of Service Attack (DDOS) attacks by clustering the bandwidths according to the network traffic history collected over the software-defined network is proposed. A dataset specific to the study was created to determine possible attacks and usages according to bandwidth. From the data in the dataset, 3 different datasets were created with the Kmeans clustering algorithm. A virtual network was created for the implementation of the model and tests were carried out on this network. Efficient use of the network is ensured by allocating bandwidth according to clusters created especially in multi-user, heavy-traffic networks. In addition, while the data is being collected, DDOS scanning is also performed to prevent possible attacks on the network.