Analysis of DDoS Attacks in Software Defined Networking using Machine Learning

The most practical configurable network is Software Defined Networking (SDN). SDN delineates the network's control layer from the data layer, while operationally integrating them. With global visibility into network status, the's logical centralized control improves network management. However, centralized management introduces new security risks. One of the most attractive victims of distributed denial of service (DDoS) attacks is the SDN control platform. Due to machine learning (ML) performance and unavoidable advantages against fingerprint vulnerabilities, this article proposes and considers a ML solution for protecting against DDoS attacks in SDN. ML techniques are evaluated in both, a theoretical format and a practical setup such that SDN controllers are disclosed to DDoS-attacks in order to reach significant deductions about ML-based cybersecurity of future networking protocols.