ENHANCING DDOS ATTACK DETECTION IN SDN ENVIRONMENTS USING GRU

Distributed Denial of Service (DDOS) attacks continue to pose a serious threat to Internet security, aiming to render specific systems or networks inaccessible. Detecting diverse types of DDOS cyber attackss with improved algorithms while managing computational costs is critical for bolstering cyber security. DDOS attacks, originating from various distributed sources across multiple network locations, exploit vulnerabilities to degrade performance, exhaust resources, or monopolize networks, impeding legitimate users. This paper introduces a novel DDOS attack detection system in SDN environments employing a Deep Learning(DL) approach with a focus on the recently released CICDDoS2019 dataset. Our proposed model, leveraging the Gated Recurrent Unit (GRU) algorithm, demonstrates significant improvements in attack detection compared to bench marking methods, enhancing confidence in securing SDN networks against evolving DDOS attackss.