Security Model for the Detection of Distributed Denial of Service Attack (DDoS) at the Access Layer in Cloud Reference Architecture

Abstract

DDoS attacks on the cloud have become an emerging problem, which calls for an intelligent protection solution for information, communication and computing resources. Cloud reference architecture as a new technology with wider access facility, the access layer in the reference architecture becomes grave to numerous security challenges, and it is mostly exploited by the attacker using compromised computers to launch DDoS attack which affects its performance by shutting down some services. Many security frameworks have been proposed in the past, but they are not enough to provide security solution for accurate detection and control of DDoS attack at the access layer. Intelligent security model based on artificial neural network was developed to detect and control DDoS attack at the access layer. DDoS dataset obtained online was used to train and test the intelligent security model using MATHLAB workspace. Simulated results implemented and generated with receive operator characteristic analyzer (Roc) showed a true positive rate of 98.2% to detect and control DDoS attack at the access layer and a false negative rate of 0.02%. The inclusion of additional parameters to determined and identifies the source of the attack, whether it is lunch from outside or inherent within the reference architecture is recommended. KEYWORDS: Reference Architecture, Access Layer, Distributed Denial of Service (DDOS), Artificial Neural Network’s (ANN’s)