CNN-Based Model for the HTTP Flood Attack Detection

Flooding Distributed Denial of Service (DDoS) attacks are a major concern for security professionals. DDoS flooding attacks are typically deliberate attempts to impede the access of legitimate users to services. Numerous researchers have worked on the detection and prevention of DDoS attacks over the past decade. As a result of the diversity of DDoS attacks, the solutions are wide-ranging. Among the most effective detection and prevention mechanisms, mitigation techniques based on artificial intelligence have become a trending topic. In this study, we focus on HTTP flooding DDoS attacks. By Implementing a detection model based on 1-D Convolutional Neural Network (CNN), our proposed solution detects the HTTP flooding attack in its earliest stages. Nearly 2,100,000 traffic data are used to test the model. The proposed CNN model produced an accuracy of 99%. Our model performed well when applied to benchmark datasets CICIDS2017 and CSE-CIC-IDS 2018.