A Deep Learning Approach for Detecting Distributed Denial of Service Attacks

Abstract

Distributed Denial of Service (DDoS) attacks remain one of the most significant threats to network security. By overwhelming a target with massive amounts of malicious traffic, these attacks disrupt essential online services and cause substantial financial losses. Traditional detection methods, such as signature-based and anomaly-based systems, struggle to detect new and evolving attack patterns. Moreover, most existing deep learning approaches focus on binary classification (normal vs attack), which limits their ability to distinguish between different types of DDoS attacks. This study proposes a deep learning approach for multi-class DDoS attack detection using a one-dimensional Convolutional Neural Network (1D CNN). The proposed model is trained and evaluated on the CICDDoS2019 dataset, which contains various types of DDoS attacks. The dataset undergoes preprocessing steps including label mapping, removal of constant and identifying columns, duplicate removal, handling of missing values, normalization, and type conversion. The model is evaluated using standard metrics: accuracy, precision, recall, and F1-score.

The experimental results demonstrate that the 1D CNN model effectively distinguishes be- tween different types of DDoS attacks. The main contributions of this work include a detailed

analysis of the CICDDoS2019 dataset, a 1D CNN model for multi-class classification, and a comparison with existing approaches.