Gradient Boost Enhanced Artificial Immune System Algorithm for Adaptive DDoS Attack Detection in IoT

Distributed Denial of Service attacks (DDoS) targeting the Internet of Things (IoT) remain a pervasive cybersecurity challenge. Biologically inspired solutions have shown promise for DDoS attack detection. For example, the human immune system has inspired various Artificial Immune System (AIS) solutions for anomaly detection. In this paper, we address the challenges of DDoS detection in IoT by proposing a Gradient boost regression and Adam-optimized Negative Selection Algorithm (GANSA). We show that the proposed algorithm is effective and can adapt to changes in network traffic patterns, thereby accurately detecting known and unknown DDoS attacks. We evaluate the proposed system against state-of-the-art machine learning DDoS detection algorithms (e.g., CNN, SVM). We show that the proposed system achieves a low false positive rate (0.0003) and near-perfect detection accuracy (0.99), F1 score (0.99), and MCC (0.97) while adapting to incoming network traffic in real-time.