An Intelligent and Lightweight ML Model for Real-Time DDoS Attack Detection in IoT Networks
N Radhika Amareshwari, K. Venkateswarlu, Kaushik Shivakumar, Amit Raj, Rakesh Kumar Donthi, Santhosh Kumar Medishetti
Abstract
The rapid expansion of Internet of Things (IoT) networks has introduced significant vulnerabilities, making them prime targets for Distributed Denial-of-Service (DDoS) attacks. Traditional detection systems such as signature-based methods and single-classifier models often fall short in identifying sophisticated and evolving threats due to their limited adaptability and high computational demands. This research proposes a lightweight Machine Learning (ML) framework based on the Random Forest (RF) algorithm to detect and mitigate DDoS attacks in IoT environments. The framework emphasizes efficient preprocessing and feature extraction to enable accurate classification while maintaining low resource consumption, making it suitable for deployment in constrained IoT devices. Using the ToN_IoT dataset and simulating the environment in NS3, the proposed system was benchmarked against conventional techniques including SVM and SNORT. Experimental results reveal that the RF-based approach outperforms the others in terms of detection accuracy, precision, recall, and F1-score, demonstrating its robustness in recognizing both known and unknown attack patterns. The lightweight and scalable nature of the proposed model ensures enhanced network security with minimal overhead, making it a practical and effective solution for real-time DDoS defense in IoT-based ecosystems.