Lightweight Intrusion Detection for IoD Infrastructure using Deep Learning

Abstract

The rapid growth of the Internet of Drones (IoD) has created new challenges for cybersecurity experts. Network intru- sion remains a major concern in cyberspace, and traditional in- trusion detection methods are limited in their ability to detect and prevent attacks. Machine learning-based approaches have shown promise in detecting network intrusions, but their accuracy is still a challenge. To address this, a machine learning approach was proposed using seven classifiers, including DT, random forest, na¨ıve bayes, Adaptive Boosting Algorithm (ADA), Adaptive Boosting Algorithm (XGB), K-Nearest Neighbors (KNN), and logistic regression. The proposed model was evaluated on the CICIDS2017 dataset, achieving high accuracies with the DT classifier having the highest accuracy of 0.99. This approach can be applied to detect and prevent network intrusions in the growing IoD network, ensuring the integrity, confidentiality, and availability of communication networks.