UAV networks DoS attacks detection using artificial intelligence based on weighted machine learning

Abstract

While Unmanned Aerial Vehicles (UAVs) have found applications across numerous industries, they still remain vulnerable to various cybersecurity challenges. Different types of cyberattacks target UAVs. Early detection of these cyberattacks is considered the most important step in ensuring the cybersecurity of UAVs. In this article, an artificial intelligence method based on machine learning was developed for detecting different types of Denial of Service (DoS) attacks targeting the UAV network. Initially in this work, feature selection methods are implemented to select the most important features. Then, machine learning methods are used to classify attacks. According to the conducted experiments, the proposed method outperformed others with an accuracy of 99.51 % and a prediction time of 0.1 s. Additionally, a novel dataset is used in this work, which offers several advantages. The dataset was created within a real-world environment rather than a simulated one. Furthermore, the data were collected within a 5G network.