Securing the Internet of Flying Things (IoFT): A Proficient Defense Approach

The Internet of Flying Things (IoFT), popularly known as drones, has recently been adopted to perform essential tasks in several mission-critical systems, such as military, medical, ambulance, firefighting, and transportation systems. However, the traffic communication of IoFT has been vulnerable to a wide range of cyberattack vectors that threaten the privacy and authentication of IoFT devices and data. This paper presents a proficient defense approach to capturing cyberattacks deliberated against the LoFT. Particularly, we characterize the performance of three supervised machine learning schemes, including the random forest classifier (RFC), the multi-layer perceptron (MLP), and the support vector machines (SVM). The models were trained and evaluated on a recent dataset for cyberattacks over IoFT, ECU-IoFT-2022, using two balanced classes (normal vs. attack). The models’ assessment showed that the based scheme for IoFT cyberattack categorization provides the best performance indicators, achieving a 99.5% classification accuracy. Besides, its performance is improved over the existing state-of-the-art models for IoFT.