Real-Time Position Falsification Attack Detection System for Internet of Vehicles

Ensuring secured and reliable dissemination of information for a mission-critical system such as the Internet of Vehicle (IoV) in real-time is of utmost importance. In this work, a False Location Detection System (FLDS) based on an optimized Ensemble Random Forest (Ens.RF) was proposed. The performance of the Ens.RF was compared with four other Machine Learning (ML) algorithms, using the Veremi dataset where five (5) different location falsification categories and one benign category were modeled. To validate the idea in this work, a performance comparison with recent work was presented. The result shows that the proposed Ens.RF outperformed other algorithms modeled in this work as well as related works with an accuracy of 99.92%