Energy-Aware Intrusion Detection Model for Internet of Vehicles Using Machine Learning Methods

Abstract

With increasing development of Internet of Things (IoT) technology, wireless communications, big data, and smart applications, vehicular communications have become ubiquitous in smart cities, smart transportation systems, and Internet of Vehicles (IoV) environments. In this paper, a new Energy-aware Intrusion Detection System (EIDS) based on intelligent two-phase contract management model is presented for vehicle-to-vehicle (V2V) strategy in the IoV environments. In this strategy, the proposed EIDS predicts safe and energy-efficient end-to-end points for communication between existing vehicles in the IoV. The contract management process shows how the vehicles are connected together with a safe condition to transfer information. For prediction phase, a regression algorithm is applied to evaluate the proposed EIDS according to NSLKDD data set in the IoV environments. Simulation experiments show that the proposed regression-based EIDS strategy can effectively improve the accuracy and precision factors with 90% and 84%, respectively, and greatly minimize execution time by 4 seconds with respect to other machine learning algorithms.