Sec-IoV

Rapid progress in Intelligent Transportation Systems (ITS) during the last few decades has resulted in the emergence of the Internet of Vehicles (IoV), in which smart vehicles communicate with each other for information sharing. The exponential increase in the number of vehicles, together with an increasing data demands from in-vehicle users, has led to a tremendous growth of upstream data in the IoV infrastructure. However, the highly dynamic topology and distributed nature of vehicular networks exposes the vehicular traffic to higher security risks. To address these challenges, in this paper, "Sec-IoV", a multi-stage model for anomaly detection is specifically proposed for securing vehicle-to-vehicle (V2V) communications in IoV setups. The proposed anomaly detection model comprises of multiple stages: (a) relevant feature set selection, (b) optimization of Support Vector Machine (SVM) parameters, and (c) classification of vehicular traffic into benign and anomalous. The first two stages are expressed using the multi-objective optimization problems, which are iteratively computed using the hybridization of a meta-heuristic approach "Artificial Bee Colony (ABC) Optimization" with a Cauchy based mutation operator. This coupling is referred to as "C-ABC". It improves the local search capability of the optimizer with faster convergence. The last stage of data classification is then performed by employing SVM with a refined set of parameters. For the extensive evaluation of the proposed model, different state-of-the-art models have been executed on OMNET++ and SUMO. The obtained results in terms of the detection rate, accuracy, and false positive rate reflect the effectiveness of the proposed Sec-IoV model against the existing state-of-the-art schemes.