Machine Learning-driven optimization for SVM-based intrusion detection system in vehicular ad hoc networks

Machine learning (ML) driven solutions have been widely used to secure wireless communications Vehicular ad hoc networks (VANETs) in recent studies. Unlike existing works, this paper applies support vector machine (SVM) for intrusion detection in VANET. The structure of SVM has many computation advantages, such as special direction at a finite sample and irrelevance between the complexity of algorithm and the sample dimension. Intrusion detection in VANET is nonconvex and combinatorial problem. Thus, three intelligence optimization algorithms are used for optimizing the accuracy value of SVM classifier. These optimization algorithms include Genetic algorithm (GA), Particle swarm optimization (PSO), and ant colony optimization (ACO). Our results demonstrate that GA outperformed other optimization algorithms.