RBF-SVM kernel-based model for detecting DDoS attacks in SDN integrated vehicular network

The development of the intelligent transport space (ITS) comes with the challenge of securing transportation data. As the vehicular network is highly dynamic, the network architecture is susceptible to distributed malicious attacks irrespective of the emergence and integration of enabling technologies. A Software Defined Network (SDN) based VANET is an improvement over the traditional VANET and may be susceptible to attacks as a result of its centralized structure, resulting in dangerous circumstances. SDN-based VANET security for 5G deployments is essential and calls for integrating artificial intelligence tools for threat detection. An intrusion detection system was suggested in this work to detect Distributed Denial of Service (DDoS) attacks in the vehicular space. The proposed solution uses the Grid Search Cross-Validation (GSCV) exhaustive parameter search technique and the Radial Basis Function Kernel of the Support Vector Machine (RBF-SVM) algorithm. The performance of other ML algorithms implemented was compared using key performance metrics. Experimental simulations to validate the proposed framework's efficacy against DDoS intrusion show that the proposed scheme demonstrated an overall accuracy of 99.40% and a mean absolute error of 0.006.

Automated summary

The development of intelligent transport systems (ITS) brings the challenge of securing transportation data. Software Defined Network (SDN) based Vehicle Ad-Hoc Networks (VANETs) are susceptible to attacks due to their centralized structure. In this research, an intrusion detection system using artificial intelligence tools, such as the Grid Search Cross-Validation (GSCV) technique and Support Vector Machine (SVM) algorithm, was proposed to detect Distributed Denial of Service (DDoS) attacks in VANETs. Experimental simulations showed that the proposed framework achieved an overall accuracy of 99.40% and a mean absolute error of 0.006 in detecting DDoS intrusion.