DRIVE: Dual rider-remora optimization for vehicular routing

The evolution of critical Internet of Things (IoT) wings research on Vehicular Ad-hoc Networks (VANETs) added transmission as the primary pre-requisite for the successful deployment of Intelligent Transportation System. The major challenge, however, with VANETs, lies in ensuring network security, particularly in the face of complex attacks. For secure communication, the focus has recently been shifted to trust-based models as simpler and cost-effective alternative to cryptography based solutions, especially in handling delay-sensitive situations. However, the trust-based models suffer from poor diversity and lack in adapting with ever-changing dynamic VANET scenarios. Some recent works reveal that Artificial Intelligence (AI) models can help with the rigidity of the trust based models and provide the near-optimal solution with adaptive decision-making. However, there is a huge gap to fill from the research perspective. Working on these gaps, in this work, an AI-based novel hybrid optimization algorithm combining Rider Optimization and Remora Optimization has been proposed that finds the optimal secure routing path using fitness parameters considering trust, energy, delay, and distance between the nodes. The performance of the proposed algorithm is further enhanced by implementing the concept of fractional calculus to improve its convergence capability for faster decisions. The results prove the proposed trust based algorithm’s adaptive efficiency in terms of energy, trust, throughput, and packet delivery ratio in varying scenarios of sparse and dense traffic (with and without attacks). Moreover, simulation results show considerable improvement with the proposed hybrid approach compared to the individual algorithms.