"ENHANCING VANET ROUTING WITH A HYBRID BEAGLE-INSPIRED AND PARTICLE SWARM OPTIMIZATION ALGORITHM"

Abstract

The dynamic topology of Vehicular Ad-hoc Networks (VANETs) presents a significant challenge for efficient data routing. In order to identify the best communication routes, this study analyzes the multi-hop packet forwarding problem as a Traveling Salesman Problem (TSP). We present a unique Beagle-Inspired Optimization Algorithm (BIOA) and its hybrids with Differential Evolution (DE) and Particle Swarm Optimization (PSO). These are compared with various metaheuristics such as Ant Colony Optimization (ACO), Wave Optimization, a Genetic Algorithm (GA), and Snake-Skin Shedding Optimization (SSSO). By successfully balancing exploration and exploitation, the hybrid BIOA-PSO algorithm delivers excellent performance, achieving an accuracy of 76.5% and an F1-score of 72.9%, according to extensive simulations on a real-world VANET dataset. An effective substitute, ACO offers a good trade-off between computational time (1.92 seconds) and performance (74.0% accuracy). The hybrids' quick stability is demonstrated by convergence and feature importance evaluations, which also show that traffic density and erratic speed are the most important variables affecting routing choices. The findings offer a useful foundation for choosing optimization methods according to particular VANET application needs.