Multi-objective auto-regressive whale optimisation for traffic-aware routing in urban VANET

Vehicular ad-hoc network (VANET) is a growing networking concept that has been used increasingly in various applications including traffic alert broadcasting. The main purpose of VANET is to provide safety to the drivers by alerting them to the dangers that may happen. This study presents a traffic-aware routing protocol in VANET by the introduction of multi-objective auto-regressive whale optimisation (ARWO) algorithm. ARWO algorithm selects the best path from the multiple paths by considering the multiple objectives, such as end-to-end delay (EED), link life time, packet delay and distance, in the fitness function. Here, the traffic density and the expected average speed of the vehicle are predicted by the exponential weighted moving average approach. The performance of ARWO protocol is compared with four existing techniques, like stable CDS-based routing protocol, fractional glow worm swarm optimisation, glow worm swarm optimisation, and Whale Optimization Algorithm (WOA), using the metrics, EED, distance, traffic density, and throughput. The simulation results show that the proposed ARWO algorithm achieves EED of 2.941, a distance of 2.15, traffic density of 0.009 and throughput of 0.1, respectively, at maximum constraints, i.e. at a maximum number of vehicles and simulation time and thus proves its efficiency against the comparative protocols.