EC-MOPSO: an edge computing-assisted hybrid cluster and MOPSO-based routing protocol for the Internet of Vehicles

In recent years, the Internet of Vehicles (IoV) has received a lot of attention due to its unique features, such as rapid topology change, specific movement patterns, and variable node density and speed. Providing an efficient routing algorithm is one of the main challenges of these networks. Roadside units (RSU) defined in vehicular Ad hoc network (VANET) architecture can play as an edge computing device and assist in the routing process. They are usually fixed along the roadside or in specific locations such as junctions. On the other hand, bioinspired metaheuristic optimization algorithms are good candidates for this field due to their dynamic nature and ability to consider several parameters simultaneously. Clustering can also be used to reduce complexity. In this paper, an edge computing-assisted cluster-based routing algorithm utilizing multi-objective particle swarm optimization (MOPSO), named EC-MOPSO, is presented for the IoV applications. The proposed method evaluates different particles based on the three objective functions of accumulative delay, number of hops, and number of cluster members. Particles suggesting lower delays, fewer hops, and more members in the same cluster are considered to be superior. RSUs (edges) are responsible for the optimization procedure. Evaluation results show that the proposed method has a significant advantage over similar works in terms of distance traveled, number of hops, latency, packet delivery ratio, and convergence time.