Dynamic positioning of UAVs to improve network coverage in VANETs

Abstract

Unmanned aerial vehicles (UAVs) are expected to become an essential component of the future wireless networks, opening up a plethora of options to improve communications between vehicles and infrastructures. In addition, Road Side Units (RSUs) are deployed in a limited number to reduce the monetary cost leads to many uncovered regions in the network. Motivated by this, we propose a collaborative network coverage enhancement scheme (CONEC) to bring the uncovered vehicles within the infrastructure's coverage. In this paper, we leverage the UAVs to facilitate seamless network coverage for ground vehicles by optimally placing the UAVs in optimal places. To this end, we utilize the Particle Swarm Optimization (PSO) algorithm to find the best UAV deployment position in the whole network while considering vehicle density, heading direction, and previous coverage information. After that, PSO is called successively to obtain the optimal number of UAVs required to provide coverage to vehicles up to predefined network coverage threshold. Furthermore, we propose additional algorithm to efficiently schedule the UAVs in each time frame to minimize the wastage of UAVs service time. The simulation results demonstrate that the proposed scheme CONEC improves the network performance of VANET in terms of packet delivery ratio (PDR), hop counts (HOPs), end-to-end delay (EED), and throughput compared to its counterpart.