Proactive UAVs Placement in VANETs

Seamless connectivity between the vehicles and the infrastructures is required for the provisioning of future vehicular applications. However, the infrastructures in vehicular ad hoc networks (VANETs) such as roadside units (RSUs) are insufficient to provide ubiquitous connectivity and coverage in urban areas. Thus the services get interrupted, which results in lower network performance in VANETs. Motivated by this in this paper, we proposed a Proactive UAV Placement (PUP) technique to assist the RSUs in delivering the services in the out-of-coverage region by placing the UAVs in the appropriate positions. We first predicted the distribution of the future vehicles by utilizing the LSTM neural network. Based on the distribution of the future vehicles, the optimization problem for optimal UAV placement is formulated and solved by utilizing the Particle Swarm Optimization (PSO) algorithm. The results demonstrated that our scheme achieves better average coverage compared to two other UAV-assisted schemes.