Vehicular Computation Offloading in UAV-enabled MEC Systems

Abstract

The UAV-enable Mobile Edge Computing (MEC) systems and Vehicular Ad-hoc Network (VANET)-supported applications are very popular topics these days. The paper considers the vehicular task offloading problems for the Software-Defined Vehicular Network (SDVN)-supported services in the UAV-enabled MEC system. In the considered problem, one UAV and one edge server (ES) are provisioned for the workload from the moving vehicles in a certain region. For each vehicle in the region, it would periodically submit requests to the UAV-enable MEC system until it leaves the region. Each request will be taken as a computation task and could be offloaded locally on the vehicle, the UAV, or the ES. Multiple communication and energy consumption models are employed to formulate the problem model. The objectives are to minimize the total time delays and the energy consumption. A greedy heuristic based dynamic scheduling framework is proposed for the problem under study. Simulated experiments are delicately designed with dynamic traffics, various road and building distributions. Experimental results show that the proposal is more effective than the compared algorithm.