Energy Efficient Task Offloading for UAV-assisted Mobile Edge Computing

Abstract

With 5G technologies developing rapidly, there emerges a variety of computation-intensive and delay-sensitive tasks. Owing to the characteristics of low cost and high mobility, unmanned aerial vehicles (UAV) is supposed to a tool to provide high quality service in wireless communication system. The UAV-assisted mobile edge computing (MEC) system is employed to provide computing offload for mobile terminal (MT), which can provide reliable wireless communication where there is no communication infrastructure or communication functions are limited. Due to the limitations of UAV's inability to work for a long time and insufficient computing power, it brings new challenges to computing offload. A task offloading scheme is studied to reduce the total cost of UAV-assisted MEC system. We propose an optimal MTs task offloading strategy, which is called Coordinate Descent based Offloading Decision (CDOD) algorithm to solve the problem that limitation insufficient computing power and unable to work continuously of UAV. Simulation results verify the effective performance of CDOD algorithm in carious experiment conditions.