MEC-Enabled Fine-Grained Task Offloading for UAV Networks in Urban Environments

Abstract

In recent years, with the continuous development of information technology, the amount of data generated and hosted by cloud service platforms in urban environments is unprecedented. Mobile edge computing (MEC) is combined with UAV networks to better realize the ability to provide nearby services to a large number of terminal devices in cities. Unmanned aerial vehicles (UAVs) are highly maneuverable and inexpensive and are good carriers for carrying MEC platforms. In UAV edge networks, we usually face the problem of fine-grained task offloading based on relevant features of urban environments. We need to address high energy consumption and task processing delays to help achieve urban sustainability goals. Therefore, we combine the software definition network (SDN) technology and, on this basis, we propose two task offloading strategies based on an improved EFO intelligent algorithm for different user scales. At the same time, we run the proposed offloading system in the UAV sensor. The experiment shows that, compared with the traditional strategy, the unloading efficiency of the proposed method can be improved by about 10%.