Integrating Mobile Edge Computing Into Unmanned Aerial Vehicle Networks: An Sdn-Enabled Architecture

Abstract

Over the past few years, we have witnessed an exponential increase in demand for seamless and continuous Internet access to online services – multimedia services, online business, gaming, scientific analysis, and so on. With the significant advances in computing and communication techniques of edge devices, the mobile edge computing (MEC) paradigm has emerged and has been demonstrated to be more feasible and efficient in providing wireless Internet access or computing in some scenarios compared to the traditional cloud computing platform. Furthermore, the promotion and popularization of unmanned aerial vehicles (UAVs) in recent years provides the promis of ubiquitous data sensing, leading to an era of aerial MEC. In this article, we first present four representative architectures for the UAV-based MEC system. Then we survey some recent research products in this area. In particular, we argue that the UAV can be regarded as a MEC unit and propose the paradigm of a multi-UAV-based MEC network. Furthermore, we employ software-defined networking to improve the network's scalability and controllability and present the framework of the software-defined UAV network. Based on the proposed framework, we introduce the concept of a “smart function table” that can be applied to execute the network operation in the proposed framework. After evaluating our proposed framework, especially concerning the cooperative flying ability and control delay, several challenging issues and future research suggestions are presented to guide this direction.