Full-Duplex Aided User Virtualization for Mobile Edge Computing in 5G Networks

Abstract

Driven by the increasing demand of intensive computing services and the resource limitation of mobile devices at the edge of mobile networks, mobile edge computing (MEC) is concerned with being an emerging paradigm towards the 5G communications. A main issue of MEC is the coordination of communication, computation, and storage. In this paper, we propose a novel MEC framework with a user virtualization scheme in the software-defined network virtualization cellular network, in which radio resources are virtualized along with computation and storage resources to cooperatively finish MEC services. Besides, we introduce the user virtualization assisted by the full-duplex communication which can extend the radio resource of wireless networks and provide the potential to increase system performance. Moreover, enabled by user virtualization, users can offload the edge computation tasks directly to the MEC server implemented at infrastructure providers or via the virtualized mobile device attributed to different mobile virtual network operators (MVNOs). Under this MEC framework, we formulate the virtual resource allocation as a joint optimization problem. A distributed resource allocation algorithm based on an alternating direction method of multipliers is proposed which can reduce computational complexity and signaling overhead. We evaluate the proposed algorithm through extensive simulations, and the results show that the total utility of MVNOs can be improved significantly which benefited from user virtualization.