QoS-Aware Mobile Edge Computing System: Multi-Server Multi-User Scenario

Abstract

Mobile/Multi-access Edge Computing (MEC) system is a kind of fog computing system to provide computation resources at the edge of Radio Access Network (RAN). In this paper, we discuss offloading decision and resource allocation of both radio resources and computation resources which are both the crucial issues in MEC system. Moreover, we study a multiserver system which is more consistent with practical MEC system in the future rather than single-server system, and load distribution is introduced into our system to leverage the computation resources of MEC servers effectively. The objective of our work is to promote QoS, and thus we formulate a cost minimization problem by modeling QoS of the end users with our self-defined cost function. A heuristic algorithm is developed and compared with other schemes. Numerical results verify that our proposed algorithm significantly enhances QoS and gains a large performance gap over other schemes.