Market-based dynamic resource allocation in Mobile Edge Computing systems with multi-server and multi-user

Abstract Mobile Edge Computing (MEC) is critical to the development of the Internet of things (IoTs) and 5G networks. However, the computation and communication resources of edge servers are limited, so it is challenging to perform resource allocation especially when the competition among edge servers is also taken into consideration. In this paper, we propose a trading model to investigate both the computation and communication resources allocation in MEC systems with multi-server and multi-user. We model the dynamic behavior of mobile users (MUs) using an evolutionary game, and then we build the deterministic and stochastic models to study the evolution of MUs where the evolutionary equilibrium is considered as the solution. We propose an evolution algorithm to obtain the evolutionary equilibrium. Furthermore, we analyze the competition among edge cloud servers (ECSs) by a noncooperative game, and propose an iteration algorithm to obtain Nash equilibrium where the ECSs can adjust the amount of resources provided to MUs and the corresponding price charged in order to attract more MUs. The existences of evolutionary equilibrium and Nash equilibrium are validated in performance evaluation.