Game Theoretical Multi-user Computation Offloading for Mobile-Edge Cloud Computing

摘要

The emergence of mobile-edge clouds eases the tension of unbalanced relation between unlimited required resources of short video applications and limited computation capability of mobile devices. In this paper, we first build a decentralized noncooperative game model for multiple users who are provided with various computation choices, which are local computing, cloud 1 computing and cloud 2 computing. The game aims to achieve the optimal profits for each user in a distributive way. The profits of users are formulated by considering the effect of media transmission time, energy cost, and computation cost. Then, we analyze the proposed game theoretical computation offloading decision making problem and show that the game achieves a Nash equilibrium. Finally, the simulation results evaluate that the proposed game theoretical computation offloading and cloud selection method outperforms the traditional local or one-cloud computation methods.