Multi-User Computation Offloading with D2D for Mobile Edge Computing

Abstract

With the emergence of mobile edge computing (MEC), mobile users are able to process various tasks by offloading large-computation-demanding tasks to MEC server located at the edge of the network. As computation offloading requires communication between mobile users and the MEC server, an efficient computation offloading scheme which decreases both task executive delay and transmission energy consumption of mobile users plays a key role in MEC. Motivated by this, we study the computation offloading scheme in a novel MEC system where mobile users can offload tasks to the MEC server or a distributed computing node (DCN). As mobile users' offloading scheme affects the delay and energy consumption each other, we show that the offloading decision-making problem of users can be formulated as a sequential game. In particular, we demonstrate that the Nash equilibrium of the game exists which manifests that the system can converge to a stable status. A multi-user and multi-destination computation offloading scheme is also proposed to achieve the Nash equilibrium. Simulation results show that the proposed computation offloading scheme can significantly decrease the task execution delay as well as the energy consumption of mobile users.