Multi-objective Optimization for Computation Offloading in Mobile-edge Computing

Mobile-edge cloud computing is a new cloud platform to provide pervasive and agile computation augmenting services for mobile devices (MDs) at anytime and anywhere by endowing ubiquitous radio access networks with computing capabilities. Although offloading computations to the cloud can reduce energy consumption at the MDs, it may also incur a larger execution delay. Usually the MDs have to pay cloud resource they used. In this paper, we utilize queuing theory to bring a thorough study on the energy consumption, execution delay and price cost of offloading process in a mobile-edge cloud system. Specifically, both wireless transmission and computing capabilities are explicitly and jointly considered when modelling the energy consumption and delay performance. Based on the theoretical analysis, the multi-objective optimization problem is formulated with the joint objectives to minimize the energy consumption, execution delay and price cost by finding the optimal offloading probability and optimal transmission power for each MD. The scalarization scheme and interior point method are applied to address the formulated problem. Through extensive simulations, the effectiveness of the proposed scheme can be demonstrated.