Energy-Efficient Collaborative Task Offloading in D2D-assisted Mobile Edge Computing Networks

Abstract

With emerging requirement of local low-latency services, Mobile Edge Computing (MEC) is a promising solution to tackle the challenge between urgent demands for computation capability and limited battery energy of mobile devices. Moreover, the sharing property of applications costs waste as for the processing of redundant data, which derives an imperative need for the collaboration among users. In this paper, by leveraging these features, we design a D2D-assisted MEC system for energy efficiency of devices with the consideration of task delay. For sake of energy minimization, a strategy that jointly optimizes resource allocation and tasks offloading assignment is proposed. Further, a low-complexity algorithm is developed to decompose the original problem into two subproblems and get the sub-optimal solution efficiently. Simulation results present the efficient and effective performance of the proposed algorithm with different application parameters. Particularly, it is shown that our proposed algorithm gets 48.41%~90.58% and 37.33%~96.63% improvement of energy consumption than those of non-collaboratively offloading scheme and randomly offloading scheme, respectively.