Computation offloading considering fronthaul and backhaul in small-cell networks integrated with MEC

Abstract

Mobile edge computing (MEC) provides a promising way to bring enhanced computing capabilities in proximity to user equipments (UEs). Various previous works have been done and they usually focus on how to segment the computation intensive task and how to offload the task to edge of the network with plenty of computation capacity. However, few of them consider integrating MEC with small cell networks (SCNs) which is regarded as the key technology in future 5G networks. In this paper, we study the computation offloading scheme in a multiuser and multi-SC scenario with MEC. We firstly formulate an energy efficient computation offloading problem, which aims to optimize the offloading energy for the tasks while constraining the latency lower than a threshold. In view of the characteristics of SCNs, the fronthaul and backhual links are also taken into account when calculating the task energy and latency. Secondly, we apply artificial fish swarm algorithm (AFSA) to solve the proposed problem in a high efficiency. To guarantee the global optimization, strong robustness and fast convergence of AFSA, fish swarm's three important behaviors such as prey behavior, swarm behavior and following behavior are utilized in our algorithm design. Finally, the simulation has been done and show the superiority of our scheme.