Stochastic Joint Bandwidth and Computational Allocation for Multi-Users and Multi-Edge-Servers in 5G D-RANs

Mobile-edge computing (MEC) is now regarded as a promising paradigm for cellular networks, which aims at decreasing end-to-end delay and improving the quality of services. This article innovatively envisions a MEC offloading model for multi-users and collaborative MEC servers for 5G DRANs, where the tasks can be exchanged feasibly between MEC base stations. Firstly, we formulate the joint computational and radio resource management problem to minimize the power consumption of collaborative MEC servers and terminals, which takes into account the constraint of the task buffer stability. And then, an online Lyapunov optimization method is proposed to solve it to get a trade-off between the average weighted power consumption and the execution delay for all tasks distributed in MEC-enabled 5G D-RANs. The numerical results demonstrate that the proposed method has lower power consumption compared with the non-collaborative case. The reason is that the tasks are more reasonably allocated to each collaborative MEC servers with the acceptable execution delay. The control parameter V and collaboration impact factors in the online algorithm are also proved to be efficient and practical to balance the power consumption and execution delay for MEC-enabled 5G D-RANs.