Abstract
The ultra-dense network (UDN) based on mobile edge computing (MEC) is an important technology, which can achieve the low-latency of 5G communications and enhance the quality of user experience. However, how to improve the task offloading efficiency is a hot topic of UDN under the constraint on the limited wireless resources. In this article, we propose a heuristic task offloading algorithm HTOA to optimize the delay and energy consumption of offloading tasks in UDN. Firstly, a convex programming model for MEC resource allocation is established, which aims to obtain the optimal allocation set of resources for offloading tasks, and optimize the execution delay of offloading tasks. Followed by, the problem of joint channel allocation and user upload power control is solved by the greedy strategy and golden section method, which aims to optimization the delay and energy consumption of task upload data. Compared with the random task offloading algorithm, numerical simulations show that the algorithm HTOA can effectively reduce the delay and energy consumption of task offloading, and perform better as the number of users increases.
Highlights
With the development of science and technology, the user equipment (UE) has connected to the wireless network are increasing explosively
We set the number of mobile edge computing (MEC) is 10, i.e., the ultra-dense network (UDN) is composed of 10 base stations, and the number of sub-channels is 10
In this paper, we propose a heuristic task offloading algorithm HTOA for investigated the joint radio resource management and task offloading in the UDN architecture
Summary
With the development of science and technology, the user equipment (UE) has connected to the wireless network are increasing explosively. The intensive deployment of SBS improving QoE of users for task execution and meeting the low-delay needs of future communication technology It brings many problems and challenges, mainly including: (1) For MEC servers with limited computing resource, how to allocate computing resource reasonably to tasks with the character of pluralistic is a challenging problem. The UDN architecture has the characteristics of heterogeneous network complexity, the finiteness of the MEC server resources, the diversity of UEs task requirements, and the scarcity of spectrum resources This characteristic causes that it has become a research hotspot, management of radio resources and task offloading strategy efficiently and effectively and with the delay and energy consumption as the research goal.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have