Research Article
Energy-Efficient Cooperative Offloading for Multi-AP MEC in IoT Networks
@INPROCEEDINGS{10.1007/978-3-030-92638-0_1, author={Zhihui Cao and Haifeng Sun and Ning Zhang and Xiang Lv}, title={Energy-Efficient Cooperative Offloading for Multi-AP MEC in IoT Networks}, proceedings={Collaborative Computing: Networking, Applications and Worksharing. 17th EAI International Conference, CollaborateCom 2021, Virtual Event, October 16-18, 2021, Proceedings, Part II}, proceedings_a={COLLABORATECOM PART 2}, year={2022}, month={1}, keywords={Mobile Edge Computing Cooperative offloading Multiple-AP Energy efficient}, doi={10.1007/978-3-030-92638-0_1} }- Zhihui Cao
Haifeng Sun
Ning Zhang
Xiang Lv
Year: 2022
Energy-Efficient Cooperative Offloading for Multi-AP MEC in IoT Networks
COLLABORATECOM PART 2
Springer
DOI: 10.1007/978-3-030-92638-0_1
Abstract
Mobile Edge Computing (MEC) technology is used for offloading local application tasks on Mobile Devices (MDs) to the edge server to decrease task processing time and reduce energy consumption in Internet of Things (IoTs) networks. In this paper, we investigate a scenario consisting of a local MD adjacent with a group of other MDs, one of which can act as the offloading cooperator. All the MDs are surrounded by multiple Access Points (APs), and each AP is deployed an MEC server providing abundant computation resources. Based on this scenario, we propose a cooperative energy-efficient offloading scheme under delay constraint. The local MD can offload part of the application task to a cooperative relay MD or the MEC server, and the relay MD can also offload part of the segment to an AP. By solving the proposed energy-efficient cooperative offloading problem under the constraint of computing delay, the most energy-efficient cooperative offloading MD and the AP as well as the task segmentation to minimize the energy consumption are determined. Numerical analysis shows that our proposed scheme significantly outperforms the benchmark schemes in the aspect of energy consumption and the supported task length in maximum.
