EEDOS: an energy-efficient and delay-aware offloading scheme based on device to device collaboration in mobile edge computing
Document Type
Article
Publication Date
1-1-2020
Abstract
Device to device (D2D) communication and mobile edge computing (MEC) are two promising technologies in fifth generation (5G) cellular mobile communication. Besides MEC, a new task offloading technique attracts the attention as D2D collaboration. However, there is lack of integrated D2D and MEC framework to address the energy and delay costs in a joint approach. This work, proposes an energy efficient and delay-aware offloading scheme (EEDOS) based on D2D collaboration in MEC. In EEDOS, mobile devices can offload their task to the MEC or an idle mobile device in their proximity. The task execution and offloading to the MEC or an idle nearby device is formulated, and the optimization problem is defined. The whole process of allocating proper offloading destination is designed in the edge server. EEDOS, classifies offloading requests according to the deadline and energy constraint of requesting device. Then, it finds the proper offloading destination by utilising the maximum matching with minimum cost graph algorithm. Through simulation, we show that EEDOS achieves 95 percent of energy efficiency in comparison of no-offloading task execution and outperforms existing studies in term of energy efficiency with an improved delay in task execution. Moreover, EEDOS is capable of performing more successful task offloading and requires less edge server resources. © 2019, Springer Science+Business Media, LLC, part of Springer Nature.
Keywords
Device to device communication, Energy efficiency, Mobile edge computing, Task offloading
Divisions
fsktm
Funders
Project No.: BK067-2015 from University of Malaya,Fundamental Research Grant Scheme (FRGS), Project: FP007-2016 from Ministry of Higher Education, Malaysia
Publication Title
Telecommunication Systems
Volume
73
Issue
2
Publisher
Kluwer (now part of Springer)