Research Area:  Fog Computing

Abstract:

With the rapid development of smart devices and compute-intensive applications, fog computing has emerged as a promising solution to accommodate the ever-increasing computational demands. Particularly, in the peak time, the computational tasks can be offloaded from the overloaded base stations to fog servers by leveraging the under-utilized computational resources at the demand side. However, there are two major obstacles hindering the wide deployment of fog computing in Internet of things, which are the lack of an effective incentive mechanism and a task offloading algorithm. In this paper, we develop a two-stage resource sharing and task offloading approach by integrating contract theory with computational intelligence. In the first stage, we propose an efficient incentive mechanism to encourage servers to share their residual computational resources by employing the contract theory. In the second stage, a distributed task offloading algorithm is proposed by leveraging the online learning capability of multi-armed bandit. Specifically, we propose a distance-aware, occurrence-aware, and task-property-aware volatile upper confidence bound algorithm to minimize the long-term delay of task offloading. Finally, extensive simulations are carried out to validate the performance of the proposed algorithm.

Keywords:  

Author(s) Name:   Zhenyu Zhou; Haijun Liao; Bo Gu; Shahid Mumtaz; Jonathan Rodriguez

Publisher name:  IEEE

DOI:   10.1109/TETCI.2019.2902869

Volume Information:  Volume: 4, Issue: 3, June 2020, Page(s): 227 - 240

Paper Link:   https://ieeexplore.ieee.org/abstract/document/8887222