An MDP-Based Wireless Energy Harvesting Decision Strategy for Mobile Device in Edge Computing.

Deze Zeng,Shengli Pan,Zhuoling Chen,Lin Gu

DOI: https://doi.org/10.1109/mnet.001.1800534

IF: 10.294

2019-01-01

IEEE Network

Abstract:Edge computing has been regarded as an ideal platform to provision computation and storage resources at the network edge thanks to its vast distribution and user proximity. Owing to such features, it is also promising to provision energy resources at the network edge via wireless energy transferring. With the consideration of the mobility of wireless devices and the diversity of charging attenuation in the edge computing environment, the device shall make the right charging decision at runtime during the movement to maximize the long-term average utility. In this article, we discover that this problem can be described as a Markov Decision Process (MDP) and accordingly design an MDP based wireless energy harvesting strategy. Based on extensive simulation-based experimental results, we observe that the proposed strategy outperforms several existing schemes and significantly improves the system utility.