.png){kind=icon}
返回
Energy-efficient Online Data Sensing and Processing Optimization in Wireless Powered Edge Computing Systems
.png){kind=icon}
PRE
AI
DOI：10.1109/ICC45855.2022.9838472.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
This paper considers a wireless powered mobile edge computing (MEC) system consisting of multiple wireless devices (WDs) and one hybrid access point (HAP) broadcasting radio frequency (RF) energy to the WDs. Relying on the harvested energy, the WDs senses data from the monitored environment and execute the task data locally or offload the task to the HAP for edge processing. Given an average power constraint at the HAP, we aim to design an energy-efficient online algorithm under random fading channels to maximize the long-term average data sensing rate of WDs while meeting the system data queue stability. We formulate the target problem as a multi-stage stochastic optimization, where the major difficulty lies in the uncertainty of future channel state and the tight couplings among control decisions over different time slots. To solve this problem, we propose a Lyapunov optimization-based online algorithm named LEESE. Specifically, LEESE equivalently transforms the multi-stage stochastic optimization into per-slot deterministic problems. For each per-slot problem, we derive the optimal closed-form solution. We show that the optimal control on WPT and data processing follows an interesting threshold-based manner decided by the battery state and data queue backlog. Numerical simulations show that the proposed LEESE algorithm can achieve more than 21.9% performance improvement over the considered benchmark methods.
Keyword:
MAXIMIZATION
期刊
I
IF:
0
论文数:
114
被引数:
0
机构
引用论文
Deep Reinforcement Learning for Online Computation Offloading in Wireless Powered Mobile-Edge Computing Networks用于无线供电的移动边缘计算网络中在线计算卸载的深度强化学习
OAAI
.png){kind=icon}
OAAI分享收藏Joint Offloading and Computing Optimization in Wireless Powered Mobile-Edge Computing Systems无线移动边缘计算系统中的联合卸载和计算优化OAAI
OAAI分享收藏
查看更多内容