.png){kind=icon}
返回
Deep Deterministic Policy Gradient Based Computation Offloading in Wireless-Powered MEC Networks
.png){kind=icon}
PRE
AI
DOI：10.1109/GCWkshps50303.2020.9367589.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
With the upsurge of Internet of Things (IoT), mobile edge computing (MEC) and wireless power transfer (WPT) are becoming promising methods to relieve IoT devices from the limited computation capacity and battery capacity. In this paper, we study the joint optimization of the local computing central processing unit (CPU) frequency and the offloading ratio in a dynamic WPT-MEC network. In the proposed network, all the IoT devices follow a harvest-then-compute protocol so as to be charged by the MEC server and then compute the tasks. Our goal is to minimize the long-term averaged maximal computing delay with stochastic task arrivals and time-varying wireless channels. It is challenging to find the optimal solution due to the uncertain network information and the non-convex objective function. To tackle this problem, we propose a deep reinforcement learning-based deep deterministic policy gradient (DDPG) framework, which can learn a centralized policy to coordinate the local computing CPU frequencies and offloading ratios of IoT devices. Simulation results show that the proposed DDPG algorithm can achieve up to 73.2% and 16.8% gain in terms of the averaged maximal computing delay compared to the local-computing-only method and the deep Q-network algorithm, respectively.
Keyword:
Mobile edge computing
Deep deterministic policy gradient
Wireless power transfer
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
I
IF:
0
论文数:
19
被引数:
0
机构
引用论文
.png){kind=icon}
分享收藏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
OAAI分享收藏A Machine Learning Approach for Task and Resource Allocation in Mobile-Edge Computing-Based Networks基于移动边缘计算的网络中任务和资源分配的机器学习方法OAAI
OAAI分享收藏
查看更多内容