.png){kind=icon}
.png){kind=icon}
Return
Deep Q-Learning for Minimum Task Drop in SWIPT-Enabled Mobile-Edge Computing
.png){kind=icon}
PRE
AI
DOI：10.1109/LWC.2024.3349529.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In this letter, we study mobile-edge computing (MEC) systems empowered with simultaneous wireless information and power transfer (SWIPT). This holds promise to meet the growing energy and computation requirements of the Internet of Things devices. To cope with the network dynamics and limited resources causing task drops, we propose a deep Q-learning computation offloading algorithm for SWIPT-enabled MEC. We formulate an optimization problem to calculate the reward for offloading decisions and penalize the task drop in order to minimize it in the long term. Simulation results demonstrate a three-fold reduction in task drop compared to an existing work.
Keywords:
Task analysis
Wireless communication
Heuristic algorithms
Indexes
Q-learning
Batteries
Optimization
Deep Q-learning
mobile-edge computing
simultaneous wireless information and power transfer
Journal
IF:
11.5
Papers:
2.7K
Citations:
1.3W