Toward A Task Offloading Framework Based on Cyber Digital Twins in Mobile Edge Computing

In the metaverse, the concept of the digital twin has been expanded from modeling industrial manufacturing to the counterpart of physical objects in cyberspace. The cyber digital twin is updated using real-time data and reasoning to improve decision-making, which imposes a high computational demand...

Full description

Saved in:
Bibliographic Details
Published in:IEEE wireless communications 2023-06, Vol.30 (3), p.157-162
Main Authors: Tan, Bin, Ai, Lihua, Wang, Min, Wang, Jiaxi
Format: Article
Language:English
Subjects:
Algorithms
Artificial intelligence
Changing environments
Computation offloading
Decision making
Digital twins
Edge computing
Electronic devices
Machine learning
Mobile computing
Mobile handsets
Multi-access edge computing
Optimization
Performance evaluation
Quality of service
Real-time systems
Resource management
Task complexity
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In the metaverse, the concept of the digital twin has been expanded from modeling industrial manufacturing to the counterpart of physical objects in cyberspace. The cyber digital twin is updated using real-time data and reasoning to improve decision-making, which imposes a high computational demand on the mobile edge. Mobile edge computing (MEC) provides computing resources for mobile devices to handle complex tasks, addressing the shortcomings of mobile devices in performance. Cyber digital twins with artificial intelligence (AI) capability have great advantages in addressing complex and changing environments. In this article, we propose a cyber digital twin-based mobile edge computing framework, which integrates artificial intelligence into mobile edge networks to enable intelligent resource management. We address the edge computation offloading task through formulating an optimization problem that minimizes the latency of a mobile user via MEC server selection and power allocation. Our solution employs a reinforcement learning-based algorithm, which we demonstrate to be effective. The experimental results show that the cyber digital twin based framework with artificial intelligence capability can further reduce task processing latency and improve the quality of service provided to users.
ISSN:1536-1284
1558-0687
DOI:10.1109/MWC.020.2200533