Research Article
Task Offloading Method for Industrial Internet of Things (IIoT) Targeting Computational Resource Management
@INPROCEEDINGS{10.1007/978-3-031-55976-1_4, author={Wenhui Wang and Xuanzhe Wang and Zhenjiang Zhang and Zeng Jianjun}, title={Task Offloading Method for Industrial Internet of Things (IIoT) Targeting Computational Resource Management}, proceedings={Smart Grid and Internet of Things. 7th EAI International Conference, SGIoT 2023, TaiChung, Taiwan, November 18-19, 2023, Proceedings}, proceedings_a={SGIOT}, year={2024}, month={3}, keywords={edge computing offloading D2D offloading time-varying task offloading deep reinforcement learning}, doi={10.1007/978-3-031-55976-1_4} }- Wenhui Wang
Xuanzhe Wang
Zhenjiang Zhang
Zeng Jianjun
Year: 2024
Task Offloading Method for Industrial Internet of Things (IIoT) Targeting Computational Resource Management
SGIOT
Springer
DOI: 10.1007/978-3-031-55976-1_4
Abstract
In the context of industrial scenarios, devices exhibit specificity and task arrival rates vary over time. Considering real-world task queuing issues and incorporating edge computing offloading and D2D offloading techniques, this paper proposes TVTAO for computational resource management to meet latency requirements. First, three offloading decisions are introduced, then offloading policy constraints are proposed to restrict devices from selecting the same task for execution during task offloading. Simulation results demonstrate that the TVTAO algorithm can reasonably make task offloading decisions and allocate computational resources, effectively reducing the average processing latency of the overall system.
Copyright © 2023–2025 ICST
