Multi-dimensional resource management with deep deterministic policy gradient for digital twin-enabled Industrial Internet of Things in 6 generation
In the era of sixth generation mobile networks (6G), industrial big data is rapidly generated due to the increasing data-driven applications in the Industrial Internet of Things (IIoT). Effectively processing such data, for example, knowledge learning, on resource-limited IIoT devices becomes a chal...
| Main Authors: | , , , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2024
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/180258 |
| _version_ | 1826118272753336320 |
|---|---|
| author | Hu, Yue Cao, Ning Lu, Hao Jiang, Yunzhe Liu, Yinqiu He, Xiaoming |
| author2 | School of Computer Science and Engineering |
| author_facet | School of Computer Science and Engineering Hu, Yue Cao, Ning Lu, Hao Jiang, Yunzhe Liu, Yinqiu He, Xiaoming |
| author_sort | Hu, Yue |
| collection | NTU |
| description | In the era of sixth generation mobile networks (6G), industrial big data is rapidly generated due to the increasing data-driven applications in the Industrial Internet of Things (IIoT). Effectively processing such data, for example, knowledge learning, on resource-limited IIoT devices becomes a challenge. To this end, we introduce a cloud-edge-end collaboration architecture, in which computing, communication, and storage resources are flexibly coordinated to alleviate the issue of resource constraints. To achieve better performance in hyper-connected experience, real-time communication, and sustainable computing, we construct a novel architecture combining digital twin (DT)-IIoT with edge networks. In addition, considering the energy consumption and delay issues in distributed learning, we propose a deep reinforcement learning-based method called deep deterministic policy gradient with double actors and double critics (D4PG) to manage the multi-dimensional resources, that is, CPU cycles, DT models, and communication bandwidths, enhancing the exploration ability and improving the inaccurate value estimation of agents in continuous action spaces. In addition, we introduce a synchronization threshold for distributed learning framework to avoid the synchronization latency caused by stragglers. Extensive experimental results prove that the proposed architecture can efficiently conduct knowledge learning, and the intelligent scheme can also improve system efficiency by managing multi-dimensional resources. |
| first_indexed | 2024-10-01T04:41:00Z |
| format | Journal Article |
| id | ntu-10356/180258 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T04:41:00Z |
| publishDate | 2024 |
| record_format | dspace |
| spelling | ntu-10356/1802582024-09-25T06:35:05Z Multi-dimensional resource management with deep deterministic policy gradient for digital twin-enabled Industrial Internet of Things in 6 generation Hu, Yue Cao, Ning Lu, Hao Jiang, Yunzhe Liu, Yinqiu He, Xiaoming School of Computer Science and Engineering Computer and Information Science 6G Industrial inter-net of things In the era of sixth generation mobile networks (6G), industrial big data is rapidly generated due to the increasing data-driven applications in the Industrial Internet of Things (IIoT). Effectively processing such data, for example, knowledge learning, on resource-limited IIoT devices becomes a challenge. To this end, we introduce a cloud-edge-end collaboration architecture, in which computing, communication, and storage resources are flexibly coordinated to alleviate the issue of resource constraints. To achieve better performance in hyper-connected experience, real-time communication, and sustainable computing, we construct a novel architecture combining digital twin (DT)-IIoT with edge networks. In addition, considering the energy consumption and delay issues in distributed learning, we propose a deep reinforcement learning-based method called deep deterministic policy gradient with double actors and double critics (D4PG) to manage the multi-dimensional resources, that is, CPU cycles, DT models, and communication bandwidths, enhancing the exploration ability and improving the inaccurate value estimation of agents in continuous action spaces. In addition, we introduce a synchronization threshold for distributed learning framework to avoid the synchronization latency caused by stragglers. Extensive experimental results prove that the proposed architecture can efficiently conduct knowledge learning, and the intelligent scheme can also improve system efficiency by managing multi-dimensional resources. This work is supported by National Natural Science Foundation of China under Grant No. 62271190. 2024-09-25T06:35:05Z 2024-09-25T06:35:05Z 2024 Journal Article Hu, Y., Cao, N., Lu, H., Jiang, Y., Liu, Y. & He, X. (2024). Multi-dimensional resource management with deep deterministic policy gradient for digital twin-enabled Industrial Internet of Things in 6 generation. Transactions On Emerging Telecommunications Technologies, 35(4), e4962-. https://dx.doi.org/10.1002/ett.4962 2161-3915 https://hdl.handle.net/10356/180258 10.1002/ett.4962 2-s2.0-85188548718 4 35 e4962 en Transactions on Emerging Telecommunications Technologies © 2024 John Wiley & Sons, Ltd. All rights reserved. |
| spellingShingle | Computer and Information Science 6G Industrial inter-net of things Hu, Yue Cao, Ning Lu, Hao Jiang, Yunzhe Liu, Yinqiu He, Xiaoming Multi-dimensional resource management with deep deterministic policy gradient for digital twin-enabled Industrial Internet of Things in 6 generation |
| title | Multi-dimensional resource management with deep deterministic policy gradient for digital twin-enabled Industrial Internet of Things in 6 generation |
| title_full | Multi-dimensional resource management with deep deterministic policy gradient for digital twin-enabled Industrial Internet of Things in 6 generation |
| title_fullStr | Multi-dimensional resource management with deep deterministic policy gradient for digital twin-enabled Industrial Internet of Things in 6 generation |
| title_full_unstemmed | Multi-dimensional resource management with deep deterministic policy gradient for digital twin-enabled Industrial Internet of Things in 6 generation |
| title_short | Multi-dimensional resource management with deep deterministic policy gradient for digital twin-enabled Industrial Internet of Things in 6 generation |
| title_sort | multi dimensional resource management with deep deterministic policy gradient for digital twin enabled industrial internet of things in 6 generation |
| topic | Computer and Information Science 6G Industrial inter-net of things |
| url | https://hdl.handle.net/10356/180258 |
| work_keys_str_mv | AT huyue multidimensionalresourcemanagementwithdeepdeterministicpolicygradientfordigitaltwinenabledindustrialinternetofthingsin6generation AT caoning multidimensionalresourcemanagementwithdeepdeterministicpolicygradientfordigitaltwinenabledindustrialinternetofthingsin6generation AT luhao multidimensionalresourcemanagementwithdeepdeterministicpolicygradientfordigitaltwinenabledindustrialinternetofthingsin6generation AT jiangyunzhe multidimensionalresourcemanagementwithdeepdeterministicpolicygradientfordigitaltwinenabledindustrialinternetofthingsin6generation AT liuyinqiu multidimensionalresourcemanagementwithdeepdeterministicpolicygradientfordigitaltwinenabledindustrialinternetofthingsin6generation AT hexiaoming multidimensionalresourcemanagementwithdeepdeterministicpolicygradientfordigitaltwinenabledindustrialinternetofthingsin6generation |