Dynamic Time Division Duplexing for Green Internet of Things
- Authors
- Tuong, Van Dat; Dao, Nhu-Ngoc; Noh, Wonjong.; Cho, Sungrae
- Issue Date
- Jan-2022
- Publisher
- IEEE Computer Society
- Keywords
- Deep reinforcement learning; dynamic time division duplex; green internet of things
- Citation
- International Conference on Information Networking, v.2022, pp 356 - 358
- Pages
- 3
- Journal Title
- International Conference on Information Networking
- Volume
- 2022
- Start Page
- 356
- End Page
- 358
- URI
- https://scholarworks.bwise.kr/cau/handle/2019.sw.cau/55501
- DOI
- 10.1109/ICOIN53446.2022.9687184
- ISSN
- 1976-7684
- Abstract
- Motivated by the promising benefit of time division duplexing (TDD) in interference mitigation, this paper investigates the dynamic TDD problem for green Internet of Things (IoT), where IoT devices and access points (APs) are distributed in an indoor network coverage area and the network traffic is time-varying. The problem is formulated in which each AP optimizes its radio frame configuration (RFC) to maximize a network utility in terms of the throughput in the long term. Considering the time-varying network traffic of uplink and downlink transmissions, we approach with deep reinforcement learning (DRL), an emerging tool in wireless network control, to train the optimal RFC policy for each AP. Simulation results reveal that our approach improves the data rate by approximately 30%, compared to those of existing schemes. © 2022 IEEE.
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - College of Software > School of Computer Science and Engineering > 1. Journal Articles
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.