| Project/Area Number |
20K11764
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Review Section |
Basic Section 60060:Information network-related
|
|---|
| Research Institution | Ibaraki University |
|---|
Principal Investigator |
Wang Xiaoyan 茨城大学, 理工学研究科(工学野), 准教授 (10725667)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
梅比良 正弘 茨城大学, 理工学研究科(工学野), 特命研究員 (00436239)
|
|---|
| Project Period (FY) |
2020-04-01 – 2024-03-31
|
| Project Status |
Completed (Fiscal Year 2023)
|
| Budget Amount *help |
¥4,290,000 (Direct Cost: ¥3,300,000、Indirect Cost: ¥990,000)
Fiscal Year 2022: ¥1,690,000 (Direct Cost: ¥1,300,000、Indirect Cost: ¥390,000)
Fiscal Year 2021: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Fiscal Year 2020: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
|
|---|
| Keywords | 周波数活用 / wireless access / federated learning / reinforcement learning / In-network learning / radio resource |
|---|
| Outline of Research at the Start |
Radio resource management (RRM) is the key enabler for full-featured 5G networks. In this research, we propose a layered in-networking learning RRM approach, and evaluate its performance via both simulations and experiments on testbeds.
|
| Outline of Final Research Achievements |
The 5G system adopts a small cell configuration, which significantly improves frequency efficiency but also leads to inter-cell interference issues. To address this problem, beamforming, a technique that focuses radio waves in specific directions, is widely considered. Achieving globally optimal beamforming control in multi-cell systems is highly challenging. This study aims to maximize the energy efficiency of the entire network in both static and dynamic downlink scenarios, proposing a low-resolution analog beamforming and transmission power control method using deep reinforcement learning.
|
| Academic Significance and Societal Importance of the Research Achievements |
本研究では、無線リソースの最適化の実現に向け、深層強化学習を用いたアナログビームフォーマと送信電力制御について検討する。提案手法は従来手法と比べ、静的および動的なシナリオにおける、大幅なエネルギー効率改善が期待でき、周波数資源の有効利用に向けて重要な役割を持っていると考えられる。
|
