| 講演抄録/キーワード |
| 講演名 |
2023-03-02 10:15
Dynamic Antenna Control in Multi-Cell HAPS System by using Deep Reinforcement Learning Evolution Algorithm ○Siyuan Yang・Mondher Bouazizi・Tomoaki Ohtsuki(Keio Univ.) RCS2022-272 |
| 抄録 |
(和) |
(まだ登録されていません) |
| (英) |
To reduce the number of low-throughput users caused by the HAPS movement, we propose a novel Deep Reinforcement Learning Evolution Algorithm (DRLEA).
The proposed method can dynamically adjust the antenna parameters based on the throughput of the users in the coverage area to reduce the number of low-throughput users by improving the users' throughput.
Besides, compared with the traditional Deep Q Network (DQN), the proposed method combined the Evolution Algorithm (EA) with Deep Reinforcement Learning (DRL) to avoid the sub-optimal solutions in each state.
We do the simulations under 4 different real user distribution scenarios and compare the proposed method with the conventional EA and RL methods.
The simulation results show that the proposed method effectively reduces the number of low throughput users after the HAPS moves. |
| キーワード |
(和) |
/ / / / / / / |
| (英) |
antenna control / DRL / HAPS / / / / / |
| 文献情報 |
信学技報, vol. 122, no. 399, RCS2022-272, pp. 143-148, 2023年3月. |
| 資料番号 |
RCS2022-272 |
| 発行日 |
2023-02-22 (RCS) |
| ISSN |
Online edition: ISSN 2432-6380 |
著作権に ついて |
技術研究報告に掲載された論文の著作権は電子情報通信学会に帰属します.(許諾番号:10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) |
| PDFダウンロード |
RCS2022-272 |