中圖分類號：TN929.5
文獻標志碼：A
DOI: 10.16157/j.issn.0258-7998.233992
中文引用格式： 馬穎，王珂，吳戈男，等. 基于強化學習的自適應編碼調制策略[J]. 電子技術應用，2023，49(5)：35-40.
英文引用格式： Ma Ying，Wang Ke，Wu Genan，et al. Adaptive coding modulation strategy based on reinforcement learning[J]. Application of Electronic Technique，2023，49(5)：35-40.

Adaptive coding modulation strategy based on reinforcement learning

Abstract： NTN (non-terrestrial network) is an important application scenario for satellite communications and low-altitude communications, marking the transition of 5G technology applications from land communications to space communications. It is foreseeable that satellite networks will be an important component of future 6G communications networks. In order to meet the quality requirements of satellite communication and maximize the system capacity, it is necessary to apply adaptive coding and modulation technology to dynamically adjust the modulation order and coding rate according to the channel state information in the changing communication environment. AI has clear potential to solve problems arising from rapidly changing channel conditions in satellite high-dynamic scenarios. This paper adopts the low-orbit satellite adaptive coding and modulation strategy based on reinforcement learning to solve the problem of the mismatch between the threshold table and the actual channel caused by the change of the satellite communication environment, which is improved by above 20% compared with the traditional ARIMA (autoregressive integrated moving average) algorithm.

Key words : reinforcement learning；6G；adaptive coding modulation；NTN