权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

Communication Environment Estimation by Deep Learning for Improving Frequency Utilization Efficiency and its Application to Adaptive Modulation Coding

提高频率利用效率的深度学习通信环境估计及其在自适应调制编码中的应用

基本信息

批准号：
22K14253
负责人：
小島駿
金额：
$ 2.91万
依托单位：
The University of Tokyo
依托单位国家：
日本
项目类别：
Grant-in-Aid for Early-Career Scientists
财政年份：
2022
资助国家：
日本
起止时间：
2022-04-01 至 2025-03-31
项目状态：
未结题

项目摘要

無線通信網は現代および将来のインフラとしてますます重要性を増しており，周波数資源の逼迫は喫緊かつ重大な社会課題である．近年では効率的な周波数資源の活用方式として，スペクトル有効周波数分割多重（SEFDM）が注目されている．この方式では信号を圧縮し送受信を行うため，圧縮信号間で干渉が生じる問題がある．従来ではこの干渉除去に莫大な計算量を要するため，適応的な制御に必要とされる通信環境情報の高速・高精度な取得が困難であった．そこで本研究では，干渉が与える影響の理論解析を通し，深層学習への最適なSEFDM信号の入力形式の同定を行い，信号に内包される特徴を最大限に抽出することで，干渉除去を行わずに低計算量でかつ高速・高精度な通信環境情報推定手法の創出を目指す．さらにSEFDMとOFDMとの併用を可能にする適応変調符号化制御への応用を行い，高効率な通信方式の確立を図る.初年度である2022年度は，SEFDMにおける圧縮による干渉の影響の理論解析を行った．周波数方向の圧縮による信号間干渉が受信信号に及ぼす影響に対して，精緻な解析・調査を実施した．その後，干渉存在下の様々な通信路（レイリーフェージング、ライスフェージング）におけるSEFDMによる影響の解析を通し，通信環境情報として信号対雑音電力比（SNR）を推定する場合に畳み込みニューラルネットワークへの受信信号の入力形式として，スぺクトログラム波形が最も高い性能を示すことを明らかにした．これらの成果の一部をまとめたものを国内会議，IEEEジャーナルへ投稿している．

In the future, wireless communication networks are not concerned about the importance of wireless communication. In recent years, the use of weekly wavenumber resources has been very popular. In recent years, the usage of weekly wavenumber resources in wireless communication networks has been very popular. In recent years, the cycle wavenumber division multiple (SEFDM) has been widely used in wireless communication networks. In this way, signals are sent to trusted banks. In recent years, it is necessary to control the communication environment in the communication environment with high speed and high precision. In this study, the general understanding of the theoretical analysis of SEFDM signals is discussed, and the most advanced SEFDM signal input form is the same in this study. In the signal package, the maximum limit is extracted from the signal, and the target is obtained by removing the low calculation volume, the high-speed and high-precision communication environment presumption method, the high-speed and high-precision communication environment presumption method, the signal package, the maximum limit limit, the maximum limit of the signal package, the extraction of the maximum limit of the signal package, the extraction of the maximum limit of the signal package, the extraction of the maximum limit of the signal package, the extraction of the maximum limit of the signal package, the extraction of the maximum limit of the signal package, the extraction of the maximum limit of the signal package, the extraction of the maximum limit of the signal, and the elimination of the high-speed and high-precision communication environment presumption method. In the early part of the year 2022, SEFDM was responsible for the analysis of the theory and theory of the operation. In the direction of the wavenumber of the week, the trusted signals and the signals in the direction of the wavenumber were carefully analyzed. After that, There is a downlink communication channel (communication channel). The communication environment signal (SNR) is presumed to be in the form of a trusted signal input force. The highest performance of the waveform shows that there is a national conference on the results of the wave, and that the contribution is due to the IEEE.