期刊
IEEE WIRELESS COMMUNICATIONS LETTERS
卷 7, 期 5, 页码 748-751出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/LWC.2018.2818160
关键词
Massive MIMO; FDD; compressed sensing; deep learning; conventional neural network
资金
- Ministry of Science and Technology of Taiwan [MOST 106-2221-E-110-019]
- ITRI, Hsinchu, Taiwan
- National Science Foundation for Distinguished Young Scholars of China [61625106]
- National Natural Science Foundation of China [61531011]
In frequency division duplex mode, the downlink channel state information (CSI) should be sent to the base station through feedback links so that the potential gains of a massive multiple-input multiple-output can be exhibited. However, such a transmission is hindered by excessive feedback overhead. In this letter, we use deep learning technology to develop CsiNet, a novel CSI sensing and recovery mechanism that learns to effectively use channel structure from training samples. CsiNet learns a transformation from CSI to a near-optimal number of representations (or codewords) and an inverse transformation from codewords to CSI. We perform experiments to demonstrate that CsiNet can recover CSI with significantly improved reconstruction quality compared with existing compressive sensing (CS)-based methods. Even at excessively low compression regions where CS-based methods cannot work, CsiNet retains effective beamforming gain.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据