| 摘要: |
| 在大规模多输入多输出系统中，由于天线数量的增加导致信道状态信息反馈带宽开销增大。为了减少反馈开销，提出了一种基于深度学习的反馈网络。该网络将卷积注意力模块和快速迭代收缩阈值算法(Fast Iterative Shrinkage Thresholding Algorithm,FISTA)进行了结合。为了贴合实际应用，考虑到了噪声情况，分析了阈值敏感度。仿真结果表明，该网络在不同环境下其性能和鲁棒性可以得到进一步提高。 |
| 关键词: 大规模多输入多输出 信道状态信息；反馈网络；AI通信；深度学习 |
| DOI：10.20079/j.issn.1001-893x.220517005 |
|
| 基金项目:国家自然科学基金资助项目 （61601403） |
|
| Massive MIMO Channel State Information Feedback Network Based on AI Communication |
| LIU Weibo,YAN Biao,SHEN Lin,DING Yuzhou |
| (College of Information Engineering，Yangzhou University，Yangzhou 225009，China) |
| Abstract: |
| In a massive multiple-input multiple-output(MIMO) system,the increase of the number of antennas leads to the increase of the channel state information(CSI) feedback bandwidth overhead.To reduce the feedback overhead,a feedback network based on deep learning(DL) is proposed.The network combines a convolutional attention module with a Fast Iterative Shrinkage Thresholding Algorithm(FISTA).In order to fit the practical application,the noise situation is considered and the threshold sensitivity is analyzed.The simulation results show that the performance and robustness of the network can be further improved in different environments. |
| Key words: massive MIMO channel state information feedback network AI communication deep learning |
