| 摘要: |
| 通过压缩信道状态信息(Channel Status Information，CSI)传输码字降低大规模多输入多输出(Multiple-Input Multiple-Output,MIMO)系统的CSI反馈开销，可以有效减少计算资源的使用和信息传输时间的消耗。针对如何使用轻量化模型准确估计低压缩比条件下CSI反馈的问题，通过设计的轻量化迭代交叉网络(Iterative Cross Network,ICNet)模型，在用户端使用设计的迭代压缩模块压缩CSI反馈，基站端使用设计的迭代重建模块估计CSI反馈，以较高的准确率和较低的时间消耗估计了低压缩比条件下的CSI反馈。在COST2100模型生成的数据样本下评估了ICNet在低压缩比条件下的鲁棒性，实验表明，在较小的1/64压缩比条件下，ICNet的归一化均方误差比次优值降低了8.48%，ICNet的参数量降低了35%左右。 |
| 关键词: 大规模MIMO CSI反馈 交叉卷积 低压缩比 轻量化估计 |
| DOI：10.20079/j.issn.1001-893x.240419003 |
|
| 基金项目:甘肃省科技计划（23YFGA0062）；甘肃省创新基金（2022A-215） |
|
| Lightweight Estimation of Low Compression Ratio CSI Feedback for Massive MIMO Systems |
| XUE Jianbin,GAO Jiamin |
| (School of Computer and Communication,Lanzhou University of Technology,Lanzhou 730050,China) |
| Abstract: |
| Reducing the channel status information(CSI) feedback overhead of massive multiple-input multiple-output(MIMO) by compressing the CSI transmission codewords can effectively reduce the use of computing resources and the consumption of information transmission time.For the problem of how to use the lightweight model to accurately estimate the CSI feedback under the condition of low compression ratio,the lightweight iterative cross network(ICNet) model is designed,the designed iterative compression module is used to compress the CSI feedback at the user end,and the designed iterative reconstruction module is used to estimate the CSI feedback at the base station,and the CSI feedback is estimated under the low compression ratio condition with high accuracy and low time consumption.The robustness of ICNet under low compression ratio conditions is evaluated under the data samples generated by COST2100 model.Experiments show that when the small compression ratio is 1/64,the normalized mean squared error of ICNet is reduced by 8.48% compared with the suboptimal value,and the number of parameters of ICNet is reduced by about 35%. |
| Key words: massive MIMO CSI feedback cross convolution low compression ratio lightweight estimation |
《电讯技术》编辑部 