Channel estimation method for IRS assisted mine communication system based on self supervised learning
-
摘要: 针对矿井复杂环境导致的多径衰落、非视距通信及真实标签获取困难的问题，提出一种基于自监督学习（SSL）的智能反射面（IRS）辅助矿井通信系统信道估计方法。根据井下Nakagami-g衰落信道模型和IRS信号传输模型搭建井下通信系统模型，通过IRS技术解决多径衰落和非视距通信问题。通过最小二乘（LS）算法进行初步信道估计，再采用SSL框架下的八度卷积（OCT）神经网络优化信道估计结果。OCT直接对高频分量和低频分量进行处理，能同时捕捉信道的粗糙特征和细微差别，提供全面的信道信息，从而更准确地估计信道状态；SSL算法使用接收信号及其带噪版本作为训练数据，通过未标注数据的内在结构提升IRS辅助信道估计的精度和效率，从而降低对人工标签的依赖。仿真结果表明：① 引入IRS技术能有效降低信道估计误差。② OCT神经网络的损失值明显低于CNN，数据拟合效果更好；OCT神经网络计算效率高，可提高通信系统信道估计的整体性能；在计算资源有限的环境下，OCT神经网络可保持较低参数量和内存使用量。③ SSL算法在所有信噪比条件下均能保持较低的归一化均方误差，验证了其在信道估计中的高效性和鲁棒性。④ 基于SSL的IRS辅助矿井通信系统信道估计方法在大规模网络中具有较好的扩展性和鲁棒性。Abstract: A channel estimation method for intelligence reflecting surface (IRS) assisted mine communication system based on self supervised learning (SSL) is proposed to address the problems of multipath fading, non line of sight communication, and difficulty in obtaining true labels caused by complex mine environments. The method builds an underground communication system model based on the Nakagami-g fading channel model and IRS signal transmission model, and solves the problems of multipath fading and non line of sight communication through IRS technology. Preliminary channel estimation is performed using the least squares (LS) algorithm, and then the channel estimation results are optimized using octave convolution (OCT) neural network under the SSL framework. OCT directly processes both high-frequency and low-frequency components, capturing both the rough features and subtle differences of the channel, providing comprehensive channel information, and thus more accurately estimating the channel state. The SSL algorithm uses received signals and their noisy versions as training data to improve the precision and efficiency of IRS assisted channel estimation through the intrinsic structure of unlabeled data, thereby reducing reliance on manual labeling. The simulation results show the following points. ① Introducing IRS technology can effectively reduce channel estimation errors. ② The loss value of OCT neural network is significantly lower than that of CNN, and the data fitting effect is better. OCT neural network has high computational efficiency and can improve the overall performance of channel estimation in communication systems. In environments with limited computing resources, OCT neural networks can maintain low parameter and memory usage. ③ The SSL algorithm can maintain a low normalized mean square error under all signal-to-noise ratio conditions, verifying its efficiency and robustness in channel estimation. ④ The channel estimation method for IRS assisted mine communication system based on SSL has good scalability and robustness in large-scale networks.
-
图 1 基于SSL的IRS辅助矿井通信系统信道估计方法整体框架
Figure 1. Overall framework of channel estimation method for intelligence reflecting surface（IRS） assisted mine communication system based on self-supervised learning（SSL）
图 5 引入IRS技术前后算法性能对比
Figure 5. Comparison of algorithm performance before and after introducing IRS technology
图 7 不同算法的NMSE对比
Figure 7. Comparison of normalized mean square error（NMSE） of different algorithms
表 1 OCT神经网络参数
Table 1. Parameters of OCT neural network
层类型 输出通道数 卷积核大小 Initial_Conv 128 （3，3） OctConv 32 （3，3） BN 64 — 残差模块（Conv） 64 （3，3） 
下载: 导出CSV
表 2 不同算法的运行时间
Table 2. The running time of different algorithms
算法 训练时间/s 测试时间/s 深度学习 CNN 119.04 $ 1.240 \times {10^{ - 3}} $ OCT 84.86 $ 8.660 \times {10^{ - 4}} $ 传统算法 LS — 1.200 LMMSE — 0.722
下载: 导出CSV
表 3 不同模型性能对比
Table 3. Performance comparison of different models
网络模型 参数量/个 内存使用量/GiB 当前 峰值 CNN 230 272 10.1 11.6 OCT 226 816 5.4 6.2 ResNet 369 280 8.3 9.0 DenseNet 821 760 7.3 8.1
下载: 导出CSV
表 4 不同规模网络中的NMSE与内存使用情况
Table 4. NMSE and memory usage in networks of different scales
网络规模 NMSE 内存使用量/GiB 小 0.114 5.4 中 0.255 8.5 大 0.381 14.6
下载: 导出CSV
-
[1] 霍振龙,肖松,孟玮,等. 矿井5G无线通信系统关键技术及装备研发与示范应用[J]. 智能矿山,2022,3(4):55-60.HUO Zhenlong,XIAO Song,MENG Wei,et al. Research,development and demonstration application of key technologies and equipment of mine 5G wireless communication system[J]. Journal of Intelligent Mine,2022,3(4):55-60. [2] 孙翠珍,毛昕蓉,马延军. 井下OFDM信道估计算法研究[J]. 工矿自动化,2014,40(9):39-43.SUN Cuizhen,MAO Xinrong,MA Yanjun. Research of underground OFDM channel estimation algorithm[J]. Industry and Mine Automation,2014,40(9):39-43. [3] 樊佳恒. 基于压缩感知的煤矿井下信道估计技术研究[D]. 徐州:中国矿业大学,2020.FAN Jiaheng. Research on channel estimation technology in coal mine based on compressive sensing[D]. Xuzhou:China University of Mining and Technology,2020. [4] BASAR E,DI RENZO M,DE ROSNY J,et al. Wireless communications through reconfigurable intelligent surfaces[J]. IEEE Access,2019,7:116753-116773. doi: 10.1109/ACCESS.2019.2935192 [5] NADEEM Q U A,ALWAZANI H,KAMMOUN A,et al. Intelligent reflecting surface-assisted multi-user MISO communication:channel estimation and beamforming design[J]. IEEE Open Journal of the Communications Society,2020,1:661-680. doi: 10.1109/OJCOMS.2020.2992791 [6] HUANG Chongwen,ZAPPONE A,ALEXANDROPOULOS G C,et al. Reconfigurable intelligent surfaces for energy efficiency in wireless communication[J]. IEEE Transactions on Wireless Communications,2019,18(8):4157-4170. doi: 10.1109/TWC.2019.2922609 [7] ARDAH K,GHEREKHLOO S,DE ALMEIDA A L F,et al. TRICE:a channel estimation framework for RIS-aided millimeter-wave MIMO systems[J]. IEEE Signal Processing Letters,2021,28:513-517. doi: 10.1109/LSP.2021.3059363 [8] KISSELEFF S,CHATZINOTAS S,OTTERSTEN B. Reconfigurable intelligent surfaces in challenging environments:underwater,underground,industrial and disaster[J]. IEEE Access,2021,9:150214-150233. doi: 10.1109/ACCESS.2021.3125461 [9] CHEN Jie,LIANG Yingchang,CHENG H V,et al. Channel estimation for reconfigurable intelligent surface aided multi-user mmWave MIMO systems[J]. IEEE Transactions on Wireless Communications,2023,22(10):6853-6869. doi: 10.1109/TWC.2023.3246264 [10] XIE Wenwu,XIAO Jian,ZHU Peng,et al. Deep compressed sensing-based cascaded channel estimation for RIS-aided communication systems[J]. IEEE Wireless Communications Letters,2022,11(4):846-850. doi: 10.1109/LWC.2022.3147590 [11] CHEN J C. Machine learning-inspired algorithmic framework for intelligent reflecting surface-assisted wireless systems[J]. IEEE Transactions on Vehicular Technology,2021,70(10):10671-10685. doi: 10.1109/TVT.2021.3110970 [12] ZHOU Gui,PAN Cunhua,REN Hong,et al. Channel estimation for RIS-aided multiuser millimeter-wave systems[J]. IEEE Transactions on Signal Processing,2022,70:1478-1492. doi: 10.1109/TSP.2022.3158024 [13] JENSEN T L,DE CARVALHO E. An optimal channel estimation scheme for intelligent reflecting surfaces based on a minimum variance unbiased estimator[C]. IEEE International Conference on Acoustics,Speech and Signal Processing,Barcelona,2020:5000-5004. [14] WANG Tianqi,WEN Chaokai,WANG Hanqing,et al. Deep learning for wireless physical layer:opportunities and challenges[J]. China Communications,2017,14(11):92-111. doi: 10.1109/CC.2017.8233654 [15] ELBIR A M,PAPAZAFEIROPOULOS A,KOURTESSIS P,et al. Deep channel learning for large intelligent surfaces aided mm-wave massive MIMO systems[J]. IEEE Wireless Communications Letters,2020,9(9):1447-1451. doi: 10.1109/LWC.2020.2993699 [16] GAO Shen,DONG Peihao,PAN Zhiwen,et al. Deep multi-stage CSI acquisition for reconfigurable intelligent surface aided MIMO systems[J]. IEEE Communications Letters,2021,25(6):2024-2028. doi: 10.1109/LCOMM.2021.3063464 [17] LIU Shicong,GAO Zhen,ZHANG Jun,et al. Deep denoising neural network assisted compressive channel estimation for mmWave intelligent reflecting surfaces[J]. IEEE Transactions on Vehicular Technology,2020,69(8):9223-9228. doi: 10.1109/TVT.2020.3005402 [18] KUNDU N K,MCKAY M R. A deep learning-based channel estimation approach for MISO communications with large intelligent surfaces[C]. IEEE 31st Annual International Symposium on Personal,Indoor and Mobile Radio Communications,London,2020:1-6. [19] LIU Chang,LIU Xuemeng,NG D W K,et al. Deep residual learning for channel estimation in intelligent reflecting surface-assisted multi-user communications[J]. IEEE Transactions on Wireless Communications,2022,21(2):898-912. doi: 10.1109/TWC.2021.3100148 [20] 李涛,蒋磊,陈博文. 独立不同分布Nakagami-m衰落信道下最大比合并性能分析[J]. 空军工程大学学报(自然科学版),2018,19(6):84-89.LI Tao,JIANG Lei,CHEN Bowen. A performance analysis of maximal ratio combining under condition of non-identically distributed nakagami-mFading channels[J]. Journal of Air Force Engineering University (Natural Science Edition),2018,19(6):84-89. [21] 张长森,张艳芳. 矿井移动通信中Nakagami衰落信道模型的研究[J]. 计算机工程与应用,2014,50(7):238-241. doi: 10.3778/j.issn.1002-8331.1207-0345ZHANG Changsen,ZHANG Yanfang. Research of Nakagami fading channel model in mine mobile communication[J]. Computer Engineering and Applications,2014,50(7):238-241. doi: 10.3778/j.issn.1002-8331.1207-0345 [22] CHEN Yunpeng,FAN Haoqi,XU Bing,et al. Drop an octave:reducing spatial redundancy in convolutional neural networks with octave convolution[C]. 2019 IEEE/CVF International Conference on Computer Vision,Seoul,2019:3435-3444. -
点击查看大图
图(7) / 表(4)
计量
- 文章访问数: 61
- HTML全文浏览量: 18
- PDF下载量: 2
- 被引次数: 0
