摘 要：信道建模是智能反射面（IRS）领域研究的热点之一，为数学上处理方便，大多数论文采用独立同分布方式信道建模。受机器学习算法启发，文章采用基于无监督期望最大化（EM）学习算法来建模 IRS 端到端等效信道。其仿真结果表明，相关瑞利衰落相比独立同分布信道可提供更多信道信息，能提供更好的中断性能。通过让两个 Nakagami-m 分布线性加权简单混合后，在两种概率分布情况下，不同 IRS 元素个数和迭代次数，呈现出不同中断概率。
关键词：IRS；信道建模；EM 算法；中断概率
DOI:10.19850/j.cnki.2096-4706.2021.12.015
中图分类号：TN929.5 文献标识码：A 文章编号：2096-4706（2021）12-0055-04
Research on IRS Channel Modeling Based on EM Algorithm
HUANG Hongjian1 , YANG Xiaonan2 , HE Zhihai 3
(1.Hainan Tropical Ocean University, Sanya 572022, China; 2.Hainan College of Economics and Business, Haikou 571127, China; 3.Hainan Vocational University, Haikou 570216, China)
Abstract: Channel modeling is one of the research hotspots in the field of Intelligent Reflecting Surface (IRS). For the convenience of mathematical processing, most papers use independent identically distributed channel modeling. Inspired by machine learning algorithm, this paper uses unsupervised expectation maximization (EM) learning algorithm to model the end-to-end equivalent channel of IRS. The simulation results show that related Rayleigh fading can provide more channel information and better outage performance than independent identically distributed channel. After the linear weighting and simple mixing of two Nakagami-m distributions, the number of IRS elements and the number of iterations show different outage probabilities under two probability distributions.
Keywords: IRS; channel modeling; EM algorithm; outage probability
参考文献：
[1] BJÖRNSON E，SANGUINETTI L. Rayleigh fading modeling and channel hardening for reconfigurable intelligent surfaces [J].IEEE Wireless Communications Letters，2021，10（4）：830-834.
[2] HAN Y，TANG W，JIN S，et al. Large intelligent surfaceassisted wireless communication exploiting statistical CSI [J].IEEE Transactions on Vehicular Technology，2019，68（8）：8238-8242.
[3] WU Q Q，ZHANG R. Towards Smart and Reconfigurable Environment：Intelligent Reflecting Surface Aided Wireless Network [J]. IEEE Communications Magazine，2020，58（1）：106-112.
[4] HE J G，NGUYEN N T，SCHROEDER R，et al. Channel Estimation and Hybrid Architectures for RIS-Assisted Communications [J].arXiv:2104.07115 [eess.SP].（2021-04-14）.https://arxiv.org/ abs/2104.07115v1.
[5] SHTAIWI E，ZHANG H L，VISHWANATH S，et al. Channel Estimation Approach for RIS Assisted MIMO Systems [J].IEEE Transactions on Cognitive Communications and Networking，2021，7 （2）：452-465.
[6] WANG Z R，LIU L，CUI S G. Channel Estimation for Intelligent Reflecting Surface Assisted Multiuser Communications [C]//2020 IEEE Wireless Communications and Networking Conference （WCNC）.Seoul：IEEE，2020：1-6.
[7] KUNDU N K，MCKAY M R. Channel Estimation for Reconfigurable Intelligent Surface Aided MISO Communications： From LMMSE to Deep Learning Solutions [J].IEEE Open Journal of the Communications Society，2021（2）：471-487.
[8] WANG Z R，LIU L，CUI S G. Channel Estimation for Intelligent Reflecting Surface Assisted Multiuser Communications： Framework，Algorithms，and Analysis [J].IEEE Transactions on Wireless Communications，2020，19（10）：6607-6620.
[9] WU Q Q，ZHANG S W，ZHENG B X，et al. Intelligent Reflecting Surface-Aided Wireless Communications：A Tutorial [J]. IEEE Transactions on Communications，2021（5）：3313-3351.
[10] SÁNCHEZ J D V，URQUIZA-AGUIAR L，PAREDES M C P，et al. Expectation-Maximization Learning for Wireless Channel Modeling of Reconfigurable Intelligent Surfaces [J].arXiv:2103.13525[cs. IT].（2021-03-24）.https://arxiv.org/abs/2103.13525.
[11] 王兆瑞，刘亮，李航，等 . 面向 6G 物联网的智能反射表 面设计 [J]. 物联网学报，2020，4（2）：84-95.
[12] 程履帮 .OFDMA 系统中基于 LMMSE 信道估计算法的改 进及其性能分析 [J]. 电子学报，2008（9）：1782-1785.
作者简介：黄宏健（1989.09—），男，汉族，海南文昌人， 办公室主任，讲师，硕士，研究方向：信号与信息处理；通讯作者： 杨晓楠（1989.04—），女，汉族，辽宁铁岭人，讲师，硕士研究生， 研究方向：人工智能与机器学习。