Intelligent reflecting surface (IRS) is a promising new technology that is able to create a favorable wireless signal propagation environment by collaboratively reconfiguring the passive reflecting elements yet with low hardware and energy cost. In IRS-aided wireless communication systems, channel modeling is a fundamental task for communication algorithm design and performance optimization, which however is also very challenging since in-depth domain knowledge and technical expertise in radio signal propagations are required, especially for modeling the high-dimensional cascaded base station (BS)-IRS and IRS-user channels (also referred to as the reflected channels). In this paper, we propose a model-driven generative adversarial network (GAN)-based channel modeling framework to autonomously learn the reflected channel distribution, without complex theoretical analysis or data processing. The designed GAN (also named as IRS-GAN) is trained to reach the Nash equilibrium of a minimax game between a generative model and a discriminative model. For the single-user case, we propose to incorporate the special structure of the reflected channels into the design of the generative model. While for the multiuser case, we extend the IRS-GAN and present a multiuser IRS-GAN (abbreviated as IRS-GAN-M), where the distributions of the reflected channels associated with different users are learned simultaneously with reduced number of network parameters (as compared to the naive scheme that assigns a dedicated IRS-GAN for each user). Moreover, theoretical analysis is presented to prove that the minimax game in the IRS-GAN-M framework has a global optimum if the generative and discriminative models are given with enough capacity. Simulation results are presented to validate the effectiveness of the proposed IRS-GAN framework.

中文翻译：

---

信道分布学习：IRS 辅助无线通信的基于模型驱动的基于 GAN 的信道建模

智能反射面（IRS）是一种很有前途的新技术，它能够通过协作重新配置无源反射元件来创造一个有利的无线信号传播环境，而且硬件和能源成本低。在 IRS 辅助的无线通信系统中，信道建模是通信算法设计和性能优化的一项基本任务，然而这也非常具有挑战性，因为需要深入的领域知识和无线电信号传播方面的技术专长，特别是对于建模高维度级联基站 (BS)-IRS 和 IRS-用户信道（也称为反射信道）。在本文中，我们提出了一种基于模型驱动的生成对抗网络（GAN）的信道建模框架，以自主学习反射信道分布，无需复杂的理论分析或数据处理。设计的 GAN（也称为 IRS-GAN）经过训练以达到生成模型和判别模型之间的极小极大博弈的纳什均衡。对于单用户情况，我们建议将反射通道的特殊结构纳入生成模型的设计中。而对于多用户情况，我们扩展了 IRS-GAN 并提出了一个多用户 IRS-GAN（缩写为 IRS-GAN-M），其中与不同用户相关的反射通道的分布是在减少网络参数数量的情况下同时学习的（与为每个用户分配专用 IRS-GAN 的简单方案相比）。而且，理论分析证明，如果生成模型和判别模型具有足够的容量，则 IRS-GAN-M 框架中的极小极大博弈具有全局最优。给出了模拟结果以验证所提出的 IRS-GAN 框架的有效性。