Reconfigurable Intelligent Surfaces (RISs) are recently attracting a wide interest due to their capability of tuning wireless propagation environments in order to increase the system performance of wireless networks. In this paper, a multiuser wireless network assisted by a RIS is studied and resource allocation algorithms are presented for several scenarios. First of all, the problem of channel estimation is considered, and an algorithm that permits separate estimation of the mobile user-to-RIS and RIS-to-base stations components is proposed. Then, for the special case of a single-user system, three possible approaches are shown in order to optimize the Signal-to-Noise Ratio with respect to the beamformer used at the base station and to the RIS phase shifts. Next, for a multiuser system with two cells, assuming channel-matched beamforming, the geometric mean of the downlink Signal-to-Interference plus Noise Ratios across users is maximized with respect to the base stations transmit powers and RIS phase shifts configurations. In this scenario, the RIS is placed at the cell-edge and some users are jointly served by two base stations to increase the system performance. Numerical results show that the proposed procedures are effective and that the RIS brings substantial performance improvements to wireless system.
翻译:最近,由于它们有能力调整无线传播环境以提高无线网络的系统性能,重新配置的智能表面(RIS)最近吸引了广泛的兴趣。本文研究了一个多用户无线网络,并针对几种情况提出了资源分配算法。首先,审议了频道估算问题,并提出了允许分别估计移动用户对RIS和RIS对基地台站组件的算法。然后,就单一用户系统的特殊情况而言,展示了三种可能的办法,以优化基站和RIS阶段转换所用光谱仪的信号到噪音比率。接着,对于拥有两个细胞的多用户系统,假设频道配对面的光谱法,对用户之间信号到干涉的下行线加上噪音比率的几何平均值,在基站传输权力和RIS阶段变换配置方面达到最大化。在这一假设中,将RIS置于单元格对信号到噪音的比率优化,一些用户通过两个基站联合服务于信号-噪音系统,以显示有效的性能改进。