In contrast to the conventional reconfigurable intelligent surfaces (RIS), intelligent omni-surfaces (IOS) are capable of full-space coverage of smart radio environments by simultaneously transmitting and reflecting the incident signals. In this paper, we investigate the ergodic spectral efficiency of IOS-aided systems for transmission over random channel links, while considering both realistic imperfect channel state information (CSI) and transceiver hardware impairments (HWIs). Firstly, we formulate the linear minimum mean square error estimator of the equivalent channel spanning from the user equipments (UEs) to the access point (AP), where the transceiver HWIs are also considered. Then, we apply a two-timescale protocol for designing the beamformer of the IOS-aided system. Specifically, for the active AP beamformer, the minimum mean square error combining method is employed, which relies on the estimated equivalent channels, on the statistical information of the channel estimation error, on the inter-user interference as well as on the HWIs at the AP and UEs. By contrast, the passive IOS beamformer is designed based on the statistical CSI for maximizing the upper bound of the ergodic spectral efficiency. The theoretical analysis and simulation results show that the transceiver HWIs have a significant effect on the ergodic spectral efficiency, especially in the high transmit power region. Furthermore, we show that the HWIs at the AP can be effectively compensated by deploying more AP antennas.

翻译：与传统可重构智能表面（RIS）不同，智能全向面（IOS）能够通过同时透射和反射入射信号实现智能无线电环境的全空间覆盖。本文研究IOS辅助系统在随机信道链路传输中的遍历频谱效率，同时考虑实际非完美信道状态信息（CSI）和收发信机硬件损伤（HWIs）。首先，我们构建了从用户设备（UE）到接入点（AP）的等效信道线性最小均方误差估计器，其中也考虑了收发信机HWIs。随后，采用双时间尺度协议设计IOS辅助系统的波束成形器。具体而言，对于有源AP波束成形器，采用基于估计等效信道、信道估计误差统计信息、用户间干扰以及AP和UE处HWIs的最小均方误差合并方法。相比之下，无源IOS波束成形器基于统计CSI设计，以最大化遍历频谱效率的上界。理论分析与仿真结果表明，收发信机HWIs对遍历频谱效率具有显著影响，尤其是在高发射功率区域。此外，我们发现通过部署更多AP天线可以有效补偿AP处的HWIs。