We analyze the performance of large intelligent surface (LIS) with hardware distortion at its RX-chains. In particular, we consider the memory-less polynomial model for non-ideal hardware and derive analytical expressions for the signal to noise plus distortion ratio after applying maximum ratio combining (MRC) at the LIS. We also study the effect of back-off and automatic gain control on the RX-chains. The derived expressions enable us to evaluate the scalability of LIS when hardware impairments are present. We also study the cost of assuming ideal hardware by analyzing the minimum scaling required to achieve the same performance with a non-ideal hardware. Then, we exploit the analytical expressions to propose optimized antenna selection schemes for LIS and we show that such schemes can improve the performance significantly. In particular, the antenna selection schemes allow the LIS to have lower number of non-ideal RX-chains for signal reception while maintaining a good performance. We also consider a more practical case where the LIS is deployed as a grid of multi-antenna panels, and we propose panel selection schemes to optimize the complexity-performance trade-offs and improve the system overall efficiency.

翻译：本文分析了接收链存在硬件失真的大规模智能表面（LIS）的性能。具体而言，我们采用无记忆多项式模型描述非理想硬件，并推导了在LIS应用最大比合并（MRC）后信噪失真比的解析表达式。我们还研究了接收链回退与自动增益控制的影响。所得表达式使我们能够评估存在硬件损伤时LIS的可扩展性。通过分析非理想硬件达到相同性能所需的最小规模扩展，我们进一步探讨了假设理想硬件所带来的代价。随后，我们利用解析表达式提出针对LIS的优化天线选择方案，并证明此类方案可显著提升性能。特别地，天线选择方案使LIS能够以更少数量的非理想接收链维持良好接收性能。我们还考虑了LIS部署为多天线面板阵列的实际场景，提出面板选择方案以优化复杂度与性能的权衡，从而提升系统整体效率。