Spectral Efficiency Analysis and Optimization in Hybrid RIS Systems Under Nonlinear Power Amplifier Effects

This paper investigates the impact of nonlinear distortions caused by high-power amplifiers (HPAs) in hybrid reconfigurable intelligent surface (RIS)-assisted communication systems. By applying the Bussgang theorem, we derive an analytical model for the nonlinear distortions and develop the corresponding expressions for the received signal and the spectral efficiency (SE) in both ideal (linear) and practical (nonlinear) cases, as well as in the passive RIS scenario. To mitigate the performance degradation induced by hardware impairments, we propose an active element selection strategy that aims to optimize the SE by appropriately selecting the active elements of the RIS. Different optimization algorithms are considered, namely a Genetic Algorithm (GA), a Greedy Algorithm, and Random Search (RS). Numerical results highlight the severe impact of nonlinearities on SE and demonstrate that the proposed selection strategy can significantly improve the system performance. Furthermore, we show that the Random Search method provides a favorable trade-off between performance and computational complexity, especially when the number of tested combinations is properly chosen.