Enhanced ELM Based Channel Estimation for RIS-Assisted OFDM Systems With Insufficient CP and Imperfect Hardware

Reconfigurable intelligent surface (RIS)-assisted orthogonal frequency division multiplexing (OFDM) systems have aroused extensive research interests due to the controllable communication environment and performance of combating multi-path interference. However, as premise RIS-assisted OFDM systems, accuracy channel estimation is severely degraded by increased possibility insufficient cyclic prefix (CP) produced extra cascaded channels RIS nonlinear distortion lead imperfect hardware. To address these issues, an enhanced extreme learning machine (ELM)-based (eELM-CE) proposed in this letter facilitate accurate estimation. Based on model-driven mode, least square (LS) employed highlight initial linear features for Then, according obtained features, ELM network constructed refine In particular, we start from perspective guiding it recognize feature, normalize data after activation function enhance ability identifying non-linear factors. Experiment results show that, compared with existing methods, method achieves a much lower normalized mean error (NMSE) given CP addition, simulation indicate that possesses robustness against parameter variations.