Channel Estimation for Extremely Large-Scale RIS-Aided Communication System

In reconfigurable intelligent surface (RIS) assisted wireless communication systems, the complex fading that occurs from the base station (BS) to the RIS and then to the user often necessitates large-scale RIS setups to mitigate these effects. Consequently, extremely large RIS (XL-RIS) configurations are becoming increasingly common. However, due to the large aperture of XL-RIS, scatterers are more likely to be within the nearfield region, where planar wave assumptions no longer hold. This discrepancy results in significant performance degradation when far-field channel estimation methods are applied to XL-RIS nearfield communications. Moreover, the increase in RIS elements leads to higher pilot overhead. To address these challenges, we first reduce the pilot overhead and obtain independent channel state information for both the BS-RIS and RIS-user links by utilizing partially sensing reflecting elements. Secondly, we incorporate near-field effects by adopting a more realistic spherical wave model. Lastly, we exploit the inherent sparsity of the near-field channel in the polar coordinate domain to enable precise reconstruction of the complete channel. Simulation results demonstrate that the proposed modified Polar-domain NOMP algorithm achieves excellent performance.