Indoor RIS-Assisted Wireless System with Location-Based Reflective Patterns

Reconfigurable intelligent surface (RIS) has emerged as a highly promising infrastructure benefiting from its capability to manipulate the propagation environment and facilitate efficient aggregation of wireless transmission signals. However, a major challenge in these systems is the significant overhead incurred by the acquisition of channel state information. This paper proposes a pre-designed beam-based transmission protocol that aims to reduce the burden of pilot overhead by designing a RIS reflective pattern codebook as an alternative to channel estimation (CE). We adopt an electromagnetic-compliant RIS model and develop a theoretical approximate expression for the channel gain, incorporating the impact of location mismatch. This approximation facilitates the construction of a location-based reflective pattern codebook, wherein the chosen locations linked with the codewords represent optimal location sampling points derived from theoretical results. By utilizing the constructed codebook, our proposed transmission protocol enables the system to search reflective patterns instead of real-time optimization. To validate our approach, extensive numerical simulations are conducted. The results demonstrate the accuracy of the approximate channel gain expression and highlight the superior coverage achieved by the proposed location-based reflective pattern codebook as well as the achievable spectral efficiency of our transmission protocol.