Channel Estimation for Dynamic Metasurface Antennas
Maryam Rezvani, Raviraj Adve
Abstract
To meet the rising demand for data rates, the next generation of wireless communication will need to deploy more antennas at the base stations (BSs) while also addressing sustainability and power efficiency concerns. Recently, metasurfaces, specifically dynamic metasurface antennas (DMAs), have been suggested as potential solutions. Specifically, researchers have demonstrated the potential throughput performance of DMAs as basestation antennas. In this paper, we study the channel estimation problem for DMAs, modeled as correlated multiple input multiple output (MIMO) systems. We first propose a system model that details the noise sources in these systems. Then, exploiting the fast response time of DMAs, we consider the possibility of multiple channel measurements during a single symbol. Adopting a minimum-mean-square-error (MMSE) based approach and assuming availability of the long-term channel correlation matrices at the receiver and transmitters, then, we formulate the channel estimation problem. We first solve this problem assuming full-control over the DMA. We then propose a more practical algorithm to map the full-control solution to the DMA structure, thereby accounting for the limitations imposed by the DMA. Our numerical results demonstrate that the proposed practical algorithm suffers only a minor performance loss compared to the full-control case.