Continuous and Discrete Phase Shifting for the RIS-aided DMA-Based MU MIMO System

The introduction of dynamic metasurface antennas (DMAs)-a metamaterial-based antenna technology that incorporates reconfigurable elements within waveguide structures called microstrip, has sparked significant interest in academia following its potential to economically replace traditional antennas. However, it is well known that DMAs suffer from inherent signal attenuation from their use of waveguides. Moreover, most previous works have focused on continuous phase shifting for DMAs, which is costly for practical applications. To compensate for the losses within the DMA microstrip, we propose using reconfigurable intelligent surfaces (RIS) to enhance signal transmission in a multiuser system that deploys DMAs at both the base station (BS) and users. Furthermore, we demonstrate how to use both discrete and continuous phases to configure the DMA weights and RIS reflection coefficients. Additionally, for the first time, we identify and propose an efficient solution to the ill-conditioning problem associated with the DMA's Lorentzian constraint when configured with discrete phases. Next, by leveraging techniques such as iterative waterfilling, tabu search, and gradient projection, we propose effective methods for optimizing transmit precoders, RIS reflection coefficients, and DMA weights at the BS and users for spectral efficiency maximization of the uplink and downlink transmissions. Lastly, the presented numerical results demonstrate the effectiveness of the proposed algorithms in achieving higher performance than the benchmark schemes.