Stacked Intelligent Metasurfaces for Multiuser Beamforming in the Wave Domain

TL;DR

This paper introduces a stacked intelligent metasurface (SIM) for multiuser beamforming in wireless communications, enabling low-cost, energy-efficient, and fast wave-domain processing that surpasses traditional digital methods.

Contribution

It proposes a novel multilayer SIM structure for wave-domain beamforming, reducing hardware costs, energy consumption, and delay in multiuser MIMO systems.

Findings

Significant reduction in hardware cost and energy use.

Enhanced sum rate performance compared to benchmarks.

Substantially decreased precoding delay.

Abstract

Reconfigurable intelligent surface has recently emerged as a promising technology for shaping the wireless environment by leveraging massive low-cost reconfigurable elements. Prior works mainly focus on a single-layer metasurface that lacks the capability of suppressing multiuser interference. By contrast, we propose a stacked intelligent metasurface (SIM)-enabled transceiver design for multiuser multiple-input single-output downlink communications. Specifically, the SIM is endowed with a multilayer structure and is deployed at the base station to perform transmit beamforming directly in the electromagnetic wave domain. As a result, an SIM-enabled transceiver overcomes the need for digital beamforming and operates with low-resolution digital-to-analog converters and a moderate number of radio-frequency chains, which significantly reduces the hardware cost and energy consumption, while substantially decreasing the precoding delay benefiting from the processing performed in the wave domain. To leverage the benefits of SIM-enabled transceivers, we formulate an optimization problem for maximizing the sum rate of all the users by jointly designing the transmit power allocated to them and the analog beamforming in the wave domain. Numerical results based on a customized alternating optimization algorithm corroborate the effectiveness of the proposed SIM-enabled analog beamforming design as compared with various benchmark schemes. Most notably, the proposed analog beamforming scheme is capable of substantially decreasing the precoding delay compared to its digital counterpart.