Leveraging Stacked Intelligent Surfaces for Near-Field Localization by Using a Multi-Port Network Model

In this paper, we leverage stacked intelligent metasurfaces (SIMs) to enable near-field localization. To this end, we employ a multi-port network model, which enables the electromagnetically consistent characterization of SIMs and is suitable for optimization as well. Near-field localization is achieved by processing the signals at a receiving array, and by exploiting the curvature of the incoming wavefront. We show how an SIM can be used to accomplish this goal. Specifically, we prove that an SIM can be optimized to perform a linear transformation that operates entirely in the electromagnetic domain, and that the proposed linear transformation of the input signal accounts for both angular and distance information simultaneously, which are required for near-field localization.