RIS-Empowered Near-Field Imaging in NLOS Scenarios

This paper addresses the near-field NLOS imaging problem within an industrial scenario, where large multiple-input multiple-output (MIMO) systems and reconfigurable intelligent surfaces (RISs) are primarily deployed for wireless communications. The proposed system features a fixed transceiver antenna array illuminating and observing a designated region of interest (ROI) to estimate its scattering coefficients. Since we account for high-frequency bandwidths (e.g., THz) and MIMO technology, the ROI is in the radiating near-field of both the RIS and the transceiver, thus augmenting the number of available degrees of freedom (DoF) (i.e., the channel matrix rank). This gives rise to the so-called holographic imaging. To boost the imaging performance, we investigate the ROI illumination and RIS configuration and provide a closed-form solution for the RIS phase profile that minimizes the mean squared error (MSE) associated with the image estimate. Numerical results show the impact of RIS configuration, transmit signal optimization, and regularization on imaging performance, proving the feasibility of imaging with the considered system.