Near-field beamforming at intelligent re-configurable surface in uplink transmission

Abstract

Intelligent re-configurable surfaces (IRS) constitute electronically steerable structures that can be used to control the propagation of electromagnetic (EM) waves. Owing to their passive nature in terms of signal reflection, IRS require large-scale implementations. This means that the near-field effect, e.g., the curvature of the propagating EM wave must be considered. In this paper, a simple two-step near-field beamforming method is proposed. In the first step, the direction from the IRS center point to the user equipment (UE) is found assuming plane wave model. In the second step, the transmitted beam is focused towards the UE based on the pre-calculated constant phase contours. All operations in the proposed method can be implemented as memory operations, i.e., no complex computations are needed. The operation of the method is verified with simulations. The results show that the proposed beamforming technique is capable of focusing the beam pattern of the IRS towards the user. More specifically, the near-field focusing of a (50 ×50)-element IRS operating at 5 GHz frequency increases the received power by 3 dB when compared to a case where the near-field effect is ignored.