A Primer on Near-Field Beamforming for Arrays and Reconfigurable Intelligent Surfaces

TL;DR

This paper explores near-field beamforming for large arrays and reconfigurable surfaces, challenging traditional far-field assumptions and identifying the conditions for effective near-field focusing and beam gain.

Contribution

It introduces a new characterization of near-field behavior, moving beyond the Fraunhofer distance to determine when finite-depth beamforming is feasible.

Findings

Finite-depth beamforming is possible within a specific near-field distance range.

The Fraunhofer distance is not the appropriate metric for near-field focusing.

Beamforming gain diminishes beyond a certain near-field distance.

Abstract

Wireless communication systems have almost exclusively operated in the far-field of antennas and antenna arrays, which is conventionally characterized by having propagation distances beyond the Fraunhofer distance. This is natural since the Fraunhofer distance is normally only a few wavelengths. With the advent of active arrays and passive reconfigurable intelligent surfaces (RIS) that are physically large, it is plausible that the transmitter or receiver is located in between the Fraunhofer distance of the individual array/surface elements and the Fraunhofer distance of the entire array. An RIS then can be configured to reflect the incident waveform towards a point in the radiative near-field of the surface, resulting in a beam with finite depth, or as a conventional angular beam with infinity focus, which only results in amplification in the far-field. To understand when these different options are viable, an accurate characterization of the near-field behaviors is necessary. In this paper, we revisit the motivation and approximations behind the Fraunhofer distance and show that it is not the right metric for determining when near-field focusing is possible. We obtain the distance range where finite-depth beamforming is possible and the distance where the beamforming gain tapers off.