Intelligent Reflecting Surface-Aided Wideband THz Communications: Modeling and Analysis

TL;DR

This paper analyzes the performance of wideband THz communications aided by intelligent reflecting surfaces, introducing a novel near-field channel model that accounts for spherical wavefronts and wideband effects, revealing limitations of conventional beamfocusing.

Contribution

It presents a generalized near-field channel model for THz IRS systems that includes spherical wavefronts and spatial-wideband effects, and analytically characterizes beam squint impact.

Findings

Conventional frequency-flat beamfocusing causes significant power loss due to beam squint.

The proposed model accurately predicts the reduction in power gain.

Numerical results validate the analysis and inform future IRS design for THz systems.

Abstract

In this paper, we study the performance of wideband terahertz (THz) communications assisted by an intelligent reflecting surface (IRS). Specifically, we first introduce a generalized channel model that is suitable for electrically large THz IRSs operating in the near-field. Unlike prior works, our channel model takes into account the spherical wavefront of the emitted electromagnetic waves and the spatial-wideband effect. We next show that conventional frequency-flat beamfocusing significantly reduces the power gain due to beam squint, and hence is highly suboptimal. More importantly, we analytically characterize this reduction when the spacing between adjacent reflecting elements is negligible, i.e., holographic reflecting surfaces. Numerical results corroborate our analysis and provide important insights into the design of future IRS-aided THz systems.