Electric Field Evaluation of Reconfigurable Intelligent Surface in Wireless Networks

TL;DR

This work addresses electromagnetic-field exposure in RIS-assisted wireless networks by developing an analytical framework for E-field evaluation under beamforming and reflective operating modes. It validates the framework with 3D ray-tracing simulations, explores RIS-size, height, frequency range, and element-pattern effects across FR1/FR2/FR3, and derives deployment guidelines that keep exposure within regulatory limits—RO modes pose minimal risk, while BO modes require careful placement (e.g., minimum $d_{\rm BR}$ and $h_{\rm RIS}$). The paper also outlines practical, low-cost measurement methodologies for real-world 5G and future networks, aiming to enable scalable E-field assessments and safe RIS deployments. Overall, the results provide concrete deployment rules and measurement pathways to realize RIS-enabled networks while ensuring public EMF safety and regulatory compliance.

Abstract

Reconfigurable intelligent surface (RIS) used as infrastructure in wireless networks has been a trend, thanks to its low cost and high flexibility. Working in many ways including reflective mirrors and phase-shifted surfaces, RIS is able to enhance the coverage in communications and provide more degrees of freedom for sensing. However, the key issue lies in how to place RIS in accordance with the regulations for electromagnetic field (EMF) exposure, which requires refined evaluations. In this paper, we first investigate the regulations in terms of E-field. Then, relevant deployment characteristics are evaluated jointly: the minimum distance from the base station (BS) to the RIS, and the minimum height of the RIS are given for a given BS power limit and as function of the number of RIS elements. The ray-tracing simulations verify the correctness of our analysis. Besides, different frequency ranges (FRs) and radiation patterns of RIS elements are investigated. The results show that the EMF exposure risk is negligible when RIS works in the reflective-only (RO) mode. However, when it works in the beamforming (BO) mode, its placement should be well specified based on our analytical framework to comply with the regulations of E-field limit in general public scenarios. Finally, we provide an E-field measurement methodology and low-cost solutions in terms of general wireless networks and 5G standalone networks, which pave the way for real-world evaluation in future work.

Figures (9)

• Figure 1: System model.
• Figure 2: Comparison of E-field considering various RIS sizes with omnidirectional element and 1 mW power at the RIS.
• Figure 3: E-field map in a $10\times10$ m evaluation area, considering $h_{u}=1.5$ m, $d_{\rm BR}=20$ m, $8\times8$ RIS, and $P_{max}=75$ dBm.
• Figure 4: Peak E-field as a function of $d_{\rm BR}$ and $h_{\rm RIS}$ for different $N_{\rm RIS}$, considering 75 dBm EIRP at BS and $G=\cos(\theta)^3$.
• Figure 5: E-field in the RO mode in the bore-sight direction for different numbers of RIS elements where we consider $f=3.5$ GHz, $h_{\rm RIS}=1.5$ m, $d_{\rm BR}=20$ m, and $P_{max}=75$ dBm.