Réseaux du futur : 5G et au-delà

11-13 mars, 2020
Telecom Paris, Institut Polytechnique de Paris, Palaiseau 

Numerical dosimetry in human model for 5G and beyond
Abdelrahman Ijjeh  1@  , Marylene Cueille  2, *@  , Jean-Lou Dubard  1, *@  , Michel Ney  3, *@  
1 : Laboratoire d'Electronique, Antennes et Télécommunications  (LEAT)
Université Côte d’Azur
2 : Laboratoire d'Electronique, Antennes et Télécommunications  (LEAT)
University Côte d'Azur
3 : IMT Atlantique  (IMT Atlantique)
Institut Mines Telecom (IMT)
* : Auteur correspondant

To design any new EM-device, it is of great importance to have a good estimation of the specific absorption rate (SAR) distribution or the Power Density (PD) it induces in users or people in its proximity. Respecting the recommended SAR limits is mandatory for any new technology to go public. Moreover, the good knowledge of SAR/PD is critical in any electromagnetic compatibility (EMC) study between EM-devices with lossy media. From numerical analysis point of view, as frequencies go higher and higher e.g. from 5G to millimetric or even terahertz applications, the computational problem becomes exhaustive in size. Note that at higher frequencies, the transmitting devices get normally smaller in size. In this article, we will study the effect of considering only the tissues (the part of the human model) adjacent to the transmitter vs. including the full human model in the computation domain. For a good spatial representation, the block-meshing scheme can be applied to finely mesh the antenna system and the tissues close to it. Numerical examples show the SAR distributions at 3.5 GHz in human head and the entire body, in addition to show the validity and CPU-time/memory gain obtained when considering part of the human model instead of the entire model. Moreover, we show the effects of human head presence on the antenna performance in its proximity.

Personnes connectées : 1