Exposure assessment of one-year-old child to 3G tablet in uplink mode and to 3G femtocell in downlink mode using polynomial chaos decomposition.

So far, the assessment of the exposure of children, in the ages 0-2 years old, to relatively new radio-frequency (RF) technologies, such as tablets and femtocells, remains an open issue. This study aims to analyse the exposure of a one year-old child to these two sources, tablets and femtocells, operating in uplink (tablet) and downlink (femtocell) modes, respectively. In detail, a realistic model of an infant has been used to model separately the exposures due to (i) a 3G tablet emitting at the frequency of 1940 MHz (uplink mode) placed close to the body and (ii) a 3G femtocell emitting at 2100 MHz (downlink mode) placed at a distance of at least 1 m from the infant body. For both RF sources, the input power was set to 250 mW. The variability of the exposure due to the variation of the position of the RF sources with respect to the infant body has been studied by stochastic dosimetry, based on polynomial chaos to build surrogate models of both whole-body and tissue specific absorption rate (SAR), which makes it easy and quick to investigate the exposure in a full range of possible positions of the sources. The major outcomes of the study are: (1) the maximum values of the whole-body SAR (WB SAR) have been found to be 9.5 mW kg(-1) in uplink mode and 65 µW kg(-1) in downlink mode, i.e. within the limits of the ICNIRP 1998 Guidelines; (2) in both uplink and downlink mode the highest SAR values were approximately found in the same tissues, i.e. in the skin, eye and penis for the whole-tissue SAR and in the bone, skin and muscle for the peak SAR; (3) the change in the position of both the 3G tablet and the 3G femtocell significantly influences the infant exposure.

The influence of the reflective environment on the absorption of a human male exposed to representative base station antennas from 300 MHz to 5 GHz.

The environment is an important parameter when evaluating the exposure to radio-frequency electromagnetic fields. This study investigates numerically the variation on the whole-body and peak spatially averaged-specific absorption rate (SAR) in the heterogeneous virtual family male placed in front of a base station antenna in a reflective environment. The SAR values in a reflective environment are also compared to the values obtained when no environment is present (free space). The virtual family male has been placed at four distances (30 cm, 1 m, 3 m and 10 m) in front of six base station antennas (operating at 300 MHz, 450 MHz, 900 MHz, 2.1 GHz, 3.5 GHz and 5.0 GHz, respectively) and in three reflective environments (a perfectly conducting wall, a perfectly conducting ground and a perfectly conducting ground + wall). A total of 72 configurations are examined. The absorption in the heterogeneous body model is determined using the 3D electromagnetic (EM) finite-difference time-domain (FDTD) solver Semcad-X. For the larger simulations, requirements in terms of computer resources are reduced by using a generalized Huygens' box approach. It has been observed that the ratio of the SAR in the virtual family male in a reflective environment and the SAR in the virtual family male in the free-space environment ranged from -8.7 dB up to 8.0 dB. A worst-case reflective environment could not be determined. ICNIRP reference levels not always showed to be compliant with the basic restrictions.

Statistical analysis of whole-body absorption depending on anatomical human characteristics at a frequency of 2.1 GHz.

In this paper, we propose identification of the morphological factors that may impact the whole-body averaged specific absorption rate (WBSAR). This study is conducted for the case of exposure to a front plane wave at a 2100 MHz frequency carrier. This study is based on the development of different regression models for estimating the WBSAR as a function of morphological factors. For this purpose, a database of 12 anatomical human models (phantoms) has been considered. Also, 18 supplementary phantoms obtained using the morphing technique were generated to build the required relation. This paper presents three models based on external morphological factors such as the body surface area, the body mass index or the body mass. These models show good results in estimating the WBSAR (<10%) for families obtained by the morphing technique, but these are still less accurate (30%) when applied to different original phantoms. This study stresses the importance of the internal morphological factors such as muscle and fat proportions in characterization of the WBSAR. The regression models are then improved using internal morphological factors with an estimation error of approximately 10% on the WBSAR. Finally, this study is suitable for establishing the statistical distribution of the WBSAR for a given population characterized by its morphology.

Analysis of RF exposure in the head tissues of children and adults.

This paper analyzes the radio frequencies (RF) exposure in the head tissues of children using a cellular handset or RF sources (a dipole and a generic handset) at 900, 1800, 2100 and 2400 MHz. Based on magnetic resonance imaging, child head models have been developed. The maximum specific absorption rate (SAR) over 10 g in the head has been analyzed in seven child and six adult heterogeneous head models. The influence of the variability in the same age class is carried out using models based on a morphing technique. The SAR over 1 g in specific tissues has also been assessed in the different types of child and adult head models. Comparisons are performed but nevertheless need to be confirmed since they have been derived from data sets of limited size. The simulations that have been performed show that the differences between the maximum SAR over 10 g estimated in the head models of the adults and the ones of the children are small compared to the standard deviations. But they indicate that the maximum SAR in 1 g of peripheral brain tissues of the child models aged between 5 and 8 years is about two times higher than in adult models. This difference is not observed for the child models of children above 8 years old: the maximum SAR in 1 g of peripheral brain tissues is about the same as the one in adult models. Such differences can be explained by the lower thicknesses of pinna, skin and skull of the younger child models.

Variability analysis of SAR from 20 MHz to 2.4 GHz for different adult and child models using finite-difference time-domain.

This paper deals with the variability of body models used in numerical dosimetry studies. Six adult anthropomorphic voxel models have been collected and used to build 5-, 8- and 12-year-old children using a morphing method respecting anatomical parameters. Finite-difference time-domain calculations of a specific absorption rate (SAR) have been performed for a range of frequencies from 20 MHz to 2.4 GHz for isolated models illuminated by plane waves. A whole-body-averaged SAR is presented as well as the average on specific tissues such as skin, muscles, fat or bones and the average on specific parts of the body such as head, legs, arms or torso. Results point out the variability of adult models. The standard deviation of whole-body-averaged SAR of adult models can reach 40%. All phantoms are exposed to the ICNIRP reference levels. Results show that for adults, compliance with reference levels ensures compliance with basic restrictions, but concerning children models involved in this study, the whole-body-averaged SAR goes over the fundamental safety limits up to 40%.

Modeling of RF head exposure in children.

This study analyzes the main parameters that should influence the specific absorption rate (SAR) in children's heads. The evolution of their head shape and the growth of specific parameters, such as the skull thickness, are analyzed. The influence of these parameters on the radio frequency (RF) exposure of children's head is studied. The SAR over 1 g in specific tissue is assessed in different children's head models based on magnetic resonance imaging (MRI) and on non-uniformly down-scaled adult heads. Comparisons with SAR data in adults are reported using a handset with a patch antenna operating at 900 MHz.