SAR Analysis at 900MHz of Layered and Homogeneous

Phantoms Incorporating Variations to the Thickness of Fat and

Muscle Tissues, at Multiple Dipole Distances

Stuart Nicol*, Christopher Penney**, Raymond Luebbers**, and Jacek Wojcik***

APREL Laboratories*, Remcom Inc**, Spectrum Sciences Institute***

ABSTRACT

This paper discusses results from a study conducted into the

effects of SAR (Specific Absorption Rate) when comparing

homogeneous and heterogeneous models against a dipole

which was positioned from 10mm up to 200mm in distance

II. HOMOGENEOUS PHANTOM

form the centre of the dipole to the tissue boundary. Typical

methodologies were employed [1, 2, 3] with regards the

The Flat Phantom test geometry from IEEE 1528 (

reference homogeneous simulation models which followed

Figure 1) is used with the dielectric parameters defined

the accepted methodologies for defining reference SAR

(eps=41.5, sig=0.97, rho=1000). The antenna is a simple

values used for experimental SAR system validation and

(two cylinder) half-wave dipole at 900MHz spaced 10, 25,

calibration [1, 2]. Further studies utilized a heterogeneous

50, 100, and 200mm from the phantom.

model based on skin, fat and muscle. The purpose of the

study was to investigate the possible effects of distance and

model layering on SAR. The intent of the study was to

establish  if  the  accepted  homogeneous  model  was

conservative at the typical GSM band.

Index Terms: Dosimetry, SAR, Validation, radio frequency

exposure, GSM, Homogeneous and Heterogeneous.

I. INTROUCTION

As part of this study it was decided to assess the conservative

SAR for homogeneous and heterogeneous phantom models

and a series of problems were created and run within the

Remcom Inc XFDTD [5] numerical software package. The

intention of this study was to investigate the effect of

multiple tissue layers on final evaluated SAR and compare

against the established conservative homogeneous model so

Figure 1: The homogeneous phantom shown with the

as to determine if the intention of the accepted measurement

dielectric shell and simulating liquid above the dipole.

standard(s) [1, 2, 7] of performing conservative SAR

analysis were being met at the GSM frequency [6].

III. LAYERED PHANTOM

Numerical problems were run with variations to the skin, fat,

and muscle thicknesses so as to identify the parameters

A flat, layered, rectangular phantom comprising of skin, fat

where the conservative SAR can be found within a layered

and muscle has been used (Figure 2).  The skin layer

heterogeneous model.

thickness has been varied from 1 to 3mm in 1mm steps. The

fat layer is set at four thicknesses values: 20mm, 30mm,

40mm, and 50mm. The fat layer is followed by a muscle

When the conservative model was identified and defined

layer of 100, 90, 80, and 70mm thickness, depending on the

further numerical problems were run where changes to the

fat thickness selected for the problem. Dielectric parameters

dipole separation distance from 10mm up to 200mm were

used are fat: eps=11.33, sig=0.11, rho=916; skin: eps=41.41,

made and conservative SAR was assessed. Further numerical

sig=0.87,  rho=1125;  muscle:  eps=55.03,  sig=0.94,

problems were then ran using the IEEE 1528 [1]

rho=1046.9.

homogeneous phantom specifications and protocols so as to

compare the heterogeneous [7, 8, 9] findings for conservative

SAR against accepted global practices. All problems were

run using a commercially available FDTD program

(Remcom Inc XFDTD) [5, 6].

1

Submitted to Asia Pacific EMC Taipei City, Taiwan 2005