Biological Effects - RF/Microwave Interaction with Biological Tissues - page 133

The human skin exhibits an emissivity close to 1.0. Determination of surface
temperature as well as temperature variations over the surface depends
directly upon the surface emissivity. The following parameters have to be accu-
rately determined prior to starting the measurements: emissivity, ambient
temperature, atmospheric temperature, relative humidity of the air, and dis-
tance [7].
Calibration is of course important. At the beginning of a set of measure-
ments, the temperature indicated by the IR camera has to be calibrated in
several points of the image using data read by thermometers. The tempera-
ture of a reference point is measured using a thermometer, for instance a ther-
mocouple. The emissivity is then altered until the temperature calculated by
the IR camera agrees with the thermometer reading. This is the emissivity
value of the reference object. The temperature of the reference object must,
however, not be too close to the ambient temperature for this to work. The
procedure for measuring a phantom of course offers more flexibility than that
for a human head. Measurements can, for instance, be calibrated by painting
part of the phantom surface in black and comparing results obtained from the
bare part of the phantom to that of the black part of the phantom [118].
It is rather usual—and recommended—to first operate with canonical
models: spheres, cylinders, and homogeneous, single-layer phantoms. This
permits the comparison between calculated and measured values.
Operating with phantoms has demerits. One is that high power is in general
necessary for the experiment. Hence, actual telecommunication mobile equip-
ment cannot be tested. Another is that three-dimensional measurements
cannot be done. Therefore, it is not efficient to measure the 1- or 10-g average
SAR by thermographic experiments. On the other hand, the use of thermo-
graphic experiments is efficient in the SAR evaluations for a medium of com-
plicated shape such as the inner ear and earlobe.

3.11.3 Computational Methods for SAR Evaluation

There are a number of studies carried out by numerical simulation, in partic-
ular about SAR values in the human head when exposed to microwave radi-
ation from hand-held devices [18]. Several papers in [116] cover this subject
while mentioning a number of references.
Exposures to radiation from various types of antennas, effect of frequency,
and separation distances, for example, have been examined. The effect of the
human body on the absorption by the human head has also been considered,
in particular the conditions under which it is necessary to take the effect of
the human body into account in the calculations [119].
The frequency-domain method of moments (MoM) and FDTD method are
widely used for analyzing complex EM problems such as biological structures.
Treating electrically large and/or penetrable structures with the MoM requires
extensive computational resources. On the other hand, accurate models of
wire and curved structure are difficult to implement in the FDTD method.