Validation of Specific Absorption Rate (SAR) Calculation in XFdtd^® 6.0.

The recently published IEEE document P1528/D1.2 titled Recommended Practice for Determining the Peak Spatial-Average Specific Absorption Rate (SAR) in the Human Head from Wireless Communications Devices: Measurement Techniques, sets standards for measuring the SAR generated by wireless devices. One section of this document regarding the calibration of the measurement system contains a table of reference SAR values. Here the calibration approach is simulated in XFdtd for comparison of the reference SAR values.

The subject under test is a flat phantom comprised of a plastic shell and a tissue-equivalent liquid. The phantom size is set in the document as 22.5x15 cm for the frequency range of 835-3000MHz. For lower frequencies a phantom of size 0.6x0.4 wavelengths is used. The material parameters for the tissue equivalent liquid vary with frequency and are given in Table 1. The plastic shell should have a relative permittivity less than 5 and a loss tangent less than 0.05. The plastic shell thickness is defined as 2mm for frequencies in the 800-3000MHz range and 6.5mm for lower frequencies. The tissue equivalent liquid shall have a minimum depth of 15cm.

The phantom is to be exposed to the fields of an appropriately sized dipole (see Table 2) which is spaced 15 mm from the shell/liquid interface for frequencies less than or equal to 1000MHz and 10 mm from the interface for higher frequencies. The local SAR and 1 and 10 gram average SAR values are to be determined for the location directly above the feed of the dipole for a 1 W input power.

For the XFdtd simulation a 1mm cubical grid was chosen for all simulation except 300MHz where a 1.5mm grid was used. The plastic shell was defined as a dielectric with a relative permittivity of 3.7 and no electrical conductivity. The phantom and shell were sized appropriately based on the requirements of the document. The dipole antenna was defined as two cylinders of the specified radius with a single FDTD cell space between them for the feed. The solid view of the XFdtd 6.0 geometry is shown in Figure 1 while Figure 2 shows the actual mesh used in the calculation. The applied excitation was a voltage source with a sinusoidal input. All calculations were run for 16 full-amplitude cycles of the sine wave. Following the simulation, the input power was adjusted from the computed value to the specified 1W. The resulting values are shown in Table 3 compared to the reference values in the P1528 document. Images of the Local SAR, 1g Averaged SAR, and 10g Averaged SAR in the first plane of the liquid (with the shell hidden) for the 1800MHz case are shown in Figures 3, 4, and 5.

Table 1. Tissue-equivalent liquid parameters.

Table 2: Dipole antenna dimensions

Table 3: Comparison of Reference and Computed (with XFdtd 6.0) SAR Results for the Flat Phantom test object