IEC IEEE 62704-1 pdf free download
IEC IEEE 62704-1 pdf free download.Determining the peak spatial-average specific absorption rate (SAR) in the human body from wireless communications devices, 30 MHz to 6 GHz – Part 1: General requirements for using the finite-difference time-domain (FDTD) method for SAR calculations.
5 Finite-difference time-domain method — basic definition
This document applies the finite-difference time-domain (FDTD) method to calculate SAR. FDTD is extensively used for bio-electromagnetic calculations, dating back to 1975 for the analysis of microwave heating of a human eye (Taflove and Brodwin [4]). In order to determine applicability of this document, it shall be clear whether a specific numerical solution of Maxwell’s equations qualifies as an implementation of FDTD.
This is not entirely straightforward since there have been many variations of the FDTD method developed over the time since Kane Yee’s original paper in 1966 [5], which may be taken as defining the fundamental FDTD method. Since then, the method has received significant attention in the scientific community. Numerous articles on the method and its extensions and applications have been published. Several text books on the FDTD method recapitulate its fundamentals, its further developments and its use ([6], [7], [8]). Details on the application of the FDTD method are given in Annex C.
This document is based in part on the extensive validation of FDTD as it applies to SAR calculation reported in the literature. Therefore, for the application of this document, the implementation of FDTD in the spatial region where SAR is being calculated shall include the following characteristics.
• The electric field components are spatially located on the edges of a Cartesian coordinate system structured mesh composed of rectangular parallelepipeds.
• The magnetic field components are spatially located on the edges of a Cartesian coordinate system structured mesh composed of rectangular parallelepipeds which is offset from the electric field mesh by a half-voxel in each direction.
• The numerical solution method is a finite-difference approximation of the Maxwell curl equations using central differences which are (at least globally) second-order accurate and would therefore include non-uniform meshes, which are also second-order accurate (Monk and Süli [9]).
• The numerical solution method solves for both electric and magnetic fields by a fully explicit leapfrog time-stepping process.
• Gauss’s laws are implicitly enforced, the fields are divergence-free, and charge is conserved.
• The time-stepping algorithm is non-dissipative. Thus, there is no spurious decay of energy due to non-physical artefacts of the algorithm, and artificial dissipation is not required for stability.
NOTE Complex numerical dispersion in non-uniform meshes leads to spurious rise or decay of field amplitudes [9]. For the applications of the scope of this document, these effects are generally negligible. The impact of the numerical dispersion is assessed within the uncertainty evaluation (Clause 7).
Artificial dissipation may be introduced in certain finite-difference algorithms in order to prevent or limit the spurious generation of energy in the computational domain which can lead to numerical instabilities. Such artificial dispersion is not required for the FDTD algorithm as defined here.
The basic requirements on the FDTD method are given in Annex A. Other methods and/or extensions of FDTD that satisfy the above definitions but provide additional features may be applied provided that they have been fully validated according to the procedures in this document; code validation and benchmarks, etc. However, these alternative methods shall provide at least the same accuracy as the FDTD method defined above. The finite integration technique (FIT) in time domain can be regarded as an extension of FDTD in this context.
IEC IEEE 62704-1 pdf download.
