edVirtus Reference

IEC/IEEE 63195-2:2022—Assessment of Power Density of Human Exposure to Radio Frequency Fields from Wireless Devices in Close Proximity to the Head and Body (Frequency Range of 6 GHz to 300 GHz)—Part 2: Computational Procedure

Preview: Learn more about IEC/IEEE 63195-2 and its computational procedure for assessing power density from wireless devices used close to the head or body.

IEC/IEEE 63195-2:2022 specifies computational procedures for conservative and reproducible calculation of radiofrequency power density incident on the human head or body from transmitting devices. It applies from 6 GHz to 300 GHz and addresses equipment used with radiating parts at distances up to 200 mm from the head or body.

The standard is an IEC/IEEE dual-logo publication. It complements the measurement procedure in Part 1 by establishing how numerical models may be built, solved, validated, and reported for conformity assessment against an applicable maximum power-density requirement. It is a method standard and does not itself determine which exposure limit has legal force.

Scope and application

The procedures cover hand-held and body-worn RF transmitting devices intended for use by a significant majority of the population. Products can contain single or multiple transmitters or antennas and can operate alone, with accessories, with other transmitting or non-transmitting equipment, or while embedded in garments.

Examples include mobile telephones, transmitters in personal computers, desktop and laptop equipment, and multi-band or multi-antenna products. The procedures do not apply to power-density assessment of fields emitted or altered by implanted devices or objects.

The applicable frequency range reflects the growing use of millimetre-wave communications and the shift from mass-averaged SAR to surface-related power-density quantities at higher frequencies. The applicable exposure framework or regulator defines the metric, averaging area, time basis, numerical limit, and compliance decision rule.

Computational methods

The standard identifies finite-difference time-domain (FDTD) and finite element methods (FEM) as computational techniques for determining electromagnetic quantities by solving Maxwell’s equations. Both methods can represent complex antennas, materials, device structures, near fields, and evaluation surfaces, but each requires careful selection of model extent, discretisation, boundary conditions, source representation, and solver controls.

A numerical result is not established merely by running commercial software. The model must be physically appropriate, sufficiently resolved, converged, and validated for the frequencies, geometries, and quantities being assessed. The user must understand both the electromagnetic problem and the limitations of the chosen implementation.

Device and source representation

The computational model must represent the transmitting device with sufficient fidelity to reproduce the fields relevant to exposure. Important features can include the antenna geometry, feed and matching structure, nearby conductors, enclosure, display, battery, ground plane, dielectric materials, and accessories. Simplification is acceptable only where its effect is understood or conservatively bounded.

Source power, phase, amplitude, modulation, antenna state, and simultaneous-transmission conditions must correspond to the compliance configuration. Normalisation from a simulated excitation to the maximum permitted device power must be justified. Where beamforming or multiple-input multiple-output operation is used, the assessed states must bound the exposure produced by the available beams and combinations.

Computational domain and resolution

The solution domain must be large enough and terminated so that artificial reflections do not materially affect the result. Spatial discretisation must resolve the shortest relevant wavelength in the model materials, small antenna features, strong field gradients, and the evaluation surface. Excessively coarse cells or elements can underestimate local maxima or distort phase and amplitude.

Convergence is demonstrated by showing that reasonable refinement of the mesh, time step, domain, boundary placement, or solver criteria does not materially change the reported conformity quantity. Numerical stability and energy balance checks can provide additional evidence that the solution is reliable.

Evaluation surface and power density

The field is evaluated on a defined surface representing the required assessment location relative to the head or body. Electric and magnetic field information is processed to determine the specified incident power-density quantity. The evaluation surface, orientation, distance, sampling density, and interpolation method must be consistent with the standard and the applicable exposure limits.

Because close-proximity fields can be highly non-uniform, the assessment searches for the maximum spatially averaged result. The averaging algorithm must preserve significant local structure and use the prescribed averaging area and surface treatment. A convenient mean over the whole model is not a substitute for the required local maximum.

Material and geometric assumptions

Unlike a detailed absorbed-power calculation in anatomical tissue, an incident power-density assessment focuses on the field arriving at the evaluation surface. Nevertheless, nearby device materials, body representation or boundary, and separation geometry can affect the field distribution. Assumptions must match the prescribed procedure and the compliance case.

Manufacturing tolerances and user configurations can alter antenna tuning or separation. The model should address variations capable of increasing exposure, either explicitly or through conservative margins. Reliance on a fixed separation distance must be supported by the product design, instructions, accessory, or regulatory classification.

Validation

Validation establishes confidence that the computational implementation predicts the relevant fields correctly. It can involve comparison with analytical solutions, canonical problems, reference models, or measurements of the actual device or a representative validation source. Agreement criteria and the range over which validation applies must be documented.

Validation should test the complete modelling chain, including geometry import, material assignment, excitation, solver, field extraction, transformation to power density, spatial averaging, and normalisation. Agreement at one frequency or position does not necessarily validate a broadband model or every antenna state.

Uncertainty and conservatism

The uncertainty analysis considers numerical discretisation, convergence, boundary conditions, material properties, geometry, source power, antenna representation, post-processing, validation discrepancy, and device variability. Components are combined and reported in a form suitable for the compliance decision rule.

Conservative assumptions can be used to bound uncertain inputs, but indiscriminate stacking of unrealistic worst cases can make a model uninformative. The preferred result is a traceable assessment in which each margin has a clear physical or regulatory basis.

Multiple transmitters and mixed-frequency devices

Simultaneous transmitters must be treated under the summation rules of the applicable exposure framework. The standard’s published scope points to IEC/IEEE 62704-1 and IEC/IEEE 62704-4 for computational SAR assessment of simultaneous transmitters below 6 GHz. A product spanning lower and higher bands may therefore require both SAR and power-density analyses.

Software controls that prevent simultaneous maximum operation or limit time-averaged power may form part of the assessment only where their operation is reliable, bounded, and acceptable to the responsible regulator. The compliance file should identify these dependencies clearly.

Relationship to measurement

Part 1 provides the corresponding measurement procedure. Computation can examine internal design details, numerous beam states, and configurations that are difficult to probe, and it can support optimisation before hardware is final. Measurement can provide an independent check on the model and a direct assessment of production hardware.

The best route depends on the device, regulatory regime, available validation evidence, and uncertainty. A hybrid approach is often valuable: measurement validates selected conditions while computation extends the assessment to a justified set of configurations.

Documentation and practical significance

A complete report identifies the equipment, frequencies, operating modes, antennas, powers, geometries, solver and version, model construction, material properties, mesh or elements, domain and boundaries, convergence evidence, validation, evaluation surface, averaging procedure, uncertainty budget, results, and compliance conclusion. Files should be retained in a form that permits technical review and reproduction.

IEC/IEEE 63195-2 provides a disciplined computational route for assessing close-proximity devices from 6 GHz to 300 GHz. For RF radiation safety, its value lies in turning numerical electromagnetic modelling into auditable conformity evidence. Its use requires specialist competence and must remain tied to the applicable exposure limits and equipment-authorisation requirements.