Showing papers by "Dragan Poljak published in 2021"

Intercomparison of Calculated Incident Power Density and Temperature Rise for Exposure From Different Antennas at 10–90 GHz

Abstract: Recently, international exposure guidelines/standards for human protection from electromagnetic fields were revised. For frequencies between 6–300 GHz, the permissible incident power density is defined as the reference level, which is derived from a new metric “absorbed/epithelial power density” based on thermal modeling. However, only a few groups computed the power density and the resultant temperature rise at frequencies greater than 6 GHz, where their exposure conditions were different. This study presents the first intercomparison study of the incident power density and the resultant temperature rise in a human body exposed to different frequency sources ranging from 10 to 90 GHz. This intercomparison aims to clarify the main causes of numerical calculation errors in dosimetry analyses through objective comparison of computation results from six organizations using their numerical methods with various body and antenna models. The intercomparison results indicate that the maximum relative standard deviation (RSD) of peak spatially averaged incident power densities for dipole and dipole array antennas is less than 22.1% and 6.3%, respectively. The maximum RSD of the heating factor, which is defined as the ratio of the peak temperature elevation at the skin surface to the peak spatially averaged incident power density in free space, for dipole and dipole array antennas is less than 43.2% and 41.2%, respectively. The deviations in the heating factors caused by different body models and dielectric/thermal parameters are within 33.1% and 19.6% at 10 and 30 GHz, respectively, when the antenna-to-skin model distance is greater than 5 mm. Under this condition (>5 mm), the deviation in the heating factors caused by different antenna models at 30 GHz does not exceed 26.3%. The fair agreement among the intercomparison results demonstrates that numerical calculation errors of dosimetry analyses caused by the definition of spatially averaged incident power density are marginal.

Stochastic-deterministic boundary element modelling of transcranial electric stimulation using a three layer head model

Abstract: This paper deals with the boundary element (BE) approach to modelling of transcranial electric stimulation as an alternative to the widely used finite element method (FEM). The advantages of the BE approach are listed in the paper and demonstrated on a computational example. The formulation is based on the quasi-static approximation of currents and voltages induced in living tissues while the head is represented by a three layered model consisting of skin, skull and brain tissues. Another contribution is the fact that the uncertainty present in the tissue conductivity values is taken into account by modelling them as uniformly distributed random variables. The stochastic collocation method (SCM) is applied for propagation of the uncertainty to the output electric scalar potential. Accordingly, stochastic moments are computed and sensitivity analysis is carried out using the ANalysis Of VAriance approach (ANOVA). The results given in the paper show the efficiency of the BE-SCM combination. Inspecting the results obtained from the proposed BE-SCM approach it is clear that the confidence intervals are appreciably larger in the interior tissues. The impact of the skull's conductivity is shown to be negligible for most of the observation points while the skin and brain conductivities have a significant impact on the output value.

Influence of Uncertainty of Body Permittivity on Achievable Radiation Efficiency of Implantable Antennas—Stochastic Analysis

Abstract: Design of medical sensor implants is very challenging due to numerous restrictions in terms of size and biological acceptability. In addition, large absorption and reflection of electromagnetic waves in body tissues significantly degrades the communication link capabilities. The usual assumption in calculating the link budget is that the permittivity and conductivity of the body tissues are known. However, biological tissues show considerable diversity in structure and consequently in dielectric properties. The first aim of this article is to determine how much the uncertainty of the body tissues’ constitutive parameters affects the prediction of the achievable radiation efficiency of the implant. The second aim is to understand the loss mechanism and thus which tissue permittivity and/or conductivity has the dominant influence on losses. The analysis is done by considering a spherical multilayer human phantom and applying spherical wave decomposition proposed in a previous article. The stochastic collocation method is used to estimate the uncertainty of power density outside the human body, and sensitivity on the uncertainty of each tissue parameter is performed using the analysis of variance (ANOVA) approach. Several test cases were considered and the results clearly indicate which constitutive parameters have dominant effect on uncertainty.

Stochastic analysis of the electromagnetic induction effect on a neuron’s action potential dynamics

Abstract: The presence of time-varying electromagnetic fields across a neuron cell may cause changes in its electrical characteristics, most notably, in the action potential dynamics. This phenomenon is examined by simulating electrophysiology of a single cortex neuron. Magnetic flux is captured by using a polynomial approximation of a memristor embedded into Hodgkin–Huxley model, equivalent electrical circuit of a neuron cell. Bifurcation analysis is carried out for two different electrical modes associated with the nature of the external neuronal stimulus. Aiming to determine the true influence of the variability of ion channels conductivity, the stochastic sensitivity analysis is undertaken post hoc. Additionally, numerical simulations are enriched with uncertainty quantification, observing values of ion channel conductivity as random variables. The aim of the study is to computationally determine the sensitivity of the action potential dynamics with respect to changes in conductivity of each ion channel so that the future experimental procedures, most often medical treatments, may be adapted to different individuals in various environmental conditions.

Absorbed Power Density at the Surface of Planar Tissue due to Radiation of Dipole Antenna

Abstract: The paper deals with an analytical/numerical assessment of the absorbed power density (S ab ) in the planar model of the human tissue due to radiation of dipole antenna. The effect of two media interface (air-body) is taken into account via the Fresnel plane wave reflection coefficient. The antenna is analyzed by solving the Pocklington integrodifferential equation and related integral expressions for the radiated electric and magnetic field, respectively. Some illustrative numerical results are given for the radiated field on the model surface and S ab pertaining to the frequency range of the fifth generation (5G) mobile communication systems.

Stochastic-Deterministic Thermal Dosimetry Below 6 GHz for 5G Mobile Communication Systems

Abstract: The article deals with stochastic-deterministic thermal dosimetry in lower portion of GHz frequency range for human exposure to radiation from 5G communication systems. Deterministic assessment of a temperature distribution in multilayered head model is based on the finite element solution of Pennes equation while the stochastic analysis uses the stochastic collocation approach. Thermal conductivity and volumetric blood perfusion rate of scalp, skull, and brain tissues are modeled as uniformly distributed random parameters, while values of other thermal input parameters are fixed to their respective average values. The stochastic mean and variance of the heating factor at 3.5 GHz are computed and confidence interval is estimated for some observation points of interest. By using “one-at-a-time” and analysis of variance principles, sensitivity analysis of input parameter set is carried out. Some illustrative numerical results stemming from deterministic and stochastic model, respectively, are presented. When observing the impact of input parameters on heating factor values in the direction of incidence, blood perfusion of scalp is dominant with exception in the interior head parts, where the impact of the brain blood perfusion prevails. For the exposure scenario of interest, the temperature elevation is within the limits given by basic restrictions.

Analysis of curved thin wires radiating over a layered medium and some engineering applications

Abstract: The paper deals with modelling of radiation from arbitrarily shaped thin wires above a multilayer and related engineering applications. The formulation is based on the set of corresponding Pockling...

Application of Automatic Differentiation in Electromagnetic Dosimetry - Assessment of the Absorbed Power Density in the mmWave Frequency Spectrum

Abstract: This paper introduces the concept of automatic differentiation in the evaluation of the absorbed power density in the mmWave frequency spectrum for the new generation of mobile telecommunication technology. Automatic differentiation has been shown to be far superior over numerical differentiation by means of speed and accuracy. To demonstrate the full capacity of the proposed method, a comprehensive analysis of computing the absorbed power density on the surface of irradiated human skin in various configurations is presented.

Unit Cube Test for Double Surface Integrals in Frequency Domain Integral Equation Formulations

Abstract: The numerical solution of the frequency domain integral equation based formulations require the solution of various double surface integrals. The evaluation of such integrals depends on the distance between the source and observation points, respectively. Many typical integrals can be solved using a numerical integration or a combination of numerical and analytical approach. The paper is on the unit cube test developed for testing various combinations of double surface integrals. Several combinations of source and observation triangles are considered including the far term and near terms, respectively. The numerical solution of integral is facilitated by Gaussian quadrature utilizing the Dunavant's rules for triangles. The extensive tests are carried out to investigate the solution convergence with respect to the increasing frequency and the various utilized discretization schemes.

Uncertainty Estimation of Achievable Radiation Efficiency of Implantable Antennas

Abstract: Designing antennas for implanted medical devices is a very complex task where one needs to tackle inherently large EM absorption in the human tissue and very limited available space for the antenna. In addition, designers need to carefully limit the power output of the devices due to health safety regulations and small battery volume. Furthermore, the initial designs assume that permittivity and conductivity of different types of human tissues are known. Unfortunately, these values have a significant variation depending on the age, sex and health conditions of the actual patient. Such variations need to be taken into account in initial designs and in this paper we analyze the influence which the uncertainty of the tissues dielectric parameters have on the predictable power density in implanted antenna scenarios.

Study on the Suitability of Numerical Integration at 5G Frequencies Using Unit Cube Test

Abstract: The assessment of human exposure to 5G mobile systems requires the use of numerical methods such as method of moments (MoM). MoM solution of the integral equation based formulation results in the fully populated system matrix whose filling time represents the most time-consuming operation for many problems. As the matrix size is directly related to the frequency of the problem, the numerical solutions in the 5G frequency range will thus put a tremendous computational cost both in terms of matrix fill time but also in the memory allocation. The paper is on the investigation of suitability of numerical integration at 5G frequencies using a unit cube test. The example of double surface integral in case of a combination of coplanar triangles is numerically solved using the Dunavant’s symmetrical quadrature rules for triangles. The results show that depending on the discretization scheme, lower integration orders could be utilized, resulting in the reduced matrix fill time without actually sacrificing the accuracy.

A Note on the Calculation of the Power Flow along a Straight Thin Wire Scatterer Horizontally Located above a Lossy Ground

Abstract: The paper deals with the calculation of active, reactive and apparent power along a horizontal straight thin wire scatterer of finite length. Once the current distribution and voltage distribution along the straight scatterer are determined by solving the Pocklington integro-differential equation and generalized telegrapher’s equation respectively, the active power, reactive power and apparent power are calculated from the well-known formulas arising from the circuit theory. Some illustrative computational examples are presented in the paper.

Assessment of Transmitted Power Density due to Radiation from Dipole Antenna of Finite Length : Part I: Theoretical background and current distribution

Abstract: The paper deals with an analytical/numerical approach to determine the transmitted power density (TPD) in flat human tissue model exposed to the radiation of dipole antenna. Human tissue is represented by muscle properties at a frequency of interest. The influence of two-media interface is taken into account via the Fresnel plane wave reflection/transmission approximation. Numerical procedures pertain to Galerkin-Bubnov Indirect Boundary Element Method (GB-IBEM). Some illustrative examples for current distribution at various frequencies and distances of dipole antenna from the interface are given.