EuCAP 2006 - European Conference on Antennas & Propagation

Session: Session 2A05A - EM Interactions with Biological Tissues and Human Body (04e)
Type: Oral Antenna
Date: Tuesday, November 07, 2006
Time: 08:30 - 12:20
Room: Risso 8
Chair: J. Wiart & Kuster

Seq   Time   Title   Abs No
1   08:30   Failure Modes and Effects Analysis (FMEA) on the RF Performance of Mobile Device Terminals
Tay, R.1; Chavannes, N.2; Futter, P.3; Ng, G.H.1; Kuster, N.2
1Motorola Mobile Devices Design Centre, SINGAPORE;
2Foundation for Research on Information Technologies in Society (IT’IS), SWITZERLAND;
33Schmid & Partner Engineering AG (SPEAG), SWITZERLAND

Introduction and Objectives

During the product design and development phase of mobile telecommunications equipment it is essential to assess and predict potential failures of the device at an early stage and in particular prior to mass production. The information on the expected product reliablity is critical to make the necessary business decision on the fade of the product. Possible sources of performance variation which could results in product not meeting its intended design specifications include design related elements (e.g., contacts between LCD bezel and PCB, loose shields, etc.) or component related issues (e.g., amount of metallic content in plastic, variation of metallic thickness due to process changes, etc.). The conventional approach of "built-test-fix" culture need to be replace with predictive engineering environment in order to compete effectively. The objective of this study was to investigate whether enhanced TCAD environments can act beyond common EM simulation and proof its value to identify failure modes and their effects to RF performance. In this paper, we present our validation with a quadband phone which was introduced into market recently. In this investigation, we embedded the assessment scheme into a regular product design and development environment. We examined the design rigorously with the chosen enhanced TCAD tool on components for their potential dimension variation, manufacturing repeatibility and tolerances for their RF performance variation.


Simulations were conducted by using the FDTD based platform SEMCAD X which provides 3-D ACIS based modeling and OGL rendering on FDTD hardware acceleration enhancements. The original detailed ProE CAD dataset consisted of > 1'000 IGES parts which were directly imported into the TCAD platform and processed without major simplifications. Minimum spatial grid resolutions of 100 µm were used to resolve critical details via the automated grid generation. The availability of hardware accelerated FDTD providing computation speeds up to 600 - 800 MCells/s (million FDTD cells per second), enables a simulation of the described device structure in < 5 minutes. Initial simulations were run to ensure an accurate representation of the phone whereas thorough comparisons to experimental data from the DASY4 scanners and from the anechoic chamber was performed. Subsequently, the availability of accelerated simulations allowed to build up a failure and effect synthesis matrix, concentrating on the design and component related items of interest.

Results and Discussion

After initial numerical results of the phone by inclusion of matching networks from the physical model were validated by measurements of impedance and radiation performance, the failure synthesis table was built up. The analysis of manufacturing defects of various connectors (LCD brackets, keypad connectors, etc.) revealed that certain underestimated connections have a significant influence on the RF performance of the device. Numerous conditions were assessed by measurements; in all cases the simulations were able to predict the experimental findings. Finally, the influence of composite materials and metal coatings was investigated for certain shielding and housing parts. The study concluded that speed and feature enhanced TCAD tools exceed the use as a common simulation tool and revealed its value as a predictive engineering tool for product development environment.

2   08:50   The Relation between SAR and The Electromagnetic Field Distribution for Heterogeneous Exposure Conditions
Neubauer, G.1; Cecil, S.1; Preiner, P.1; Mitrevski, N.1; Vermeeren, G.2; Joseph, W.2; Martens, L.2; Kuehn, S.3; Kuster, N.3
1ARC Seibersdorf Research, AUSTRIA;
2Ghent University, BELGIUM;
3Foundation for Research on Information Technologies in Society, SWITZERLAND

In the frame of this part of the EUREKA project BASEXPO the relation between the specific absorption rate and the electric field strength is investigated for real life exposure conditions arising next to common RF sources, e.g. mobile communication base stations. The first results show that the deviation of the whole body specific absorption rate for heterogeneous exposure conditions from the plane wave conditions is in most cases below 50 %.

3   09:10   Electromagnetic, Thermal and Convection Dosimetry for in Vitro Microwave Exposure System
Lévęque, P.1; Collin, A.1; Cueille, M.1; Perrin, A.2
2Molecular and Cellular Biophysics (CRSSA), FRANCE

In vitro exposure systems must be well controlled to determine the specific absorption rate (SAR, W/kg) distribution in solutions containing suspended or plated cells. The dosimetry is based on electromagnetic, thermal and convection simulations, to take into account thermal effects if necessary. The aim of our work is to predict SAR distribution in several exposure systems and to develop a numerical model to complete the dosimetry taking into account thermal and hydrodynamic phenomena.

4   09:30   Public Exposure to RF from Installed Sources: Site Measurements and Personal Exposimetry
Thuróczy, G.; Molnár, F.; Szabó, J.; Jánossy, G.; Nagy, N.; Kubinyi, G.; Bakos, J.
National Research Institute for Radiobiology and Radiohygiene, HUNGARY

The aim of present study is to measure the RF exposure at the location accessible to public with site measurements and the exposure to RF with personal dosimeter. The site measurements were spot measurements (n=292 sites) with spectrum analyzer and broadband RF antenna in three axis and the resultant was calculated according to the sum of E-field strength vector components. The area and type of the measurement sites were also classified. In the other part of the study the applicability of the RF Dosimeter (RF Personal ExposiMeter - PEM) for human exposure assessment in the real urban environment was investigated. The RF exposure levels, at the living area of general public were collected and evaluated according to the EU Recommendation (1999/519/EC) and ICNIRP reference levels [1,2]. At the present state of data collection and evaluation, the median value of exposure in GSM band was 0.025 mikrowatt/cm2 at outdoor, 0.013 mikrowatt/cm2 at indoor sites respectively. Within 300 m of the base station no clear expression could be found between the exposure levels and distances similarly to other recent studies. The results from personal exposure showed that one third of the participants spent 40-70 % percent of 24h recording time above the detection limits (0.05 V/m) and half of subjects spent less than 10 %. The highest exposure was detected during the travelling period and the lowest in the bed at home.

5   09:50   Modelling of the Interactions between Implanted Auditory Prosthesis and 900 MHz Radiofrequency: Application to Cochlear Implants
Ravazzani, P.1; Franzoni, C.2; Parazzini, M.1; Tognola, G.1
1Istituto di Ingegneria Biomedica, ITALY;
2Universitŕ degli Studi di Genova, ITALY

The rapid diffusion of environmental electromagnetic radiations, in particular in the radiofrequency range, is determining increasing interest on their possible consequent health effects. A particular attention is deserved to those subjects who are considered to be more sensitive such as people who use cochlear implants. Cochlear implants (CI) are electronic devices that restore partial hearing to deaf or severely hard of hearing people. Aim of this study was to analyze the possible functional interactions of RF microwaves with CI. Results of this study show that the electrode array inside a cochlea exposed to an external radiofrequency EMF source at 900 MHz produced peaks on the amplitude of the electric and magnetic fields near the implant and a consequent local increase of the absorbed power and tissue temperature. Moreover, peaks of current density in correspondence of the electrodes were found, with average values of about 4 µA.

6   10:40   Assesment of Human Exposure to Realistic Radio Frequency Source by Means of Analytical and Computational Methodologies
Faraone, A.1; Bit-Babik, G.1; Zaridze, R.2
1Motorola Labs, UNITED STATES;
2Tbilisi State University, GEORGIA

An overview of various analytical and computational methodologies for the assessment of the human exposure to radio-frequency (RF) sources is provided. Different methodologies find application in different exposure conditions, for instance in the near or far field of RF sources, or in indoor or outdoor environments. Exposure metrics can be the specific absorption rate (SAR), which is the basic restriction defined in international safety standards, or derived incident field levels. The thermal load produced by RF energy deposition is also addressed by referring to the numerical solution of the Pennes' bio-heat equation. Suitability of various methodologies to address different exposure conditions is discussed also in relation to present trends in safety and compliance standards.

7   11:00   Development of a Japanese 7-Month Pregnant Woman Model and Evaluation of SAR Generated by Mobile Radio Terminals
Togashi, Toshihiro1; Nagaoka, T.2; Saito, K.1; Takahashi, M.1; Ito, K.1; Watanabe, S.2; Ueda, T.3; Ito, H.1; Osada, H.4
1Chiba University, JAPAN;
2National Institute of Information and Communications Technology, JAPAN;
3University of Tsukuba, JAPAN;
4Juntendo University Shizuoka Hospital, JAPAN

In recent years, mobile radio terminals such as cellular phones and equipments, which use electromagnetic (EM) waves such as the induction heater (IH) cookers, have been increasing, and a large number of people use these equipments, including pregnant women. The influence of the EM waves on the human body is dependent on the frequency. The EM waves mainly contribute to the heating effect, which is generated by the EM energy absorption, above 100 kHz. The specific absorption rate (SAR) has been usually used as a primary dosimetric parameter of the EM waves exposure in the standards. Therefore, the World Health Organization (WHO) considers the problem under the point of view of the SAR in pregnant women. In particular, WHO cares about the SAR in the brain of the fetus.

In addition, in the world, females such as police women, who are pregnant, might wear wireless radio terminals in their abdominal region for their work. It is therefore necessary to evaluate the interaction between the EM waves and the fetus under various situations.
This study presents the development of the numerical model of a Japanese 7-month pregnant woman aimed at evaluating the SAR. The main process is divided into two steps. At first, we prepared 361 images from magnetic resonance imaging (MRI), which were taken in the abdominal region of a Japanese volunteer woman, who was 7-month pregnant. In addition, the intrinsic organs including fetus, amniotic fluid, placenta and uterus of a pregnant woman were identified and their three-dimensional models were realized, by combining the two-dimensional identified imaging data. Secondly, the abdominal region of the non-pregnant woman model was dilated to match the shape of the abdominal region for a pregnant woman based on original MR images (abdominal region) of the pregnant woman. The developed models including the fetus were placed in the abdominal region of the deformed model. Then, a detailed correction of the developed pregnant woman model was performed under the supervision of medical doctors.
In addition, the SAR distributions in the abdominal region of the pregnant woman are calculated by use of the developed model in the case a mobile radio terminal is used close to her abdomen.

As a result, it has been confirmed that the SAR in the anterior part of the developed model is the highest and the SAR in the fetus is low.

8   11:20   SAR Distributions in the Abdomen of a Pregnant Woman Generated in a Bird Cage Coil for the MRI System
Saito, K.1; Kikuchi, S.1; Takahashi, M.1; Ito, K.1; Ikehira, H.2
1Chiba University, JAPAN;
2National Institute of Radiological Sciences, JAPAN

MRI (magnetic resonance imaging) is an effective method for diagnosis of diseases. The MRI system is made of some important units including RF (radio frequency) devices. The RF coil is one of the important parts in the RF unit.

During the imaging, the RF coil radiates EM (electromagnetic) pulse to the human body and in response receives the NMR (nuclear magnetic resonance) signals emitted from the body. Therefore, it is necessary to estimate the SAR (specific absorption rate) in the human body due to the radiated EM energy from RF coils. Especially, in this study, the SAR distributions in the abdomen of pregnant woman are investigated by the FDTD (Finite Difference Time Domain) calculations. Here, we assumed the imaging of a pregnant woman in early period and employed a realistic women model, developed at the National Institute of Information and Communications Technology (NICT), Japan. In addition, a bird cage coil is employed as one of the typical RF coils for MRI system. The operating frequency of the coil is around 64 MHz, which is used in the 1.5 T MRI system.

Before the SAR calculations, the capacitors on the bird cage coil were adjusted so that the resonance frequency of the coil might be around 64 MHz. In the SAR calculations, the realistic women model is inserted into the bird cage coil. As a result of the calculations, relatively high SAR values are observed around the skin, muscle, etc which have a high electrical conductivity and are located close to the surface of the body. Meanwhile, the SAR in the uterus and the ovary is extremely low compared with the above mentioned regions. As a further study, the SAR distributions in a 7-month pregnant woman model, which is developed by NICT, Japan and our laboratory, will be investigated. In this investigation, it is considered that the SAR of the fetus, especially around the head is important.

Moreover, it is necessary to calculate the SAR by the other RF coils used in conventional MRI systems.

9   11:40   SAR Assessment using Tissues Homogeneization Technique
Wiart, J.1; Pradier, A.1; Lacroux, F.1; Lautru, D.2; Wong, M.F.1; Fouad Hanna, V.2
1France Telecom R&D, RESA/FACE, FRANCE;
2Université Pierre et Marie Curie, FRANCE

This paper presents a new approach which allows reducing the computation time as well as the memory requirement. Dealing with SAR assessment the resolution often used nowadays is the millimetre resolution of Magnetic Resonance Imaging (MRI) techniques. Such resolution complies with the numerical constraints up to 5GHz but requires large memory and induces important computation time. Using homogenized tissues and larger grid for the electric field calculation and non homogenized tissues for the conductivity and density the Specific Absorption Rate (SAR) can by estimated with a limited reduction of accuracy. This Tissues Homogenization Technique (THT) has been analyzed with a 1D multilayer structure exposed to a plane wave and to a dipole operated at 900MHz and applied to a 3D child head model exposed to a plane wave. This THT approach allows relaxing the constraints on the grid size of simulations performed with the finite difference in time domain with a limited impact on the accuracy of simulations.

10   12:00   Radiation Effectiveness of GSM Cellular Phone has Little Effects on Average Exposure Dose in GSM Network
Kuehn, S.1; Sulser, C.2; Spaet, D.1; Kelsh, M.3; Sheppard, A.4; Kuster, N.1
1IT’IS Foundation and Integrated Systems Laboratory, SWITZERLAND;
3Exponent Inc., UNITED STATES;
4Asher Sheppard Consulting, UNITED STATES

Epidemiologic research on mobile phones is usually based on the assumption that the exposure is a function of the peak spatial SAR and the average power control (PWC) level, whereby the latter of is directly dependent on the radiation efficiency and the network coverage. Our objective was to investigate which design features of the phone are relevant for possible exposure proxies. We show that the effect of the phone’s radiation efficiency is negligible during movement within a GSM network, since the time-averaged exposure is dominated by the peak power settings during handovers. Hence the peak spatial SAR assessed at maximum output power dominates the cumulative exposure and is a reliable exposure proxy in epidemiologic studies.

11   12:20   RF Dosimetry Using Japanese Anatomical Models
Onishi, T.1; Ebara, H.1; Pongpaibool, P.2; Hamada, L.2; Kiminami, K.1; Uebayashi, S.1; Watanabe, S.2
1NTT DoCoMo, Inc., JAPAN;
2National Institute of Information and Communications Technology (NiCT), JAPAN

Recently, wireless radio devices have been used not only for voice but also for data communication and video phone in practical situations. In addition, the Wireless Local Area Network (LAN) and Bluetooth devices mounted on the PC or the wireless devices have been general. The International Electrotechnical Commission (IEC) has developed on a SAR measurement procedure that uses a flat shape phantom in order to better deal with such varied. Some studies on the SAR calculation of an anatomical model with respect to other regions than the temporal one and comparisons to the SAR calculation results of the flat phantom have been reported. In their studies, the gVisible Human (VH)h model has been mainly used as the anatomically based whole-body realistic human model, which was developed at Brooks Air Force Base. The question is whether the usage of the VH model is enough or not to consider all cases, e.g. sex, size of the body and so on.

Therefore, we employed two Japanese anatomical models, which were developed by National Institute of Information and Communications Technology (NICT), Tokyo, Japan. These two models are realistic high-resolution whole-body voxel models for Japanese adult males and females of average height and weight. The models are segmented into 51 anatomic regions.

In this presentation, we show the calculated results of SAR on three anatomical models by the FDTD method when the antenna exists close to the trunk of the body. The results are normalized to the SAR of the flat shape phantom as IEC has proposed. Then, multilayered flat shape phantoms are introduced to evaluate the effects of tissue thickness and tissue compositions. Finally, the factor, which is the SAR ratio of between the multilayered and the flat phantoms, is obtained with a statistical approach.