Electromagnetic Absorption by the Human Body from 1 to 15 Ghz
Total Page:16
File Type:pdf, Size:1020Kb
Electromagnetic Absorption by the Human Body from 1 to 15 GHz Gregory Connor Richard Melia A thesis submitted for the Ph.D. degree The University of York Department of Electronics August 2013 Abstract Microwave radiation is emitted by a wide variety of computing, communications and other tech- nologies. In many transport, industrial and medical contexts, humans are placed in close prox- imity to several of these sources of emission in reflective, enclosed cavities. Pseudo-reverberant conditions are created, in which absorption by human bodies can form a significant, even the dominant loss mechanism. The amount of energy stored, and hence the field intensities in these environments depend on the nature of electromagnetic absorption by the human body, so quantifying human absorption at these frequencies is necessary for accurate modelling of both electromagnetic interference and communications path loss in such situations. The research presented here aims to quantify absorption by the body, for the purpose of simulating its effect on the environments listed above. For this purpose, nine volunteer participants are enlisted in a preliminary study in which their height and mass are taken and their electromagnetic absorption cross section is measured in a reverberation chamber. The preliminary study is unable to gather enough data to provide precise measurements during the time that a participant is willing to sit motionless in the chamber. Issues also exist due to power loss in some parts of the equipment. A number improvements are made to both the experimental equipment and methodology, and the study is repeated with a sample of 60 adult volunteer participants. The results are compared to the preliminary data and found to match, once unwanted absorption in the latter has been subtracted. The results are also validated using data from absorption by a spherical phantom of known absorptive properties. The absorption cross section of the body is plotted and its behaviour is compared to several biometric parameters, of which the body’s surface area is found to have a dominant effect on absorption. This is then normalised out to give an absorption efficiency of the skin, which is again compared to several biometric parameters; the strongest correlation is found to be with an estimate for average thickness of the subcutaneous fat layer. These data are used to model the effect of 400 passengers on the Q-factor of an airliner’s cabin. Absorption by the passengers is shown to be the dominant loss mechanism in the cabin, showing the importance of accounting for human absorption when modelling electromagnetic propagation and interference in situations that include human occupants. The relationship be- tween subcutaneous fat and absorption efficiency is suggested for further research, as it promises development of new tools to study body composition, with possible medical applications. Contents List of Figures iv List of Tables ix Acknowledgements x Declaration xi List of Symbols xii List of Abbreviations xiv 1 Introduction 2 1.1 Background...................................... 3 1.2 Aims........................................... 4 1.3 Funding ......................................... 4 1.4 The Absorption Cross Section of the Human Body . ........ 5 1.4.1 UnitsofAbsorption: ACSandSAR . 5 1.4.2 PhantomsoftheHumanBody . 5 1.4.3 Measurements of Absorption by the Body at Microwave Frequencies . 6 1.5 Summary ........................................ 9 2 Theory 10 2.0.1 Overview .................................... 11 2.1 Electromagnetic Absorption by the Human Body . ......... 11 2.1.1 Absorption by tissues: dielectric relaxation and the response of dielectrics totime-varyingfields.............................. 11 2.1.2 Absorption at the surface of the body: The effects of layering . 12 2.1.3 Absorptionbythewholebody . 17 2.2 Methods used in electromagnetic modelling . .......... 19 2.2.1 Fullwavemethods ............................... 19 2.2.2 MieScattering ................................. 19 2.2.3 PowerBalanceModelling . 20 2.3 Construction of an EM environment for measuring human absorption . 23 2.3.1 ReverberationChamberTheory. 23 2.3.2 CalculatingAbsorptionCrossSection . ....... 29 2.3.3 Unstirred energy in a Reverberation Chamber . ........ 30 2.3.4 Coherent Backscattering in a Reverberation Chamber . .......... 31 i 2.3.5 AntennaEfficiency ............................... 32 2.4 Summary ........................................ 33 3 Development of a Methodology for Measuring ACS 34 3.1 Overview ........................................ 35 3.2 Measurement of Absorption Cross Section in a Reverberation Chamber: Initial method ......................................... 35 3.2.1 Accounting for the radiation efficiencies of the antennas .......... 37 3.2.2 Removal of unstirred energy using vector average subtraction . 39 3.2.3 Achievable accuracy of the initial ACS measurement . .......... 39 3.3 Validation of the ACS measurement using a spherical phantom .......... 40 3.3.1 Error analysis of the Mie sphere calculation . ......... 40 3.3.2 Comparison of a measurement of the spherical phantom’s ACS to a two- layerMiesimulation .............................. 42 3.4 Improvement and Optimisation of the Human ACS Measurement......... 45 3.4.1 Equipment ................................... 45 3.4.2 Continuousstirring. 45 3.4.3 Optimisation .................................. 46 3.4.4 Measurementusingoneantenna . 58 3.5 Finalised methodology for measuring ACS ...................... 65 3.5.1 Uncertainty in the estimation of the surface area of the human body . 66 3.5.2 Errorsduetolossesintheantennas . ..... 70 3.5.3 Errors due to subject position within the reverberation chamber . 70 3.5.4 Errorsduetosubjectposture . 72 3.5.5 Errorsduetosubjectclothing. ..... 72 3.5.6 Error analysis of the optimised measurement technique .......... 74 3.6 Comparison of the initial and optimised measurement techniques . 76 3.6.1 Comparing the stepped and stirred measurements . ........ 76 3.6.2 Controllingforthestool . 78 3.6.3 Controlling for the change in antenna position . ......... 81 3.6.4 Conclusions of the comparison between measurements . .......... 88 3.7 Summary ........................................ 88 4 Results of the ACS Measurement 90 4.1 Overview ........................................ 91 4.2 Campaign1....................................... 91 4.2.1 Apparatusandmeasurementprotocols . ...... 91 4.2.2 Samplepopulation .............................. 92 4.2.3 Error analysis: measurement of physical parameters . ........... 92 4.2.4 ResultsofCampaign1measurements . 93 4.3 Campaign2....................................... 95 4.3.1 Measurementprotocols . 95 4.3.2 Apparatus.................................... 96 4.3.3 Error analysis: measurement of physical parameters . ........... 96 4.3.4 Subjects common to both measurement campaigns . ....... 96 4.3.5 Physical characteristics of the experimental sample ............. 98 ii 4.3.6 ResultsoftheCampaign2measurements . 100 4.3.7 Correlation of windowed ACS with biometric data . ........102 5 Data Analysis 107 5.1 Overview ........................................ 108 5.2 PopulationAnalysis .............................. 108 5.2.1 Campaign1sample ..............................108 5.2.2 Campaign2sample ..............................108 5.3 Comparison of measured ACS to literature values . ...........114 5.4 Variation of ACS with Biometric Parameters . .........114 5.5 AbsorptionEfficiency.............................. 118 5.6 The aircraft cabin: An example of the effects of human absorption on the Q- factorofanenclosedenvironment. 128 5.6.1 Q-factorofanairlinercabin. 129 5.6.2 Additionofseatsandpassengers . 131 5.7 Summary ........................................133 6 Conclusions 134 6.1 Development of an Experiment to Measure Human Absorption ..........135 6.2 Measurement of Absorption by a Sample Population of HumanSubjects . .136 6.3 Absorption Efficiency of the Surface of the Body . .........137 6.4 Example Application: Passengers on an Aircraft . ...........138 6.5 FurtherWork..................................... 139 A Terminology and Conventions for modelling Electromagnetic Systems 141 A.1 PerfectDielectrics .............................. 141 A.1.1 RefractiveIndex ............................... 143 A.2 LossyDielectrics ................................ 143 A.2.1 ComplexPermittivity . 143 A.2.2 LossFactorandLossTangent. 144 A.2.3 ComplexPropagationConstant . 144 B Equipment used in Campaign 1 145 Bibliography 146 iii List of Figures 1.1 ACS of NORMAN phantom standing on a conducting groundplane under plane wave incidence from (i)front, polarized horizontally (ii) above, polarized front to back (iii) above, polarized right to left (iv) front, 450 from normal, f/b (v) front, 0 45 , r/l. (αe = direction of polarization) (Findlay & Dimbylow, 2008) . ... 7 1.2 Mean ACS from the NORMAN phantom simulations in Figure 1.1 compared to Uusitupa’s equivalent study of the 72.2 kg VF Male phantom in freespace. 8 2.1 Log-Log plot of Penetration Depth from 100 MHz to 40 GHz of Dry Skin, Infil- trated Fat, and Muscle by Gabriel’s Cole-Cole model . ......... 13 2.2 Three dielectric materials, the basic case of multiple dielectric boundaries . 14 2.3 Comparison of the Pena & Pal 2 layer Mie code to the Matzler 1 layer Mie code 20 2.4 Examples of well-stirred and poorly-stirred RC measurements: Experimental data showing real vs imaginary parts of the S21 coefficient over one rotation of the stirrer in 200 steps. The black diamond marks the origin. ..........