SURROGATE MODELING OF INDOOR DOWN-LINK HUMAN EXPOSURE BASED ON SPARSE POLYNOMIAL CHAOS EXPANSION

Human exposure induced by wireless communication systems increasingly draws the public attention. Here, an indoor down-link scenario is concerned and the exposure level is statistically analyzed. The electromagnetic field (EMF) emitted by a WiFi box is measured and electromagnetic dosimetry features are evaluated from the whole-body specific absorption rate as computed with a Finite-Difference Time-Domain (a.k.a. FDTD) code. Due to computational cost, a statistical analysis is performed based on a surrogate model, which is constructed by means of so-called sparse polynomial chaos expansion (PCE), where the inner cross validation (ICV) is used to select the optimal hyperparameters during the model construction and assess the model performance. However, the ICV error is optimized and the model assessment tends to be overly optimistic with small data sets. The method of cross-model validation is used and outer cross validation is carried out for the model assessment. The effects of the data preprocessing are investigated as well. Based on the surrogate model, the global sensitivity of the exposure to input parameters is analyzed from Sobol' indices.

[1]  Joe Wiart,et al.  Statistical analysis and surrogate modeling of indoor exposure induced from a WLAN source , 2017, 2017 11th European Conference on Antennas and Propagation (EUCAP).

[2]  J. Wiart,et al.  Influence of Plane-Wave Incidence Angle on Whole Body and Local Exposure at 2100 MHz , 2011, IEEE Transactions on Electromagnetic Compatibility.

[3]  Anne Dutfoy,et al.  A generalization of the Nataf transformation to distributions with elliptical copula , 2009 .

[4]  Kunal Roy,et al.  A Primer on QSAR/QSPR Modeling: Fundamental Concepts , 2015 .

[5]  B. Sudret,et al.  An adaptive algorithm to build up sparse polynomial chaos expansions for stochastic finite element analysis , 2010 .

[6]  Dongbin Xiu,et al.  The Wiener-Askey Polynomial Chaos for Stochastic Differential Equations , 2002, SIAM J. Sci. Comput..

[7]  Joe Wiart,et al.  A new surrogate modeling technique combining Kriging and polynomial chaos expansions - Application to uncertainty analysis in computational dosimetry , 2015, J. Comput. Phys..

[8]  M Martínez-Búrdalo,et al.  FDTD assessment of human exposure to electromagnetic fields from WiFi and bluetooth devices in some operating situations. , 2009, Bioelectromagnetics.

[9]  Joe Wiart,et al.  Specific absorption rate assessment near a base-station antenna (2,140 MHz): some key points , 2008, Ann. des Télécommunications.

[10]  Joe Wiart,et al.  Statistical model of the human RF exposure in small cell environment , 2019, Ann. des Télécommunications.

[11]  Joe Wiart,et al.  Children exposure to femtocell in indoor environments estimated by sparse low-rank tensor approximations , 2018, Ann. des Télécommunications.

[12]  P. Dimbylow,et al.  SAR in children from exposure to wireless local area networks (WLAN) , 2012, 2012 Asia-Pacific Symposium on Electromagnetic Compatibility.

[13]  Y. C. Pati,et al.  Orthogonal matching pursuit: recursive function approximation with applications to wavelet decomposition , 1993, Proceedings of 27th Asilomar Conference on Signals, Systems and Computers.

[14]  G. D. Wyss,et al.  A user`s guide to LHS: Sandia`s Latin Hypercube Sampling Software , 1998 .

[15]  Yoshua Bengio,et al.  Model Selection for Small Sample Regression , 2002, Machine Learning.

[16]  Bruno Sudret,et al.  Global sensitivity analysis using polynomial chaos expansions , 2008, Reliab. Eng. Syst. Saf..

[17]  J. R. Quinlan.  Learning With Continuous Classes , 1992 .

[18]  Joe Wiart,et al.  SURROGATE MODELING OF STOCHASTIC FUNCTIONS-APPLICATION TO COMPUTATIONAL ELECTROMAGNETIC DOSIMETRY , 2018, International Journal for Uncertainty Quantification.

[19]  Luc Martens,et al.  Use of artificial intelligence to model exposure to radiofrequency electromagnetic fields based on sensor network measurements , 2018 .

[20]  Bruno Sudret,et al.  Adaptive sparse polynomial chaos expansion based on least angle regression , 2011, J. Comput. Phys..

[21]  J. Freidman,et al.  Multivariate adaptive regression splines , 1991 .

[22]  P Ravazzani,et al.  Exposure assessment of one-year-old child to 3G tablet in uplink mode and to 3G femtocell in downlink mode using polynomial chaos decomposition , 2016, Physics in medicine and biology.

[23]  R. Tibshirani,et al.  Least angle regression , 2004, math/0406456.

[24]  T. Ishigami,et al.  An importance quantification technique in uncertainty analysis for computer models , 1990, [1990] Proceedings. First International Symposium on Uncertainty Modeling and Analysis.

[25]  W. V. Harper,et al.  Sensitivity/uncertainty analysis of a borehole scenario comparing Latin Hypercube Sampling and deterministic sensitivity approaches , 1983 .

[26]  David Hinkley,et al.  Bootstrap Methods: Another Look at the Jackknife , 2008 .

[27]  Marta Parazzini,et al.  Study of the Influence of the Orientation of a 50-Hz Magnetic Field on Fetal Exposure Using Polynomial Chaos Decomposition , 2015, International journal of environmental research and public health.

[28]  M. Parazzini,et al.  Radio Frequency Electromagnetic Fields Exposure Assessment in Indoor Environments: A Review , 2019, International journal of environmental research and public health.

[29]  Alvaro Augusto Almeida de Salles,et al.  SAR simulations of EMF exposure due to tablet operation close to the user's body , 2015, 2015 SBMO/IEEE MTT-S International Microwave and Optoelectronics Conference (IMOC).

[30]  P. Dimbylow,et al.  SAR in a child voxel phantom from exposure to wireless computer networks (Wi-Fi) , 2010, Physics in medicine and biology.

[31]  Knut Baumann,et al.  Reliable estimation of prediction errors for QSAR models under model uncertainty using double cross-validation , 2014, Journal of Cheminformatics.

[32]  Niels Kuster,et al.  The Virtual Family—development of surface-based anatomical models of two adults and two children for dosimetric simulations , 2010, Physics in medicine and biology.

[33]  Joe Wiart,et al.  Stochastic dosimetry for the assessment of the fetal exposure to 4G LTE tablet in realistic scenarios , 2017, 2017 XXXIInd General Assembly and Scientific Symposium of the International Union of Radio Science (URSI GASS).

[34]  Harald Martens,et al.  Reducing over-optimism in variable selection by cross-model validation , 2006 .

[35]  Sylvain Arlot,et al.  A survey of cross-validation procedures for model selection , 2009, 0907.4728.

[36]  Joe Wiart,et al.  Surrogate modeling based on resampled polynomial chaos expansions , 2018, Reliab. Eng. Syst. Saf..

[37]  Joe Wiart,et al.  Radio-Frequency Human Exposure Assessment: From Deterministic to Stochastic Methods , 2016 .

[38]  Seymour Geisser,et al.  The Predictive Sample Reuse Method with Applications , 1975 .

[39]  Ulrich Jakobus,et al.  Human exposure assessment in the near field of GSM base-station antennas using a hybrid finite element/method of moments technique , 2003, IEEE Transactions on Biomedical Engineering.

[40]  Clive Parini,et al.  Spherical near-field antenna measurements , 2014, Theory and Practice of Modern Antenna Range Measurements, 2nd Expanded Edition, Volume 2.

[41]  O. Fujiwara,et al.  Dominant factors influencing whole‐body average SAR due to far‐field exposure in whole‐body resonance frequency and GHz regions , 2007, Bioelectromagnetics.

[42]  E. Angelini,et al.  Influence of pregnancy stage and fetus position on the whole-body and local exposure of the fetus to RF-EMF , 2014, Physics in medicine and biology.