{"id":61110,"date":"2023-12-04T10:32:21","date_gmt":"2023-12-04T08:32:21","guid":{"rendered":"https:\/\/www.iemn.fr\/?p=61110"},"modified":"2023-12-04T10:32:21","modified_gmt":"2023-12-04T08:32:21","slug":"these-mohammed-mallik-emf-exposure-reconstruction-using-ai","status":"publish","type":"post","link":"https:\/\/www.iemn.fr\/en\/agenda\/these-mohammed-mallik-emf-exposure-reconstruction-using-ai.html","title":{"rendered":"THESE : Mohammed MALLIK \u2013 \u00ab\u00a0EMF Exposure Reconstruction using AI\u00a0\u00bb"},"content":{"rendered":"<div id='layer_slider_1'  class='avia-layerslider main_color avia-shadow  avia-builder-el-0  el_before_av_heading  avia-builder-el-first  container_wrap sidebar_right'  style='height: 261px;'  ><div id=\"layerslider_58_zpiu9e7t5hv9\" data-ls-slug=\"homepageslider\" class=\"ls-wp-container fitvidsignore ls-selectable\" style=\"width:1140px;height:260px;margin:0 auto;margin-bottom: 0px;\"><div class=\"ls-slide\" data-ls=\"duration:6000;transition2d:5;\"><img loading=\"lazy\" decoding=\"async\" width=\"2600\" height=\"270\" src=\"https:\/\/www.iemn.fr\/wp-content\/uploads\/2019\/01\/sliders_news1.jpg\" class=\"ls-bg\" alt=\"\" srcset=\"https:\/\/www.iemn.fr\/wp-content\/uploads\/2019\/01\/sliders_news1.jpg 2600w, https:\/\/www.iemn.fr\/wp-content\/uploads\/2019\/01\/sliders_news1-300x31.jpg 300w, https:\/\/www.iemn.fr\/wp-content\/uploads\/2019\/01\/sliders_news1-768x80.jpg 768w, https:\/\/www.iemn.fr\/wp-content\/uploads\/2019\/01\/sliders_news1-1030x107.jpg 1030w, https:\/\/www.iemn.fr\/wp-content\/uploads\/2019\/01\/sliders_news1-1500x156.jpg 1500w, https:\/\/www.iemn.fr\/wp-content\/uploads\/2019\/01\/sliders_news1-705x73.jpg 705w\" sizes=\"auto, (max-width: 2600px) 100vw, 2600px\" \/><ls-layer style=\"font-size:14px;text-align:left;font-style:normal;text-decoration:none;text-transform:none;font-weight:700;letter-spacing:0px;border-style:solid;border-color:#000;background-position:0% 0%;background-repeat:no-repeat;width:180px;height:30px;left:0px;top:231px;line-height:32px;color:#ffffff;border-radius:6px 6px 6px 6px;padding-left:50px;background-color:rgba(0, 0, 0, 0.57);\" class=\"ls-l ls-ib-icon ls-text-layer\" data-ls=\"minfontsize:0;minmobilefontsize:0;\"><i class=\"fa fa-quote-right\" style=\"color:#ffffff;margin-right:0.8em;font-size:1em;transform:translateY( -0.125em );\"><\/i>ACTUALITES<\/ls-layer><\/div><\/div><\/div><div id='after_layer_slider_1'  class='main_color av_default_container_wrap container_wrap sidebar_right'  ><div class='container av-section-cont-open' ><div class='template-page content  av-content-small alpha units'><div class='post-entry post-entry-type-page post-entry-61110'><div class='entry-content-wrapper clearfix'>\n\n<style type=\"text\/css\" data-created_by=\"avia_inline_auto\" id=\"style-css-av-lpqmnlhn-fb97c036b4a7d29d4b7e51421fe2fb43\">\n#top .av-special-heading.av-lpqmnlhn-fb97c036b4a7d29d4b7e51421fe2fb43{\nmargin:0 0 10px 0;\npadding-bottom:4px;\n}\nbody .av-special-heading.av-lpqmnlhn-fb97c036b4a7d29d4b7e51421fe2fb43 .av-special-heading-tag .heading-char{\nfont-size:25px;\n}\n.av-special-heading.av-lpqmnlhn-fb97c036b4a7d29d4b7e51421fe2fb43 .av-subheading{\nfont-size:15px;\n}\n<\/style>\n<div  class='av-special-heading av-lpqmnlhn-fb97c036b4a7d29d4b7e51421fe2fb43 av-special-heading-h2  avia-builder-el-1  el_after_av_layerslider  el_before_av_hr  avia-builder-el-first'><h2 class='av-special-heading-tag'  itemprop=\"headline\"  >THESE : Mohammed MALLIK \u2013 \u00ab\u00a0EMF Exposure Reconstruction using AI\u00a0\u00bb <\/h2><div class=\"special-heading-border\"><div class=\"special-heading-inner-border\"><\/div><\/div><\/div>\n\n<style type=\"text\/css\" data-created_by=\"avia_inline_auto\" id=\"style-css-av-18u73nj-dad6a947580930e400fc42ba200e80f1\">\n#top .hr.av-18u73nj-dad6a947580930e400fc42ba200e80f1{\nmargin-top:5px;\nmargin-bottom:5px;\n}\n.hr.av-18u73nj-dad6a947580930e400fc42ba200e80f1 .hr-inner{\nwidth:100%;\n}\n<\/style>\n<div  class='hr av-18u73nj-dad6a947580930e400fc42ba200e80f1 hr-custom  avia-builder-el-2  el_after_av_heading  el_before_av_textblock  hr-left hr-icon-no'><span class='hr-inner inner-border-av-border-thin'><span class=\"hr-inner-style\"><\/span><\/span><\/div>\n<section  class='av_textblock_section av-jriy64i8-2f4600354c0449b610997916bbd9b6bc'   itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><div class='avia_textblock'  itemprop=\"text\" >\n<style type=\"text\/css\" data-created_by=\"avia_inline_auto\" id=\"style-css-av-13ewzjw-68e036126b913e5028f77311dc66b825\">\n.av_font_icon.av-13ewzjw-68e036126b913e5028f77311dc66b825{\ncolor:#bfbfbf;\nborder-color:#bfbfbf;\n}\n.av_font_icon.av-13ewzjw-68e036126b913e5028f77311dc66b825 .av-icon-char{\nfont-size:60px;\nline-height:60px;\n}\n<\/style>\n<span  class='av_font_icon av-13ewzjw-68e036126b913e5028f77311dc66b825 avia_animate_when_visible av-icon-style- avia-icon-pos-left avia-icon-animate'><span class='av-icon-char' aria-hidden='true' data-av_icon='\ue8c9' data-av_iconfont='entypo-fontello' ><\/span><\/span>\n<p><strong>Mohammed MALLIK<br \/>\n<\/strong><\/p>\n<p>Soutenance : 6 D\u00e9cembre 2023 \u00e0 14h00<strong><br \/>\n<\/strong>Amphith\u00e9\u00e2tre de l\u2019IRCICA- Villeneuve d\u2019Ascq<\/p>\n<\/div><\/section>\n<section  class='av_textblock_section av-jtefqx33-628129dba2299b2ecd65ebfc92eac29d'   itemscope=\"itemscope\" itemtype=\"https:\/\/schema.org\/BlogPosting\" itemprop=\"blogPost\" ><div class='avia_textblock'  itemprop=\"text\" ><div  class='hr av-kjh3zw-4dff888f744b728a1aca9b3a0971493a hr-default  avia-builder-el-6  avia-builder-el-no-sibling'><span class='hr-inner'><span class=\"hr-inner-style\"><\/span><\/span><\/div>\n<h5><strong><span style=\"color: #800000;\">Jury :<\/span><\/strong><\/h5>\n<ul>\n<li>Reviewer: Professor Fran\u00e7ois\u00a0 SEPTIER,\u00a0 Universite Bretagne Sud, France<br \/>\nReviewer: Professor Julien\u00a0 SARRAZIN ,Sorbonne University, France;<br \/>\nExaminer: Professor Claude\u00a0 OESTGES , Catholic University of Louvain , Belgium;<br \/>\nExaminer: Professor Mr. Alexandre\u00a0 CAMINADA, Polytechnic School of University of C\u00f4te d\u2019Azur, France.<br \/>\nExaminer: Professor Margot\u00a0 DERUICK , Universiteit Gent, Belgium;<br \/>\nExaminer: Madam Emmanuelle\u00a0 CONIL, ANFR, France;<br \/>\nDirector: Professor Davy P.\u00a0 GAILLOT , University of Lille , France;<br \/>\nCo-supervisor:\u00a0Professor Laurent\u00a0 CLAVIER, IMT Nord Europe, France;<br \/>\nCo-supervisor:Professor Joe Wiart, Telecom Paris, France.<\/li>\n<\/ul>\n<h5>Summary:<\/h5>\n<p>Le sujet de l\u2019exposition aux champs \u00e9lectromagn\u00e9tiques a fait l\u2019objet d\u2019une grande attention \u00e0 la lumi\u00e8re du d\u00e9ploiement actuel du r\u00e9seau cellulaire de cinqui\u00e8me g\u00e9n\u00e9ration (5G). Malgr\u00e9 cela, il reste difficile de reconstituer avec pr\u00e9cision le champ \u00e9lectromagn\u00e9tique dans une r\u00e9gion donn\u00e9e, faute de donn\u00e9es suffisantes. Les mesures in situ sont d\u2019un grand int\u00e9r\u00eat, mais leur viabilit\u00e9 est limit\u00e9e, ce qui rend difficile la compr\u00e9hension compl\u00e8te de la dynamique du champ. Malgr\u00e9 le grand int\u00e9r\u00eat des mesures localis\u00e9es, il existe encore des r\u00e9gions non test\u00e9es qui les emp\u00eachent de fournir une carte d\u2019exposition compl\u00e8te. La recherche a explor\u00e9 des strat\u00e9gies de reconstruction \u00e0 partir d\u2019observations provenant de certains sites localis\u00e9s ou de capteurs distribu\u00e9s dans l\u2019espace, en utilisant des techniques bas\u00e9es sur la g\u00e9ostatistique et les processus gaussiens. En particulier, des initiatives r\u00e9centes se sont concentr\u00e9es sur l\u2019utilisation de l\u2019apprentissage automatique et de l\u2019intelligence artificielle \u00e0 cette fin. Pour surmonter ces probl\u00e8mes, ce travail propose de nouvelles m\u00e9thodologies pour reconstruire les cartes d\u2019exposition aux CEM dans une zone urbaine sp\u00e9cifique en France. L\u2019objectif principal est de reconstruire des cartes d\u2019exposition aux ondes \u00e9lectromagn\u00e9tiques \u00e0 partir de donn\u00e9es provenant de capteurs r\u00e9partis dans l\u2019espace. Nous avons propos\u00e9 deux m\u00e9thodologies bas\u00e9es sur l\u2019apprentissage automatique pour estimer l\u2019exposition aux ondes \u00e9lectromagn\u00e9tiques. Pour la premi\u00e8re m\u00e9thode, le probl\u00e8me de reconstruction de l\u2019exposition est d\u00e9fini comme une t\u00e2che de traduction d\u2019image \u00e0 image. Tout d\u2019abord, les donn\u00e9es du capteur sont converties en une image et l\u2019image de r\u00e9f\u00e9rence correspondante est g\u00e9n\u00e9r\u00e9e \u00e0 l\u2019aide d\u2019un simulateur bas\u00e9 sur le trac\u00e9 des rayons. Nous avons propos\u00e9 un r\u00e9seau adversarial cGAN conditionn\u00e9 par la topologie de l\u2019environnement pour estimer les cartes d\u2019exposition \u00e0 l\u2019aide de ces images. Le mod\u00e8le est entra\u00een\u00e9 sur des images de cartes de capteurs tandis qu\u2019un environnement est donn\u00e9 comme entr\u00e9e conditionnelle au mod\u00e8le cGAN. En outre, la cartographie du champ \u00e9lectromagn\u00e9tique bas\u00e9e sur le Generative Adversarial Network est compar\u00e9e au simple Krigeage. Les r\u00e9sultats montrent que la m\u00e9thode propos\u00e9e produit des estimations pr\u00e9cises et constitue une solution prometteuse pour la reconstruction des cartes d\u2019exposition. Cependant, la production de donn\u00e9es de r\u00e9f\u00e9rence est une t\u00e2che complexe car elle implique la prise en compte du nombre de stations de base actives de diff\u00e9rentes technologies et op\u00e9rateurs, dont la configuration du r\u00e9seau est inconnue, par exemple les puissances et les faisceaux utilis\u00e9s par les stations de base. En outre, l\u2019\u00e9valuation de ces cartes n\u00e9cessite du temps et de l\u2019expertise. Pour r\u00e9pondre \u00e0 ces questions, nous avons d\u00e9fini le probl\u00e8me comme une t\u00e2che d\u2019imputation de donn\u00e9es manquantes. La m\u00e9thode que nous proposons prend en compte l\u2019entra\u00eenement d\u2019un r\u00e9seau neuronal infini pour estimer l\u2019exposition aux champs \u00e9lectromagn\u00e9tiques. Il s\u2019agit d\u2019une solution prometteuse pour la reconstruction des cartes d\u2019exposition, qui ne n\u00e9cessite pas de grands ensembles d\u2019apprentissage. La m\u00e9thode propos\u00e9e est compar\u00e9e \u00e0 d\u2019autres approches d\u2019apprentissage automatique bas\u00e9es sur les r\u00e9seaux UNet et les r\u00e9seaux adversaires g\u00e9n\u00e9ratifs conditionnels, avec des r\u00e9sultats comp\u00e9titifs.<\/p>\n<h5>Abstract:<\/h5>\n<p>The topic of exposure to electromagnetic fields has received much attention in light of the current deployment of the fifth generation (5G) cellular network. Despite this, it remains difficult to precisely reconstruct the electromagnetic field in a given region due to lack of sufficient data. In situ measurements are of great interest, but their viability is limited, making it difficult to fully understand the field dynamics. Despite the great interest in localized measurements, there are still untested regions that prevent them from providing a complete exposure map. The research explored reconstruction strategies from observations from certain localized sites or spatially distributed sensors, using techniques based on geostatistics and Gaussian processes. In particular, recent initiatives have focused on the use of machine learning and artificial intelligence for this purpose. To overcome these problems, this work proposes new methodologies to reconstruct EMF exposure maps in a specific urban area in France. The main objective is to reconstruct maps of exposure to electromagnetic waves from data coming from sensors distributed in space. We proposed two methodologies based on machine learning to estimate exposure to electromagnetic waves. For the first method, the exposure reconstruction problem is defined as an image-to-image translation task. First, the sensor data is converted into an image and the corresponding reference image is generated using a ray tracing based simulator. We proposed a cGAN adversarial network conditioned by the topology of the environment to estimate exposure maps using these images. The model is trained on sensor map images while an environment is given as conditional input to the cGAN model. Furthermore, electromagnetic field mapping based on the Generative Adversarial Network is compared to simple Kriging. The results show that the proposed method produces accurate estimates and is a promising solution for exposure map reconstruction. However, producing reference data is a complex task as it involves taking into account the number of active base stations of different technologies and operators, whose network configuration is unknown, e.g. the powers and beams used by the base stations. Additionally, evaluating these maps requires time and expertise. To answer these questions, we defined the problem as a missing data imputation task. The method we propose takes into account the training of an infinite width neural network to estimate exposure to electromagnetic fields. This is a promising solution for exposure map reconstruction, which does not require large training sets. The proposed method is compared with other machine learning approaches based on UNet and conditional generative adversarial networks, with competitive results.<\/p>\n<\/div><\/section>","protected":false},"excerpt":{"rendered":"","protected":false},"author":20,"featured_media":61112,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[87,65,84,187,318],"tags":[],"class_list":["post-61110","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-agenda-en","category-agenda","category-agenda-en-en","category-annonces-these","category-these-2023"],"_links":{"self":[{"href":"https:\/\/www.iemn.fr\/en\/wp-json\/wp\/v2\/posts\/61110","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.iemn.fr\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.iemn.fr\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.iemn.fr\/en\/wp-json\/wp\/v2\/users\/20"}],"replies":[{"embeddable":true,"href":"https:\/\/www.iemn.fr\/en\/wp-json\/wp\/v2\/comments?post=61110"}],"version-history":[{"count":0,"href":"https:\/\/www.iemn.fr\/en\/wp-json\/wp\/v2\/posts\/61110\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.iemn.fr\/en\/wp-json\/wp\/v2\/media\/61112"}],"wp:attachment":[{"href":"https:\/\/www.iemn.fr\/en\/wp-json\/wp\/v2\/media?parent=61110"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.iemn.fr\/en\/wp-json\/wp\/v2\/categories?post=61110"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.iemn.fr\/en\/wp-json\/wp\/v2\/tags?post=61110"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}