pr\u00e9sentent<\/p>\n\n\n\n
Autoris\u00e9 \u00e0 pr\u00e9senter ses travaux en vue de l\u2019obtention du Doctorat de l’Institut Polytechnique de Paris, pr\u00e9par\u00e9 \u00e0 T\u00e9l\u00e9com SudParis en :<\/p>\n\n\n\n
le MARDI 9 JUILLET 2024 \u00e0 15h00<\/p>\n\n\n\n
\u00e0<\/p>\n\n\n\n
Amphi 2
19 pl. Marguerite Perey 92120, Palaiseau, France<\/p>\n\n\n\n
Membres du jury :<\/strong><\/p>\n\n\n\n Mme Catherine LEPERS<\/strong>, Professeure, T\u00e9l\u00e9com SudParis, FRANCE – Directeur de these R\u00e9sum\u00e9 :<\/strong><\/p>\n\n\n\n Avec la mont\u00e9e en puissance des capacit\u00e9s de calcul n\u00e9cessaires pour les m\u00e9thodes plus d\u00e9velopp\u00e9es d\u2019Apprentissage Machine (ML), la Pr\u00e9diction des Incidents R\u00e9seau (NFP:Network Fault Prediction) connait un regain d\u2019int\u00e9r\u00eat scientifique. La capacit\u00e9 de pr\u00e9dire les incidents des \u00e9quipements r\u00e9seau est de plus en plus fr\u00e9quemment identifi\u00e9e comme un moyen efficace d\u2019am\u00e9liorer la fiabilit\u00e9 du r\u00e9seau. Cette capacit\u00e9 pr\u00e9dictive peut \u00eatre utilis\u00e9e pour att\u00e9nuer ou mettre en \u0153uvre une maintenance pr\u00e9dictive en pr\u00e9vision des cas d’incidents r\u00e9seau imminents. Cela pourrait contribuer \u00e0 la mise en oeuvre de r\u00e9seaux sans d\u00e9faillance et sans pertes, et permettre aux applications critiques d’\u00eatre ex\u00e9cut\u00e9es sur des r\u00e9seaux de plus grandes dimensions et h\u00e9t\u00e9rog\u00e8nes. Dans ce manuscrit, nous nous proposons de contribuer au domaine du NFP en nous focalisant sur la pr\u00e9diction des alertes r\u00e9seau. Dans un premier temps, nous pr\u00e9sentons une \u00e9tude de l’\u00e9tat de l’art complet du NFP en utilisant des m\u00e9thodes d’apprentissage machine (ML) enti\u00e8rement d\u00e9di\u00e9e aux r\u00e9seaux de t\u00e9l\u00e9communications. Ensuite, nous \u00e9tablissons de futures directions de recherche dans le domaine. Dans un deuxi\u00e8me temps, nous proposons et \u00e9tudions un couple de m\u00e9triques (R\u00e9duction des co\u00fbts de maintenance, et mesure des gains de Qualit\u00e9 de Service) de performances de ML adapt\u00e9es au NFP dans le cadre de la maintenance des r\u00e9seaux. Dans un troisi\u00e8me temps, nous d\u00e9crivons l’architecture compl\u00e8te de traitement des donn\u00e9es, incluant l\u2019infrastructure r\u00e9seau et logicielle, et la cha\u00eene de pr\u00e9traitement des donn\u00e9es n\u00e9cessaires au ML qui ont \u00e9t\u00e9 mis en \u0153uvre chez SPIE ICS, soci\u00e9t\u00e9 d\u2019int\u00e9gration de r\u00e9seaux et de syst\u00e8mes. Nous d\u00e9crivons \u00e9galement avec pr\u00e9cision le mod\u00e8le du probl\u00e8me d\u2019alarme et d’incidents. Dans un quatri\u00e8me temps, nous \u00e9tablissons une comparaison des diff\u00e9rentes m\u00e9thodes de ML appliqu\u00e9es \u00e0 notre jeu de donn\u00e9es. Nous consid\u00e9rons des m\u00e9thodes conventionelles de ML, bas\u00e9s sur des arbres de d\u00e9cision, des perceptrons multicouches et des S\u00e9parateurs \u00e0 Vastes Marges. Nous testons la g\u00e9n\u00e9ralisation des performances des mod\u00e8les par rapport aux diff\u00e9rents types d’\u00e9quipements, ainsi que les g\u00e9n\u00e9ralisations en ML des mod\u00e8les de ML et des param\u00e8tres propos\u00e9s. Ensuite, nous \u00e9tudions avec succ\u00e8s les architectures de ML \u00e0 entr\u00e9e s\u00e9quentielle – R\u00e9seaux de neurones convolutifs et Long Short Term Memory – dans le cas de donn\u00e9es SNMP s\u00e9quentielles sur notre ensemble de donn\u00e9es. Finalement, nous \u00e9tudions l’impact sur la performance de pr\u00e9diction des variables de temps X, Y et Z de la mod\u00e9lisation du probl\u00e8me, d\u00e9finies en tant que la fen\u00eatre temporelle des donn\u00e9es d’entr\u00e9e Y, la fen\u00eatre temporelle tampon X entre le moment de la pr\u00e9diction et la fen\u00eatre temporelle de l’\u00e9tat cible Z.<\/p>\n\n\n\n Abstract :<\/strong><\/p>\n\n\n\n With the improvement of computation power necessary for advanced applications of Machine Learning (ML), Network Fault Prediction (NFP) experiences a renewed scientific interest. The ability to predict network equipment failure is increasingly identified as an effective means to improve network reliability. This predictive capability can be used, to mitigate or to enact predictive maintenance on incoming network failures. This could contribute to establishing zero-failure networks and allow safety-critical applications to run over higher dimension and heterogeneous networks. In this PhD thesis, we propose to contribute to the NFP field by focusing on network alarm prediction. First, we present a comprehensive survey on NFP using Machine Learning (ML) methods entirely dedicated to telecommunication networks, and determine new directions for research in the field. Second, we propose and study a set of Machine Learning performance metrics (maintenance cost reduction and Quality of Service improvement) adapted to NFP in the context of network maintenance. Third, we describe the complete data processing architecture, including the network and software infrastructure, and the necessary data preprocessing pipeline that was implemented at SPIE ICS, Networks and Systems Integrator. We also describe the alarm or failure prediction problem model precisely. Fourth, we establish a benchmark of the different ML solutions applied to our dataset. We consider Decision Tree-based methods, Multi-Layer Perceptron and Support Vector Machines. We test the generalization of performance prediction across equipment types as well as normal ML generalization of the proposed models and parameters. Then, we apply sequential – Convolutional Neural Networks (CNN) and Long Short Term Memory (LSTM) – ML architectures with success on our sequential SNMP dataset. Finally, we study the impact on prediction performance of the time variables Y, X and Z of the problem model defined as the input data sequence timeframe Y, the buffer X between the time of prediction and the targeted timeframe Z.<\/p>\n","protected":false},"excerpt":{"rendered":" L’Ecole doctorale : Ecole Doctorale de l’Institut Polytechnique de Paris et le Laboratoire de recherche SAMOVAR – Services r\u00e9partis, Architectures, Mod\u00e9lisation, Validation, Administration des R\u00e9seaux pr\u00e9sentent l\u2019AVIS DE SOUTENANCE de Monsieur Killian MURPHY Autoris\u00e9 \u00e0 pr\u00e9senter ses travaux en vue de l\u2019obtention du Doctorat de l’Institut Polytechnique de Paris, pr\u00e9par\u00e9 \u00e0 T\u00e9l\u00e9com SudParis en : […]<\/p>\n","protected":false},"author":4,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"ocean_post_layout":"","ocean_both_sidebars_style":"","ocean_both_sidebars_content_width":0,"ocean_both_sidebars_sidebars_width":0,"ocean_sidebar":"","ocean_second_sidebar":"","ocean_disable_margins":"enable","ocean_add_body_class":"","ocean_shortcode_before_top_bar":"","ocean_shortcode_after_top_bar":"","ocean_shortcode_before_header":"","ocean_shortcode_after_header":"","ocean_has_shortcode":"","ocean_shortcode_after_title":"","ocean_shortcode_before_footer_widgets":"","ocean_shortcode_after_footer_widgets":"","ocean_shortcode_before_footer_bottom":"","ocean_shortcode_after_footer_bottom":"","ocean_display_top_bar":"default","ocean_display_header":"default","ocean_header_style":"","ocean_center_header_left_menu":"","ocean_custom_header_template":"","ocean_custom_logo":0,"ocean_custom_retina_logo":0,"ocean_custom_logo_max_width":0,"ocean_custom_logo_tablet_max_width":0,"ocean_custom_logo_mobile_max_width":0,"ocean_custom_logo_max_height":0,"ocean_custom_logo_tablet_max_height":0,"ocean_custom_logo_mobile_max_height":0,"ocean_header_custom_menu":"","ocean_menu_typo_font_family":"","ocean_menu_typo_font_subset":"","ocean_menu_typo_font_size":0,"ocean_menu_typo_font_size_tablet":0,"ocean_menu_typo_font_size_mobile":0,"ocean_menu_typo_font_size_unit":"px","ocean_menu_typo_font_weight":"","ocean_menu_typo_font_weight_tablet":"","ocean_menu_typo_font_weight_mobile":"","ocean_menu_typo_transform":"","ocean_menu_typo_transform_tablet":"","ocean_menu_typo_transform_mobile":"","ocean_menu_typo_line_height":0,"ocean_menu_typo_line_height_tablet":0,"ocean_menu_typo_line_height_mobile":0,"ocean_menu_typo_line_height_unit":"","ocean_menu_typo_spacing":0,"ocean_menu_typo_spacing_tablet":0,"ocean_menu_typo_spacing_mobile":0,"ocean_menu_typo_spacing_unit":"","ocean_menu_link_color":"","ocean_menu_link_color_hover":"","ocean_menu_link_color_active":"","ocean_menu_link_background":"","ocean_menu_link_hover_background":"","ocean_menu_link_active_background":"","ocean_menu_social_links_bg":"","ocean_menu_social_hover_links_bg":"","ocean_menu_social_links_color":"","ocean_menu_social_hover_links_color":"","ocean_disable_title":"default","ocean_disable_heading":"default","ocean_post_title":"","ocean_post_subheading":"","ocean_post_title_style":"","ocean_post_title_background_color":"","ocean_post_title_background":0,"ocean_post_title_bg_image_position":"","ocean_post_title_bg_image_attachment":"","ocean_post_title_bg_image_repeat":"","ocean_post_title_bg_image_size":"","ocean_post_title_height":0,"ocean_post_title_bg_overlay":0.5,"ocean_post_title_bg_overlay_color":"","ocean_disable_breadcrumbs":"default","ocean_breadcrumbs_color":"","ocean_breadcrumbs_separator_color":"","ocean_breadcrumbs_links_color":"","ocean_breadcrumbs_links_hover_color":"","ocean_display_footer_widgets":"default","ocean_display_footer_bottom":"default","ocean_custom_footer_template":"","ocean_post_oembed":"","ocean_post_self_hosted_media":"","ocean_post_video_embed":"","ocean_link_format":"","ocean_link_format_target":"self","ocean_quote_format":"","ocean_quote_format_link":"post","ocean_gallery_link_images":"on","ocean_gallery_id":[],"footnotes":""},"categories":[286,543],"tags":[],"class_list":["post-6569","post","type-post","status-publish","format-standard","hentry","category-fractualites-ennews-fr","category-seminaire-istec","entry"],"_links":{"self":[{"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/posts\/6569","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/users\/4"}],"replies":[{"embeddable":true,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/comments?post=6569"}],"version-history":[{"count":1,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/posts\/6569\/revisions"}],"predecessor-version":[{"id":6570,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/posts\/6569\/revisions\/6570"}],"wp:attachment":[{"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/media?parent=6569"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/categories?post=6569"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/tags?post=6569"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
Mme Christine TREMBLAY<\/strong>, Professeure, Ecole Technologie Sup\u00e9rieure, Montr\u00e9al, CANADA – Examinateur
M. Thomas CLAUSEN<\/strong>, Professeur, Ecole Polytechnique, FRANCE – Examinateur
M. Christophe GRAVIER<\/strong>, Professeur, Telecom St-Etienne, FRANCE – Examinateur
M. C\u00e9dric WARE<\/strong>, Professeur, Telecom Paris, FRANCE – Examinateur
M. Antoine LAVIGNOTTE<\/strong>, Directeur d’\u00e9tudes, T\u00e9l\u00e9com SudParis, FRANCE – Co-encadrant de these
Mme Sandrine VATON<\/strong>, Professeure, IMT Atlantique, FRANCE – Rapporteur
M. Francesco MUSUMECI<\/strong>, Professeur associ\u00e9, Politecnico di Milano, ITALIE – Rapporteur<\/p>\n\n\n\n\u00ab Maintenance pr\u00e9dictive d’\u00e9quipements r\u00e9seau par apprentissage machine \u00bb<\/h2>\n\n\n\n
pr\u00e9sent\u00e9 par Monsieur Killian MURPHY<\/h2>\n\n\n\n