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 5 NOVEMBRE 2024 \u00e0 11h00<\/p>\n\n\n\n
\u00e0<\/p>\n\n\n\n
Sala U0.1, Edifici U
Universitat Oberta de Catalunya (UOC), C\/ del Per\u00fa, 52, Barcelona, 08018, Spain<\/p>\n\n\n\n
Membres du jury :<\/strong><\/p>\n\n\n\n M. Joaquin GARCIA-ALFARO<\/strong>, Full professor, T\u00e9l\u00e9com SudParis, FRANCE – Directeur de these R\u00e9sum\u00e9 :<\/strong><\/p>\n\n\n\n Cette th\u00e8se examine l’int\u00e9gration de techniques de s\u00e9curit\u00e9 avanc\u00e9es dans des IDS (Syst\u00e8mes de d\u00e9tection d’intrusions, ou IDS en anglais) dans les r\u00e9seaux de type Internet des objets (IoT). Ces r\u00e9seaux sont de plus en plus susceptibles aux cybermenaces, en raison de leur nature interconnect\u00e9e et des ressources limit\u00e9es. Les techniques de s\u00e9curit\u00e9 traditionnelles, comme la d\u00e9tection bas\u00e9e sur les signatures, ne peuvent identifier que les attaques connues, tandis que la d\u00e9tection bas\u00e9e sur l’analyse comportementale peut identifier des attaques inconnues, mais g\u00e9n\u00e8re souvent des taux \u00e9lev\u00e9s de fausses alarmes. Nous proposons un syst\u00e8me robuste et l\u00e9ger pour r\u00e9soudre les probl\u00e8mes d’int\u00e9grit\u00e9 des donn\u00e9es, ainsi que pour s\u00e9curiser les informations de provenance et am\u00e9liorer l’efficacit\u00e9 des IDS comportementaux. La solution propos\u00e9e introduit une nouvelle approche de type zero-watermarking, en utilisant l’information de provenance des donn\u00e9es \u00e9chang\u00e9es. Tout d’abord, nous effectuons une revue syst\u00e9matique des IDS r\u00e9cents bas\u00e9s sur l’apprentissage automatique pour les r\u00e9seaux IoT, en identifiant les principaux d\u00e9fis tels que les taux de d\u00e9tection, les faux positifs, la d\u00e9tection en temps r\u00e9el, la surcharge de calcul et la consommation d’\u00e9nergie. Nous soulignons la n\u00e9cessit\u00e9 de recherches approfondies couvrant tous les types d’attaques et les technologies IoT r\u00e9centes. De plus, nous soulignons le potentiel des algorithmes de type zero-watermarking en tant que solution efficace en termes de ressources pour la mise en \u0153uvre des IDS comportementaux. Deuxi\u00e8mement, nous examinons l’int\u00e9gration de la provenance des donn\u00e9es, en abordant les vuln\u00e9rabilit\u00e9s et le besoin de fiabilit\u00e9, de qualit\u00e9, de tra\u00e7abilit\u00e9 et de s\u00e9curit\u00e9 des donn\u00e9es. Nous identifions des lacunes de recherche importantes et soulignons la n\u00e9cessit\u00e9 d’une exploration plus approfondie pour am\u00e9liorer la s\u00e9curit\u00e9, en fournissant des orientations de recherche pour am\u00e9liorer les techniques de s\u00e9curit\u00e9 de provenance existantes dans les r\u00e9seaux IoT. En outre, nous nous concentrons sur l’int\u00e9grit\u00e9 des donn\u00e9es et la transmission s\u00e9curis\u00e9e des informations de provenance dans les r\u00e9seaux IoT gr\u00e2ce \u00e0 notre approche de type zero-watermarking. Nous pr\u00e9sentons une solution algorithmique et une mod\u00e9lisation pour divers sc\u00e9narios, validant les capacit\u00e9s de s\u00e9curit\u00e9 de notre approche par une analyse de s\u00e9curit\u00e9 formelle et des simulations num\u00e9riques. Nos r\u00e9sultats d\u00e9montrent que le sch\u00e9ma propos\u00e9 est l\u00e9ger, efficace en termes de calcul et consomme moins d’\u00e9nergie que les solutions existantes. De plus, nous proposons une nouvelle approche pour am\u00e9liorer les performances des syst\u00e8mes de d\u00e9tection d’intrusions r\u00e9seau bas\u00e9s sur les anomalies en int\u00e9grant le zero-watermarking avec des techniques bas\u00e9es sur l’apprentissage automatique. En utilisant une approche \u00e0 deux couches, combinant un mod\u00e8le d’apprentissage automatique pour la classification initiale, plus l’incorporation de notre proposition de zero-watermarking sur la provenance des donn\u00e9es pour la classification secondaire, nous obtenons une am\u00e9lioration en termes de pr\u00e9cision, ainsi que nous r\u00e9duisons consid\u00e9rablement les taux de fausses alarmes. Une \u00e9valuation \u00e0 l’aide des jeux de donn\u00e9es publiques confirme l’efficacit\u00e9 de notre approche en termes de performances de classification et d’efficacit\u00e9 de calcul. Nous avons \u00e9valu\u00e9 nos contributions en utilisant des simulations num\u00e9riques, exp\u00e9rimentation et une analyse comparative avec des solutions existantes. Enfin, nous avons identifi\u00e9 plusieurs possibilit\u00e9s pour les travaux futurs visant \u00e0 \u00e9largir et \u00e0 am\u00e9liorer davantage les connaissances existantes sur le terrain, en soulignant des directions de recherche prometteuses pour l’avancement des techniques de watermarking, la provenance des donn\u00e9es et les techniques de d\u00e9tection d’intrusion pour les r\u00e9seaux IoT. This thesis investigates the integration of advanced security techniques into Intrusion Detection Systems (IDS) for Internet of Things (IoT) networks. Such networks are increasingly susceptible to various cyber threats due to their interconnected nature and constrained resources. Traditional security techniques, like signature-based detection, can only identify known attacks, while anomaly detection can identify unknown attacks but often generates high false alarm rates. This makes advanced IDS crucial. Additionally, we also tackle the challenge of efficiently and securely transmitting provenance information in IoT networks. The goal is to ensure the integrity and reliability of data within IoT environments. We propose a robust and lightweight system to address data integrity issues, secure provenance information, and enhanced effectiveness of anomaly-based IDS. The proposed solution introduces a novel zero-watermarking approach, utilizing data provenance information as a lightweight methodology. Firstly, we conduct a systematic review of recent Machine Learning (ML)-based IDS for IoT networks. We identify the following challenges: management of detection rates (mainly false positives), assurance of real-time detection, and reduction of both computational overhead and energy consumption. We highlight the necessity for extensive research covering all attack types and recent IoT technologies. We also emphasize the potential of watermarking algorithms as a resource-efficient solution for IDS implementation. Secondly, we examine the integration of IoT and data provenance, addressing vulnerabilities and the need for data trustworthiness, quality, traceability, and security. We identify significant research gaps and emphasize the need for further exploration to enhance network security, providing research directions to improve existing provenance security techniques in IoT. We also focus on addressing data integrity and secure transmission of provenance information in IoT networks through zero-watermarking. We present algorithmic modeling for various scenarios, validating the security capabilities of our approach through formal security analysis and numerical simulations. Our findings validate that the proposed scheme is lightweight, computationally efficient, and consumes less energy compared to existing solutions. Finally, we propose a novel approach to enhance the performance of anomaly-based Network Intrusion Detection Systems (NIDS) by integrating zero-watermarking with ML-based techniques. Using a two-layer approach, we combine ML-based model for initial classification and data provenance-based zero-watermarking for secondary classification. By doing so, we achieve high accuracy and significantly reduce the rate of false alarms. Evaluation using public datasets confirms the effectiveness of our approach in terms of classification performance and computational efficiency. We evaluated all the proposed approaches using numerical simulations, experiments and comparative analysis with existing solutions. We also identified several possibilities for future work to further extend and improve the existing field knowledge, highlighting promising research directions for advancing watermarking techniques, data provenance and IDS in IoT networks.<\/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 Omair FARAJ 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":[286,603],"tags":[],"class_list":["post-6618","post","type-post","status-publish","format-standard","hentry","category-fractualites-ennews-fr","category-seminaire-scn","entry"],"_links":{"self":[{"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/posts\/6618","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=6618"}],"version-history":[{"count":1,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/posts\/6618\/revisions"}],"predecessor-version":[{"id":6619,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/posts\/6618\/revisions\/6619"}],"wp:attachment":[{"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/media?parent=6618"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/categories?post=6618"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/samovar.telecom-sudparis.eu\/index.php\/wp-json\/wp\/v2\/tags?post=6618"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
M. Michal CHORAS<\/strong>, Full professor, Bydgoszcz University of Science and Technology (PBS), POLOGNE – Rapporteur
M. Mohamed MOSBAH<\/strong>, Full professor, Bordeaux INP, FRANCE – Rapporteur
M. Alexandre VIEJO<\/strong>, Associate Professor, Universitat Rovira i Virgili, ESPAGNE – Examinateur
M. David MEG\u00edAS JIM\u00e9NEZ<\/strong>, Full professor, Universitat Oberta de Catalunya (UOC), ESPAGNE – CoDirecteur de these
M. Carlos BORREGO IGLESIAS<\/strong>, Associate Professor, Autonomous University of Barcelona. , ESPAGNE – Examinateur<\/p>\n\n\n\n\u00ab Zero-Watermarking pour l’int\u00e9grit\u00e9 des donn\u00e9es, la provenance s\u00e9curis\u00e9e et la d\u00e9tection des intrusions dans les r\u00e9seaux IoT \u00bb<\/h2>\n\n\n\n
pr\u00e9sent\u00e9 par Monsieur Omair FARAJ<\/h2>\n\n\n\n
Abstract :<\/strong><\/p>\n\n\n\n