Browsing by Author "Medjek, Faiza"

Now showing 1 - 7 of 7
  • Thumbnail Image
    Item
    Etude des modèles des "GRID Services" et implémentation d'une plateforme de développement / gridification basée sur les "Web services"
    (Université Abderrahmane Mira de Béjaia, 2009) Medjek, Faiza; Badache, Nadjib; El Maouhab, Aouaouche
    Les applications de calcul scientifique qui nécessitent des ressources de calcul et de stockage importantes nécessitent des environnements de grilles de calcul. Ces environnements de grilles se basent sur des approches orientées services. Dans ce contexte, les infrastructures « web services », dans le domaine du calcul scientifique peuvent être utilisées pour le développement et la « gridification1 » des applications. Par ailleurs, les architectures orientées services qui utilisent les infrastructures des Grid services permettent les invocations des services à travers les réseaux et les exécutions parallèles selon les workflows2 qui les définissent. L‘orchestration des activités/processus désignés par les workflows est une condition essentielle. Notre objectif est de résoudre le problème d‘interaction (intégration) des différents environnements qui répondent aux besoins de gridification d‘applications scientifiques (géo-spatiales, physiques, bioinformatique, etc.) basées sur la notion de workflows et de permettre l‘accès transparent aux applications en faisant abstraction des différents langages de programmation. La finalité étant de masquer la complexité à utiliser la grille et de permettre le parallélisme des applications en cas de besoin et ce en exploitant la force des workflows.
  • Thumbnail Image
    Item
    Evaluation of the impacts of Sybil attacks against RPL under mobility
    (CERIST, 2014-06) Medjek, Faiza; Tandjaoui, Djamel; Djedjig, Nabil
    The Routing Protocol for Low-Power and Lossy Networks (RPL) is the routing protocol standardized for constrained environments such as 6LoWPAN networks, and is considered as the routing protocol of the Internet of Things (IoT). However, this protocol is subject to several attacks that have been analyzed on static case. Nevertheless, IoT will likely present dynamic and mobile applications. In this paper, we introduce potential security threats on RPL, in particular Sybil attacks when the Sybil nodes are mobile. In addition, we present an analysis and a discussion on how network performances can be affected. Our analysis shows, under Sybil attacks while nodes are mobile, that the performances of RPL are highly affected compared to the static case. In fact, we notice a decrease in the rate of packet delivery, and an increase in control messages overhead. As a result, energy consumption at constrained nodes increases. Our proposed attacks demonstrate that Sybil mobile nodes can easily disrupt RPL and overload the network with fake messages making it unavailable. Based on the obtained results we provide some recommendations to tackle this issue.
  • Thumbnail Image
    Item
    Fault-tolerant AI-driven Intrusion Detection System for the Internet of Things
    (Elsevier, 2021-09) Medjek, Faiza; Tandjaoui, Djamel; Djedjig, Nabil; Romdhani, Imed
    Internet of Things (IoT) has emerged as a key component of all advanced critical infrastructures. However, with the challenging nature of IoT, new security breaches have been introduced, especially against the Routing Protocol for Low-power and Lossy Networks (RPL). Artificial-Intelligence-based technologies can be used to provide insights to deal with IoT’s security issues. In this paper, we describe the initial stages of developing, a new Intrusion Detection System using Machine Learning (ML) to detect routing attacks against RPL. We first simulate the routing attacks and capture the traffic for different topologies. We then process the traffic and generate large 2-class and multi-class datasets. We select a set of significant features for each attack, and we use this set to train different classifiers to make the IDS. The experiments with 5-fold cross-validation demonstrated that decision tree (DT), random forests (RF), and K-Nearest Neighbours (KNN) achieved good results of more than 99% value for accuracy, precision, recall, and F1-score metrics, and RF has achieved the lowest fitting time. On the other hand, Deep Learning (DL) model, MLP, Naïve Bayes (NB), and Logistic Regression (LR) have shown significantly lower performance.
  • Thumbnail Image
    Item
    Multicast DIS attack mitigation in RPL-based IoT-LLNs
    (Elsevier, 2021-09) Medjek, Faiza; Tandjaoui, Djamel; Djedjig, Nabil; Romdhani, Imed
    The IPv6 Routing Protocol for Low-Power and Lossy Networks (RPL) was standardised by the IETF ROLL Working Group to address the routing issues in the Internet of Things (IoT) Low-Power and Lossy Networks (LLNs). RPL builds and maintains a Destination Oriented Directed Acyclic Graph (DODAG) topology using pieces of information propagated within the DODAG Information Object (DIO) control message. When a node intends to join the DODAG, it either waits for DIO or sends a DODAG Information Solicitation (DIS) control message Multicast to solicit DIOs from nearby nodes. Nevertheless, sending Multicast DIS messages resets the timer that regulates the transmission rate of DIOs to its minimum value, which leads to the network’s congestion with control messages. Because of the resource-constrained nature of RPL-LLNs, the lack of tamper resistance, and the security gaps of RPL, malicious nodes can exploit the Multicast DIS solicitation mechanism to trigger an RPL-specification-based attack, named DIS attack. The DIS attack can have severe consequences on RPL networks, especially on control packets overhead and power consumption. In this paper, we use the Cooja–Contiki simulator to assess the DIS attack’s effects on both static and dynamic PRL networks. Besides, we propose and implement a novel approach, namely RPL-MRC, to improve the RPL’s resilience against DIS Multicast. RPL-MRC aims to reduce the response to DIS Multicast messages. Simulation results demonstrate how the attack could damage the network performance by significantly increasing the control packets overhead and power consumption. On the other hand, the RPL-MRC proposed mechanism shows a significant enhancement in reducing the control overhead and power consumption for different scenarios.
  • Thumbnail Image
    Item
    Services Web et orchestration appliqués à la gestion des noms de domaines 'DZ'
    (Université des Sciences et de la Technologie Houari-Boumediene (USTHB), 2012) Bouider, Sahima; El Maouhab, Aouaouche; Medjek, Faiza
    Il est communément reconnu que le concept de « Service » facilite l’intégration des systèmes logiciels en fournissant une vue globale du système. Les Services Web définissent une manière standard d'interagir avec des applications distantes en utilisant les technologies du Web. L'infrastructure des Services Web répond aux problèmes d’interopérabilité et d’intégration technique des applications. En joignant l’aspect de la composition des Services Web (orchestration ou chorégraphie) via un des langages de composition tel que le BPEL (Business Process Execution Language) a permis d’utiliser les compétences de plusieurs services, afin de résoudre un problème qu’aucun ne saurait résoudre individuellement. Notre objectif est d’utiliser l’Architecture Orientée Services, notamment, les Services Web et leurs composition dans l’implémentation d’un prototype de système de gestion des noms de domaines ‘.dz’ qui sont attribués par délégation. Cette délégation se fait sur des serveurs de gestion de noms de domaine "DNS" et l'enregistrement effectif se fait à travers des entités d'enregistrement appelées registraires qui gèrent les noms de domaines pour le compte de ses clients.
  • Thumbnail Image
    Item
    Trust management in IoT routing protocol
    (CERIST, 2015-04-19) Djedjig, Nabil; Tandjaoui, Djamel; Medjek, Faiza
    The Routing Protocol for Low-Power and Lossy Networks (RPL) is the routing protocol standardized for constrained environments such as 6LoWPAN networks, and is considered as the routing protocol of the Internet of Things (IoT). However, this protocol is subject to several internal and external attacks. In this paper, we investigate a trust management protocol in RPL. Our idea of trust management in RPL is to establish a dynamic trust relationship between the different nodes involved in routing. In fact, RPL organizes a logical representation of the network topology using control messages. In our proposed protocol, we strengthen RPL by adding a new trustworthiness metric during RPL construction and maintenance. This metric allows a node to decide whether or not to trust the other nodes during the construction of the topology.
  • Thumbnail Image
    Item
    Trust-aware and cooperative routing protocol for IoT security
    (2020-06) Djedjig, Nabil; Tandjaoui, Djamel; Medjek, Faiza; Romdhani, Imed
    The resource-constrained nature of IoT objects makes the Routing Protocol for Low-power and Lossy Networks (RPL) vulnerable to several attacks. Although RPL specification provides encryption protection to control messages, RPL is still vulnerable to internal attackers and selfish behaviours. To address the lack of robust security mechanisms in RPL, we design a new Metric-based RPL Trustworthiness Scheme (MRTS) that introduces trust evaluation for secure routing topology construction. Extensive simulations show that MRTS is efficient in terms of packet delivery ratio, energy consumption, nodes’ rank changes, and throughput. In addition, a mathematical modelling analysis shows that MRTS meets the requirements of consistency, optimality, and loop-freeness and that the proposed trust-based routing metric has the isotonicity and monotonicity properties required for a routing protocol. By using game theory concepts, we formally describe MRTS as a strategy for the iterated Prisoner’s Dilemma and demonstrate its cooperation enforcement characteristic. Both mathematical analysis and evolutionary simulation results show clearly that MRTS, as a strategy, is an efficient approach in promoting the stability and the evolution of the Internet of Things network.