Client selection for federated edge learning in UAV networks / Ilyes Akram Ahmed Chaouch

Public ISBD Titre : Client selection for federated edge learning in UAV networks Type de document : document multimédia Auteurs : Ilyes Akram Ahmed Chaouch, Auteur ; Mohamed Khaled Ali Oudnani, Auteur ; Yousra Cheriguene, Directeur de thèse Editeur : Laghouat : Université Amar Telidji - Département d'informatique Année de publication : 2025 Importance : 41 p. Accompagnement : 1 disque optique numérique (CD-ROM) Note générale : Option : Networks, distributed systems and applications Langues : Anglais (eng) Mots-clés : FL  UAV Networks  Client selection  Mobility-Aware selection  Edge intelligence Résumé : One promising way to enable distributed intelligence at the edge while protecting data privacy is to integrate Federated Learning (FL) with Unmanned Aerial Vehicles (UAV) networks.Using FL enables each UAV to cooperatively train a global model without sharing raw data, especially in UAV swarms used for surveillance, monitoring, or emergency response missions. But choosing the best clients (UAVs) for every training cycle is made extremely difficult by the dynamic and diverse character of UAV environments. These difficulties are brought on by things like fluctuating connectivity, shifting patterns of movement, and energy limitations. In this work, we investigate the problem of client selection for Federated Edge Learning in UAV networks. We first present a taxonomy of existing selection strategies, considering criteria such as model performance and UAV mobility. Then, we propose an adaptive client selection framework that integrates both mobility-awareness and distance with speed to enhance learning efficiency and model accuracy. Extensive simulations demonstrate that our method significantly improves convergence speed and reduces communication overhead, while maintaining high model performance in dynamic UAV scenarios. note de thèses : Mémoire de master en informatique Client selection for federated edge learning in UAV networks [document multimédia] / Ilyes Akram Ahmed Chaouch, Auteur ; Mohamed Khaled Ali Oudnani, Auteur ; Yousra Cheriguene, Directeur de thèse . - Laghouat : Université Amar Telidji - Département d'informatique, 2025 . - 41 p. + 1 disque optique numérique (CD-ROM). Option : Networks, distributed systems and applications Langues : Anglais (eng) Mots-clés : FL  UAV Networks  Client selection  Mobility-Aware selection  Edge intelligence Résumé : One promising way to enable distributed intelligence at the edge while protecting data privacy is to integrate Federated Learning (FL) with Unmanned Aerial Vehicles (UAV) networks.Using FL enables each UAV to cooperatively train a global model without sharing raw data, especially in UAV swarms used for surveillance, monitoring, or emergency response missions. But choosing the best clients (UAVs) for every training cycle is made extremely difficult by the dynamic and diverse character of UAV environments. These difficulties are brought on by things like fluctuating connectivity, shifting patterns of movement, and energy limitations. In this work, we investigate the problem of client selection for Federated Edge Learning in UAV networks. We first present a taxonomy of existing selection strategies, considering criteria such as model performance and UAV mobility. Then, we propose an adaptive client selection framework that integrates both mobility-awareness and distance with speed to enhance learning efficiency and model accuracy. Extensive simulations demonstrate that our method significantly improves convergence speed and reduces communication overhead, while maintaining high model performance in dynamic UAV scenarios. note de thèses : Mémoire de master en informatique

Réservation

Réserver ce document

Exemplaires (1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
MF 01-84	MF 01-84	CD	BIBLIOTHEQUE DE FACULTE DES SCIENCES	théses (sci)	Disponible

Contribution to the UAV-based mobile edge computing communication strategies / Yousra Cheriguene

Public ISBD Titre : Contribution to the UAV-based mobile edge computing communication strategies Type de document : document multimédia Auteurs : Yousra Cheriguene, Auteur ; Chaker Abdelaziz Kerrache, Directeur de thèse ; Fatima Zahra Bousbaa, Directeur de thèse Editeur : Laghouat : Université Amar Telidji - Département d'informatique Année de publication : 2024 Importance : 120 p. Langues : Anglais (eng) Mots-clés : Unmanned aerial vehicle  UAV  Federated learning  FL  Edge computing  MEC Résumé : In recent years, the surge in data from smart devices and the Internet of Things (IoT) has propelled the need for efficient data storage, transportation, and analysis. This has led to a groundbreaking era in wireless communication and computing frameworks with the integration of Unmanned Aerial Vehicles (UAVs) in Mobile Edge Computing (MEC). This thesis explores UAV-assisted MEC, highlighting opportunities and addressing key challenges, specifically focusing on implementing Federated Learning (FL). FL, a decentralized machine learning (ML) method, allows users to collectively train an ML model without revealing their private local datasets. The thesis first optimizes UAV participant selection for edge FL, enhancing accuracy by considering factors like energy consumption, communication quality, and local dataset diversity. Empirical studies demonstrate the superior performance of the proposed selection scheme over the random selection benchmark. The second contribution introduces a novel client selection method improving convergence by prioritizing UAVs with high reliability and excluding malicious UAVs. The strategy outperforms baseline methods, as verified by evaluations under diverse attack scenarios. The third contribution presents an energy-efficient inter-UAV multicast routing protocol, COCOMA, and its enhancement, COCOMA+. Extensive simulations confirm their efficacy in establishing an efficient communication backbone with Quality-of-Service (QoS) attributes and reducing total emission energy. The thesis concludes by outlining open challenges and initiating discussions on future research directions in UAV-assisted MEC, fostering continued advancements and innovation. note de thèses : Thése de doctorat en informatique Contribution to the UAV-based mobile edge computing communication strategies [document multimédia] / Yousra Cheriguene, Auteur ; Chaker Abdelaziz Kerrache, Directeur de thèse ; Fatima Zahra Bousbaa, Directeur de thèse . - Laghouat : Université Amar Telidji - Département d'informatique, 2024 . - 120 p. Langues : Anglais (eng) Mots-clés : Unmanned aerial vehicle  UAV  Federated learning  FL  Edge computing  MEC Résumé : In recent years, the surge in data from smart devices and the Internet of Things (IoT) has propelled the need for efficient data storage, transportation, and analysis. This has led to a groundbreaking era in wireless communication and computing frameworks with the integration of Unmanned Aerial Vehicles (UAVs) in Mobile Edge Computing (MEC). This thesis explores UAV-assisted MEC, highlighting opportunities and addressing key challenges, specifically focusing on implementing Federated Learning (FL). FL, a decentralized machine learning (ML) method, allows users to collectively train an ML model without revealing their private local datasets. The thesis first optimizes UAV participant selection for edge FL, enhancing accuracy by considering factors like energy consumption, communication quality, and local dataset diversity. Empirical studies demonstrate the superior performance of the proposed selection scheme over the random selection benchmark. The second contribution introduces a novel client selection method improving convergence by prioritizing UAVs with high reliability and excluding malicious UAVs. The strategy outperforms baseline methods, as verified by evaluations under diverse attack scenarios. The third contribution presents an energy-efficient inter-UAV multicast routing protocol, COCOMA, and its enhancement, COCOMA+. Extensive simulations confirm their efficacy in establishing an efficient communication backbone with Quality-of-Service (QoS) attributes and reducing total emission energy. The thesis concludes by outlining open challenges and initiating discussions on future research directions in UAV-assisted MEC, fostering continued advancements and innovation. note de thèses : Thése de doctorat en informatique

Detection of large scale influence operations using machine learning techniques / Yasser Hacini

Public ISBD Titre : Detection of large scale influence operations using machine learning techniques Type de document : document multimédia Auteurs : Yasser Hacini, Auteur ; Yousra Cheriguene, Directeur de thèse Editeur : Laghouat : Université Amar Telidji - Département d'informatique Année de publication : 2025 Importance : 32 p. Accompagnement : 1 disque optique numérique (CD-ROM) Note générale : Option : Data science and artificial intelligence Langues : Anglais (eng) Mots-clés : Artificial intelligence  Propaganda  Word embedding  WEAT Résumé : With the ever-increasing popularity of social media, and the rising prevalence of fifth generation warfare that relies heavily on non-kinetic techniques such as propaganda, social engineering, hacking, etc. There has been a significant increase of large scale influence operation, which has become significantly easier to carry out with the recent advance- ments in generative AI and large language models, this has increased prejudice between communities, which has in turn decreased the tolerance seen between each other. In this dissertation we propose a technique which allows for the enumeration of both explicit and implicit biases found in one or multiple communities using a technique known as WEAT (Word embedding association tests), we mainly use it to find problematic associations made by these communities and how these associations change over time in response to outside influence. Using our technique, we were able to achieve an average P value of 0.033 and have been able to show clear problematic associations made by different communi- ties, we also detected sudden shifts in associations which correlated with related outside events note de thèses : Mémoire de master en informatique Detection of large scale influence operations using machine learning techniques [document multimédia] / Yasser Hacini, Auteur ; Yousra Cheriguene, Directeur de thèse . - Laghouat : Université Amar Telidji - Département d'informatique, 2025 . - 32 p. + 1 disque optique numérique (CD-ROM). Option : Data science and artificial intelligence Langues : Anglais (eng) Mots-clés : Artificial intelligence  Propaganda  Word embedding  WEAT Résumé : With the ever-increasing popularity of social media, and the rising prevalence of fifth generation warfare that relies heavily on non-kinetic techniques such as propaganda, social engineering, hacking, etc. There has been a significant increase of large scale influence operation, which has become significantly easier to carry out with the recent advance- ments in generative AI and large language models, this has increased prejudice between communities, which has in turn decreased the tolerance seen between each other. In this dissertation we propose a technique which allows for the enumeration of both explicit and implicit biases found in one or multiple communities using a technique known as WEAT (Word embedding association tests), we mainly use it to find problematic associations made by these communities and how these associations change over time in response to outside influence. Using our technique, we were able to achieve an average P value of 0.033 and have been able to show clear problematic associations made by different communi- ties, we also detected sudden shifts in associations which correlated with related outside events note de thèses : Mémoire de master en informatique

Réservation

Réserver ce document

Exemplaires (1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
MF 03-07	MF 03-07	CD	BIBLIOTHEQUE DE FACULTE DES SCIENCES	théses (sci)	Disponible

Digital twin–assisted urgency-aware task offloading in UAV–HAP edge networks / Hamida Tobbiche

Public ISBD Titre : Digital twin–assisted urgency-aware task offloading in UAV–HAP edge networks Type de document : document multimédia Auteurs : Hamida Tobbiche, Auteur ; Fatima Reche, Auteur ; Yousra Cheriguene, Directeur de thèse Editeur : Laghouat : Université Amar Telidji - Département d'informatique Année de publication : 2026 Importance : 59 p. Accompagnement : 1 disque optique numérique (CD-ROM) Note générale : Option : Networks, systems and distributed applications Langues : Anglais (eng) Mots-clés : Mobile edge computing  UAV–HAP networks  Digital twin  Task offloading  Contextual bandits  Online learning  Disaster response Résumé : One promising approach for enabling distributed intelligence at the aerial edge is the integration of Mobile Edge Computing (MEC) with hierarchical Unmanned Aerial Vehicle (UAV) and High-Altitude Platform (HAP) networks. Such architectures provide flexible and real-time computation offloading capabilities, making them particularly suitable for post-disaster rescue and emergency response operations where terrestrial infrastructure may be unavailable. However, optimal task offloading and resource allocation in these environments remain challenging due to dynamic wireless conditions, uncertain aerial channels, fluctuating link quality, and the strict latency requirements of safety-critical applications. This work investigates task offloading and resource scheduling in hierarchical UAV HAPassisted edge networks. After reviewing existing offloading and selection strategies, we propose an integrated online decision framework, termed DT-MAB, which combines a telemetry-driven Digital Twin (DT) layer with a Contextual Multi-Armed Bandit (MAB) learning algorithm. The framework employs an Exponentially Weighted Moving Average (EWMA)-based DT model to predict short term channel conditions and support proactive decision making. To ensure operational reliability, the system incorporates a hard safety filter that enforces communication constraints and an urgency-aware emergency bypass mechanism that prioritizes time-critical tasks without exploration delay. Extensive simulations conducted under complementary operating conditions demonstrate the effectiveness of the proposed approach. Under tight deadline and heavy workload conditions, DT-MAB achieves a 1.55 reduction in the Urgency-Weighted Deadline Violation Rate (UW-DVR) compared with the best static offloading policy. During severe channel degradation events, the framework detects deteriorating link conditions within 2–3 time slots, maintains zero reliability violations throughout the degradation period, and achieves a critical-task deadline violation rate of only 7.1%. These results demonstrate that combining predictive Digital Twin intelligence with contextual online learning and safetyaware decision making provides a robust, adaptive, and deployable solution for next generation UAV–HAP edge computing systems operating in mission-critical environments. note de thèses : Mémoire de master en informatique Digital twin–assisted urgency-aware task offloading in UAV–HAP edge networks [document multimédia] / Hamida Tobbiche, Auteur ; Fatima Reche, Auteur ; Yousra Cheriguene, Directeur de thèse . - Laghouat : Université Amar Telidji - Département d'informatique, 2026 . - 59 p. + 1 disque optique numérique (CD-ROM). Option : Networks, systems and distributed applications Langues : Anglais (eng) Mots-clés : Mobile edge computing  UAV–HAP networks  Digital twin  Task offloading  Contextual bandits  Online learning  Disaster response Résumé : One promising approach for enabling distributed intelligence at the aerial edge is the integration of Mobile Edge Computing (MEC) with hierarchical Unmanned Aerial Vehicle (UAV) and High-Altitude Platform (HAP) networks. Such architectures provide flexible and real-time computation offloading capabilities, making them particularly suitable for post-disaster rescue and emergency response operations where terrestrial infrastructure may be unavailable. However, optimal task offloading and resource allocation in these environments remain challenging due to dynamic wireless conditions, uncertain aerial channels, fluctuating link quality, and the strict latency requirements of safety-critical applications. This work investigates task offloading and resource scheduling in hierarchical UAV HAPassisted edge networks. After reviewing existing offloading and selection strategies, we propose an integrated online decision framework, termed DT-MAB, which combines a telemetry-driven Digital Twin (DT) layer with a Contextual Multi-Armed Bandit (MAB) learning algorithm. The framework employs an Exponentially Weighted Moving Average (EWMA)-based DT model to predict short term channel conditions and support proactive decision making. To ensure operational reliability, the system incorporates a hard safety filter that enforces communication constraints and an urgency-aware emergency bypass mechanism that prioritizes time-critical tasks without exploration delay. Extensive simulations conducted under complementary operating conditions demonstrate the effectiveness of the proposed approach. Under tight deadline and heavy workload conditions, DT-MAB achieves a 1.55 reduction in the Urgency-Weighted Deadline Violation Rate (UW-DVR) compared with the best static offloading policy. During severe channel degradation events, the framework detects deteriorating link conditions within 2–3 time slots, maintains zero reliability violations throughout the degradation period, and achieves a critical-task deadline violation rate of only 7.1%. These results demonstrate that combining predictive Digital Twin intelligence with contextual online learning and safetyaware decision making provides a robust, adaptive, and deployable solution for next generation UAV–HAP edge computing systems operating in mission-critical environments. note de thèses : Mémoire de master en informatique

Federated learning intrusion detection system for internet of things / Yacine Chettouh

Public ISBD Titre : Federated learning intrusion detection system for internet of things Type de document : document multimédia Auteurs : Yacine Chettouh, Auteur ; Mohamed Houssam Eddine Guenou, Auteur ; Yousra Cheriguene, Directeur de thèse Editeur : Laghouat : Université Amar Telidji - Département d'informatique Année de publication : 2026 Importance : 55 p. Accompagnement : 1 disque optique numérique (CD-ROM) Note générale : Option : Networks, systems and distributed applications Langues : Anglais (eng) Résumé : Nowadays, the Internet of Things (IoT) is widely used across critical infrastructures, connecting billions of embedded edge devices. In this context, securing these decentralized networks faces significant limitations due to the severe resource constraints and privacy requirements of IoT nodes, making traditional centralized Intrusion Detection Systems (IDS) impractical. While Federated Learning (Federated Learning (FL)) offers a privacy-preserving alternative, standard protocols like FedAvg face problems such as severe network congestion and high communication overhead caused by the continuous transmission of massive, uncompressed model updates over low-bandwidth links. To address these problems, therefore, this dissertation proposes a new communication-efficient Federated Learning Intrusion Detection System (FL-IDS). Leveraging a FedProx optimization core, the framework integrates an on-device Adaptive Gradient Quality Gate and top-k parameter sparsification to filter redundant updates and drastically reduce payload sizes. Furthermore, an emergency staleness override guarantees stable participation across the heterogeneous network. The performance of the proposed architecture is analyzed thoroughly and extensively through simulation using the CICIoT2023 dataset. The results demonstrate that our framework achieves a massive 80% reduction in cumulative communication bandwidth compared to standard FedAvg, while successfully maintaining a high global detection accuracy of 91.5% and a macro F1-score of 0.786. Hence, this work provides a significant improvement in IoT security protocols, presenting a reliable, resource-aware solution for modern edge environments while establishing a basis for future research in the field. note de thèses : Mémoire de master en informatique Federated learning intrusion detection system for internet of things [document multimédia] / Yacine Chettouh, Auteur ; Mohamed Houssam Eddine Guenou, Auteur ; Yousra Cheriguene, Directeur de thèse . - Laghouat : Université Amar Telidji - Département d'informatique, 2026 . - 55 p. + 1 disque optique numérique (CD-ROM). Option : Networks, systems and distributed applications Langues : Anglais (eng) Résumé : Nowadays, the Internet of Things (IoT) is widely used across critical infrastructures, connecting billions of embedded edge devices. In this context, securing these decentralized networks faces significant limitations due to the severe resource constraints and privacy requirements of IoT nodes, making traditional centralized Intrusion Detection Systems (IDS) impractical. While Federated Learning (Federated Learning (FL)) offers a privacy-preserving alternative, standard protocols like FedAvg face problems such as severe network congestion and high communication overhead caused by the continuous transmission of massive, uncompressed model updates over low-bandwidth links. To address these problems, therefore, this dissertation proposes a new communication-efficient Federated Learning Intrusion Detection System (FL-IDS). Leveraging a FedProx optimization core, the framework integrates an on-device Adaptive Gradient Quality Gate and top-k parameter sparsification to filter redundant updates and drastically reduce payload sizes. Furthermore, an emergency staleness override guarantees stable participation across the heterogeneous network. The performance of the proposed architecture is analyzed thoroughly and extensively through simulation using the CICIoT2023 dataset. The results demonstrate that our framework achieves a massive 80% reduction in cumulative communication bandwidth compared to standard FedAvg, while successfully maintaining a high global detection accuracy of 91.5% and a macro F1-score of 0.786. Hence, this work provides a significant improvement in IoT security protocols, presenting a reliable, resource-aware solution for modern edge environments while establishing a basis for future research in the field. note de thèses : Mémoire de master en informatique

Multicast routing in swarm of UAVs / Soumia Djellikh

Permalink

---

1 (1 - 6 / 6) Par page : 25 50 100 200