Doctorat
Permanent URI for this collectionhttps://dspace.univ-boumerdes.dz/handle/123456789/59
Browse
30 results
Search Results
Item Application of medical informatics and data analysis methods for automatic medical diagnosis(Université M'Hamed Bougara Boumerdès : Faculté de Technologie, 2025) Hammachi, Radhouane; Messaoudi, Noureddine(Directeur de thèse)With the increased size and complexity of data, interest has rapidly emerged in adopting artificial intelligence (AI) and deep learning (DL) to create data-driven models for automating medical diagnosis, and for neuromuscular disorders (NMDs) in particular. Therefore, this thesis aims to address the gaps in this context. To provide clinicians with a more objective and accurate methods for assessing muscle fatigue, a convolutional neural network (CNN)-based DL model was proposed to classify simulated surface electromyography (EMG) signals into different maximum voluntary contraction levels, achieving and accuracy of 88.88%. To ensure transparency and clinicians trust, the interpretability of Multi-Layer Perceptron (MLP) and Residual Neural Network (ResNet)-based DL models that achieved 95.67% and 98.37% testing accuracies, respectively, for myopathy diagnosis, was investigated. Shapley additive explanation (SHAP) for feature-based interpretation, and Gradient-weighted class activation mapping (Grad-CAM) for visual interpretation of raw signals, were employed, providing clear insights into the decision-making process. Furthermore, given the recent emergence and proved ability of quantum machine learning to handle high-dimensional data and solve complex tasks, a study was introduced to explore its potential in myopathy diagnosis. Quantum support vector machines (QSVMs) with variational quantum circuit-based kernels were proposed, and their performance was compared with classical methods. A hybrid QSVM model trained on deep features demonstrated promising classification ability, with training and testing accuracies of 96.7% and 85.1%, respectively. The results obtained in our research shed new light on the application of medical informatics in the field of healthcare and the EMG-based NMDs diagnosis in particular, indicating promising potential for future adoption of automated medical decision-makingItem Multivariate statistical process monitoring using kernel statistical techniques(Universite M'Hamed Bougara Boumerdès : Institut de Génie Eléctrique et Eléctronique, 2025) Kaib, Mohammed Tahar Habib; Harkat, Mohamed Faouzi(Directeur de thèse)Fault Detection and Diagnosis (FDD) is an important part of industrial plants because monitoring systems are responsible for capturing faults as soon as they occur to avoid major casualties in equipment, operators, and the environment......Item Etude et conception d'une antenne patch à bande rejetée(Université M'Hamed Bougara Boumerdès : Faculté de Technologie, 2025) Fortas, Ibrahim; Ayad, Mouloud(Directeur de thèse)Les communications sans fil connaissent une expansion rapide, nécessitant des solutions innovantes face à la hausse des besoins en débits élevés et à la saturation du spectre. La technologie ultra-large bande (ULB) se présente comme une alternative prometteuse. Dans le cadre de cette thèse, deux nouvelles antennes ULB ont été proposées : une antenne élémentaire patch à doubles ellipses alimentée par une ligne coplanaire (CPW) et une antenne microruban réseau de dipôles logarithmiquement périodique (MLPDA) alimenté par deux lignes microrubans. Etant donné que les systèmes ULB peuvent générer des interférences avec les applications existantes (WiMAX, WLAN, bande X, etc.), l'intégration de mécanismes de bandes rejetées s'avère essentielle. Pour y remédier, des cellules à métamatériaux sous forme de résonateurs en boucle ouverte ont été placées près du patch de la première antenne, tandis que des stubs rectangulaires ont été connectés à la ligne microruban de la seconde antenne. Ces techniques permettent de rejeter les bandes WLAN et Satellite DL (bande X) pour l’antenne patch, ainsi que les bandes WiMAX et WLAN pour l’antenne MLPDA. Les performances des antennes ont été analysées via des simulations sous CST Studio, puis validées par des mesures expérimentales. Les résultats obtenus montrent une bonne concordance entre simulations et expérimentations, confirmant l’efficacité des solutions proposées pour répondre aux exigences des communications sans fil modernes en termes de large bande et de rejet d’interférencesItem Interaction analysis and coupled vibrations control in rotary drilling systems(Universite M'Hamed Bougara Boumerdès : Institut de Génie Eléctrique et Eléctronique, 2025) Meddah, Sabrina; Idir, Abdelhakim(Directeur de thèse)This work focus on coupled vibration analyses of torsional-axial and torsional-lateral interactions control with a PID classic and an FOPID controller optimized by PSO for rotary drilling systems, oil and gaze field, in different Scenario. Where, the vibrations encountered during drilling operations are strongly coupled, resulting in complex effects on drilling performance. These vibrations can be categorized into three types based on their propagation directions within the drilling systems: axial, lateral, and torsional. While many researchers have individually studied each type of vibration, the robustness of developed controllers depends on considering the coupling effects of the other vibrations that are often overlooked. To ensure the robustness of such controllers, it is imperative to analyze the interaction effects of the control systems, specifically focusing on the torsional-axial and torsional-lateral interactions. The primary objective of this thesis is to investigate the interactions among these three types of vibrations and subsequently propose controllers for the coupled cases based on the interaction analysis results. The proposed contribution of interaction analysis was demonstrated through simulation studies under MATLAB, and FOPID controller was designed then optimized using PSO technique and RN. The deep analyses of obtained results demonstrated improved system controller robustness compared to previous studiesItem Sélection des paramètres pertinents pour la reconnaissance des expressions faciales en mode indépendant des personnes(Université M'Hamed Bougara Boumerdès : Faculté de Technologie, 2025) Boukhobza, Fatima Zohra; Rouabah, Khaled(Directeur de thèse)La reconnaissance automatique des expressions faciales est une tâche essentielle dans de nombreux domaines tels que la psychologie, l'interaction homme-machine, et la surveillance. Un système de REF se divise principalement en deux phases : la phase d'apprentissage et la phase de reconnaissance. Dans un premier temps, l'apprentissage vise à modéliser les différentes expressions faciales en extrayant des caractéristiques discriminantes. Dans un second temps, la phase de reconnaissance permet de classifier une nouvelle image faciale en fonction de son expression émotionnelle à l'aide de techniques telles que les classifieurs K-Nearest Neighbors (KNN), Support Vector Machines (SVM), ou encore les réseaux de neurones artificiels (ANN). Toutefois, les systèmes REF opérant en mode indépendant de la personne, c'est-à-dire sans personnalisation préalable aux individus, présentent généralement des taux de reconnaissance plus faibles. Cette baisse de performance est souvent due à la variabilité des visages selon les personnes (âge, sexe, morphologie), créant des interférences lors de la reconnaissance des expressions. Afin de surmonter cette difficulté, la sélection de paramètres pertinents devient un enjeu majeur. Cette thèse vise à améliorer les performances de ces systèmes en sélectionnant un ensemble de paramètres pertinents pour la reconnaissance des expressions, qui permettent de distinguer efficacement les expressions tout en minimisant l'influence des variations entre individus. En s'appuyant sur des techniques de sélection de caractéristiques telles que les méthodes de type filter, wrapper ou embedded, ce travail vise à concevoir un système plus robuste et performant dans des contextes où les expressions faciales doivent être reconnues de manière fiable et indépendante de l'identité des sujetsItem Contrôle des systèmes FACTS dans un réseau electrique connecté à une source d’energie renouvelable(Universite M'Hamed Bougara Boumerdès : Faculté des Hydrocarbures et de la Chimie, 2025) Boukarana, Leila; Fellag, Sid Ali(Directeur de thèse)Le présent travail de recherche traite la problématique du contrôle des réseaux électriques par les systèmes FACTS (Flexible AC Transmission System) avec l’intégration d’une source d’énergie renouvelable. Le contrôle du power system porte sur : la compensation de l’énergie réactive des lignes, maintien de la tension au niveau des jeux de barres, l’amélioration du transit des puissances dans les lignes de transport et l’injection de l’énergie photovoltaïque sur le réseau électrique. Plusieurs approches et techniques sont analysées et évaluées à travers des simulationsItem Contribution to the design of new antenna structure more efficient for 5G communications systems(Université M'Hamed Bougara Boumerdès : Faculté de Technologie, 2025) Khodja, Khalida; Atia, Salima(Directeur de thèse)This thesis explores key advancements and challenges in telecommunications and antenna design, focusing on the evolution towards 5G networks and the utilization of millimeter-wave (MMW) frequencies. The first chapter provides a comprehensive overview of 5G networks, emphasizing the unique propagation characteristics and transformative potential of MMW technology. It investigates the technological innovations required to optimize MMW spectrum for ultra-high-speed data transmission. The second chapter presents a comprehensive overview of the evolution of wave guiding techniques including their fundamental principles and common applications, detailing their historical development, advantages, and drawbacks. It traces the advancements from traditional hollow waveguides to more recent innovations designed to meet the increasing demands of high-frequency communication systems, particularly in the MMW band. It also explores the limitations of these conventional techniques that have spurred the development of novel waveguide technologies. Among the various emerging techniques, this chapter highlights the Ridge Gap Waveguide (RGW) technology as the most promising solution for MMW applications and discusses in detail its main characteristics, while displaying its key advantages. Actually, the RGW's ability to overcome many of the challenges faced by traditional waveguides are emphasized, showing that the RGW's unique ridge structure offers a significant improvement in performance and versatility, which makes it a superior candidate for next-generation communication systems. Additionally, this chapter addresses the current drawbacks of RGWs and it concludes with a critical evaluation of RGW technology in the context of its application to advanced communication network. This overview establishes a foundational understanding of wave guiding techniques and positions RGW technology as a leading candidate for addressing the demands of modern high-frequency communication systems. The third chapter introduces a novel antenna design tailored for the Ka-band frequency range, featuring both dual-band and dual-beam radiation capabilities. This dual-band functionality is realized through a carefully engineered radiating structure that accommodates the different wavelength requirements of each band, ensuring optimal performance and minimal interference. In addition to its dualband capability, the antenna features a dual-beam radiation pattern; this design innovation allows for simultaneous coverage of two separate spatial regions, enhancing the system's flexibility and efficiency. Chapter “four” introduces a miniaturized, high-gain, and highly efficient antenna designed for operation at 60 GHz, leveraging the innovative Double Printed Ridge Gap Waveguide (D-PRGW) technology. The proposed antenna utilizes D-PRGW technology to achieve exceptional performance while maintaining a compact size factor. This design innovation allows for a significant reduction in antenna dimensions without compromising gain or efficiency. By employing a dual-ridge configuration, the antenna effectively mitigates signal losses and enhances power handling capabilities, making it wellsuited for high-frequency applications where space constraints are a major concern. The chapter provides a detailed analysis of the antenna's design, including its geometric parameters, simulation results and experimental measurements that demonstrate the antenna's excellent performance metrics, such as gain, beam width, and efficiencyItem Contribution to improving the efficiency of a wireless power transfer system using artificial intelligence techniques(Université M'Hamed Bougara Boumerdès : Faculté de Technologie, 2024) Bennia, Fatima; Boudouda, Aimad(Directeur de thèse)Wireless Power Transfer (WPT) technology is an innovative method for powering devices without physical wires, which has been used here to provide power to bioimplantable devices. The main design constraints are to achieve maximum transfer efficiency while keeping the implant size small enough to be suitable for the living subject's body. Magnetic Resonant Coupling Wireless Power Transfer (MRCWPT), which uses pairs of inductor coils in the external and implant circuits, is a method actively researched for this type of power transmission. The objective of this thesis is to design and optimize a high-efficiency WPT receiving coil for biomedical applications. Traditionally, optimizing WPT systems based on mathematical equations or numerical models is often time-consuming and may not yield optimal designs. To address these limitations, this thesis introduces a novel approach that integrates a machine-learning model with metaheuristic methods for design and optimization. The primary goal is to maximize the transfer efficiency for an implantable coil with dimensions of 20 mm and a transfer distance of 30 mm, operating at a frequency of 13.56 MHz. To achieve this goal, we firstly identified the critical geometric coil parameters that significantly influence the WPT system's efficiency. A model-based Artificial Neural Network (ANN) was then constructed and trained on a comprehensive dataset generated through Finite Element Method (FEM) simulations. This model predicts efficiency based on geometric coil parameters, eliminating the need for complex calculations. Subsequently, two metaheuristic algorithms: the Genetic Algorithm (GA) and the Coyote Optimization Algorithm (COA), were employed to find the optimal parameters that maximize efficiency. The proposed ANN model demonstrates exceptional accuracy, exceeding 97%. Furthermore, this WPT coil design approach significantly enhances transfer efficiency by up to 76% while drastically reducing computation time compared to conventional methodsItem Contribution to chaotic encryption methods for digital data(Université M'Hamed Bougara Boumerdès : Faculté de Technologie, 2024) Bourekouche, Hadjer; Belkacem, Samia(Directeur de thèse)In this thesis, we investigate the development process of chaos-based image encryption algorithms from various perspectives, including the serious challenge of generating secure random number sequences for use as dynamic encryption keys. First, at the aim of improving the randomness and non-periodicity qualities of the basic pseudo-random number generators (PRNG) used as key-stream generators, we exploit the unique attributes of the logistic map (LM), logistic-sine system(LSS), linear feedback shift registers (LFSR), and nonlinear feedback shift register (NLFSR) to design new key-stream generators (namely: LSS-LFSR-PRNG, LM-NLFSR-PRNG, and LSS-NLFSR-PRNG). Therefore, our generators succeed in generating unlimited, random, and nonlinear sequences by passing the totality of the National Institute of Standard and Technology (NIST) statistical tests, and displayed strong cryptographic security, resulting in high entropy, high key sensitivity, and large key space exceeding 2^100. The second goal highlights the importance of selecting an appropriate chaos-based architecture for confusion and diffusion. The dimensions of the chaos-based confusion-diffusion architecture vary depending on the specific chaotic map being used. Hence, we design three confusion-diffusion algorithms of various levels (1D LM-based cryptosystem, 1D LM-Chebyshev-based cryptosystem, and 3D intertwining logistic map-cosine (ILM) based cryptosystem), to discuss and demonstrate the impact of choosing the appropriate dimension of the chaotic map on the vulnerability of a cryptosystem. It has been proven that higher-dimensional chaotic maps, such as 3D-ILM, can enhance the ability to resist exhaustive and statistical attacks by achieving desirable values of the number of pixels change rate (NPCR) and unified average changing intensity (UACI), while these maps are unable to maintain encryption speed. The third goal of this thesis is to improve the core of the mathematical model of chaos-based cryptosystems by boosting the chaotic complexity and chaotic range of basic one-dimensional chaotic maps. Where, we propose a new nonlinear chaotification system capable of producing 1D enhanced discrete chaotic maps (enhanced tangent-Logistic map T-LM, enhanced tangent-Sine map T-SM, and enhanced tangent-Chebyshev system T-CH), by applying tangent nonlinear transforms to the outputs of the existing chaotic maps. This strategy improves the performance of basic 1D chaotic maps by exhibiting better dynamical behavior, Lyapunov exponent, bifurcation, and larger chaotic intervalsItem Biomedical security : performance study and analysis(Université M'Hamed Bougara Boumerdès : Faculté de Technologie, 2024) Benyoucef, Aicha; Hamadouche, M'Hamed(Directeur de thèse)This thesis investigates the development of a robust approach for securing medical data through watermarking techniques, with a specific focus on the application of QR code encryption. The research addresses the pressing need for improved security measures in medical data transmission and storage, considering the vulnerability of patient information to unauthorized access and manipulation. Through a comprehensive literature review and analysis of existing methods, the thesis identifies key challenges in medical image watermarking, including limitations in payload capacity, imperceptibility, and robustness against attacks. To address these challenges, the research proposes a novel watermarking approach that leverages QR code encryption to enhance both security and capacity within medical images. The methodology involves embedding QR code representations of Medical Imaging Test Reports (MITR) into the non-interest regions of medical images using Discrete Wavelet Transform (DWT) and Singular Value Decomposition (SVD) techniques. Evaluation of the proposed method is conducted using performance metrics such as Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), and Normalization Coefficient (NC). The results demonstrate significant improvements in payload capacity, imperceptibility, and security against various attacks compared to existing watermarking methods. The pro- posed approach offers a balance between security requirements and practical considerations, making it suitable for real-world applications in medical data transmission and storage. Overall, this research contributes to advancing the field of medical image watermarking and lays the foundation for future developments in biomedical security
- «
- 1 (current)
- 2
- 3
- »