Khalid Minaoui

Severe conditions in hard-to-reach regions, where terrestrial networks are limited, make data backhaul from these areas challenging. In this context, advances in the Space-IoT have created new opportunities for data collection and monitoring using LEO satellites. To address this issue, this paper presents an orbit-aware S&F architecture for data collection from an intelligent ground terminal located in remote areas using a 3U CubeSat orbiting at 500 km altitude in a Sun-synchronous orbit. The designed ground terminal integrates an ESP32 microcontroller and a 433 MHz LoRa module and is enhanced with an embedded satellite pass prediction, Doppler pre-correction, and an adaptive LoRa strategy. The CubeSat utilizes a TOTEM SDR receiver, onboard data buffering, and an S-band downlink with variable coding for throughput enhancement. The orbital model for satellite prediction is validated using Ansys STK software version 12, while UHF and S-band links are evaluated using time-varying link-budget analysis. The adaptive transmission is compared with fixed configurations in terms of usable contact time and data delivered per pass. Results indicate that the proposed system demonstrates the feasibility of a Store-and-Forward mission co-design that combines precise orbit prediction, Doppler compensation, and adaptive transmission for future Space-IoT applications.

Voice-based analysis is attracting growing interest as a noninvasive means of identifying early markers of autism spectrum disorder (ASD). While pretrained audio models such as YAMNet and VGGish provide complementary views of children’s speech, most existing studies rely on a single representation and do not explore how these embeddings may be combined in a structured manner. This work introduces the multibranch attention and fusion network (MBAFNet), an architecture designed to make fuller use of heterogeneous embeddings by processing each stream through its own convolutional encoder, extracting temporal cues at multiple scales, and modeling cross-representation interactions through a self-attention layer. A gating mechanism then regulates the relative contribution of each embedding before classification. Experiments conducted on the CASD-SC corpus under a subject-disjoint five-fold evaluation protocol show that MBAFNet achieves 94.17% accuracy, outperforming all evaluated baselines and previously reported state-of-the-art approaches. These findings indicate that carefully designed selective fusion can reveal complementary information contained in pretrained embeddings and supports the development of more robust speech-based ASD assessment frameworks, while broader validation remains necessary before screening-oriented use.

Computer security issues have a widespread impact across all sectors and are increasingly influencing the healthcare industry. Since 2019, several reported attacks have been specifically aimed at exploiting vulnerabilities in medical network protocols like 'DICOM' and 'HL7,' as well as file formats, particularly those related to digital imaging and medical communications. While numerous articles delve into subjects such as security and blockchain, they often disregard the significant role that human errors play in compromising security. Human errors within hospital computer systems can lead to substantial implications for both patient care and data security. This article delves into the various forms of human errors that can occur within such systems and assesses their potential consequences on the overall effectiveness and security of healthcare information management. The study underscores the necessity of effective mitigation strategies to reduce the associated risks of these errors and enhance the dependability of hospital computer systems.

We present an operational framework designed to convert hourly ERA5 reanalysis data into informed cloud-seeding decisions over the Moroccan-Iberian region. Seven machine-learning and deep-learning models were compared across three forecast horizons (H =+1 h, +3 h, and +6 h). The predictor set combines atmospheric variables, short-term temporal memory, neighbouring spatial information, and physically based indicators related to moisture transport and circulation structure. Among the tested approaches, the Enhanced Mixture-of-Experts model showed the most robust behaviour for short-range forecasting at the operational horizon H =+1 h and was therefore retained for the decision stage. In the independent test period, the model reached R² = 0.6495, KGE = 0.6013, and F1 = 0.8109. The forecast outputs feed into a two-stage eligibility filter that combines rainfall probability with atmospheric suitability, narrowing the candidate windows to those where seeding is physically plausible. Eligible cases are then passed to a reinforcement-learning controller, which derives continuous seeding actions from the prevailing atmospheric state under physically realistic constraints. We further assessed the framework across years with contrasting hydroclimatic regimes, including two comparatively dry years (2022 and 2024) alongside a wetter year (2026), and observed consistent behaviour from both the forecasting and decision components. The seeding maps exhibit a coherent spatial structure, with intervention potential concentrated along the northern Moroccan Atlantic coast and the western Iberian margin, while the drier southern sectors remain largely inactive. Taken together, this work integrates rainfall forecasting and operational decision support within a unified framework, offering a scalable and physically interpretable foundation for weather-modification strategies in regions marked by strong climatic variability.

Coastal upwelling off Morocco’s Atlantic shores fuels some of the world’s most productive marine ecosystems and underpins its vital fisheries. In this study, we introduce a robust, fully automated framework for detecting and characterizing upwelling dynamics in a region defined by steep thermal gradients and pronounced latitudinal variation. We perform an unsupervised segmentation of the ocean surface using a Gaussian Mixture Model (GMM), selecting the optimal number of clusters via the Bayesian Information Criterion (BIC). After identifying individual upwelling events, we perform a monthly climatological analysis to map their location, seasonal frequency, and intensity along the Moroccan coast. The results reveal clear seasonal cycles and persistent upwelling hotspots. This automated approach makes it easier to monitor Morocco’s upwelling system. It provides the data needed to support fisheries management and advance ecosystem research.

Introduction: In the context of digital education, integrating heterogeneous learning- related data sources remains a significant challenge due to schema variability, semantic mismatches, and platform diversity. This study aims to systematically evaluate current research on data integration within e-learning environments, focusing on identifying challenges, effective techniques, performance metrics, and future directions. Methods: A Systematic Literature Review (SLR) was conducted using Kitchenham’s guidelines, targeting empirical studies published between 2015 and 2024. The review applied rigorous inclusion, exclusion, and quality assessment criteria, resulting in 142 high-quality studies. Extracted data covered integration techniques, evaluation metrics, and educational contexts such as Learning Management Systems (LMS) and Student Information Systems (SIS). Results: Semantic modeling, ontology-based integration, and service-oriented architectures (SOAs) emerged as dominant techniques. Performance evaluations frequently relied on metrics such as accuracy, scalability, interoperability, and semantic completeness. Persistent challenges include real-time data handling, adaptability, and achieving semantic interoperability in largescale educational systems. Discussion: The review highlights a growing shift toward AI-driven and semantic approaches, reflecting the need for scalable and intelligent integration. However, the lack of standardized evaluation metrics and limited real-world deployment restrict generalizability. There is also a need for solutions addressing both technical and user-centric factors, including personalization and quality of experience. Conclusion: This review synthesizes current research on educational data integration, outlining key methods, challenges, and research gaps. Future efforts should prioritize real-time, adaptive frameworks supported by standardized evaluation protocols and validated through practical implementation in diverse educational settings.