Research Articles

Permanent URI for this collection

https://dir.muni.ac.ug/handle/20.500.12260/17

Browse

Now showing 1 - 20 of 81

Medicinal and aromatic plants as climate-smart crops: case studies on Pelargonium graveolens and Viola odorata under Egyptian conditions
(Springer Nature, 2026-03-04) Hamed, Sobhy A.; Abo-Karima, Mohamed K.; Ali, Guma; Elmessery, Wael M.; Elwakeel, Abdallah Elshawadfy; Ahmed, Atef Fathy; AL-Harbi, Mohammad S.; Abouelatta, Ahmed M.
Medicinal and aromatic plants (MAPs) represent high-value agricultural commodities that provide economic returns through essential oil production while potentially contributing to climate change mitigation via photosynthetic carbon sequestration and oxygen release. Despite their recognized economic importance, few studies have systematically quantified the net environmental performance of MAP cultivation and processing within integrated climate mitigation frameworks. This study evaluated the carbon footprint, oxygen production, and CO₂ absorption of two commercially important MAPs—Pelargonium graveolens (geranium) and Viola odorata (violet)—cultivated under Egyptian field conditions, using life cycle assessment methodology with system boundaries from field operations through extraction. Primary data were collected from commercial farms (geranium: 37 feddans aggregated; violet: 1 feddan) over complete growing cycles. Geranium (6-month season) demonstrated net climate-positive performance with a negative carbon footprint of − 375 kg CO₂-eq. per feddan per season, producing 54,324 m³ of oxygen and absorbing 155,632 kg CO₂ during growth, with photosynthetic uptake exceeding all process emissions (fuel, irrigation electricity, fertilizers, and composting). In contrast, violet (12-month annual cycle) exhibited a positive footprint of + 15,972 kg CO₂-eq. per feddan annually, despite generating 11,148 m³ oxygen and absorbing 12,700 kg CO₂, primarily due to its fuel-intensive solvent extraction process that accounts for 97.3% of total emissions. Monte Carlo uncertainty analysis (N = 10,000 simulations) confirmed geranium’s robustness as a net carbon sink (probability 67.4%) while violet remained a consistent carbon source under current extraction practices. Scenario modeling demonstrated that substituting fossil fuel with solar thermal energy or biogas-derived heat for violet distillation could reduce net emissions by 50–100%, potentially shifting the crop from carbon source to near-neutral status. These findings indicate that MAPs can function as climate-smart crops when cultivation practices are coupled with renewable energy integration in post-harvest processing. The study provides quantitative evidence for prioritizing low-emission extraction technologies and precision irrigation management in MAP value chains to maximize both economic and environmental sustainability outcomes.
Prevalence and health risk factors of nomophobia among students in private colleges
(Modestum DOO, Serbia, 2026-03-31) Mahajan, Sunita; Thapa, Pramila Pudasaini; Sharma, Prakash; Tsirkas, Panagiotis; Ali, Guma; Diamanti, Konstantina; Adamopoulos, Ioannis Pantelis
Nomophobia is the fear of being out of smartphone contact. This study examines its prevalence and potential links to socio-demographic and risk factors. Modern technologies have led to nomophobia, a psychosocial risk factor causing technostress. This fear of new technologies is influenced by ergonomics, which studies how humans physically react to and fit with devices. Technostress is a result of altered behaviors resulting from the use of modern technologies at work and home. The primary goal of this research was to assess the prevalence of nomophobia among college students with specific objectives and research questions. Researchers used a quantitative cross-sectional design to assess nomophobia among 231 higher secondary students. Participants completed a semi-structured, self-administered questionnaire, and the study maintained ethical considerations. Researchers analyzed the data using SPSS version 26. The respondents had a mean age of 17.18 years. The study found that 49.8% used smartphones for more than 1-3 hours daily, while 28.1% checked their phones for notifications a few times daily. Findings revealed that 32% of respondents experienced mild nomophobia, 34.2% had a moderate level, and 33.8% suffered from severe nomophobia. Sixty-seven-point-five percent of respondents used smartphones primarily for social media. The study found no significant association between socio-demographic factors and nomophobia levels. However, daily smartphone usage showed a substantial correlation with nomophobia severity. These results indicate that most respondents experienced some degree of nomophobia, with its prevalence likely increasing. Effective management of nomophobia requires early detection and proactive intervention strategies.
AHNet: Design and execution of adaptive hybrid network for credit risk prediction using spatio-temporal attention-based convolutional autoencoder features in the banking sector
(Springer Nature, 2026-01-07) Ahmad, Ahmad Y. A. Bani; Shukla, Madhu; Jayaprakash, B; Bharathi, B; Ali, Guma; Yogapriya, J
Over the past ten years, pattern recognition experts have become very interested in market financial predictions. To support investment decision-making processes, it is crucial to develop an intelligent financial forecasting model in the financial markets. However, multivariate financial time series prediction is still difficult. Time series data are typically used for market analysis, and the high degree of fluctuation in this type of data necessitates the use of highly effective classification tools that are prevalent in the state of the art, such as Convolutional Neural Networks (CNNs) systems like AlexNet, residual network, Inception, and so forth. Researchers must start from scratch when training new tools as a result. These procedures could take a long time. Therefore, it is crucial to address the many issues related to using conventional methods for financial forecasting. The benchmark resources are used to gather the necessary financial data for verification in the first step. Next, the collected data are offered to the Spatio-Temporal Attention-based Convolutional Autoencoder (STA-CAE)-based feature extraction phase. The financial prediction phase receives the essential features used for validation extracted in this phase. Here, a novel Adaptive Hybrid Network (AHNet) is employed to perform the prediction procedures. The developed AHNet is an integrated version of Ridge Regression with Stacked Residual Recurrent Neural Network (RR-SResRNN). Moreover, the parameters of AHNet are optimized utilizing the Improved Random Array-based Secretary Bird Optimization Algorithm (IRA-SBOA) that helps to improve the prediction efficiency of the developed prediction technique. The efficiency of the created technique is compared to classical frameworks by executing various tests after collecting financial prediction outcomes from the AHNet model.
Prevalence and clinical risk factors of stroke among hypertensive patients: A Cross-sectional study
(Peninsula Press, 2026-03-09) Adamopoulos, Ioannis; Eslahi, Aida Vafae; Syrou, Niki; Mishra, Harshit; Tsirkas, Panagiotis; Ali, Guma
Background: Stroke continues to be a primary cause of death and disability which mainly affects patients who have high blood pressure. The research set out to determine how often strokes occur together with their associated clinical risk factors in this specific patient group. Methods: Our research applied a quantitative method to perform a cross-sectional study which included 1,024 patients who had hypertension. Our research team conducted an analysis of 15 variables which included numerical data about patient age and their blood pressure readings and cholesterol measurements and categorical data about their gender and their heart disease status. Our research team conducted statistical analyses to identify which factors most strongly predicted the development of stroke in our patient population. Variables significant only in univariate analysis (p<0.05) but not retained after adjustment are displayed in the Univariate‑only section. ORs are plotted on a logarithmic scale, follows conventions consistent with Stata 17, data analysis through descriptive statistics and machine learning approaches while using Python and Excel as their main software applications. Results: Our research findings demonstrated that 25% of patients with hypertension developed strokes while age emerged as the most critical factor which increased stroke risk. The highest occurrence of 45% appeared in patients who were 70 years old or older. The research showed that glucose levels above 126 mg/dL together with obesity defined by a BMI of 30 kg/m² or higher served as important predictive factors which achieved statistical significance through p-values of 0.002 and 0.01. The presence of heart disease was also linked to increased stroke risk (p 0.001), emphasizing the need for comprehensive assessments in this demographic. Conclusion: The research findings reveal an urgent requirement for focused treatment programs which target adjustable risk elements including patient age and high blood sugar levels and excess body weight in people with hypertension. Our research shows that active control of these risk elements will lower the chance of stroke development which proves that patients need to follow specific lifestyle changes and maintain their health through scheduled checkups to achieve better results for their condition. Article
A Survey on securing smart finance using artificial intelligence and blockchai
(Peninsula Publishing Press, 2026-01-10) Ali, Guma; Otim, Emmanuel; Mijwil, Maad M.; Buruga, Bosco Apparatus; Eslahi, Aida Vafae; Adamopoulos, Ioannis
The rapid digitalization of financial services has given rise to smart finance ecosystems that integrate FinTech platforms, Internet of Things (IoT) devices, cloud infrastructures, and decentralized applications. While these systems enhance automation, operational efficiency, and financial inclusion, their highly distributed, data-intensive architectures introduce critical security, privacy, and trust challenges. In this context, artificial intelligence (AI) and blockchain have emerged as complementary technologies capable of addressing these challenges through intelligent decision-making, advanced threat detection, data integrity, and transparent operations. This survey provides a comprehensive review of recent research on securing smart finance systems using AI- and blockchain-based approaches. The survey comprehensively analyzed research published between 2023 and 2026 using the Scopus database, focusing on the keywords “AI,” “blockchain,” and “smart finance.” The analysis reveals extensive use of AI-driven security mechanisms, including credit scoring and risk assessment, transaction monitoring and fraud detection, anti-money laundering (AML) and know-your-customer compliance, identity verification, cyber threat detection, smart contract security analysis, behavioral biometrics, insurance fraud detection, and market risk prediction. In parallel, the survey examines blockchain-enabled security solutions, including secure payment and settlement systems, cross-border remittances, AML and counter-terrorism financing frameworks, digital identity management, smart contracts, asset tokenization, decentralized finance, auditability, and secure interbank communication. The integration of AI and blockchain offers significant advantages, including improved fraud detection accuracy, enhanced transparency and traceability, stronger data integrity, automated compliance, real-time threat response, and increased system resilience. Despite these benefits, key challenges persist, particularly in scalability, privacy preservation, interoperability, regulatory and ethical compliance, energy efficiency, explainability, and post-quantum security. The survey concludes by outlining future research directions and design guidelines for developing secure, scalable, and trustworthy smart finance systems that effectively leverage the integration between AI and blockchain.
Blockchain and deep Q-learning for trusted cloud-enabled drone network in smart forestry: A Survey
(Imam Ja’afar Al-Sadiq University, 2025-12-14) Ali, Guma; Wamusi, Robert; Mijwil, Maad M.; Al-Hamzawi, Hassan A. Hameed; Al Sailawi, Ali S. Abed; Salau, Ayodeji Olalekan
The convergence of drone technology, cloud computing, and intelligent decision-making is revolutionizing precision forestry. However, deploying large-scale drone networks in smart forestry faces challenges such as trust, security, data integrity, and autonomous coordination. This survey examines how combining Blockchain technology with deep Q-learning (DQL) can address these issues within cloud-enabled drone networks. Drawing on 102 peer-reviewed sources published between 2022 and 2025 from reputable platforms such as ACM Digital Library, Frontiers, Wiley Online Library, PLoS, Nature, Springer, ScienceDirect, MDPI, IEEE Xplore Digital Library, Taylor & Francis, Sage, and Google Scholar, this work highlights recent advancements in secure and intelligent drone ecosystems. Blockchain provides a decentralized, tamper-resistant framework for validating transactions and securing data exchange among autonomous drones, ensuring the integrity, confidentiality, and authenticity of environmental data. This is critical in forestry, where data manipulation and unauthorized access pose significant risks. Complementing this, DQL enables drones to make autonomous decisions by interpreting real-time environmental data and learning from past experiences, allowing drones to adjust their flight paths, optimize resource utilization, and enhance data collection in dynamic forest environments, such as wildfires or illegal logging operations. Together, Blockchain and DQL create a resilient, scalable architecture that supports secure, real-time, and intelligent forest monitoring. This framework lays the groundwork for developing autonomous and trustworthy drone networks that promote sustainable and climate-smart forestry management.
Fuzzy-PID control design and performance analysis for PMSM drives in electric vehicles
(Lamintang Education and Training (LET) Centre, 2025-12-28) Kalyankolo, Umaru; Nafuna, Ritah; Mugabe, Rodney; Nansukusa, Yudaya; Asikuru, Salaama; Ochima, Noah; Mutaburura, Pison; Kalyankolo, Zaina
The increasing demand for high performance and energy efficient electric vehicles has driven research into advanced motor control strategies for Permanent Magnet Synchronous Motors. This study investigates the design and performance evaluation of a Fuzzy PID controller as the speed regulator to address the limitations of the typical PID controllers in EV propulsion and a field-oriented control strategy is used. A conventional PID controller is initially implemented and tuned using the Ziegler-Nichols closed loop method. A Fuzzy Inference System is developed and then integrated with the PID controller to form a hybrid Fuzzy PID controller capable of adjusting the PID gains in real time. The performance of both controllers is evaluated under various test scenarios including speed variations, load disturbances, and parameter changes. Simulation results demonstrate that the Fuzzy PID controller significantly reduced overshoot by 0.5%, reduced rise time by 32.04%, improved settling time by 8.04%, and therefore enhanced system stability and responsiveness compared to the typical PID controller. These improvements validate the effectiveness of fuzzy logic in managing the uncertainties associated with PMSM control in EV applications.
Within reach? Sustainable energy infrastructure financing for “hardest to reach” communities
(Frontiers, 2025-11-28) Pailman, Whitney; Caprotti, Federico; Yaguma, Penlope; Hastie, Helena; Oemmelen, Katharina; Opio, Innocent Miria; Sheridan, David
Providing energy access in “hard to reach” under- or unelectrified contexts like informal settlements or remote rural regions requires rethinking how we develop and finance energy access business models. While terminologies like “hardest to reach,” “reaching the last mile” or “leaving no-one behind” have increasingly been used within energy access and broader development discourses, different country and regional contexts present unique and practical challenges for deploying electrification models in these areas. These challenges are also intrinsically linked to the viability gap, which results from a disjuncture between end-users' ability to pay and revenues required to cover the cost of service. “Hard to reach” areas can comprise geographically remote regions like rural villages or urban informal settlements where households and businesses are precluded from grid electricity and other key infrastructure services due to financial, socio-technical and socio-political barriers despite being directly “under the grid.” In this paper we argue that contextual grounding is needed when exploring the intricacies of delivering energy access in contexts that traditionally lack formal service provision, security of tenure and material certainty. We furthermore argue that it is necessary to critically engage with discourses that characterize geographic remoteness as “un-electrifiable.” Notwithstanding the increased focus on leaving no-one behind in the international agenda, more pragmatic grounding is needed to understand and draw lessons from energy access in dynamic contexts. Drawing on the authors' current and prior experience working on research projects on off-grid energy and other infrastructures across sub-Saharan Africa, the paper compares the geographic contexts of urban informality and geographically remote contexts through six case studies from Kenya, Rwanda, Uganda, South Africa, the Kingdom of Eswatini (formerly Swaziland) and Madagascar. It explores the intricacies and practicalities of providing energy access in urban informal settlements, remote rural villages or displacement settings, and provides lessons for policy and practice.
Blockchain and quantum machine learning approach for securing smart water management systems: A Scoping review.
(Peninsula Publishing Press, 2025-09-01) Ali, Guma; Mijwil, Maad M.; Adamopoulo, Ioannis; Dhoska, Klodian
Smart water management systems (SWMS) increasingly rely on Internet of Things (IoT) devices to enhance water distribution, detect leaks, and support sustainable resource use, but this reliance also heightens exposure to cyberattacks, data manipulation, and privacy risks. Conventional security approaches often fall short due to the decentralized design and real-time demands of these systems. This scoping review analyzes 266 studies published between January 2022 and December 2025 to assess how integrating Blockchain and quantum machine learning (QML) can strengthen the security, privacy, and reliability of SWMS. The review examines Blockchain-enabled water management, quantum computing applications, and QML-based security frameworks, using thematic analysis to categorize emerging architectures and challenges. Findings of the focused studies show growing adoption of Blockchain for secure data logging, access control, and tamper-proof auditing. At the same time, QML demonstrates strong potential in anomaly detection, predictive maintenance, and optimizing distribution networks. Although these technologies offer a promising foundation for resilient water infrastructure, most research remains conceptual, with limited real-world deployment or scalability assessments. Integrating Blockchain with QML could create robust, privacy-preserving SWMS frameworks. However, significant barriers persist, including the computational intensity of quantum models, interoperability issues with existing IoT infrastructures, and the absence of standardized protocols. Addressing these gaps is essential for practical implementation. This review underscores the need for scalable hybrid designs, applied validation, and cross-disciplinary standards to advance secure, efficient, and sustainable smart water management solutions.
Development, performance evaluation and prediction of optimal operational conditions for a double-row sugarcane harvester using deep learning
(Springer Nature, 2025-12-01) Elwakeel, Abdallah Elshawadfy; Elden, Abdallah Zein; Ahmed, Saad F.; Issa, Sali; Li, Changyou; Ali, Khaled Abdeen Mousa; Hanafy, Waleed M.; Ali, Guma; Alzahrani, Fawaz; Fathy, Atef
Sugarcane is a vital global crop, serving as a primary source of sugar, biofuel, and renewable energy. Advancements in harvesting are critical to meeting rising demand, enhancing profitability, and supporting eco-friendly agricultural practices in the sugarcane sector. Based on the current challenges of sugarcane harvesting in developed countries, the current study aimed to develop a semiautomatic whole-stalk sugarcane harvester (SWSH) for harvesting two rows of sugarcane stalks at a time and to be front-mounted on a classic four-wheel agricultural tractor. Then performance evaluation and prediction of optimal operational conditions for a double-row sugarcane harvester using Feedforward Neural Network (FNN) and Deep Neural Network (DNN) at different levels of forward speeds (3, 3.5, 4.5, and 5 km/h), row spacing (71, 78.89, and 88.75 cm), cutting heights (0, 2, and 4 cm), and numbers of knives (2 and 4) of the cutting systems. The obtained results showed that the cutting efficiency of the developed SWSH reached 100%. Where the higher cutting efficiency was observed at a cutting height equal to zero (ground level), forward speed of 3 km/hand row spacing of 71 cm using both 2 and 4 knives. The minimum total operating cost of the developed SWSH was about 4.42 USD/ha, and it was detected when using a forward speed of 4.5 km/h, row spacing of 88.75 cm, a cutting height of 4 cm, and two knives only on the cutting disk. Furthermore, at a row spacing of 88.75 cm, the maximum field capacity of the developed SWSH was 0.554 ha/h, observed at a forward speed of 4.5 km/h.
Performance evaluation of a triple-sided solar dryer in terms of energy-exergy analysis, sustainable indicators and CFD simulation during drying tilapia fish strips
(Springer Nature, 2025-11-17) Ghanem, Tarek Hussien M.; Oraiath, Awad Ali Tayoush; Nsasrat, Loai S.; Ali, Guma; Younis, Omar Shahat; Eldin, Abdalla Zain; Elmolakab, Khaled Mohamad; Alhumedi, M.; Ahmed, Atef Fathy; Tantawy, Aml Abubakr; Elwakeel, Abdallah Elshawadfy
Fixed flat-plate solar collectors suffer from low energy efficiency during mornings and evenings due to suboptimal solar incidence angles, reducing thermal output. While tracking systems improve efficiency by following the sun’s path, their high initial costs, mechanical complexity, and need for advanced control systems limit widespread adoption. These drawbacks demonstrate the importance of cost-effective, efficient alternatives that balance performance and simplicity in solar thermal applications. Thus, a triple-sided solar dryer (TSSD) integrated with intelligent airflow gating was developed to overcome these issues. This study evaluates the performance of a TSSD for drying tilapia strips at three thicknesses (4, 8, and 12 mm) using computational fluid dynamics (CFD), energy-exergy analysis, and sustainability indicators. According to the CFD simulations, they were employed to analyze airflow patterns, temperature distribution, and velocity profiles inside the TSSC and drying room (DR) during a day from 8 a.m. to 5 p.m. Additionally, the CFD was used to estimate the highest air temperature inside the drying to choose the appropriate speed of the air exhaust fan. The simulation analysis indicated that the highest air temperatures were 188.67, 124.4, and 96.51 °C, at three corresponding air velocities of the exhaust fan (1.0, 1.5, and 2.0 m/s), respectively, under a solar intensity of 872 W/m². Where the best velocity of the air exhaust fan was 2 m/s, it provided a uniform drying temperature of 96.51 °C, at solar noon (less than 100 °C). On the other hand, the energy-exergy analysis and sustainable indicators were estimated over two consecutive drying days (8 a.m.–5 p.m.) to assess the thermal behavior of the TSSC and TSSD. The energy analysis showed that the TSSC attained a maximum input energy of 1752.72 W and a useful energy of 810.31 W. Its energy efficiencies ranged from 40.79% to 57.21%. Meanwhile, the maximum drying efficiency was 8.19%, 8.51%, and 8.46% for tilapia strip thicknesses of 4, 8, and 12 mm, respectively. Furthermore, the exergy efficiency ranged from 7.28% to 32.83% (TSSC) and from 66.5% to 87.19% (DR). Additionally, sustainability indicators, such as improvement potential (IP) ranging from 1.19 to 7.22 W, waste exergy ratio (WER) between 0.67 and 0.93, and sustainability index (SI) from 1.08 to 1.49, showed that the system is both environmentally friendly and effective in its operations. The results show that the TSSD is an effective, eco-friendly, and affordable option compared to traditional solar drying systems, providing the best heat performance, better energy-exergy efficiency, and less harm to the environment for drying tilapia.
An Editorial vision for civil engineering: navigating intelligence and innovation
(Mesopotamian Academic Press, 2025-11-01) Mijwil, Maad M.; Adamopoulos, Ioannis; Ali, Guma
The integration of Artificial Intelligence in civil engineering has seen a major advancement over the applications of primitive data-driven models, reaching advanced hybrid physics-informed models. This evolution represents a significant change of paradigm from the basic predictive analytics to what is now called "structural cognition." But in this cutting-edge paradigm, the evolutionary learning of AI beyond the structural prediction is further developed, with the ability to learn underlying causal relationships in complicated engineering systems. This not only allows them to diagnose potential problems, but also to proactively propose well-targeted remedial measures, giving engineers a greater understanding of the situation as well as better, better-informed decision-making with respect to infrastructure sustainability and safety [1], [2]. A critical element to this evolution is the emergence of "perceptive infrastructure." This idea is further greatly supported by similar inventions like wavelet-diffusion architectures, and it is very useful in the design of the most robust vision-based monitoring systems. The systems are especially well-suited to demanding environments, such as low-light and other bad weather conditions where conventional surveillance techniques tend to break down. By equipping infrastructure with the ability to sense and make real-time sense of their environment with high fidelity, these technologies are ushering in a new era of self-sensing, self-reporting and even self-predicting civil assets, heading toward constant intelligent self-diagnosis [1]. Figure 1 illustrates artificial intelligence tools in civil engineering.
The impact of microplastics on global public health, distribution, and contamination: a systematic review and meta-analysis
(Springer Nature, 2025-10-27) Adamopoulos, Ioannis; Valamontes, Antonios; Karantonis, John T.; Syrou, Niki; Mpourazanis, George; Tsirkas, Panagiotis; Mpourazanis, Pantelis; Ali, Guma; Mijwil, Maad M.; Tornjanski, Vesna; Frantzana, Aikaterini; Vogiatzis, Romanos; Dounias, George; Lamnisos, Demetris
Purpose: Microplastics (MPs), defined as plastic particles smaller than 5 mm, are an emerging environmental contaminant of concern due to their widespread presence in air, water, food, and human tissue. Although the evidence has identified MPs in the bloodstream, placentas, and excrement, suggesting chronic internal exposure, MPs' health outcomes and biological mechanisms remain inadequate. Methods: This systematic review and meta-analysis follow PRISMA 2020 guidelines, and the Coherence risk of bias tools to evaluate MPs' exposure pathways and health risks. A comprehensive search across PubMed, Scopus, Web of Science, Embase, Cochrane, and other databases identified 2441 initial records. 78 studies met the inclusion criteria. Ingestion, inhalation, and dermal contact pathways were analysed, and observational and toxicological studies were assessed. Results: Ingestion accounted for ~ 74% of MP exposure pathways, followed by inhalation (22%) and dermal absorption (4%). MPs were detected in 33–65% of human faecal and blood samples across national surveys. Meta-analysis-random-effects models and test of overall effect size z = 1.18, p-value = 0,24 were used to account for inter-study variance. Homogeneity: Q = 1.16 and Heterogeneity: H-squared = 1,00. Conclusions: Animal studies suggest MPs induce oxidative stress, inflammation, immune disruption, and endocrine effects. However, causality in human populations remains unverified due to inconsistent study designs and exposure quantification methods. Subgroup analysis revealed stronger associations with reproductive, gastrointestinal, and neurotoxic markers in rodent models. MPs exposure is recognised as a ubiquitous and potentially hazardous public health concern. This study underscores the pressing need for long-term studies and effective public risk strategies to mitigate human exposure and safeguard the ecosystem.
Post-quantum secure blockchain-based federated learning framework for enhancing smart grid security.
(the University of Information Technology and Communications (UoITC), 2025-10-10) Mijwil, Maad M.; Ali, Guma; Kabiito, Simon Peter; Dhoska, Klodian; Adamopoulos, Ioannis
Emerging technologies have accelerated the digitalization of smart grids, improving demand-side management, sustainability, and operational efficiency. The attack surface is widened by this interconnection, though, leaving vital smart grid data and systems vulnerable to online attacks. Single points of failure, privacy violations, and a lack of robustness against sophisticated attacks persist in centralized data processing. Traditional cryptographic techniques are further threatened by the development of quantum computing, which raises significant security risks for smart grids. With a focus on post-quantum cryptography (PQC) resilience, this study examines 206 peer-reviewed research articles on blockchain-based federated learning (BFL) in smart grids that were published between January 2023 and July 2025. It assesses the advantages, limitations, and compromises of the current BFL models in this field. The paper suggests a unique post-quantum secure BFL (PQS-BFL) framework that integrates federated learning (FL), lightweight PQC protocols, and a scalable blockchain architecture to solve the vulnerabilities that have been uncovered. This design enables decentralized, private, and impenetrable cooperation among grid nodes. The results demonstrate that the system mitigates quantum-resilient attacks and inference threats while improving data integrity, key management, and secure model aggregation. A path for creating safe, scalable PQS-BFL solutions for upcoming smart energy systems is provided in the paper's conclusion, along with an overview of the main research issues. This study shows that using PQC, blockchain, and FL to secure next-generation smart grids is both feasible and important.
Fusion of blockchain, IoT, artificial intelligence, and robotics for efficient waste management in smart cities
(Association of Talent under Liberty in Technology, 2025-08-09) Ali, Guma; Asiku, Denis; Mijwil, Maad M.; Adamopoulos, Ioannis; Dudek, Marek
Rapid urbanization and population growth are accelerating waste generation in cities worldwide, posing serious environmental and socio-economic challenges. Traditional waste management systems, often centralized and infrastructure-deficient, struggle with inefficiencies, unscheduled collection, and a lack of real-time data. These limitations hinder progress toward smart and sustainable urban environments. Blockchain, the Internet of Things (IoT), Artificial Intelligence (AI), and Robotics are reshaping waste collection, sorting, and recycling. This review examines how these technologies integrate to create secure, efficient, and sustainable waste management in smart cities. An analysis of 184 studies published between January 2022 and July 2025 reveals key shortcomings in conventional waste management systems and showcases the benefits of smart waste management solutions. The results showed that cities are already using IoT-enabled smart bins, AI-driven route optimization, Blockchain for waste tracking, and robotic sorting. However, challenges such as data privacy concerns, limited Blockchain scalability, system interoperability gaps, sensor reliability issues, and high computational demands limit broader adoption. The review outlines future research priorities, including AI-powered waste forecasting, swarm robotics, real-time edge computing, and enhanced cybersecurity. By providing a roadmap for technological innovation and integration, this study supports policymakers, urban planners, and industry leaders in developing intelligent, cost-effective, and environmentally resilient waste management systems.
Blockchain and federated learning in edge-fog-cloud computing environments for smart logistics
(Mesopotamian Academic Press, 2025-07-22) Ali, Guma; Adebo, Thomas; Mijwil, Maad M.; Al-Mahzoum, Kholoud; Sallam, Malik; Salau, Ayodeji Olalekan; Adamopoulos, Ioannis; Bala, Indu; Al-jubori, Aseed Yaseen Rashid
The rapid growth of smart logistics, driven by IoT devices and data-intensive applications, necessitates secure, scalable, and efficient computing frameworks. As the edge-fog-cloud (EFC) paradigm supports real-time data processing, it faces significant security threats and attacks, including privacy risks, data breaches, and unauthorized access. To address these security threats and attacks, blockchain and federated learning (FL) have gained popularity as potential solutions in EFC computing environments for smart logistics. This survey reviews the current landscape in EFC computing environments for smart logistics, highlighting the existing benefits and challenges identified in 134 research studies published between January 2023 and June 2025. The applications of blockchain and FL demonstrate their ability to enhance data security and privacy, improve real-time tracking and monitoring, and ensure inventory and supply chain optimization. Although these technologies offer promising solutions, challenges such as scalability issues, data quality, interoperability and standardization hinder their effective implementation. The survey suggests future research directions focused on developing advanced blockchain and FL, integrating emerging technologies, developing policies and regulations, fostering collaborative research, and ensuring cross-industry adoption and interoperability. Integrating blockchain and FL within the EFC model offers a transformative path toward building secure, intelligent, and resilient logistics systems.
Influence of physical shape and salting on tomato drying performance using mixed mode solar and open-air methods in semi-cloudy weather
(Springer Nature, 2025-07-20) Elwakeel, Abdallah Elshawadfy; Ali, Guma; Eldin, Abdalla Zain; Alsebiey, Mohamed Mahmoud; Tantawy, Aml Abubakr; AL-Harbi, Mohammad S.; Ahmed, Atef Fathy; Metwally, Khaled A.
SD Solar drying is increasingly recognized as a sustainable and energy-efficient solution for preserving agricultural products, offering a practical alternative to fossil fuel-dependent methods and traditional open sun drying (OSD). However, its overall performance is highly influenced by environmental variability and system design. This study provides a detailed evaluation of a newly developed direct solar dryer (DDSD) for tomato dehydration, conducted under real and fluctuating climatic conditions in Aswan, Egypt, from February 22 to 27, 2025. During the trial period, solar irradiance ranged widely from 88 to 826 W/m2 due to intermittent cloud cover, while ambient temperatures fluctuated between 22 and 34 °C—conditions representative of actual field environments. Tomato samples were prepared in three physical forms—halves, quarters, and 6 mm slices—and subjected to two pretreatment methods (salted and unsalted) to assess their effects on drying kinetics. The DDSD demonstrated significantly better performance than OSD, reducing drying durations by 25–39.6%. The most efficient results were achieved for salted 6 mm slices, which dried in just 9 h—substantially faster than the 29 h for unsalted halves in DDSD and 48 h in OSD. These samples also exhibited the highest effective moisture diffusivity (Deff) (5.92 × 10⁻⁹ m2/s), reflecting enhanced internal moisture transport. Among 12 drying models evaluated, the Logistic model most accurately described the drying behavior in the DDSD, with an excellent statistical fit (R2 = 0.999524, χ2 = 6.74 × 10⁻5, RMSE = 0.006868). Economically, the DDSD, integrated with a photovoltaic (PV) system, required a modest initial investment of $520 and achieved a payback period of just 1.82 years for salted slices due to faster processing and increased throughput. From an environmental perspective, the system is projected to offset approximately 105.68 metric tons of CO₂ emissions over a 20-year lifespan, with an energy payback time of only 1.10 years and potential revenue of $1321.04 from carbon credits. These findings underscore the DDSD’s potential as a cost-effective, environmentally sustainable, and technically efficient solution for agricultural drying in solar-rich regions.
Enhancing cybersecurity in smart education with deep learning and computer vision: A Survey.
(Mesopotamian Academic Press, 2025-06-26) Ali, Guma; Aziku, Samuel; Mijwil, Maad M.; Al-Mahzoum, Kholoud; Sallam, Malik; Salau, Ayodeji Olalekan; Bala, Indu; Dhoska, Klodian; Melekoglu, Engin
The rapid digital transformation of education, driven by the widespread adoption of smart devices and online platforms, has ushered in the era of smart education. While this shift enhances learning experiences, it also introduces significant cybersecurity risks that threaten the confidentiality, integrity, and availability of educational resources, student data, and institutional systems. This survey examines how deep learning (DL) and computer vision (CV) techniques can enhance cybersecurity in smart education environments. By reviewing 202 peer-reviewed research papers published between January 2022 and June 2025 across leading publishers such as ACM Digital Library, Frontiers, Wiley Online Library, IGI Global, Nature, Springer, ScienceDirect, MDPI, IEEE Xplore Digital Library, Taylor & Francis, Sage, BMC, and Google Scholar, the study explores the integration of these advanced technologies to address emerging threats. It highlights the use of DL in intrusion detection, anomaly detection, and biometric authentication to protect digital learning platforms. It also examines how CV techniques, such as facial recognition, behavioral analysis, and emotion detection, enhance security and foster adaptive learning environments. The survey also addresses key challenges, including data quality, model interpretability, computational costs, and ethical considerations. By identifying research gaps and proposing future directions, this survey offers valuable insights for researchers, educators, and policymakers aiming to develop robust, scalable, and ethical AI-driven cybersecurity solutions for smart education.
Spectral realization of the nontrivial zeros of the Riemann Zeta function via a Hermitian operator framework
(Mesopotamian Academic Press, 2025-05-18) Valamontes, Antonios; Adamopoulos, Ioannis; Ali, Guma
We present a spectral construction of a Hermitian operator whose spectrum coincides exactly with the imaginary parts of the nontrivial zeros of the Riemann zeta function The operator, denoted H∞, is defined on a discrete geometric space modeled by a 20-vertex dodecahedral graph, incorporating a discrete Laplacian, an entropy-based coherence potential, and a prime-indexed infinite-order algebraic term derived from Infinity Algebra. We show that H∞ is self-adjoint, spectrally complete, and compatible with the analytic continuation and functional symmetry of ζ(s). A spectral determinant constructed from its eigenvalues matches the Hadamard product representation of ζ 1 + it , and no extraneous roots appear off the critical line. Numerical approximations from a truncated version of the operator validate this correspondence. The construction yields a functional-analytic framework that supports a spectral- theoretic resolution of the Riemann Hypothesis
Model predictive control for quad active bridge DC-DC converter for more electric aircraft applications
(Springer Nature, 2025-04-21) Adam, Ahmed Hamed Ahmed; Chen, Jiawei; Xu, Minghan; Kamel, Salah; Ali, Guma
The isolated multi-port converters quad-active bridge (QAB) presents a unique opportunity to connect multiple sources and loads operating at different power and voltage levels, offering galvanic isolation and shared magnetics as advantages. However, the high number of modulation variables, dynamic response, and overall modeling complexity of QAB converters pose challenges to controller design. Traditional linear controllers often struggle with voltage overshooting and undershooting under abrupt load changes and exhibit limited dynamic performance and coupling among different ports. To address these challenges, this paper introduces a moving discretized control set-model predictive control (MDCS-MPC) strategy for QAB converters. The developed approach predicts phase shift values through the converter model, ensuring fast dynamic performance and eliminating steady-state errors in control variables. The prediction model’s embedded circuit parameters and operating modes enhance performance across various power and terminal voltage ranges. An adaptive step is implemented for quick transitions, significantly reducing computational demands. These analytical findings and the MDCS-MPC strategy are verified through Matlab simulation results and experimental results obtained from the Hardware-in-the-Loop (HIL) real-time Typhoon 602 platform. Both experimental and simulation results demonstrate the effectiveness of the developed strategy, showing superior dynamic response, robustness, and reduced computational requirements. Furthermore, the voltage achieves a very fast dynamic response and exhibits no significant voltage overshoot or undershoot.

Browse

Recent Submissions