alexandria engineering journal最新文献

{"title":"ERINDA: A novel framework for Enhancing the Resilience of Industrial Networks against DDoS Attacks with adaptive recovery","authors":"Thuraya N.I. Alrumaih ,&nbsp;Mohammed J.F. Alenazi","doi":"10.1016/j.aej.2025.02.042","DOIUrl":"10.1016/j.aej.2025.02.042","url":null,"abstract":"<div><div>The increasing threat of distributed denial-of-service (DDoS) attacks targeting the availability of critical infrastructure systems controlled by industrial control systems (ICS). DDoS attacks endanger the high-reliability requirements ICSs by overloading network and system resources, causing malfunction or ceasing operations. Recognizing the severe consequences of service disruptions in these environments, we present a novel framework for Enhancing the Resilience of Industrial Networks against DDoS Attacks (ERINDA), designed to minimize downtime and maintain functionality. It consists of a two-phase approach that combines proactive and reactive strategies to efficiently mitigate DDoS attacks while minimizing service disruptions. First, network traffic is continuously monitored to detect any anomalies indicating potential attacks. Second, response mechanisms are activated upon an actual attack identification to quickly neutralize the threat and restore the integrity of the network. Experimental results, obtained using ns-3 network simulations mimicking a small-scale industrial network topology, demonstrate that, by integrating real-time monitoring, situation reporting, and rapid adaptive response mechanisms, ERINDA improves key performance metrics. Under a DDoS attack, ERINDA recovered approximately 88 % of normal throughput at 25 % channel utilization, compared to a 77 % reduction without ERINDA. Furthermore, ERINDA consistently restored packet delivery ratio and round-trip delay values close to normal operational conditions across various traffic loads.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"121 ","pages":"Pages 248-262"},"PeriodicalIF":6.2,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143519663","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"STIGANet: Integrating DGCNS and attention mechanisms for real-time 3D pose estimation in sports","authors":"Qi Liu ,&nbsp;Zhenzhou Wang ,&nbsp;Han Zhang ,&nbsp;Changqing Miao","doi":"10.1016/j.aej.2025.02.058","DOIUrl":"10.1016/j.aej.2025.02.058","url":null,"abstract":"<div><div>In modern sports training and competitions, precise action analysis and feedback are essential for optimizing athletes’ performance. Traditional methods, however, are time-consuming, labor-intensive, and prone to subjective judgment, leading to inconsistencies and inaccuracies. Existing AI-based approaches struggle with high-speed movements, complex backgrounds, and real-time processing. To address these limitations, we propose the Spatio-Temporal Interweaved Graph and Attention Network (STIGANet) for accurate 3D human pose estimation. STIGANet combines Dynamic Graph Convolutional Networks (DGCN), a Spatio-Temporal Cross-Attention Mechanism (STCA), Spatio-Temporal Interweaved Attention (STIA), and a Deformable Transformer Encoder, enabling effective capture and fusion of spatial and temporal features in human actions. The model improves pose estimation accuracy and robustness in dynamic, real-time sports environments. On the Human3.6M and MPI-INF-3DHP datasets, STIGANet achieves superior performance with MPJPEs of 38.2 mm and 45.3 mm, respectively, outperforming existing methods. These findings highlight the model’s potential for real-time sports action analysis. Overall, this work enhances sports action analysis by combining graph convolutional networks with attention mechanisms, offering a robust framework for real-time insights during sports training and rehabilitation.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"121 ","pages":"Pages 236-247"},"PeriodicalIF":6.2,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143519664","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Knowledge graph construction and talent competency prediction for human resource management","authors":"Bowen Yang ,&nbsp;Zhixuan Shen","doi":"10.1016/j.aej.2025.02.043","DOIUrl":"10.1016/j.aej.2025.02.043","url":null,"abstract":"<div><div>Job matching and talent recommendation are essential yet challenging tasks in human resource management. Traditional methods, such as rule-based matching and collaborative filtering, often struggle with issues like data sparsity, cold-start problems, and the dynamic nature of user preferences, limiting their effectiveness in real-world applications. To address these challenges, we propose a hybrid model that integrates Graph Convolutional Networks (GCN), Reinforcement Learning (RL), and Deep Collaborative Filtering (DCF). The GCN module captures complex multi-relational structures between jobs and candidates, the RL module dynamically optimizes recommendation strategies based on feedback, and the DCF module enhances personalized recommendation capabilities. Experimental results demonstrate that the proposed model outperforms traditional methods in key metrics such as Precision@10, Recall@10, NDCG@10, and CTR, while achieving broader coverage. This research provides a novel and effective solution for improving job matching and talent recommendation, offering practical significance for applications in human resource management.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"121 ","pages":"Pages 223-235"},"PeriodicalIF":6.2,"publicationDate":"2025-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143520178","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Exploring the drawbacks of main nozzle design in air-jet weaving: Current challenges and future directions","authors":"George Joseph Thomas,&nbsp;Hermann Finckh,&nbsp;Hans-Juergen Bauder,&nbsp;Goetz Theodor Gresser","doi":"10.1016/j.aej.2025.02.055","DOIUrl":"10.1016/j.aej.2025.02.055","url":null,"abstract":"<div><div>Air-jet weaving is the most productive weaving technique. It uses compressed air for weft transportation. The main nozzle plays a crucial role in the efficiency of the process and the quality of the woven product. The basic design of the main nozzle consisting of separate nozzles for different wefts has remained unchanged since its inception. This design has certain drawbacks that affect the performance and efficiency of the process. This study throws light into these drawbacks with the help of numerical simulations. A generic main nozzle with eight nozzles is chosen for the study. The orientation of the nozzles with respect to the weft channel affects the performance of the nozzles. The difference in the performance of the nozzles is analysed to understand the deficits of the nozzle and its impact on the process. The paper also proposes a novel design concept for the main nozzle that can overcome the identified limitations. In addition, an economic analysis is conducted to quantify the potential savings with the proposed nozzle system with a focus on the German market. The authors also suggest future research directions to realise and test the new concept that can improve the efficiency and productivity of air-jet weaving.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"121 ","pages":"Pages 213-222"},"PeriodicalIF":6.2,"publicationDate":"2025-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143519666","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cascaded learning empowered classification of UAVs using radio frequency under wireless interference","authors":"Md Habibur Rahman ,&nbsp;Jung-In Baik ,&nbsp;Md Abdul Aziz ,&nbsp;Rana Tabassum ,&nbsp;Mohammad Abrar Shakil Sejan ,&nbsp;Hyoung-Kyu Song","doi":"10.1016/j.aej.2025.02.031","DOIUrl":"10.1016/j.aej.2025.02.031","url":null,"abstract":"<div><div>Autonomous UAVs are increasingly valuable in disaster response, imaging, agriculture, defense, and public services. However, they pose security risks if misused near sensitive areas like airports and power plants. Rapid UAV identification is essential for safety, and deep learning (DL) algorithms offer effective solutions for automatic drone detection and classification across diverse scenarios. To leverage recent advances in DL technology, this paper proposes a novel DL-based cascaded model that combines convolutional neural networks (CNN) with bidirectional long short-term memory (BiLSTM) networks for the precise classification of UAVs. The DroneRC and CARDRF datasets are used in this simulation study. Before training the ML models, raw RF data is preprocessed using the short-time Fourier transform, and the power spectral density approach is employed to extract the most relevant features. The feature extraction of RF signals from various drones is performed using grayscale values instead of RGB channels. Additionally, to evaluate the model’s effectiveness in robustly classifying drones, it is trained in the presence of wireless interference, such as WiFi and Bluetooth signals. To assess efficacy, various DL algorithms (CNN, LSTM, CNN-LSTM, KNN, SVM) were configured identically for comparison. According to the results, the suggested model has a low error rate and good accuracy.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"121 ","pages":"Pages 201-212"},"PeriodicalIF":6.2,"publicationDate":"2025-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143519665","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Leveraging sparrow search optimization with deep learning-based cybersecurity detection in industrial internet of things environment","authors":"Fatma S. Alrayes ,&nbsp;Nadhem Nemri ,&nbsp;Wahida Mansouri ,&nbsp;Asma Alshuhail ,&nbsp;Wafa Sulaiman Almukadi ,&nbsp;Ali M. Al-Sharafi ,&nbsp;Jawhara Aljabri ,&nbsp;Faisal Mohammed Nafie","doi":"10.1016/j.aej.2025.02.004","DOIUrl":"10.1016/j.aej.2025.02.004","url":null,"abstract":"<div><div>The Industrial Internet of Things (IIoT) is a current research field that links digital services and equipment to physical methods. The IIoT was employed to make vast amounts of data from many sensors, and the device has met numerous issues. The IIoT has tackled several methods of cyberattacks that threaten its ability to distribute organizations with steady operations. Such threats result in reputational and financial spoils for businesses and the stealing of delicate data. Therefore, numerous network intrusion detection systems (IDSs) are proposed to address and defend against threats in IIoT environments. However, the data needed to develop an intelligent IDS is often complex and difficult to handle, resulting in significant challenges in detecting new and existing attack vectors. These complexities arise due to the dynamic nature of network traffic and growing cyberattack strategies, which require adaptive and advanced detection techniques. This study presents a Leveraging Sparrow Search Optimization Algorithm for a Deep Learning-Based Cybersecurity (LSSOA-DLBC) approach in the IIoTs environment. The LSSOA-DLBC approach aims to identify cybersecurity in the IIoT environment automatically. In the LSSOA-DLBC model, the first data normalization phase utilizes Z-score normalization. For the feature selection (FS) method, the LSSOA-DLBC model utilizes a sine cosine algorithm (SCA). Besides, the deep belief network (DBN) model automatically identifies cybersecurity in the IIoT environment. Eventually, the sparrow search algorithm (SSA) model is implemented to optimize the hyperparameter tuning of the DBN model. The experimental outcome of the LSSOA-DLBC methodology is examined on the benchmark dataset. The performance validation of the LSSOA-DLBC methodology portrays a superior accuracy value of 99.29 % over existing approaches concerning distinct evaluation metrics.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"121 ","pages":"Pages 128-137"},"PeriodicalIF":6.2,"publicationDate":"2025-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143509657","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Navigating tuberculosis control: A mathematical approach to disease dynamics and vaccination strategies","authors":"Kareem Alanazi ,&nbsp;Takasar Hussain ,&nbsp;Muhammad Ozair ,&nbsp;Sirda Shafiq ,&nbsp;Marium Siddique ,&nbsp;Kottakkaran Sooppy Nisar ,&nbsp;M. Abdalla ,&nbsp;Asim Anwar","doi":"10.1016/j.aej.2025.02.053","DOIUrl":"10.1016/j.aej.2025.02.053","url":null,"abstract":"<div><div>Tuberculosis has become one of the world’s most serious health issues. We generated a deterministic mathematical model for analyzing how vaccination affects tuberculosis dynamics with slow and fast propagation of disease in a community. The possibility of both endemic and disease-free stable states is explored. The epidemiological threshold, also known as the reproduction number <span><math><mrow><mrow><mo>(</mo><msub><mrow><mover><mrow><mi>R</mi></mrow><mrow><mo>ˆ</mo></mrow></mover></mrow><mrow><mn>0</mn></mrow></msub><mo>)</mo></mrow><mo>,</mo></mrow></math></span> has been calculated. It has been shown that only reducing the reproduction number below unity is no longer sufficient for preventing tuberculosis (TB) from entering the population due to backwards bifurcation. Numerical simulations are used to validate the conclusions of analytical study. The threshold proportion of vaccinated individuals has been calculated which must be attained for the complete eradication of the disease. It has been demonstrated that the population’s burden of tuberculosis will be decreased by decreasing effective interaction with tuberculosis infected individuals and raising the proportion of vaccination vulnerable individuals with a significant vaccine efficiency.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"121 ","pages":"Pages 183-192"},"PeriodicalIF":6.2,"publicationDate":"2025-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143512478","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predicting economic resilience: A machine learning approach to rural development","authors":"Shun Du ,&nbsp;Yajuan Xu ,&nbsp;Lei Wang","doi":"10.1016/j.aej.2025.02.049","DOIUrl":"10.1016/j.aej.2025.02.049","url":null,"abstract":"<div><div>Economic resilience is the ability of a country to resist, adapt to, and recover from economic shocks while maintaining sustainable growth and long-term stability. Analyzing resilience requires understanding the dynamics among economic, social, and governance indicators and anticipating the influence which they pose on resilience. In this research, models based on autoregressive integrated moving average (ARIMA), which purely time series model, and machine learning (ML) models, which comprise SVR (support vector regression), and ANN (artificial neural network) were applied in order to predict a resilience indicator and to compare their performance. Initially, the correlation matrix showed weak correlation among the features. Across the models, ANN showed superior performance over all metrics: mean squared error (MSE), root mean squared error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE), while its close competitor SVR. ARIMA failed to understand the intricate interactions and proved a case for why advanced ML models are more suitable for predicting resilience. From a policy standpoint, these results imply the necessity to adopt approaches that are grounded in ML techniques, notably ANN, toward more accurate and precise predictions in decision-making processes.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"121 ","pages":"Pages 193-200"},"PeriodicalIF":6.2,"publicationDate":"2025-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143511879","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-layered security system: Integrating quantum key distribution with classical cryptography to enhance steganographic security","authors":"Arman Sykot,&nbsp;Md Shawmoon Azad,&nbsp;Wahida Rahman Tanha,&nbsp;B.M. Monjur Morshed,&nbsp;Syed Emad Uddin Shubha,&nbsp;M.R.C. Mahdy","doi":"10.1016/j.aej.2025.02.056","DOIUrl":"10.1016/j.aej.2025.02.056","url":null,"abstract":"<div><div>This research introduces a novel cryptographic system that enhances the security of steganographic images by integrating Quantum Key Distribution (QKD) with traditional encryption. Using the E91 QKD protocol, it generates a shared secret key between parties, leveraging quantum entanglement to provide superior protection against eavesdropping. The quantum key is then hashed with the Secure Hash Algorithm (SHA) to produce a fixed-length, high-entropy key for symmetric encryption. The Advanced Encryption Standard (AES) is employed to encrypt steganographic images, which conceal critical data within digital images, adding an extra layer of security through obscurity. Experiments with images of varying resolutions (64 × 64 to 512 × 512 pixels) show consistent encryption performance. The encrypted images exhibit high randomness, with an average entropy of 7.9929. Differential attack resilience is demonstrated by a Number of Pixels Change Rate (NPCR) averaging 99.6928% and a Unified Average Changing Intensity (UACI) of 56.1549%. The average E91 protocol key generation time is 5.78 s with a key rate of 7.43 bps. Encryption and decryption times for test images are 0.00149 and 0.00175 s, respectively. This research combines quantum and classical cryptography with steganography, providing a robust security framework highly resistant to both quantum and classical threats.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"121 ","pages":"Pages 167-182"},"PeriodicalIF":6.2,"publicationDate":"2025-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143512477","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Advancements in rainfall-runoff prediction: Exploring state-of-the-art neural computing modeling approaches","authors":"Dani Irwan ,&nbsp;Ali Najah Ahmed ,&nbsp;Saerahany Legori Ibrahim ,&nbsp;Izihan Ibrahim ,&nbsp;Moamin A. Mahmoud ,&nbsp;Gan Jacky ,&nbsp;Aiman Nurhakim ,&nbsp;Mervyn Chah ,&nbsp;Pavitra Kumar ,&nbsp;Mohsen Sherif ,&nbsp;Ahmed El-Shafie","doi":"10.1016/j.aej.2025.02.060","DOIUrl":"10.1016/j.aej.2025.02.060","url":null,"abstract":"<div><div>Rainfall-runoff (RR) is a vital process as it is a key component of the Earth’s water cycle, which is required for the survival of life on our planet. It is responsible for water resource management as it will alter the water quality and availability for living things and environmental requirements. Most of the previous research, in this domain, focused on short-term modelling using data from a specific region. However, fewer studies have been conducted to predict water availability for longer periods. There is an urgent need to explore a model that can predict RR in diverse locations for varied periods and climate circumstances. In this context, predictive models for RR prediction in literature are reviewed in this study. The findings are highlighted, and the discussion of the results are condensed. The review has been carried out for 80 articles that were published within last 21 years (2003–2023) on the competency of the predictive models used in RR prediction in the analysis of the input variables and the data size of the time series. The publications include relevant information such as the model limitation and the suggestions for further research that will be useful to researchers who intend to perform similar studies in RR predictions in the future. In addition, researchers from previous studies found that the hybrid deep learning (DL) models are greater than the hybrid machine learning (ML) models, DL models and standalone ML models. In this study, four new models are suggested to forecast the RR.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"121 ","pages":"Pages 138-149"},"PeriodicalIF":6.2,"publicationDate":"2025-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143509658","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}