The focusing ability of the lens antennas can significantly contribute to perform precise radio localization with low complexity in a massive antennas array (MAA) system. Thus, by combining lens antennas with a MAA, we propose an approach for radio source localization through angle of arrival (AoA) estimation in a multipath propagation scenario. We discuss two simplified antenna selection techniques to select and process the focused subsets of antennas. Then, we implement a low complexity time of arrival estimation scheme to detect the first arrival path (FAP) that contains the actual AoA information. Finally, the FAP signal is processed to estimate the AoA very efficiently, by applying one of the traditional subspace based methods. To simplify further the system architecture with only slight degradation in performance, we also propose a mixed analog-digital AoA estimation structure based on the application of the sum-difference signal patterns technique. The simulated root mean squared error performance curves as a function of the signal-to-noise-ratio demonstrate that the proposed approach is capable of providing good AoA estimation results while considerably reducing the overall system complexity as compared to the MAA system without the lens.

In this paper, the efficacy of machine learning (ML) techniques for predicting the academic success of students is investigated. In issues pertaining to higher education, as well as machine learning, deep learning, and its linkages to educational data, predicting student achievement is essential. The choice of courses and the development of effective future study plans for students can be easier with the help of the capacity to forecast a student's success. In addition to predicting student achievement, it makes it easier for instructors and administrators to keep an eye on children so that they can offer support and integrate trainings for the greatest outcomes. In this study, we define the idea of predicting the student performance in education and its several iterations. We discuss a number of ML approaches, such as the Fuzzy C-Means, the Multi-Layer Perceptron (MPL), the Logistic Regression (LR), and the Random Forest (RF) algorithms, for predicting student achievement in the classroom. The models for forecasting student performance that are now in use and those that have been proposed in this paper are carefully investigated. The paper examines different combinations of the algorithms including FCM – MLP, FCM – LR, and FCM – RF, and provides the detailed results of each combination. These strategies are assessed using quantitative standards including accuracy, detection rate, and false alarm rate.

This paper studies the performance analysis of machine learning (ML) and data mining techniques for anomaly detection in credit cards. As the usage of digital money or plastic money grows in developing nations, so does the risk of fraud. To counter these scams, we need a sophisticated fraud detection method that not only identifies the fraud but also detects it before it occurs efficiently. We have introduced the notion of credit card fraud and its many variants in this research. Numerous ML fraud detection approaches are studied in this paper including Principal Component Analysis (PCA) data mining and the Fuzzy C-Means methodologies, as well as the Logistic Regression (LR), Decision Tree (DT), and Naive Bayes (NB) algorithms. The existing and proposed models for credit card fraud detection have been thoroughly reviewed, and these strategies have been compared using quantitative metrics including accuracy rate and characteristics curves. This paper discusses the shortcomings of existing models and proposes an efficient technique to analyze the fraud detection.

The fifth generation (5G) mobile communication technology though, is being deployed in various parts of the world to improve the wireless systems in terms of infrastructure and quality of services, in view of mobile communication history, the 5G may be unable to handle the deluge mobile network traffic of the future digital society. In general, new wireless technologies are introduced every 10 years, and services mature every 20 years, and therefore sixth generation (6G) mobile communication technologies, which have a great potential to impact on the market, are expected to be commercialized around 2030. In order to secure technological leadership in mobile communication, which is an essential infrastructure for everyday life and industrial sites, advanced countries around the world have begun to develop 6G technology. The expected main features of the 6G technology include: providing 10 to 100 times more performance than 5G, maximizing performance by utilizing the terahertz band which was not utilized by existing communication networks before, use artificial intelligence techniques for smart networking, and combining the previously isolated technologies. This paper provides an overview on the profile, advancements, implications, architecture, expected applications, and challenges of the futuristic 6G wireless communication systems described by innovative researchers and major global organizations.

This paper presents a simple approach to providing location-based services (LBSs) for the remotely located educational and health institutes on demand, particularly in emergency situations i.e., COVID-19, using an antenna array. The proposed approach consists of the sum and difference patterns of the signal obtained from the antenna array at the base station of a wireless network. It provides the location of the targeting institute i.e., education and health, in terms of the angle-of-arrival (AoA) which can be used to steer the radiation beam in the targeted direction in order to provide the desired services i.e., emergency wireless communication link. In this way, fast and high bandwidth-based communication and networking can be possible for the mentioned organizations. We show antenna array design and location finding results in this paper. The obtained results in terms of antenna parameters and AoA show that the proposed approach is efficient, less complicated, and can be implemented in the next-generation wireless networks i.e., AI-enabled IoT and 5G systems, particularly for the educational and health institutes.

Most STEM (Science, Technology, Engineering, and Mathematics) subjects rely heavily on graphs and charts, which remain largely inaccessible to students who are blind or visually impaired. While text can often be made accessible through screen readers, complex visual structures such as charts are much harder to interpret non-visually. This study presents a proof-of-concept system that applies the DONUT (Document Understanding Transformer) model to STEM charts. The model was trained and evaluated on the Benetech STEM dataset and tested on multiple images, demonstrating promising results in extracting key information such as chart type, chart ID, and x–y coordinate values. Although no user-centered trials or formal educational studies have yet been conducted, this work establishes an initial technical foundation for converting chart data into accessible formats. By enabling interpretation of chart types and data trends, the proposed system has the potential to improve accessibility in STEM education for blind and visually impaired learners, pending further validation and integration with assistive technologies.