SJEE

Electric Machines Winding Applications – DC Machine GeoGebra Winding Application

2025-12-04T22:52:17+00:00

Understanding the winding configurations of electric machines is essential for students of electrical engineering, yet it remains one of the most complex and abstract areas of study. At the beginning of this paper, a brief overview of existing software solutions for electric machine winding design is provided, along with a summary of previously developed educational tools created at the Faculty of Technical Sciences Čačak. These earlier software tools, focused on both DC and AC machine windings, laid the foundation for the creation of more advanced and interactive learning resources. Building on this foundation, and leveraging the experience gained through the development of more than 50 GeoGebra applications at the Faculty, this paper presents a new interactive GeoGebra-based application specifically developed to support the teaching and learning of electric machine windings, with a focus on DC machine armatures. The developed application enables users to define winding parameters, calculate winding steps, generate winding tables, visualize developed winding diagrams, and simulate commutator and brush placement across eight different winding types. These features are designed with a strong emphasis on educational value, offering an intuitive and didactic workflow that serves as an effective teaching aid in mastering topics related to electrical machines. The new tool introduces interactivity, parameter validation, and animation—providing students with a hands-on experience that enhances both comprehension and retention, while also improving the effectiveness of both classroom and distance learning in electrical machines education.

An Optimised Sidechain Based Biometric Attendance Solution

2025-12-04T22:52:18+00:00

A Biometric Authentication System (BAS) is the best choice when there is a need for end-users to have a higher level of security and reliability. However, capturing and verifying user biometrics requires dedicated biometric hardware and software systems with complex mechanisms at the back end. Most of the biometric systemsare client-server architecture based, which has got the single pointof failure, dependability and reliabilityproblems.Capturing, storing and verifyingbiometric templates must be highly secure and reliable. When asingle chain basedblockchain system is used at the backend, the computational resources and verification time increase significantly due to its large block chain size. This results in more delays, inefficiencies, andrequirestransaction gas costs, this shows the need for block chain optimising solutions. For block chain biometric based access control systems, speed, performance, accuracy, and security are the primary requirements. In this work, we present a Side Chain based Blockchain Transaction Optimisation Solution (SBTOS) for a BAS, which is a state-of-the-art mechanism which combines the strength of blockchain and biometric technology with optimised performance for the access control systems. SBTOS mainly focuses on optimising the blockchain when used at the back end, particularly when the database size is huge.

Machine Learning-Driven Prediction of Optimal Control Flow Graph Traversal Strategy

2025-12-04T22:52:19+00:00

Control flow graphs model possible program execution paths and thus are essential for static program analysis. Compilers use control flow graphs as a basis for their intermediate representations, allowing them to apply optimizations. As each method is represented by its control flow graph, the number of control flow graphs that a compiler needs to generate and process depends on the program being compiled. For reference, modern programs that run on the JVM consist of hundreds of thousands of methods. Thus, efficient control flow graph traversal is crucial to provide fast compilation. Prior work has shown that breadth-first and depth-first search algorithms yield different results depending on the control flow graph structure; however, the relationship between control flow graph features and the optimal traversal algorithm in terms of traversal speed remains underexplored. In this work, we construct a dataset of over 200,000 control flow graphs gathered from modern state-of-the-art JVM benchmark suites. Using this dataset, we train a set of ensemble-based machine learning models that predict optimal graph traversal algorithms for a given control flow graph using a set of lightweight graph features. Our models identify the key features that yield accurate predictions and demonstrate that the most informative features can be extracted efficiently during the graph construction process itself.

Design of SIMO SC-FDMA-DMWT Transceiver using Maximal-Ratio Combining in LoS Fading Channels

2025-12-04T22:52:19+00:00

In this paper single-input multiple-output (SIMO) transceivers employing single-carrier Frequency Division Multiple Access with Discrete Multi-Wavelet Transform (SC-FDMA-DMWT) for 3GPP Long Term Evolution (LTE) uplink channels are considered. While SC-FDMA-DMWT traditionally underperforms compared to Orthogonal Frequency Division Multiplexing (OFDM) in Rayleigh fading, line-of-sight (LoS) components significantly enhance its resilience to fading. This study evaluates Maximal-Ratio Combining (MRC) effectiveness for SC-FDMA-DMWT through systematic comparison with Zero-Forcing (ZF) and Minimum Mean Square Error (MMSE) equalizers under varying LOS power levels and antenna correlations Monte Carlo simulations perform bit processing at the level of 2×10⁶ per configuration for SNR that ranges from -5 to 25dB using realistic LTE channel models in both Rician and Rayleigh fading cases. Performance measures are BER analysis and diversity gain calculation. Results demonstrate that MRC-based SC-FDMA-DMWT achieves an 8-9dB BER improvement at 10⁻⁴ compared to conventional OFDM in LoSdominant channels. The Frequency correlation impact becomes negligible with MRC implementation, indicating optimal signal combining performance. %85-75 transmit power is reduced in Dual-antenna SIMO structures while preserving the natural low PAPR characteristics of the SC-FDMA scheme. Centralized 12- subcarrier assignments perform 4-6 dB better than decentralized 48-subcarrier settings. These results confirm that SIMO SC-FDMA-DMWT with MRC is a promising candidate for next-generation LTE uplink systems, particularly in macro and small cell deployments under dominant LoS propagation conditions.

Co-Optimized E-Mode AlGaN/GaN HEMT with Composite P-GaN Recessed Cap and Etched Doped Buffer for Simultaneous DC and RF Performance Enhancement

2025-12-04T22:52:20+00:00

This study presents a theoretical analysis of the DC and RF characteristics of enhancement mode (E-mode) AlGaN/GaN High Electron Mobility Transistor (HEMT) utilizing symbiotic integration of advanced techniques e.g. composite gate structure with slightly etched buffer. Enhancement-mode (Emode) GaN HEMTs often face a fundamental trade-off: achieving high positive threshold voltage (Vth) typically degrades RF metrics such as lower cut-off frequency (f_T) and deteriorated transconductance (g_m). To overcome these issues with practical implications, this expedite structure is introduced. The performance analysis is done using TCAD ATHENA to etch, doped, diffusion, and deposition of the architecture as well as TCAD ATLAS to characterize the DC and RF performance. Conventional methods solely focus on individual techniques such as P-GaN caps, recessed gate, buffer etching or doping. This work proposes a synergistically engineered E-mode AlGaN/GaN HEMT which combines these key features: (1) a P-GaN cap, (2) a recessed gate structure, and (3) a slightly etched and carbon-doped AlGaN buffer layer. This combined implementation is only reported in this paper. The device achieves a V_th of 2.43 V with higher g_m around 325 mS/mm. This ensures the device remains off at normal secure condition with better gate control characteristics. The lower on-resistance of 139.4 mΩ indicates it can conduct more current for a given gate voltage, reducing power loss when the device is on. This architecture also attains a breakdown voltage (V_br) around 1103 V which points out a better threshold performance metrics. Under small signal analysis at 1 MHz, key findings have achieved include maximum current of 630 mA/mm, a cut-off frequency of approximately 40 GHz. This simultaneous optimization of DC and RF metrics addresses limitations of prior E-mode designs and makes the device suitable for advanced RF and microwave applications.

Exploring Discrete Wavelet Transforms for Bimodal Speech Recognition

2025-12-04T22:52:21+00:00

Discrete Wavelet Transforms (DWTs) provide time–frequency representations that are well suited for nonstationary signals such as speech. This study presents a comparison of four wavelet families (Daubechies, Symlets, Coiflets, and Biorthogonal) for bimodal automatic speech recognition across two speech modes (normal and whispered). Experiments use the Whi-Spe database comprising ten speakers (five female and five male). A Dynamic Time Warping (DTW) back-end performs sequence alignment and recognition. Results are reported via summary tables, histograms, and confusion matrices and reveal systematic differences among the wavelet families, identifying the most effective transform for bimodal recognition. These findings provide practical guidance for selecting wavelet-based front ends in whisper-robust automatic speech recognition (ASR) systems.

Terminal Synergetic Control Based Cheetah Optimizer for Knee-Exoskeleton Systems

2025-12-04T22:52:21+00:00

The knee-exoskeleton is mechanical devices which are designed to help people rehabilitate impaired limb mobility and replace the use of physiotherapists. The scope of this study is rehabilitation assistance for the lower limb (i.e. knee of the leg). Due to the high level of complexity and nonlinearity, various control algorithms have been developed to the knee-exoskeleton system to handle these challenges. This study presents a tracking control design of the angular position for the lower limb exoskeleton knee system based on a terminal synergetic control (TSC) strategy. In addition, the cheetah optimizer (CO) algorithm is introduced and embedded in the design to adjust the design parameters of the controller for further optimization of its performance based on the root mean of square errors (RMSE). The superiority of the proposed control method is shown in comparison to the conventional synergetic control (SC) method via computer simulations using MATLAB. The simulations results show that the TSC can improve the response of the system. The numerical value reveals that the RMSE is reduced by 14.9%. In addition, the simulation results validate the efficacy of the proposed approach in the presence of external disturbances where the RMSE is reduced by 39.1%.

Development and Application of Advanced Control Strategies for Nonlinear Coupled MIMO Systems

2025-12-04T22:52:22+00:00

This paper presents the design and implementation of advanced control algorithms for a nonlinear coupled Multi-Input Multi-Output (MIMO) system, focusing on the hardware structure of Quadruple Conical Tank System (QCTS). Nonlinear MIMO systems, characterized by complex interactions between multiple inputs and outputs, pose significant challenges for control engineering. The QCTS, with its four interconnected conical tanks, serves as an exemplary testbed for evaluating advanced control strategies. The paper elaborates on the theoretical and practical implementation of Proportional-Integral-Derivative (PID) control, Fuzzy Logic Control (FLC), and Model Predictive Control (MPC), emphasizing their capabilities in managing multi-variable systems with constraints. Additionally, a comparative analysis of MPC with traditional control methods such as PID and FLC is presented. The practical implementation is demonstrated through hardware experiments. The hardware experimental results highlight the strengths and limitations of each control strategy, providing insights into their applicability for complex nonlinear MIMO systems. The findings underscore the superior performance of MPC in handling multi-variable interactions and constraints, making it a robust choice for advanced control applications in industrial processes.