Previous Issue
Volume 14, December-1
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
6.1
2.6
Journals
Electronics
Volume 14
Issue 24
electronics-logo

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

Electronics, Volume 14, Issue 24 (December-2 2025) – 21 articles

  • Issues are regarded as officially published after their release is announced to the table of contents alert mailing list.
  • You may sign up for e-mail alerts to receive table of contents of newly released issues.
  • PDF is the official format for papers published in both, html and pdf forms. To view the papers in pdf format, click on the "PDF Full-text" link, and use the free Adobe Reader to open them.
Order results
Result details
Section
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
26 pages, 5762 KB  
Article
Design and Implementation of a Low-Cost IoT-Based Robotic Arm for Product Feeding and Sorting in Manufacturing Lines
by Serdar Yilmaz, Canan Akay and Feyzi Kaysi
Electronics 2025, 14(24), 4801; https://doi.org/10.3390/electronics14244801 - 5 Dec 2025
Abstract
The convergence of Internet of Things (IoT), embedded microcontrollers, and robotics has significantly transformed industrial and service applications under the Industry 5.0 paradigm. IoT-enabled automation not only reduces human intervention but also improves system efficiency, safety, and adaptability across multiple domains. The growing [...] Read more.
The convergence of Internet of Things (IoT), embedded microcontrollers, and robotics has significantly transformed industrial and service applications under the Industry 5.0 paradigm. IoT-enabled automation not only reduces human intervention but also improves system efficiency, safety, and adaptability across multiple domains. The growing integration of automation technologies in manufacturing lines has significantly reduced human intervention while improving productivity and operational safety. Robotic arms play a crucial role in modern industrial environments, particularly for repetitive, hazardous, or precision-demanding tasks. This study presents a cost-effective robotic arm system for product selection, sorting and processing in automated production lines. The system operates in both automatic and manual modes and utilizes an ESP32-based controller, radio frequency identification (RFID) modules, and low-cost sensors to identify and transport products on a conveyor. A mobile, IoT-enabled interface provides remote real-time monitoring and control, while integrated safety mechanisms, current-voltage protections, and emergency stop circuitry enhance operational reliability. Using cost-effective components to reduce total cost, the system has been successfully validated through experiments to reduce labor dependency and operational errors, proving its scalability and economic viability for industrial automation. Compared to similar systems, this study presents an Industry 5.0 approach for low-cost IoT-based automated production lines. Full article
►▼ Show Figures

Figure 1

18 pages, 1356 KB  
Article
OptiPerformer as a Platform for Optical Fiber System Simulation in Distance and In-Class Learning
by Seweryn Lipiński
Electronics 2025, 14(24), 4800; https://doi.org/10.3390/electronics14244800 - 5 Dec 2025
Abstract
Simulation-based laboratories have become an essential component of modern engineering education, particularly in courses where access to physical equipment is limited. This paper presents a structured methodology for teaching the fundamentals of optical fiber communication systems using OptiPerformer 18, i.e., a freely available [...] Read more.
Simulation-based laboratories have become an essential component of modern engineering education, particularly in courses where access to physical equipment is limited. This paper presents a structured methodology for teaching the fundamentals of optical fiber communication systems using OptiPerformer 18, i.e., a freely available optical communication simulation platform. The novelty of this work lies in integrating a complete set of parameter-driven laboratory exercises, covering eye-diagram analysis, chromatic dispersion and dispersion compensation, Gaussian pulse propagation, and BER/Q-factor evaluation, into both distance and face-to-face teaching, and validating their effectiveness across four academic years involving more than 200 students. Representative simulation results generated with OptiPerformer are provided to illustrate the learning process and to demonstrate how key transmission impairments and system-level behaviors can be visualized and quantitatively analyzed without specialized hardware. The pedagogical effectiveness of the approach is assessed through student surveys and final grades, showing consistently high learning outcomes and strong student engagement in both remote and in-person settings. These findings indicate that the proposed simulation-based laboratory framework offers a scalable, hardware-independent, and conceptually rich alternative to traditional fiber-optic laboratory classes, supporting deeper understanding of transmission physics and enhancing analytical and problem-solving skills essential in modern optical communication engineering. Full article
(This article belongs to the Special Issue Converging Platform Technologies: Collaborative Innovations and Future Directions)
►▼ Show Figures

Figure 1

22 pages, 1610 KB  
Article
A Novel Automatic Detection and Positioning Strategy for Buried Cylindrical Objects Based on B-Scan GPR Images
by Yubao Liu, Zhenda Zeng, Hang Ye, Xinyu Sun, Zhiqiang Zou and Dongguo Zhou
Electronics 2025, 14(24), 4799; https://doi.org/10.3390/electronics14244799 - 5 Dec 2025
Abstract
This paper presents DeepMask-GPR, a novel deep learning framework for automatic detection and geometric estimation of buried cylindrical objects in ground-penetrating radar (GPR) B-scan images. Built upon Mask R-CNN, the proposed method integrates hyperbola detection, apex localization, and real-world coordinate mapping in an [...] Read more.
This paper presents DeepMask-GPR, a novel deep learning framework for automatic detection and geometric estimation of buried cylindrical objects in ground-penetrating radar (GPR) B-scan images. Built upon Mask R-CNN, the proposed method integrates hyperbola detection, apex localization, and real-world coordinate mapping in an end-to-end architecture. A curvature-enhanced dual-channel input improves the visibility of weak hyperbolic patterns, while a quadratic regression loss guides the network to recover precise geometric parameters. DeepMask-GPR eliminates the need for raw signal data or manual post-processing, enabling robust and scalable deployment in field scenarios. On two public datasets, DeepMask-GPR achieves consistently higher TPR/IoU for spatial localization than baselines. On an in-house B-scan set, it attains low MAE/RMSE for radius estimation. Full article
(This article belongs to the Special Issue Applications of Image Processing and Sensor Systems)
14 pages, 4959 KB  
Article
Human Pose Intelligent Detection Algorithm Based on Spatiotemporal Hybrid Dilated Convolution Model
by Lili Zhang, Shenxi Dai, Lihuang She and Shuwei Huo
Electronics 2025, 14(24), 4798; https://doi.org/10.3390/electronics14244798 - 5 Dec 2025
Abstract
Three-dimensional human pose estimation (3D HPE) refers to converting the input image or video into the coordinates of the keypoints of the 3D human body in the coordinate system. At present, the mainstream implementation scheme of a 3D HPE task is to take [...] Read more.
Three-dimensional human pose estimation (3D HPE) refers to converting the input image or video into the coordinates of the keypoints of the 3D human body in the coordinate system. At present, the mainstream implementation scheme of a 3D HPE task is to take the 2D pose estimation result as the intermediate process and then return it to the 3D pose. The general difficulty of this scheme is how to effectively extract the features between 2D joint points and return them to 3D coordinates in a highly nonlinear 3D space. In this paper, we propose a new algorithm, called TSHDC, to solve the above dilemma by considering the temporal and spatial characteristics of human joint points. By introducing the self-attention mechanism and the temporal convolutional network (TCN) into the 3D HPE task, the model can use only 27 frames of temporal receptive field to make the model have fewer parameters and faster convergence speed when the accuracy is not much different from the SOTA-level algorithm (+6.8 mm). The TSHDC model is deployed on the embedded platform JetsonTX2, and by deploying TensorRT, the model inference speed can be greatly improved (13.7 times) with only a small loss of accuracy (5%). The comprehensive experimental results on representative benchmarks show that our method outperforms the state-of-the-art methods in quantitative and qualitative evaluation. Full article
(This article belongs to the Special Issue Advanced Technologies and Future Trends in Visual Recognition and Signal Processing)
13 pages, 6172 KB  
Communication
An Automatic Optimization Approach to the Non-Periodic Folded-Waveguide Slow-Wave Structure for the High Efficiency Traveling Wave Tube
by Zheng Wen and Jun Zhang
Electronics 2025, 14(24), 4797; https://doi.org/10.3390/electronics14244797 - 5 Dec 2025
Abstract
An automatic optimization approach to the non-periodic (NP) folded-waveguide slow-wave structure (FW-SWS) is proposed for the high efficiency traveling wave tube (TWT). Considering the beam-wave synchronism condition, the data of the beam velocity distribution are analyzed and utilized for automatic optimization. For concise [...] Read more.
An automatic optimization approach to the non-periodic (NP) folded-waveguide slow-wave structure (FW-SWS) is proposed for the high efficiency traveling wave tube (TWT). Considering the beam-wave synchronism condition, the data of the beam velocity distribution are analyzed and utilized for automatic optimization. For concise expression, a W-band concentric arc NP FW-SWS TWT is automatically optimized as an example, where the beam voltage is set as 6000 V, the beam current is 0.12 A, the magnet field is 0.5 T, and the input power is 0.4 W. Without any training data or previous given datasets, the output power (electronic efficiency) can be optimized to reach 238.7 W (33.1%) at 94 GHz by the automatic optimization approach code within 22.7 h. Full article
25 pages, 1205 KB  
Article
Toward Sustainable Interconnected Metrological Networks: Synchronized Multi-Resource Coordination
by Quan Wang, Xia Han, Xiaodong Yin, Gang Chen, Wenqing Yin, Xiwen Chen, Jun Zhang and Zhuo Chen
Electronics 2025, 14(24), 4796; https://doi.org/10.3390/electronics14244796 - 5 Dec 2025
Abstract
Advances in low-power electronics and wireless communication have fueled the proliferation of interconnected metrological networks, increasing the need for traceable, networked measurement systems. This expansion, however, has created a surge in heterogeneous calibration tasks, while a scarcity of qualified experts and reference standards [...] Read more.
Advances in low-power electronics and wireless communication have fueled the proliferation of interconnected metrological networks, increasing the need for traceable, networked measurement systems. This expansion, however, has created a surge in heterogeneous calibration tasks, while a scarcity of qualified experts and reference standards imposes severe resource constraints on remote calibration. Existing scheduling methods, though effective in homogeneous environments, typically lack integration of high-precision time-synchronization with heterogeneous resource coordination, limiting their use in time-critical metrology. To address this gap, we propose a multi-resource synchronized scheduling framework for remote calibration. We formulate the problem as a dual-container model that concurrently optimizes task mapping and temporal dependencies between edge instruments and cloud services. A two-stage heuristic algorithm is developed to efficiently map and schedule tasks in distributed client-server architectures by leveraging critical path analysis and hierarchical scheduling strategies. Simulations across diverse workloads and scales show our method outperforms existing baselines, achieving superior scheduling efficiency, scalability, and calibration accuracy. Full article
26 pages, 4431 KB  
Article
Yolov8n-RCP: An Improved Algorithm for Small-Target Detection in Complex Crop Environments
by Jiejie Xing, Yan Hou, Zhengtao Li, Jiankun Zhu, Ling Zhang and Lina Zhang
Electronics 2025, 14(24), 4795; https://doi.org/10.3390/electronics14244795 - 5 Dec 2025
Abstract
Traditional methods for picking small-target crops like pepper are time-consuming, labor-intensive, and costly, whereas deep learning-based object detection algorithms can rapidly identify mature peppers and guide mechanical arms for automated picking. Aiming at the low detection accuracy of peppers in natural field environments [...] Read more.
Traditional methods for picking small-target crops like pepper are time-consuming, labor-intensive, and costly, whereas deep learning-based object detection algorithms can rapidly identify mature peppers and guide mechanical arms for automated picking. Aiming at the low detection accuracy of peppers in natural field environments (due to small target size and complex backgrounds), this study proposes an improved Yolov8n-based algorithm (named Yolov8n-RCP, where RCP stands for RVB-CA-Pepper) for accurate mature pepper detection. The acronym directly reflects the algorithm’s core design: integrating the Reverse Bottleneck (RVB) module for lightweight feature extraction and the Coordinate Attention (CA) mechanism for background noise suppression, dedicated to mature pepper detection in complex crop environments. Three key optimizations are implemented: (1) The proposed C2F_RVB module enhances the model’s comprehension of input positional structure while maintaining the same parameter count (3.46 M) as the baseline. By fusing RepViTBlocks (for structural reparameterization) and EMA multi-scale attention (for color feature optimization), it improves feature extraction efficiency—specifically, reducing small target-related redundant FLOPs by 18% and achieving a small-pepper edge IoU of 92% (evaluated via standard edge matching with ground-truth annotations)—thus avoiding the precision-complexity trade-off. (2) The feature extraction network is optimized to retain a lightweight architecture (suitable for real-time deployment) while boosting precision. (3) The Coordinate Attention (CA) mechanism is integrated into the feature extraction network to suppress low-level feature noise. Experimental results show that Yolov8n-RCP achieves 96.4% precision (P), 91.1% recall (R), 96.2% mAP0.5, 84.7% mAP0.5:0.95, and 90.74 FPS—representing increases of 3.5%, 6.1%, 4.4%, 8.1%, and 11.58FPS, respectively, compared to the Yolov8n baseline. With high detection precision and fast recognition speed, this method enables accurate mature pepper detection in natural environments, thereby providing technical support for electrically driven automated pepper-picking systems—a critical application scenario in agricultural electrification. Full article
►▼ Show Figures

Figure 1

29 pages, 4247 KB  
Article
Zone-AGF: An O-RAN-Based Local Breakout and Handover Mechanism for Non-5G Capable Devices in Private 5G Networks
by Antoine Hitayezu, Jui-Tang Wang and Saffana Zyan Dini
Electronics 2025, 14(24), 4794; https://doi.org/10.3390/electronics14244794 - 5 Dec 2025
Abstract
The growing demand for ultra-reliable and low-latency communication (URLLC) in private 5G environments, such as smart campuses and industrial networks, has highlighted the limitations of conventional Wireline access gateway function (W-AGF) architectures that depend heavily on centralized 5G core (5GC) processing. This paper [...] Read more.
The growing demand for ultra-reliable and low-latency communication (URLLC) in private 5G environments, such as smart campuses and industrial networks, has highlighted the limitations of conventional Wireline access gateway function (W-AGF) architectures that depend heavily on centralized 5G core (5GC) processing. This paper introduces a novel Centralized Unit (CU)-based Zone-Access Gateway Function (Z-AGF) architecture designed to enhance handover performance and enable Local Breakout (LBO) within Non-Public Networks (NPNs) for non-5G capable (N5GC) devices. The proposed design integrates W-AGF functionalities with the Open Radio Access Network (O-RAN) framework, leveraging the F1 Application Protocol (F1AP) as the primary interface between Z-AGF and CU. By performing local breakout (LBO) locally at the Z-AGF, latency-sensitive traffic is processed closer to the edge, reducing the backhaul load and improving end-to-end latency, throughput, and jitter performance. The experimental results demonstrate that Z-AGF achieves up to 45.6% latency reduction, 69% packet loss improvement, 85.6% reduction of round-trip time (RTT) for local communications under LBO, effective local offloading with quantified throughput compared to conventional W-AGF implementations. This study provides a scalable and interoperable approach for integrating wireline and wireless domains, supporting low-latency, highly reliable services within the O-RAN ecosystem and accelerating the adoption of localized next-generation 5G services. Full article
(This article belongs to the Special Issue Wireless Sensor Network: Latest Advances and Prospects)
►▼ Show Figures

Figure 1

19 pages, 4240 KB  
Article
An Impedance Measurement Method for Renewable Energy Power Station
by Ze Wei, Tao Xu, Jianan Mu, Lin Cheng, Ning Chen, Luming Ge, Xiong Du and Guoning Wang
Electronics 2025, 14(24), 4793; https://doi.org/10.3390/electronics14244793 - 5 Dec 2025
Abstract
The large-scale integration of renewable energy grid-connected converters into the grid has given rise to many broadband oscillation accidents, primarily due to impedance mismatching with the grid. Consequently, accurate measurement of both the grid-connected converter and the grid impedance is a prerequisite for [...] Read more.
The large-scale integration of renewable energy grid-connected converters into the grid has given rise to many broadband oscillation accidents, primarily due to impedance mismatching with the grid. Consequently, accurate measurement of both the grid-connected converter and the grid impedance is a prerequisite for system stability assessment. However, conventional impedance measurement methods are constrained by the breakdown voltage of semiconductor switches, thus rendering them unsuitable for high-voltage, high-capacity applications. This paper aims to enable impedance measurement in large-capacity, high-voltage applications by presenting a newly developed method that overcomes the voltage limitations of conventional approaches. First, a cascaded H-bridge (CHB) topology is adopted to fulfill the impedance measurement requirements in large-capacity, high-voltage renewable energy station applications. Subsequently, a quasi-proportional-resonant (PR) controlled perturbation injection strategy is proposed to achieve rapid current injection across the 10–1000 Hz frequency range. Finally, the effectiveness and accuracy of the proposed impedance measurement method in capturing harmonic impedance are demonstrated through a hardware-in-the-loop (HIL) experiment conducted on an RTDS platform. Full article
(This article belongs to the Special Issue Wind and Renewable Energy Generation and Integration)
►▼ Show Figures

Figure 1

17 pages, 691 KB  
Article
Balancing Specialization and Generalization Trade-Off for Speech Recognition Models
by Sebastian Cygert, Piotr Despot-Mładanowicz and Andrzej Czyżewski
Electronics 2025, 14(24), 4792; https://doi.org/10.3390/electronics14244792 - 5 Dec 2025
Abstract
Recently, using foundation models pretrained on massive volumes of data that are finetuned for the downstream task has become a standard practice in many machine learning applications, including automatic speech recognition (ASR). In some scenarios, we are interested in optimizing performance for the [...] Read more.
Recently, using foundation models pretrained on massive volumes of data that are finetuned for the downstream task has become a standard practice in many machine learning applications, including automatic speech recognition (ASR). In some scenarios, we are interested in optimizing performance for the target domain (specialization)while preserving the general capabilities of the pretrained model. In this work, we study this effect for various finetuning strategies that aim to preserve pretrained model capabilities. We identify model merging as a promising strategy that performs well across diverse scenarios. However, our findings show that leveraging a small number of data points from the task we are interested in preserving the accuracy of significantly improves the balance between specialization and generalization. In this context, we demonstrate that combining a simplest finetuning strategy with a memory buffer yields highly competitive results, surpassing other more complicated approaches. Our analysis highlights the need for further research into methods that effectively utilize memory buffers, especially in low-resource scenarios. To encourage further exploration in this area, we have open-sourced our code. Full article
(This article belongs to the Section Artificial Intelligence)
►▼ Show Figures

Figure 1

19 pages, 1433 KB  
Article
Dependability Analysis for the Blockchain Oracle System: A Quantitative Modeling Approach
by Jing Bai
Electronics 2025, 14(24), 4791; https://doi.org/10.3390/electronics14244791 - 5 Dec 2025
Abstract
Blockchain oracles, as data intermediaries between on-chain and off-chain environments, have opened up a wide range of application scenarios for blockchain technology. The dependability of a blockchain oracle system will affect the dependability of blockchain systems. However, the dynamic and heterogeneous nature of [...] Read more.
Blockchain oracles, as data intermediaries between on-chain and off-chain environments, have opened up a wide range of application scenarios for blockchain technology. The dependability of a blockchain oracle system will affect the dependability of blockchain systems. However, the dynamic and heterogeneous nature of blockchain oracle systems poses challenges to assessing their dependability. Furthermore, how to comprehensively analyze the dependability of blockchain oracle systems from multiple dimensions of transient availability, steady-state availability, and reliability is also a challenge. In order to solve these challenges, this paper proposes three models based on a semi-Markov process (SMP): (1) the SMP model for steady-state availability analysis; (2) the hierarchical model for transient analysis; and (3) the SMP model with absorption states for reliability analysis. Then, we derive the formulas for calculating the dependability metrics, which can be used to evaluate the dependability of blockchain oracle systems composed of any number of oracle nodes. Finally, based on the comparative experiments to verify the approximate accuracy of the proposed model and formulas, we analyze the impact of system parameters and the number of oracle nodes on the dependability metrics. The experimental results reveal that the key factor affecting availability is the failure time and recovery time of the threshold oracle, while the key factor affecting MTTF is the failure time of the threshold oracle. Full article
(This article belongs to the Special Issue Data Privacy Protection in Blockchain Systems)
►▼ Show Figures

Figure 1

24 pages, 2207 KB  
Article
Power Quality Optimization in PV Grid Systems Using Hippopotamus-Driven MPPT and SyBel Inverter Control
by Sudharani Satti and Godwin Immanuel Dharmaraj
Electronics 2025, 14(24), 4790; https://doi.org/10.3390/electronics14244790 - 5 Dec 2025
Abstract
In grid-connected photovoltaic systems, improving power quality is necessary for assuring constant energy delivery, consistent voltages, and current, as well as being compliant with the standards of the grid. Yet, today’s PV control systems have to deal with serious problems, for example, slow [...] Read more.
In grid-connected photovoltaic systems, improving power quality is necessary for assuring constant energy delivery, consistent voltages, and current, as well as being compliant with the standards of the grid. Yet, today’s PV control systems have to deal with serious problems, for example, slow MPPT reactions to changes in irradiation, significant harmonic distortion, weak reaction to voltage changes, and being unable to adapt well to different situations. For this reason, these problems lead to less efficient electricity, unstable connections to the power grid, and an altered quality of electricity, as solar power and load levels vary in real conditions. A way to solve these problems is introduced in this paper: (1) the Hippopotamus-based Solar Power MPPT Tracker and (2) a SyBel embedded controller for controlling the inverter. This kind of optimization mimics nature to control the duty cycle and enables the boost converter to deliver maximum power while responding quickly and maintaining accurate tracking. Meanwhile, the SyBel controller makes use of a hybrid technique by using SNN, DBN, and synergetic logic to sensibly manage the inverter switches and increase the power quality. The framework is novel because it uses biological optimization plus deep learning-based embedded control to instantly handle error reduction and harmonic suppression. The whole process records energy from solar panels, follows the maximum power point, changes its schedule as needed, and uses sophisticated controls in the inverter. We found that the proposed MPPT tracker achieves an impressive tracking efficiency of 98.6%, surpassing PSO, FLC, and ANFIS, and lowering the time required for tracking by 72%. The SyBel inverter controller provides outstanding results, keeping the voltage THD at 1.2% and current THD at 1.3%, which matches power quality standards. Full article
►▼ Show Figures

Figure 1

21 pages, 1603 KB  
Article
Behavior-Rule Inference Based on Hyponymy–Hypernymy Knowledge Tree
by Huanlai Zhou, Jianyu Guo, Haitao Jia, Kaishi Wang, Lei Guo and Long Qi
Electronics 2025, 14(24), 4789; https://doi.org/10.3390/electronics14244789 - 5 Dec 2025
Abstract
Behavior-rule reasoning aims to infer the corresponding applicable rules from specific behaviors and is a type of inductive reasoning that goes from special cases to general ones. This paper proposes a behavior-rule inference model based on hyponymy–hypernymy knowledge trees, which maps behaviors to [...] Read more.
Behavior-rule reasoning aims to infer the corresponding applicable rules from specific behaviors and is a type of inductive reasoning that goes from special cases to general ones. This paper proposes a behavior-rule inference model based on hyponymy–hypernymy knowledge trees, which maps behaviors to corresponding rules through deep learning by processing textual behavior sequences. The primary contributions of this work are threefold: we present a systematic framework that adapts K-BERT to the legal domain by integrating domain-specific hyponymy–hypernymy knowledge trees, addressing the unique challenges of legal text understanding; we conduct comprehensive optimization of key components, including context length, loss function, and base model selection, providing empirical guidelines for applying pre-trained models to legal reasoning tasks; and we propose a practical evaluation metric (tolerance) that mimics real-world legal decision-making processes, providing extensive analysis on the effectiveness of different knowledge types in legal inference. Full article
►▼ Show Figures

Figure 1

20 pages, 3306 KB  
Article
Identification of Static Eccentricity and Load Current Unbalance via Space Vector Stray Flux in Permanent Magnet Synchronous Generators
by Ilyas Aladag, Taner Goktas, Muslum Arkan and Bulent Yaniktepe
Electronics 2025, 14(24), 4788; https://doi.org/10.3390/electronics14244788 - 5 Dec 2025
Abstract
Permanent Magnet Synchronous Generators (PMSGs) have become increasingly important in industrial applications such as wind turbine systems due to their high efficiency and power density. However, their operational reliability can be affected by asymmetries such as static eccentricity (SE) and load current unbalance [...] Read more.
Permanent Magnet Synchronous Generators (PMSGs) have become increasingly important in industrial applications such as wind turbine systems due to their high efficiency and power density. However, their operational reliability can be affected by asymmetries such as static eccentricity (SE) and load current unbalance (UnB), which exhibit similar spectral features and are therefore difficult to differentiate using conventional techniques such as Motor Current Signature Analysis (MCSA). Stray flux analysis provides an alternative diagnostic approach, yet single-point measurements often lack the sensitivity required for accurate fault discrimination. This study introduces a diagnostic methodology based on the Space Vector Stray Flux (SVSF) for identifying static eccentricity (SE) and load current unbalance (UnB) faults in PMSG-based systems. The SVSF is derived from three external stray flux sensors placed 120° electrical degrees apart and analyzed through symmetrical component decomposition, focusing on the +5fs positive-sequence harmonic. Two-dimensional Finite Element Analysis (FEA) conducted on a 36-slot/12-pole PMSG model shows that the amplitude of the +5fs harmonic increases markedly under static eccentricity, while it remains nearly unchanged under load current unbalance. To validate the simulation findings, comprehensive experiments have been conducted on a dedicated test rig equipped with high-sensitivity fluxgate sensors. The experimental results confirm the robustness of the proposed SVSF method against practical constraints such as sensor placement asymmetry, 3D axial flux effects, and electromagnetic interference (EMI). The identified harmonic thus serves as a distinct and reliable indicator for differentiating static eccentricity from load current unbalance faults. The proposed SVSF-based approach significantly enhances the accuracy and robustness of fault detection and provides a practical tool for condition monitoring in PMSG. Full article
(This article belongs to the Special Issue Energy Saving Management Systems: Challenges and Applications)
►▼ Show Figures

Graphical abstract

22 pages, 396 KB  
Review
Towards a Unified Digital Ecosystem: The Role of Platform Technology Convergence
by Asif Mehmood, Mohammad Arif and Faisal Mehmood
Electronics 2025, 14(24), 4787; https://doi.org/10.3390/electronics14244787 - 5 Dec 2025
Abstract
The rapid evolution of platform technologies is transforming industries, interoperability, and innovation. Despite numerous studies on individual technologies, no prior review unifies AI, IoT, blockchain, and 5G with cross-sector standards, governance, and technical enablers to provide a comprehensive view of platform convergence. This [...] Read more.
The rapid evolution of platform technologies is transforming industries, interoperability, and innovation. Despite numerous studies on individual technologies, no prior review unifies AI, IoT, blockchain, and 5G with cross-sector standards, governance, and technical enablers to provide a comprehensive view of platform convergence. This narrative review synthesizes conceptual and technical literature from 2015–2025, focusing on how converging platform technologies interact across sectors. The review organizes findings by technological enablers, cross-domain integration mechanisms, sector-specific applications, and emergent trends, highlighting systemic synergies and challenges. The study demonstrates that AI, IoT, blockchain, cloud-edge architectures, and advanced communication networks collectively enable interoperable, secure, and adaptive ecosystems. Key enablers include standardized protocols, edge–cloud orchestration, and cross-platform data sharing, while challenges involve cybersecurity, regulatory compliance, and scalability. Sectoral examples span healthcare, finance, manufacturing, smart cities, and autonomous systems. Platform convergence offers transformative potential for sustainable and intelligent systems. Critical research gaps remain in unified architectures, privacy-preserving AI and blockchain mechanisms, and dynamic orchestration of heterogeneous systems. Emerging technologies such as quantum computing and federated learning are poised to further strengthen collaborative ecosystems. This review provides actionable insights for researchers, policymakers, and industry leaders aiming to harness platform convergence for innovation and sustainable development. Full article
(This article belongs to the Special Issue Converging Platform Technologies: Collaborative Innovations and Future Directions)
►▼ Show Figures

Figure 1

2 pages, 129 KB  
Correction
Correction: Zhao et al. A Performing Arts ICH-Driven Interaction Design Framework for Rehabilitation Games. Electronics 2025, 14, 3739
by Jing Zhao, Xinran Zhang, Yiming Ma, Yi Liu, Siyu Huo, Xiaotong Mu, Qian Xiao and Yuhong Han
Electronics 2025, 14(24), 4786; https://doi.org/10.3390/electronics14244786 - 5 Dec 2025
Abstract
In the original publication [...] Full article
24 pages, 3486 KB  
Article
Zero-Shot Industrial Anomaly Detection via CLIP-DINOv2 Multimodal Fusion and Stabilized Attention Pooling
by Junjie Jiang, Zongxiang He, Anping Wan, Khalil AL-Bukhaiti, Kaiyang Wang, Peiyi Zhu and Xiaomin Cheng
Electronics 2025, 14(24), 4785; https://doi.org/10.3390/electronics14244785 - 5 Dec 2025
Abstract
Industrial visual inspection demands high-precision anomaly detection amid scarce annotations and unseen defects. This paper introduces a zero-shot framework leveraging multimodal feature fusion and stabilized attention pooling. CLIP’s global semantic embeddings are hierarchically aligned with DINOv2’s multi-scale structural features via a Dual-Modality Attention [...] Read more.
Industrial visual inspection demands high-precision anomaly detection amid scarce annotations and unseen defects. This paper introduces a zero-shot framework leveraging multimodal feature fusion and stabilized attention pooling. CLIP’s global semantic embeddings are hierarchically aligned with DINOv2’s multi-scale structural features via a Dual-Modality Attention (DMA) mechanism, enabling effective cross-modal knowledge transfer for capturing macro- and micro-anomalies. A Stabilized Attention-based Pooling (SAP) module adaptively aggregates discriminative representations using self-generated anomaly heatmaps, enhancing localization accuracy and mitigating feature dilution. Trained solely in auxiliary datasets with multi-task segmentation and contrastive losses, the approach requires no target-domain samples. Extensive evaluation across seven benchmarks (MVTec AD, VisA, BTAD, MPDD, KSDD, DAGM, DTD-Synthetic) demonstrates state-of-the-art performance, achieving 93.4% image-level AUROC, 94.3% AP, 96.9% pixel-level AUROC, and 92.4% AUPRO on average. Ablation studies confirm the efficacy of DMA and SAP, while qualitative results highlight superior boundary precision and noise suppression. The framework offers a scalable, annotation-efficient solution for real-world industrial anomaly detection. Full article
►▼ Show Figures

Figure 1

14 pages, 2530 KB  
Article
Arrester Fault Recognition Model Based on Thermal Imaging Images Using VMamba
by Lin Lin, Jiantao Li, Jianan Wang, Yong Luo and Yueyue Liu
Electronics 2025, 14(24), 4784; https://doi.org/10.3390/electronics14244784 - 5 Dec 2025
Abstract
The intelligent fault detection of power plant equipment in industrial settings often grapples with challenges such as insufficient real-time performance and interference from complex backgrounds. To address these issues, this paper proposes an image recognition and classification model based on the VMamba architecture. [...] Read more.
The intelligent fault detection of power plant equipment in industrial settings often grapples with challenges such as insufficient real-time performance and interference from complex backgrounds. To address these issues, this paper proposes an image recognition and classification model based on the VMamba architecture. At the core of our feature extraction module, we have improved and optimized the two-dimensional state space (SS2D) algorithm to replace the traditional convolution operation. Rooted in State-Space Models (SSMs), the SS2D module possesses a global receptive field by design, enabling it to effectively capture long-range dependencies and establish comprehensive contextual relationships between local and global features. Crucially, unlike the self-attention mechanism in Vision Transformers (ViT) that suffers from quadratic computational complexity, VMamba achieves this global modeling with linear complexity, significantly enhancing computational efficiency. Furthermore, we employ an enhanced PAN-FPN multi-scale feature fusion strategy integrated with the Squeeze-and-Excitation (SE) attention mechanism. This combination optimizes the spatial distribution of feature representations through channel-wise attention weighting, facilitating the effective integration of cross-level spatial features and the suppression of background noise. This study thus presents a solution for industrial equipment fault diagnosis that achieves a superior balance between high accuracy and low latency. Full article
(This article belongs to the Special Issue Perception and Intelligent Control for Cyber–Physical Systems: From Robotics to Autonomous Ecosystems)
►▼ Show Figures

Figure 1

17 pages, 12946 KB  
Article
A Comparative Analysis of LLM-Based Customer Representation Learning Techniques
by Sangyeop Lee, Jong Seo Kim, Kisoo Kim, Bojung Ko, Junho Moon and Minsik Park
Electronics 2025, 14(24), 4783; https://doi.org/10.3390/electronics14244783 - 5 Dec 2025
Abstract
Recent advances in large language models (LLMs) have enabled the effective representation of customer behaviors, including purchases, repairs, and consultations. These LLM-based customer representation models apply to predicting future behavior of the customer or clustering customers with similar representations by latent vectors. Since [...] Read more.
Recent advances in large language models (LLMs) have enabled the effective representation of customer behaviors, including purchases, repairs, and consultations. These LLM-based customer representation models apply to predicting future behavior of the customer or clustering customers with similar representations by latent vectors. Since these representation technologies depend on data, this paper examines whether training a recommendation model (BERT4Rec) from scratch or fine-tuning a pre-trained LLM (ELECTRA) is more effective for our customer data. To address this, a three-step approach is conducted: (1) defining a sequence of customer behaviors into textual inputs for LLM-based representation learning, (2) extracting customer representation as latent vectors by training or fine-tuning representation models on a dataset of 14 million customers, and (3) training classifiers to predict purchase outcomes for eight products. Our focus is on comparing two primary approaches in step (2): training BERT4Rec from scratch versus fine-tuning pre-trained ELECTRA. The average AUC and F1-score of classifiers across eight products reveal that both methods achieve gaps of only 0.012 in AUC and 0.007 in F1-score. On the other hand, the fine-tuned ELECTRA achieves a 0.27 improvement in the top 10% lift for targeted marketing strategies. This result is particularly meaningful given that buyers of products constitute only about 0.5% of the entire dataset. Beyond the three-step approach, we make an effort to interpret latent space in two-dimensional and attention shifts in fine-tuned ELECTRA. Furthermore, we compare its efficiency advantages against fine-tuned LLaMA2. These findings provide practical insights for optimizing LLM-based representation models in industrial applications. Full article
(This article belongs to the Special Issue Machine Learning for Data Mining)
►▼ Show Figures

Figure 1

3 pages, 134 KB  
Editorial
Machine Learning in Electronic and Biomedical Engineering, Part 2
by Laura Falaschetti and Claudio Turchetti
Electronics 2025, 14(24), 4782; https://doi.org/10.3390/electronics14244782 - 5 Dec 2025
Abstract
The fast evolution of machine learning methods in both electronic and biomedical engineering continues to transform how data is interpreted, validated, and translated into actionable support systems [...] Full article
(This article belongs to the Special Issue Machine Learning in Electronic and Biomedical Engineering, Volume II)
16 pages, 3895 KB  
Article
Design and Simulation of T-Shaped Buncher for High Power Ridgetron Accelerator
by Danyang Li, Yu Yang and Zhibin Zhu
Electronics 2025, 14(24), 4781; https://doi.org/10.3390/electronics14244781 - 5 Dec 2025
Abstract
In this paper, a T-type buncher for a ridgetron accelerator is designed to further enhance the beam capture efficiency of a high-power ridgetron irradiation accelerator and reduce beam loss in the accelerator system. By incorporating a branch, the T-shaped buncher can reduce the [...] Read more.
In this paper, a T-type buncher for a ridgetron accelerator is designed to further enhance the beam capture efficiency of a high-power ridgetron irradiation accelerator and reduce beam loss in the accelerator system. By incorporating a branch, the T-shaped buncher can reduce the required space compared with conventional coaxial bunchers at the same operating frequency. The physical design and electromagnetic field simulation of the T-type buncher were carried out using the CST frequency-domain solver and eigenmode solver. Subsequently, the bunching performance and its impact on beam transport in the ridgetron accelerator were further evaluated using the PIC solver. The results show that, within an input power range of 120–200 W, a 5 ns input pulse can be compressed to less than 0.6 ns, while the energy spread is maintained between 22% and 26%. At an input power of 140 W, the application of the buncher reduces beam loss after the first deflection by approximately 40%. When a 5 ns electron bunch is compressed to 2 ns (non-FWHM), the beam current increases by a factor of approximately 3.13 compared with the injection without the buncher. These results clearly demonstrate the effectiveness of the T-shaped buncher in improving beam capture efficiency and overall accelerator performance, providing a valuable reference for further power enhancement of ridgetron accelerators. Full article
(This article belongs to the Topic Advanced Electromagnetic Modeling and Simulation for Multidisciplinary Engineering Systems)
►▼ Show Figures

Figure 1

Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-21
Previous Issue
Back to TopTop