Electronics

35 pages, 390 KB

Open AccessArticle

A Survey of RISC-V Secure Enclaves and Trusted Execution Environments

by Marouene Boubakri and Belhassen Zouari

Electronics 2025, 14(21), 4171; https://doi.org/10.3390/electronics14214171 (registering DOI) - 25 Oct 2025

RISC-V has emerged as a compelling alternative to proprietary instruction set architectures, distinguished by its openness, extensibility, and modularity. As the ecosystem matures, attention has turned to building confidential computing foundations, notably Trusted Execution Environments (TEEs) and secure enclaves, to support sensitive workloads. [...] Read more.

RISC-V has emerged as a compelling alternative to proprietary instruction set architectures, distinguished by its openness, extensibility, and modularity. As the ecosystem matures, attention has turned to building confidential computing foundations, notably Trusted Execution Environments (TEEs) and secure enclaves, to support sensitive workloads. These efforts explore a variety of design directions, yet reveal important trade-offs. Some approaches achieve strong isolation guarantees, but fall short in scalability or broad adoption. Others introduce defenses, such as memory protection or side-channel resistance, although often with significant performance costs that limit deployment in constrained systems. Lightweight enclaves address embedded contexts, but lack the advanced security features demanded by complex applications. In addition, early-stage development, complex programming models, and limited real-world validation hinder their usability. This survey reviews the current landscape of RISC-V TEEs and secure enclaves, analyzing their architectural principles, strengths, and weaknesses. To the best of our knowledge, this is the first work to present such a consolidated view. Finally, we highlight open challenges and research opportunities, aiming toward establishing a cohesive and trustworthy RISC-V trusted computing ecosystem. Full article

(This article belongs to the Special Issue Secure Hardware Architecture and Attack Resilience)

► Show Figures

Figure 1

8 pages, 4158 KB

Open AccessArticle

A Wideband Multi-Linear Polarization Reconfigurable Antenna with Artificial Magnetic Conductor

by Shixing Yu, Kaisheng Yang and Yingmeng Zhang

Electronics 2025, 14(21), 4170; https://doi.org/10.3390/electronics14214170 (registering DOI) - 25 Oct 2025

Abstract

This paper presents a wideband multi-linear polarization reconfigurable antenna featuring five linear polarization states. We use the semi-ellipsoidal dipoles as the main radiators to broaden the operating bandwidth; the states of linear polarizations are switched by controlling the ON/OFF of PIN diodes between [...] Read more.

This paper presents a wideband multi-linear polarization reconfigurable antenna featuring five linear polarization states. We use the semi-ellipsoidal dipoles as the main radiators to broaden the operating bandwidth; the states of linear polarizations are switched by controlling the ON/OFF of PIN diodes between feeding pads and dipoles to excite a specific pair of dipoles. A 7 × 7 AMC array is added below the antenna to obtain a small height of 0.14 λ₀ (λ₀ is the free space wavelength at the operating frequency). Prototypes of the designed antenna are fabricated, and experimental results illustrate that the proposed antenna yields an impedance bandwidth of 50% (from 2.25 GHz to 3.75 GHz) for all polarization states, stable radiation patterns, and low cross-polarization within the operating band. In addition, the maximum gain reaches 8.1 dBi. The proposed five linear-polarized switching antenna with wide band and low-profile features can be applied in reconfigurable conformal array antennas, thus flexibly realizing linear polarization reconfiguration of conformal arrays in radar and military platforms. Full article

► Show Figures

Figure 1

24 pages, 2881 KB

Open AccessArticle

Wear Leveling in SSDs Considered Harmful: A Case for Capacity Variance

by Ziyang Jiao and Biyuan Yang

Electronics 2025, 14(21), 4169; https://doi.org/10.3390/electronics14214169 (registering DOI) - 25 Oct 2025

Abstract

The trend of decreasing endurance of flash memory makes the overall lifetime of SSDs more sensitive to the effects of wear leveling. Under these circumstances, we observe that existing wear-leveling techniques exhibit anomalous behavior under workloads without clear access skew or under dynamic [...] Read more.

The trend of decreasing endurance of flash memory makes the overall lifetime of SSDs more sensitive to the effects of wear leveling. Under these circumstances, we observe that existing wear-leveling techniques exhibit anomalous behavior under workloads without clear access skew or under dynamic access patterns and produce high write amplification, as high as 5.4×, negating its intended benefits. We argue that wear leveling is an artifact for maintaining the fixed-capacity abstraction of a storage device, and it becomes unnecessary if the exported capacity of the SSD is to gracefully reduce. We show that this idea of capacity variance extends the lifetime of the SSD, allowing up to 2.94× more writes under real workloads. Full article

(This article belongs to the Special Issue Advances in Semiconductor Devices and Applications)

► Show Figures

Figure 1

19 pages, 1781 KB

Open AccessArticle

HiSeq-TCN: High-Dimensional Feature Sequence Modeling and Few-Shot Reinforcement Learning for Intrusion Detection

by Yadong Pei, Yanfei Tan, Wei Gao, Fangwei Li and Mingyue Wang

Electronics 2025, 14(21), 4168; https://doi.org/10.3390/electronics14214168 (registering DOI) - 25 Oct 2025

Abstract

Intrusion detection is essential to cybersecurity. However, the curse of dimensionality and class imbalance limit detection accuracy and impede the identification of rare attacks. To address these challenges, this paper proposes the high-dimensional feature sequence temporal convolutional network (HiSeq-TCN) for intrusion detection. The [...] Read more.

Intrusion detection is essential to cybersecurity. However, the curse of dimensionality and class imbalance limit detection accuracy and impede the identification of rare attacks. To address these challenges, this paper proposes the high-dimensional feature sequence temporal convolutional network (HiSeq-TCN) for intrusion detection. The proposed HiSeq-TCN transforms high-dimensional feature vectors into pseudo-temporal sequences, enabling the network to capture contextual dependencies across feature dimensions. This enhances feature representation and detection robustness. In addition, a few-shot reinforcement strategy adaptively assigns larger loss weights to minority classes, mitigating class imbalance and improving the recognition of rare attacks. Experiments on the NSL-KDD dataset show that HiSeq-TCN achieves an overall accuracy of 99.44%, outperforming support vector machines, deep neural networks, and long short-term memory models. More importantly, it significantly improves the detection of rare attack types such as remote-to-local and user-to-root attacks. These results highlight the potential of HiSeq-TCN for robust and reliable intrusion detection in practical cybersecurity environments. Full article

(This article belongs to the Special Issue Applications of Deep Learning in Cyber Threat Detection)

► Show Figures

Figure 1

26 pages, 573 KB

Open AccessArticle

Mutual V2I Multifactor Authentication Using PUFs in an Unsecure Multi-Hop Wi-Fi Environment

by Mohamed K. Elhadad and Fayez Gebali

Electronics 2025, 14(21), 4167; https://doi.org/10.3390/electronics14214167 (registering DOI) - 24 Oct 2025

Abstract

Secure authentication in vehicular ad hoc networks (VANETs) remains a fundamental challenge due to their dynamic topology, susceptibility to attacks, and scalability constraints in multi-hop communication. Existing approaches based on elliptic curve cryptography (ECC), blockchain, and fog computing have achieved partial success but [...] Read more.

Secure authentication in vehicular ad hoc networks (VANETs) remains a fundamental challenge due to their dynamic topology, susceptibility to attacks, and scalability constraints in multi-hop communication. Existing approaches based on elliptic curve cryptography (ECC), blockchain, and fog computing have achieved partial success but suffer from latency, resource overhead, and limited adaptability, leaving a gap for lightweight and hardware-rooted trust models. To address this, we propose a multi-hop mutual authentication protocol leveraging Physical Unclonable Functions (PUFs), which provide tamper-evident, device-specific responses for cryptographic key generation. Our design introduces a structured sequence of phases, including pre-deployment, registration, login, authentication, key establishment, and session maintenance, with optional multi-hop extension through relay vehicles. Unlike prior schemes, our protocol integrates fuzzy extractors for error tolerance, employs both inductive and game-based proofs for security guarantees, and maps BAN-logic reasoning to specific attack resistances, ensuring robustness against replay, impersonation, and man-in-the-middle attacks. The protocol achieves mutual trust between vehicles and RSUs while preserving anonymity via temporary identifiers and achieving forward secrecy through non-reused CRPs. Conceptual comparison with state-of-the-art PUF-based and non-PUF schemes highlights the potential for reduced latency, lower communication overhead, and improved scalability via cloud-assisted CRP lifecycle management, while pointing to the need for future empirical validation through simulation and prototyping. This work not only provides a secure and efficient solution for VANET authentication but also advances the field by offering the first integrated taxonomy-driven evaluation of PUF-enabled V2X protocols in multi-hop Wi-Fi environments. Full article

(This article belongs to the Special Issue Privacy and Security Vulnerabilities in 6G and Beyond Networks)

► Show Figures

Figure 1

23 pages, 1659 KB

Open AccessArticle

A Multi-View-Based Federated Learning Approach for Intrusion Detection

by Jia Yu, Guoqiang Wang, Nianfeng Shi, Raghav Saxena and Brian Lee

Electronics 2025, 14(21), 4166; https://doi.org/10.3390/electronics14214166 (registering DOI) - 24 Oct 2025

Abstract

Intrusion detection aims to identify the unauthorized activities within computer networks or systems by classifying events into normal or abnormal categories. As modern scenarios often involve multi-source data, multi-view fusion deep learning methods are employed to leverage diverse viewpoints for enhancing security threat [...] Read more.

Intrusion detection aims to identify the unauthorized activities within computer networks or systems by classifying events into normal or abnormal categories. As modern scenarios often involve multi-source data, multi-view fusion deep learning methods are employed to leverage diverse viewpoints for enhancing security threat detection. This paper introduces a novel intrusion detection approach using multi-view fusion within a federated learning framework, proposing an integrated AE Neural SVM (AE-NSVM) model that combines auto-encoder (AE) multi-view feature extraction and Support Vector Machine (SVM) classification. This approach simultaneously learns representative features from multiple views and classifies network samples into normal or seven attack categories while employing federated learning across clients to ensure adaptability and robustness in diverse network environments. The experimental results obtained from two benchmark datasets validate its superiority: on TON_IoT, the CAE-NSVM model achieves a highest F1-measure of 0.792 (1.4% higher than traditional pipeline systems); on UNSW-NB15, it delivers an F1-score of 0.829 with a 73% reduced training time and an 89% faster inference compared to baseline models. These results demonstrate the advantages of multi-view fusion in federated learning for balancing accuracy and efficiency in distributed intrusion detection systems. Full article

(This article belongs to the Special Issue Advances in Data Security: Challenges, Technologies, and Applications)

► Show Figures

Figure 1

21 pages, 3305 KB

Open AccessArticle

Automated Road Data Collection Systems Using UAVs: Comparative Evaluation of Architectures Based on Arduino Portenta H7 and ESP32-CAM

by Jorge García-González, Carlos Quiterio Gómez Muñoz, Diego Gachet Páez and Javier Sánchez-Soriano

Electronics 2025, 14(21), 4165; https://doi.org/10.3390/electronics14214165 (registering DOI) - 24 Oct 2025

Abstract

This article presents the design, implementation, and comparative evaluation of two Unmanned Aerial Vehicles (UAV)-based architectures for automated road data acquisition and processing. The first system uses Arduino Portenta H7 boards to perform real-time inference at the edge, reducing connectivity dependency. The second [...] Read more.

This article presents the design, implementation, and comparative evaluation of two Unmanned Aerial Vehicles (UAV)-based architectures for automated road data acquisition and processing. The first system uses Arduino Portenta H7 boards to perform real-time inference at the edge, reducing connectivity dependency. The second employs ESP32-CAM modules that transmit raw data for remote server-side processing. We experimentally validated and compared both systems in terms of power consumption, latency, and detection accuracy. Results show that the Portenta-based system consumes 36.2% less energy and achieves lower end-to-end latency (10,114 ms vs. 11,032 ms), making it suitable for connectivity-constrained scenarios. Conversely, the ESP32-CAM system is simpler to deploy and allows rapid model updates at the server. These findings provide a reference for Intelligent Transportation System (ITS) deployments requiring scalable, energy-efficient, and flexible road monitoring solutions. Full article

(This article belongs to the Special Issue Advances in Computer Vision for Autonomous Driving)

► Show Figures

Figure 1

39 pages, 1463 KB

Open AccessReview

Exploring Authentication Protocols for Secure and Efficient Internet of Medical Things Systems

by Seungbin Lee, Kyeong A Kang, Soowang Lee and Jiyoon Kim

Electronics 2025, 14(21), 4164; https://doi.org/10.3390/electronics14214164 (registering DOI) - 24 Oct 2025

Abstract

The Internet of Medical Things (IoMT) comprises the application of traditional Internet of Things (IoT) technologies in the healthcare domain. IoMT ensures seamless data-sharing among hospitals, patients, and healthcare service providers, thereby transforming the medical environment. The adoption of IoMT technology has made [...] Read more.

The Internet of Medical Things (IoMT) comprises the application of traditional Internet of Things (IoT) technologies in the healthcare domain. IoMT ensures seamless data-sharing among hospitals, patients, and healthcare service providers, thereby transforming the medical environment. The adoption of IoMT technology has made it possible to provide various medical services such as chronic disease care, emergency response, and preventive treatment. However, the sensitivity of medical data and the resource limitations of IoMT devices present persistent challenges in designing authentication protocols. Our study reviews the overall architecture of the IoMT and recent studies on IoMT protocols in terms of security requirements and computational costs. In addition, this study evaluates security using formal verification tools with Scyther and SVO Logic. The security requirements include authentication, mutual authentication, confidentiality, integrity, untraceability, privacy preservation, anonymity, multi-factor authentication, session key security, forward and backward secrecy, and lightweight operation. The analysis shows that protocols satisfying a multiple security requirements tend to have higher computational costs, whereas protocols with lower computational costs often provide weaker security. This demonstrates the trade-off relationship between robust security and lightweight operation. These indicators assist in selecting protocols by balancing the allocated resources and required security for each scenario. Based on the comparative analysis and a security evaluation of the IoMT, this paper provides security guidelines for future research. Moreover, it summarizes the minimum security requirements and offers insights that practitioners can utilize in real-world settings. Full article

(This article belongs to the Special Issue Advanced IoT Security Solutions for Healthcare and Critical Infrastructures)

65 pages, 3348 KB

Open AccessSystematic Review

The Role of Graph Neural Networks, Transformers, and Reinforcement Learning in Network Threat Detection: A Systematic Literature Review

by Thilina Prasanga Doremure Gamage, Jairo A. Gutierrez and Sayan K. Ray

Electronics 2025, 14(21), 4163; https://doi.org/10.3390/electronics14214163 (registering DOI) - 24 Oct 2025

Abstract

Traditional network threat detection based on signatures is becoming increasingly inadequate as network threats and attacks continue to grow in their novelty and sophistication. Such advanced network threats are better handled by anomaly detection based on Machine Learning (ML) models. However, conventional anomaly-based [...] Read more.

Traditional network threat detection based on signatures is becoming increasingly inadequate as network threats and attacks continue to grow in their novelty and sophistication. Such advanced network threats are better handled by anomaly detection based on Machine Learning (ML) models. However, conventional anomaly-based network threat detection with traditional ML and Deep Learning (DL) faces fundamental limitations. Graph Neural Networks (GNNs) and Transformers are recent deep learning models with innovative architectures, capable of addressing these challenges. Reinforcement learning (RL) can facilitate adaptive learning strategies for GNN- and Transformer-based Intrusion Detection Systems (IDS). However, no systematic literature review (SLR) has jointly analyzed and synthesized these three powerful modeling algorithms in network threat detection. To address this gap, this SLR analyzed 36 peer-reviewed studies published between 2017 and 2025, collectively identifying 56 distinct network threats via the proposed threat classification framework by systematically mapping them to Enterprise MITRE ATT&CK tactics and their corresponding Cyber Kill Chain stages. The reviewed literature consists of 23 GNN-based studies implementing 19 GNN model types, 9 Transformer-based studies implementing 13 Transformer architectures, and 4 RL-based studies with 5 different RL algorithms, evaluated across 50 distinct datasets, demonstrating their overall effectiveness in network threat detection. Full article

(This article belongs to the Special Issue AI-Enhanced Security: Advancing Threat Detection and Defense)

19 pages, 3886 KB

Open AccessArticle

3D Human Motion Prediction via the Decoupled Spatiotemporal Clue

by Mingrui Xu, Zheming Gu and Erping Li

Electronics 2025, 14(21), 4162; https://doi.org/10.3390/electronics14214162 (registering DOI) - 24 Oct 2025

Abstract

Human motion exhibits high-dimensional and stochastic characteristics, posing significant challenges for modeling and prediction. Existing approaches typically employ coupled spatiotemporal frameworks to generate future poses. However, the intrinsic nonlinearity of joint interactions over time, compounded by high-dimensional noise, often obscures meaningful motion features. [...] Read more.

Human motion exhibits high-dimensional and stochastic characteristics, posing significant challenges for modeling and prediction. Existing approaches typically employ coupled spatiotemporal frameworks to generate future poses. However, the intrinsic nonlinearity of joint interactions over time, compounded by high-dimensional noise, often obscures meaningful motion features. Notably, while adjacent joints demonstrate strong spatial correlations, their temporal trajectories frequently remain independent, adding further complexity to modeling efforts. To address these issues, we propose a novel framework for human motion prediction via the decoupled spatiotemporal clue (DSC), which explicitly disentangles and models spatial and temporal dependencies. Specifically, DSC comprises two core components: (i) a spatiotemporal decoupling module that dynamically identifies critical joints and their hierarchical relationships using graph attention combined with separable convolutions for efficient motion decomposition; and (ii) a pose generation module that integrates local motion denoising with global dynamics modeling through a spatiotemporal transformer that independently processes spatial and temporal correlations. Experiments on the widely used human motion datasets H3.6M and AMASS demonstrate the superiority of DSC, which achieves 13% average improvement in long-term prediction over state-of-the-art methods. Full article

(This article belongs to the Collection Computer Vision and Pattern Recognition Techniques)

► Show Figures

Figure 1

48 pages, 15781 KB

Open AccessArticle

Autonomous AI Agents for Multi-Platform Social Media Marketing: A Simultaneous Deployment Study

by Joongho Ahn and Moonsoo Kim

Electronics 2025, 14(21), 4161; https://doi.org/10.3390/electronics14214161 - 24 Oct 2025

Abstract

This exploratory proof-of-concept study investigated the simultaneous deployment of autonomous, persona-driven Artificial Intelligence (AI) agents across multiple social media platforms using the ElizaOS framework. We developed three platform-specific agents with seven-layer character architectures and deployed them on Twitter/X, Discord, and Telegram for 18 [...] Read more.

This exploratory proof-of-concept study investigated the simultaneous deployment of autonomous, persona-driven Artificial Intelligence (AI) agents across multiple social media platforms using the ElizaOS framework. We developed three platform-specific agents with seven-layer character architectures and deployed them on Twitter/X, Discord, and Telegram for 18 days. The system processed 5389 interactions while gathering feedback from 28 volunteer participants. Addressing three research questions, we found that: (1) automation effectiveness was platform-dependent, with direct support platforms (Telegram, Discord) rated more useful than broadcast-oriented Twitter/X; (2) character design impact depended primarily on platform-persona alignment rather than architectural sophistication; and (3) technical performance showed platform-specific patterns, with median storage times ranging from 9.0 milliseconds (Twitter/X) to 61.5 milliseconds (Telegram) and high variability across all platforms. A notable finding was what we term the “Discord Paradox”—high quality ratings (4.05/5) but lowest preference (8.7%), suggesting platform familiarity and accessibility influence adoption more than agent quality. While the deployment demonstrated technical feasibility and revealed distinct user dynamics across platforms, the findings indicate that platform-specific optimization may be more effective than universal approaches. This exploratory study advances understanding of multi-platform agent deployment for marketing automation, identifying behavioral patterns and platform-specific dynamics that offer testable hypotheses for future systematic research. Full article

(This article belongs to the Special Issue AI Applications of Multi-Agent Systems)

► Show Figures

Figure 1

31 pages, 1876 KB

Open AccessArticle

Hybrid Genetic Algorithm and Deep Reinforcement Learning Framework for IoT-Enabled Healthcare Equipment Maintenance Scheduling

by Francesco Nucci and Gabriele Papadia

Electronics 2025, 14(21), 4160; https://doi.org/10.3390/electronics14214160 - 24 Oct 2025

Abstract

We study the predictive maintenance scheduling for IoT-enabled medical equipment in multi-facility healthcare networks. The problem involves skill matching, time windows, and risk-aware priorities. We model a multi-skill Technician Routing and Scheduling Problem with IoT-predicted failure intervals and minimize a composite cost for [...] Read more.

We study the predictive maintenance scheduling for IoT-enabled medical equipment in multi-facility healthcare networks. The problem involves skill matching, time windows, and risk-aware priorities. We model a multi-skill Technician Routing and Scheduling Problem with IoT-predicted failure intervals and minimize a composite cost for technician activation and labor, travel/time, risk exposure within the failure window, and lateness beyond it. We propose a hybrid solver coupling a Genetic Algorithm (GA) for rapid exploration and feasible schedule generation with a Proximal Policy Optimization (PPO) agent warm-started via behavior cloning on GA elites and refined online in a receding-horizon manner. An optional, permissioned blockchain records tamper-evident maintenance events off the control loop for auditability. Across four case studies (10–30 facilities), the hybrid approach reduces total cost by 2.09–10.31% versus pure GA, by 0.57–2.65% versus pure Deep Reinforcement Learning (DRL), and by 0.93–2.86% versus OR-Tools VRP heuristic baseline. In controlled early-stopping runs guided by admissible GA/DRL time splits, we realized average wall-time savings up to 47.5% while keeping solution costs within 0.5% of full-budget runs and maintaining low or zero lateness and risk exposure. These results indicate that GA seeding improves sample efficiency and stability for DRL in complex, data-driven maintenance settings, yielding a practical, adaptive, and auditable scheduler for healthcare operations. Full article

(This article belongs to the Special Issue Innovative Systems and Solution in Healthcare Based on AI, Blockchain and IoT)

► Show Figures

Figure 1

24 pages, 4193 KB

Open AccessArticle

Reconfigurable Circularly Polarized Phased Array

by Eduardo S. Silveira, Fúlvio F. Oliveira, Bernardo M. Fabiani, Juner M. Vieira, Daniel B. Ferreira and Daniel C. Nascimento

Electronics 2025, 14(21), 4159; https://doi.org/10.3390/electronics14214159 - 24 Oct 2025

Abstract

This paper presents the design, construction, and tests of a polarization-reconfigurable phased array antenna. The proposed array allows the polarization at the main lobe maximum direction to be electronically reconfigured between right-hand (RHCP) and left-hand circular polarization (LHCP). Single-fed microstrip antennas, each with [...] Read more.

This paper presents the design, construction, and tests of a polarization-reconfigurable phased array antenna. The proposed array allows the polarization at the main lobe maximum direction to be electronically reconfigured between right-hand (RHCP) and left-hand circular polarization (LHCP). Single-fed microstrip antennas, each with four tunable varicap diodes, are employed in the phased array to achieve a low axial ratio (AR) at the steering angles. Special attention is given to the microstrip antenna design and varicap modeling, which involves the use of measured data and search algorithms running in an electromagnetic/circuit co-simulation environment. To illustrate the proposed approach, a six-element linear phased array at 2.2 GHz has been built and tested in an anechoic chamber. The experimental results demonstrate an AR below 1 dB in both RHCP and LHCP states over a wide range of steering angles, and even in a multibeam configuration, validating our design method. Full article

(This article belongs to the Section Microwave and Wireless Communications)

► Show Figures

Figure 1

20 pages, 8631 KB

Open AccessArticle

Sensor Phase Information Compensation Method Based on MQPSO

by Fengcai Cao, Ruyu Luo, Wenhao Li, Yonghong Tian and Jian Li

Electronics 2025, 14(21), 4158; https://doi.org/10.3390/electronics14214158 - 23 Oct 2025

Abstract

In the source location of underground explosions, the phase non-consistency among sensors can cause significant errors in the extraction of the time difference in the arrival of seismic waves, seriously affecting the accuracy of source location. To address the above-mentioned problem, this paper [...] Read more.

In the source location of underground explosions, the phase non-consistency among sensors can cause significant errors in the extraction of the time difference in the arrival of seismic waves, seriously affecting the accuracy of source location. To address the above-mentioned problem, this paper proposes a phase compensation method based on the Multi-strategy Quantum behaved Particle Swarm Optimization (MQPSO) algorithm. First, this method calibrates the phases of vibration sensors to obtain the phase differences among sensors. Second, it uses the MQPSO intelligent optimization algorithm to correct the phase differences among vibration sensors. Finally, simulations and field tests are carried out for verification. The experimental results show that after adopting the phase compensation method with MQPSO, the range of phase differences in sensors is reduced by an average of 91% compared with the uncompensated state. This fully verifies that the phase compensation method of MQPSO can effectively complete the phase consistency calibration of sensors, providing important support for the source location of underground explosions. Full article

(This article belongs to the Section Circuit and Signal Processing)

► Show Figures

Figure 1

23 pages, 4862 KB

Open AccessArticle

Development of High-Power DC Solid-State Power Controllers Using SiC FETs for Aircraft Electrical Systems

by Xin Zhao, Chuanyou Xu, Ke Ma, Xuanlyu Wu, Xiliang Chen, Xiangke Li and Xiaohua Wu

Electronics 2025, 14(21), 4157; https://doi.org/10.3390/electronics14214157 - 23 Oct 2025

Abstract

The growing demand for improved interruption performance characteristics in emerging aircraft high-voltage direct current (HVDC) electrical networks motivates the rapid development of solid-state power controllers (SSPCs). This article presents a comprehensive design procedure for a 270 V 300 A SSPC utilizing discrete SiC [...] Read more.

The growing demand for improved interruption performance characteristics in emerging aircraft high-voltage direct current (HVDC) electrical networks motivates the rapid development of solid-state power controllers (SSPCs). This article presents a comprehensive design procedure for a 270 V 300 A SSPC utilizing discrete SiC cascode devices. Due to the high fault current and limited power of single switches, parallel SiC FETs are essential for interrupting high fault currents in SSPCs. Consequently, the challenge of current balancing among parallel devices is addressed in this paper by adopting a passive current balancing strategy based on an irregular-shaped busbar. Furthermore, considering the voltage spikes arising from the power loop parasitic inductance and TVS characteristics during fault interruption, an RC-TVS-based transient voltage mitigation circuit is proposed to suppress these peak voltages. Moreover, thermal models for overload and short-circuit conditions were developed to optimize the thermal management system to ensure reliable operation of the SSPC. Finally, a prototype of 270 V/300 A SSPC has been built to validate the key characteristics of the proposed high power SSPC. Full article

(This article belongs to the Special Issue Compatibility, Power Electronics and Power Engineering)

► Show Figures

Figure 1

24 pages, 4829 KB

Open AccessArticle

Validating a Wearable VR Headset for Postural Sway: Comparison with Force Plate COP Across Standardized Sensorimotor Tests

by David Saucier, Kaitlyn McDonald, Michael Mydlo, Rachel Barber, Emily Wall, Hunter Derby, Jennifer C. Reneker, Harish Chander, Reuben F. Burch and James L. Weinstein

Electronics 2025, 14(21), 4156; https://doi.org/10.3390/electronics14214156 - 23 Oct 2025

Abstract

This study seeks to determine the efficacy of a novel, virtual reality (VR)-based sensorimotor assessment tool, VIST Neuro-ID, in comparison to the gold standard. This was achieved through computing common postural sway metrics, as well as comparing these metrics across population groups including [...] Read more.

This study seeks to determine the efficacy of a novel, virtual reality (VR)-based sensorimotor assessment tool, VIST Neuro-ID, in comparison to the gold standard. This was achieved through computing common postural sway metrics, as well as comparing these metrics across population groups including sex and age (50–60 vs. 61–75). Sensorimotor assessments were conducted within the VIST Neuro-ID VR software while participants stood on a force plate. A proxy for center-of-pressure measurement was developed using the six-degree-of-freedom data collected from the head-mounted display used with the VR system. Moderate-to-high (r = 0.542–0.906) Pearson’s correlations were found between VIST Neuro-ID and the force plate for all eight postural sway metrics that were computed. Both systems detected significant differences (p < 0.05) across age groups for all metrics, except for two-dimensional path length from the force plate. Several significant differences were found across sexes, including AP and resultant sway velocity from the force plate, and resultant and AP root-mean-square from the HTC Vive Pro Eye. This indicates potential for VR to be used to collect vital postural sway metrics needed for assessing patient function, while also highlighting potential to identify balance patterns related to aging. Full article

(This article belongs to the Special Issue Latest Advancements in Wearable Technologies: Sensors, Devices, Smart Wear and Exoskeletons)

► Show Figures

Figure 1

15 pages, 2174 KB

Open AccessArticle

BoxingPro: An IoT-LLM Framework for Automated Boxing Coaching via Wearable Sensor Data Fusion

by Man Zhu, Pengfei Huang, Xiaolong Xu, Houpeng He and Lijie Zhang

Electronics 2025, 14(21), 4155; https://doi.org/10.3390/electronics14214155 - 23 Oct 2025

Abstract

The convergence of Internet of Things (IoT) and Artificial Intelligence (AI) has enabled personalized sports coaching, yet a significant gap remains: translating low-level sensor data into high-level, contextualized feedback. Large Language Models (LLMs) excel at reasoning and instruction but lack a native understanding [...] Read more.

The convergence of Internet of Things (IoT) and Artificial Intelligence (AI) has enabled personalized sports coaching, yet a significant gap remains: translating low-level sensor data into high-level, contextualized feedback. Large Language Models (LLMs) excel at reasoning and instruction but lack a native understanding of physical kinematics. This paper introduces BoxingPro, a novel framework that bridges this semantic gap by fusing wearable sensor data with LLMs for automated boxing coaching. Our core contribution is a dedicated translation methodology that converts multi-modal time-series data (IMU) and visual data (video) into structured linguistic prompts, enabling off-the-shelf LLMs to perform sophisticated biomechanical reasoning without extensive retraining. Our evaluation with professional boxers showed that the generated feedback achieved an average expert rating of over 4.0/5.0 on key criteria like biomechanical correctness and actionability. This work establishes a new paradigm for integrating sensor-based systems with LLMs, with potential applications extending far beyond boxing to any domain requiring physical skill assessment. Full article

(This article belongs to the Special Issue Techniques and Applications in Prompt Engineering and Generative AI)

► Show Figures

Figure 1

32 pages, 2523 KB

Open AccessArticle

VClass Engager: A Generative AI-Based System for Enhancing Student Engagement in Synchronous Online Classes

by Ali Alammary

Electronics 2025, 14(21), 4154; https://doi.org/10.3390/electronics14214154 - 23 Oct 2025

Abstract

Student disengagement remains a major barrier to effective learning in synchronous online classrooms, where lack of interaction, limited feedback, and screen fatigue often contribute to passive participation. Despite the growing use of generative AI, there is a notable lack of empirical research investigating [...] Read more.

Student disengagement remains a major barrier to effective learning in synchronous online classrooms, where lack of interaction, limited feedback, and screen fatigue often contribute to passive participation. Despite the growing use of generative AI, there is a notable lack of empirical research investigating the application of generative AI in addressing engagement challenges in synchronous online sessions. This study introduces VClass Engager, a novel experimental system that utilizes generative AI to foster student participation and deliver instant, personalized feedback during live virtual sessions. The system integrates several features, including instant analysis of student answers to chat questions, real-time AI-generated feedback, and a leaderboard that displays students’ cumulative scores to promote sustained engagement. To assess its effectiveness, the system was evaluated across multiple courses. Engagement was measured by tracking participation in three in-class formative questions, and response quality was analyzed using a cumulative link mixed model (CLMM). In addition, a post-session survey captured students’ perceptions regarding usability, motivational impact, and feedback quality. Results demonstrated statistically significant increases in student participation and response quality in sessions using VClass Engager compared to baseline sessions. Survey responses revealed high levels of satisfaction with the system’s ease of use and motivational aspects. By combining AI and gamification, this study provides early empirical evidence for a promising approach to enhancing engagement in synchronous online learning. Full article

(This article belongs to the Special Issue New Trends in Intelligent User Interfaces and Human-Computer Interactions with Large Language Models)

► Show Figures

Figure 1

28 pages, 3758 KB

Open AccessArticle

A Lightweight, Explainable Spam Detection System with Rüppell’s Fox Optimizer for the Social Media Network X

by Haidar AlZeyadi, Rıdvan Sert and Fecir Duran

Electronics 2025, 14(21), 4153; https://doi.org/10.3390/electronics14214153 - 23 Oct 2025

Abstract

Effective spam detection systems are essential in online social media networks (OSNs) and cybersecurity, and they directly influence the quality of decision-making pertaining to security. With today’s digital communications, unsolicited spam degrades user experiences and threatens platform security. Machine learning-based spam detection systems [...] Read more.

Effective spam detection systems are essential in online social media networks (OSNs) and cybersecurity, and they directly influence the quality of decision-making pertaining to security. With today’s digital communications, unsolicited spam degrades user experiences and threatens platform security. Machine learning-based spam detection systems offer an automated defense. Despite their effectiveness, such methods are frequently hindered by the “black box” problem, an interpretability deficiency that constrains their deployment in security applications, which, in order to comprehend the rationale of classification processes, is crucial for efficient threat evaluation and response strategies. However, their effectiveness hinges on selecting an optimal feature subset. To address these issues, we propose a lightweight, explainable spam detection model that integrates a nature-inspired optimizer. The approach employs clean data with data preprocessing and feature selection using a swarm-based, nature-inspired meta-heuristic Rüppell’s Fox Optimization (RFO) algorithm. To the best of our knowledge, this is the first time the algorithm has been adapted to the field of cybersecurity. The resulting minimal feature set is used to train a supervised classifier that achieves high detection rates and accuracy with respect to spam accounts. For the interpretation of model predictions, Shapley values are computed and illustrated through swarm and summary charts. The proposed system was empirically assessed using two datasets, achieving accuracies of 99.10%, 98.77%, 96.57%, and 92.24% on Dataset 1 using RFO with DT, KNN, AdaBoost, and LR and 98.94%, 98.67%, 95.04%, and 94.52% on Dataset 2, respectively. The results validate the efficacy of the suggested approach, providing an accurate and understandable model for spam account identification. This study represents notable progress in the field, offering a thorough and dependable resolution for spam account detection issues. Full article

(This article belongs to the Special Issue Explainable Artificial Intelligence: Concepts, Techniques, Analytics and Applications)

► Show Figures

Figure 1

25 pages, 2466 KB

Open AccessArticle

Methods for Predicting the Repair Stack in an Electronics Module Manufacturing Company

by Krzysztof Górecki, Wojciech Kowalke and Przemysław Ptak

Electronics 2025, 14(21), 4152; https://doi.org/10.3390/electronics14214152 - 23 Oct 2025

Abstract

This article addresses the problem of predicting the workload of the repair department in a company manufacturing electronic modules. The number of modules needing repair is called a repair stack. A deterministic algorithm and a machine learning-based algorithm are proposed to predict the [...] Read more.

This article addresses the problem of predicting the workload of the repair department in a company manufacturing electronic modules. The number of modules needing repair is called a repair stack. A deterministic algorithm and a machine learning-based algorithm are proposed to predict the repair stack for subsequent weeks based on historical data, current yield data, and planned production. These methods allow for estimation of the repair stack and appropriate selection of repair department staff to ensure the ongoing repair of defective products. The proposed algorithms are described and the results of their practical verification based on historical data from a large enterprise are presented. The practical utility of both algorithms is demonstrated and the impact of selected factors on their accuracy is analyzed. It is shown that using the proposed algorithms, it is possible to predict the repair stack for the coming week with a relative error not exceeding a few percentages on the basis of historical data from the previous 8 weeks. These algorithms were successfully implemented in industrial practice. Full article

(This article belongs to the Section Industrial Electronics)

► Show Figures

Figure 1

16 pages, 3663 KB

Open AccessArticle

MSRDSN: A Novel Deep Learning Model for Fault Diagnosis of High-Voltage Disconnectors

by Shijian Zhu, Peilong Chen, Xin Li, Qichen Deng, Yuxiang Liao and Jiangjun Ruan

Electronics 2025, 14(21), 4151; https://doi.org/10.3390/electronics14214151 - 23 Oct 2025

Abstract

The operational state of high-voltage disconnectors plays a critical role in ensuring the safety, stability, and power supply reliability of electrical systems. To enable accurate identification of the operational status of high-voltage disconnectors, this paper proposes a fault diagnosis method based on a [...] Read more.

The operational state of high-voltage disconnectors plays a critical role in ensuring the safety, stability, and power supply reliability of electrical systems. To enable accurate identification of the operational status of high-voltage disconnectors, this paper proposes a fault diagnosis method based on a Multi-Scale Residual Depthwise Separable Convolutional Neural Network (MSRDSN). First, wavelet transform is applied to vibration signals to perform multi-scale analysis and enhance detail resolution. Then, a novel network architecture, referred to as RDSN, is constructed to extract discriminative high-level features from vibration signals by integrating residual learning blocks and depthwise separable convolution blocks. Furthermore, a combined loss function is introduced to optimize the RDSN, which simultaneously maximizes inter-class distance, minimizes intra-class distance, and reduces feature redundancy. Experimental results show that the proposed method achieves a top accuracy of 99.44% on a balanced dataset, outperforming the sub-optimal approach by 1.11%. This study offers a novel and effective solution for fault diagnosis in high-voltage disconnectors. Full article

(This article belongs to the Special Issue Advances in Condition Monitoring, Diagnosis, and Prognostics for Power Equipment)

► Show Figures

Figure 1

25 pages, 22171 KB

Open AccessArticle

Physics-Informed Co-Optimization of Fuel-CellFlying Vehicle Propulsion and Control Systems with Onboard Catalysis

by Yifei Bao, Chaoyi Chen, Hao Zhang and Nuo Lei

Electronics 2025, 14(21), 4150; https://doi.org/10.3390/electronics14214150 - 23 Oct 2025

Abstract

Fuel-cell flying vehicles suffer from limited endurance, while ammonia, decomposed onboard to supply hydrogen, offers a carbon-free, high-density solution to extend flight missions. However, the system’s performance is governed by a multi-scale coupling between propulsion and control systems. To this end, this paper [...] Read more.

Fuel-cell flying vehicles suffer from limited endurance, while ammonia, decomposed onboard to supply hydrogen, offers a carbon-free, high-density solution to extend flight missions. However, the system’s performance is governed by a multi-scale coupling between propulsion and control systems. To this end, this paper introduces a novel optimization paradigm, termed physics-informed gradient-enhanced multi-objective optimization (PI-GEMO), to simultaneously optimize the ammonia decomposition unit (ADU) catalyst composition, powertrain sizing, and flight control parameters. The PI-GEMO framework leverages a physics-informed neural network (PINN) as a differentiable surrogate model, which is trained not only on sparse simulation data but also on the governing differential equations of the system. This enables the use of analytical gradient information extracted from the trained PINN via automatic differentiation to intelligently guide the evolutionary search process. A comprehensive case study on a flying vehicle demonstrates that the PI-GEMO framework not only discovers a superior set of Pareto-optimal solutions compared to traditional methods but also critically ensures the physical plausibility of the results. Full article

(This article belongs to the Special Issue Eco-Safe Intelligent Mobility Development and Application)

► Show Figures

Figure 1

21 pages, 4796 KB

Open AccessArticle

Real-Time Lightweight Vehicle Object Detection via Layer-Adaptive Model Pruning

by Yu Zhang, Junhui Zhang, Feng Du, Wenjie Kang, Cen Wang and Guofei Li

Electronics 2025, 14(21), 4149; https://doi.org/10.3390/electronics14214149 - 23 Oct 2025

Abstract

With the rapid advancement in autonomous driving technology, vehicle object detection has become a crucial component of perception systems, where accuracy and inference speed directly influence driving safety. To address the limitations of existing lightweight detection models in small-object perception and deployment efficiency, [...] Read more.

With the rapid advancement in autonomous driving technology, vehicle object detection has become a crucial component of perception systems, where accuracy and inference speed directly influence driving safety. To address the limitations of existing lightweight detection models in small-object perception and deployment efficiency, this study proposes an enhanced YOLOv8n-based framework, termed YOLOv8n-ALM. The proposed model integrates Mixed Local Channel Attention (MLCA), a Task-Aligned Dynamic Detection Head (TADDH), and Layer-Adaptive Magnitude-based Pruning (LAMP). Specifically, MLCA enhances the representation of salient regions, TADDH aligns classification and regression tasks while leveraging DCNv2 for improved spatial adaptability, and LAMP compresses the network to accelerate inference. Experiments conducted on the KITTI dataset demonstrate that YOLOv8n-ALM improves mAP@0.5 by 2.2% and precision by 5.8%, while reducing parameters by 65.33% and computational load by 29.63%. These results underscore the proposed method’s capability to achieve real-time, compact, and accurate vehicle detection, demonstrating strong potential for deployment in intelligent vehicles and embedded systems. Full article

(This article belongs to the Special Issue Deep Learning-Based Object Detection and Tracking)

► Show Figures

Figure 1

20 pages, 4429 KB

Open AccessArticle

ANT-KT: Adaptive NAS Transformers for Knowledge Tracing

by Shuanglong Yao, Yichen Song, Ye Liu, Ji Chen, Deyu Zhao and Xing Wang

Electronics 2025, 14(21), 4148; https://doi.org/10.3390/electronics14214148 - 23 Oct 2025

Abstract

Knowledge Tracing aims to assess students’ mastery of knowledge concepts in real time, playing a crucial role in providing personalized learning services in intelligent tutoring systems. In recent years, researchers have attempted to introduce Neural Architecture Search (NAS) into knowledge tracing tasks to [...] Read more.

Knowledge Tracing aims to assess students’ mastery of knowledge concepts in real time, playing a crucial role in providing personalized learning services in intelligent tutoring systems. In recent years, researchers have attempted to introduce Neural Architecture Search (NAS) into knowledge tracing tasks to automatically design more efficient network structures. However, existing NAS-based methods for Knowledge Tracing suffer from excessively large search spaces and slow search efficiency, which significantly constrain their practical applications. To address these limitations, this paper proposes an Adaptive Neural Architecture Search framework based on Transformers for KT, called ANT-KT. Specifically, we design an enhanced encoder that combines convolution operations with state vectors to capture both local and global dependencies in students’ learning sequences. Moreover, an optimized decoder with a linear attention mechanism is introduced to improve the efficiency of modeling long-term student knowledge state evolution. We further propose an evolutionary NAS algorithm that incorporates a model optimization efficiency objective and a dynamic search space reduction strategy, enabling the discovery of high-performing yet computationally efficient architectures. Experimental results on two large-scale real-world datasets, EdNet and RAIEd2020, demonstrate that ANT-KT significantly reduces time costs across all stages of NAS while achieving performance improvements on multiple evaluation metrics, validating the efficiency and practicality of the proposed method. Full article

(This article belongs to the Section Artificial Intelligence)

► Show Figures

Figure 1

17 pages, 21481 KB

Open AccessArticle

Machine Learning-Based State-of-Charge Prediction for Electric Bus Fleet: A Critical Analysis

by Simone Volturno, Andrea Di Martino and Michela Longo

Electronics 2025, 14(21), 4147; https://doi.org/10.3390/electronics14214147 - 23 Oct 2025

Abstract

The transportation sector is undergoing a rapid energy transition. Electric Vehicles (EVs) are gradually replacing conventional ones in many different sectors, but battery management still represents a critical limitation of this process. Consequently, research in this area is expanding, aiming to develop solutions [...] Read more.

The transportation sector is undergoing a rapid energy transition. Electric Vehicles (EVs) are gradually replacing conventional ones in many different sectors, but battery management still represents a critical limitation of this process. Consequently, research in this area is expanding, aiming to develop solutions that enhance performance while minimizing environmental impact. This study addresses the application of Machine Learning (ML) techniques to estimate the battery State of Charge (SoC) for a full-electric bus fleet operating public service. The methodology is built based on the available driving data disclosed from the fleet monitoring system. The ML methods are assessed starting from model-based (MB) observers assumed as reference and performances are compared upon this basis. The datasets are retrieved from a public repository or derived from real cases, particularly referring to an electric bus fleet operating for an urban public service. The most critical limitation is the absence of the electrical input data coming from the battery, typically required by model-based approaches. Despite this, the proposed ML model achieved sufficient accuracy levels (RMSE < 0.3%) comparable to those of traditional observers. These outcomes demonstrate the potential of data-driven approaches to provide scalable and more straightforward tools for battery monitoring. Full article

(This article belongs to the Special Issue Feature Papers in Artificial Intelligence)

► Show Figures

Figure 1

17 pages, 2813 KB

Open AccessArticle

Acoustic Emission from GaN-on-Sapphire Structures

by Bartlomiej K. Paszkiewicz, Bogdan Paszkiewicz and Andrzej Dziedzic

Electronics 2025, 14(21), 4146; https://doi.org/10.3390/electronics14214146 - 23 Oct 2025

Abstract

This paper presents a study on the propagation of acoustic waves in gallium nitride (GaN) layers deposited on sapphire substrate. The influence of GaN layer thickness and the configuration of interdigital transducers (IDTs) on the generation and propagation of different surface wave modes, [...] Read more.

This paper presents a study on the propagation of acoustic waves in gallium nitride (GaN) layers deposited on sapphire substrate. The influence of GaN layer thickness and the configuration of interdigital transducers (IDTs) on the generation and propagation of different surface wave modes, including Rayleigh, Sezawa, and Love waves, was analyzed. Experimental measurements in the 100 MHz–6 GHz range were complemented by numerical simulations using the finite element method (FEM). The results demonstrated a strong dependence of wave characteristics on technological parameters, particularly the quality of the GaN–sapphire interface. The data obtained can be utilized for optimizing the design of acoustic sensors, resonators, and RF filters. Full article

(This article belongs to the Section Electronic Materials, Devices and Applications)

► Show Figures

Figure 1

16 pages, 4632 KB

Open AccessArticle

Equivalent Modeling Method for Thermal Calculation of Transformer Windings Based on Minimum Thermal Resistance

by Chengxiang Liu, Yan Li, Zhanyang Yu, Chunhui Zhang, Yedong Mao, Jingmeng Li and Ge Xu

Electronics 2025, 14(21), 4145; https://doi.org/10.3390/electronics14214145 - 23 Oct 2025

Abstract

In the accurate calculation of the temperature rise and hot spots of transformer windings, considering the inter-turn insulation will lead to a sharp increase in the workload of detailed modeling and a large number of mesh refinements. To address this issue, this paper [...] Read more.

In the accurate calculation of the temperature rise and hot spots of transformer windings, considering the inter-turn insulation will lead to a sharp increase in the workload of detailed modeling and a large number of mesh refinements. To address this issue, this paper proposes a calculation method for the equivalent thermal conductivity based on the minimum thermal resistance principle. This method can accurately calculate the equivalent thermal conductivities in the axial and radial directions of the windings. By using the disk windings model under the equivalent thermal conductivity, a temperature field analysis is performed on the actual windings with inter-turn insulation. To validate the method, models considering inter-turn insulation and equivalent models are constructed, and detailed analyses are performed using the empirical formula method and the equivalent method based on the minimum thermal resistance principle, respectively. By comparing the simulation results of the equivalent model and the model considering inter-turn insulation, it is found that the equivalent method based on the minimum thermal resistance principle not only significantly reduces the number of mesh elements but also achieves significantly improved accuracy in the temperature field analysis of the turn-divided model compared to the empirical formula method. Moreover, the winding temperature rise and hot spot positions before and after the equivalence are nearly identical and closer to the experimental results, demonstrating the validity of this method. Full article

► Show Figures

Figure 1

29 pages, 674 KB

Open AccessReview

An Overview of Existing Applications of Artificial Intelligence in Histopathological Diagnostics of Leukemias: A Scoping Review

by Mieszko Czapliński, Grzegorz Redlarski, Paweł Kowalski, Piotr Mateusz Tojza, Adam Sikorski and Arkadiusz Żak

Electronics 2025, 14(21), 4144; https://doi.org/10.3390/electronics14214144 - 23 Oct 2025

Abstract

Artificial intelligence applications in histopathological diagnostics are rapidly expanding, with particular promise in complex hematological malignancies where diagnostic accuracy remains challenging and subjective. This study undertakes a scoping review to systematically map the extent of research on artificial intelligence applications in histopathological diagnostics [...] Read more.

Artificial intelligence applications in histopathological diagnostics are rapidly expanding, with particular promise in complex hematological malignancies where diagnostic accuracy remains challenging and subjective. This study undertakes a scoping review to systematically map the extent of research on artificial intelligence applications in histopathological diagnostics of leukemias, examine geographic distribution and methodological approaches, and assess the current state of AI model performance and clinical readiness. A comprehensive search was conducted in the Scopus database covering publications from 2018 to 2025 (as of 12 July 2025), using five targeted search strategies combining AI, histopathology, and leukemia-related terms. Following a three-stage screening protocol, 418 publications were selected from an initial pool of over 75,000 records across multiple countries and research domains. The analysis revealed a marked increase in research output, peaking in 2024 with substantial contributions from India (26.3%), China (17.9%), USA (13.8%), and Saudi Arabia (11.1%). Among 43 documented datasets ranging from 80 to 42,386 images, studies predominantly utilized convolutional neural networks and deep learning approaches. AI models demonstrated high diagnostic accuracy, with 25 end-to-end models achieving an average accuracy of 97.72% compared to 96.34% for 20 classical machine learning approaches. Most studies focused on acute lymphoblastic leukemia detection and subtype classification using blood smear and bone marrow specimens. Despite promising diagnostic performance, significant gaps remain in clinical translation, standardization, and regulatory approval, with none of the reviewed AI systems currently FDA-approved for routine leukemia diagnostics. Future research should prioritize clinical validation studies, standardized datasets, and integration with existing diagnostic workflows to realize the potential of AI in hematopathological practice. Full article

(This article belongs to the Special Issue Edge AI for Biomedical Applications: Innovations in Sensing, Computing and Security)

► Show Figures

Figure 1

13 pages, 2749 KB

Open AccessArticle

Analysis of the Optimal Receiver System for Underwater Electromagnetic Detection

by Bo Tang, Linsen Zhang and Siwei Tan

Electronics 2025, 14(21), 4143; https://doi.org/10.3390/electronics14214143 - 22 Oct 2025

Abstract

Given the characteristics of underwater electromagnetic detection systems, starting from typical applications, this paper analyzes the impact of random variables on the optimal receiver. The mathematical expression of the optimal receiver is derived using the Generalized Likelihood Ratio Test (GLRT), and the test [...] Read more.

Given the characteristics of underwater electromagnetic detection systems, starting from typical applications, this paper analyzes the impact of random variables on the optimal receiver. The mathematical expression of the optimal receiver is derived using the Generalized Likelihood Ratio Test (GLRT), and the test statistics are determined. Expressions for the receiver threshold, detection probability, and false alarm probability are derived, and the system block diagram of the optimal receiver is obtained. Through simulation, the working characteristics of the generalized likelihood ratio receiver and the matched filter are compared. It is verified that the performance of this receiver is close to that of a matched filter under different conditions. Since the maximum likelihood estimation of random parameters is used, this receiver is considered suboptimal. However, when the signal-to-noise ratio (SNR) is high, this receiver can replace the matched filter as the optimal receiver. Full article

(This article belongs to the Section Circuit and Signal Processing)

► Show Figures

Figure 1

20 pages, 2508 KB

Open AccessArticle

An Attention-Enhanced Network for Person Re-Identification via Appearance–Gait Fusion

by Zelong Yu, Yixiang Cai, Hanming Xu, Lei Chen, Mingqian Yang, Huabo Sun and Xiangyu Zhao

Electronics 2025, 14(21), 4142; https://doi.org/10.3390/electronics14214142 - 22 Oct 2025

Abstract

The objective of person re-identification (Re-ID) is to recognize a given target pedestrian across different cameras. However, perspective variations, resulting from differences in shooting angles, often significantly impact the accuracy of person Re-ID. To address this issue, this paper presents an attention-enhanced person [...] Read more.

The objective of person re-identification (Re-ID) is to recognize a given target pedestrian across different cameras. However, perspective variations, resulting from differences in shooting angles, often significantly impact the accuracy of person Re-ID. To address this issue, this paper presents an attention-enhanced person Re-ID algorithm based on appearance–gait information interaction. Specifically, appearance features and gait features are first extracted from RGB images and gait energy images (GEIs), respectively, using two ResNet-50 networks. Then, a multimodal information exchange module based on the attention mechanism is designed to build a bridge for information exchange between the two modalities during the feature extraction process. This module aims to enhance the feature extraction ability through mutual guidance and reinforcement between the two modalities, thereby improving the model’s effectiveness in integrating the two types of modal information. Subsequently, to further balance the signal-to-noise ratio, importance weight estimation is employed to map perspective information into the importance weights of the two features. Finally, based on the autoencoder structure, the two features are weighted and fused under the guidance of importance weights to generate fused features that are robust to perspective changes. The experimental results on the CASIA-B dataset indicate that, under conditions of viewpoint variation, the method proposed in this paper achieved an average accuracy of 94.9%, which is 1.1% higher than the next best method, and obtained the smallest variance of 4.199, suggesting that the method proposed in this paper is not only more accurate but also more stable. Full article

(This article belongs to the Special Issue Artificial Intelligence and Microsystems)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Electronics, Volume 14, Issue 21 (November-1 2025) – 39 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI