Next Issue
Volume 15, September-1
Previous Issue
Volume 15, August-1
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
7.0
2.9
Journals
Electronics
Volume 15
Issue 16
electronics-logo

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

Electronics, Volume 15, Issue 16 (August-2 2026) – 268 articles

Cover Story (view full-size image): Gallium Nitride (GaN) HEMTs enable faster, denser, and more efficient power conversion than silicon or SiC, but their steep switching transients make an accurate, experimentally validated device model essential for electromagnetic compatibility (EMC) analysis. This work develops a functional electrothermal SPICE model of a commercial 650 V GaN HEMT, benchmarking the manufacturer's model against custom static (I-V) and dynamic (C-V) measurements. To close the observed gaps, a fully automated pipeline couples LTspice with a genetic algorithm to optimize the model over 25–100 °C, cutting the mean I-V error below 7% and the reverse-transfer capacitance error from 95.4% to 2.89%. The outcome is a compact, temperature-aware model ready for converter-level EMC validation. View this paper
 
  • 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:
16 pages, 1676 KB  
Article
Color Adversarial Patch Generation for Physical-Domain Palmprint Recognition Attacks
by Yue Liu, Qi Xiong, Lu Leng, Cheonshik Kim, Jun Miao and Lu Wang
Electronics 2026, 15(16), 3759; https://doi.org/10.3390/electronics15163759 - 21 Aug 2026
Viewed by 293
Abstract
Physical-domain adversarial attacks have been extensively studied in face recognition and object detection, yet the field of palmprint recognition remains largely unexplored. Existing methods generate grayscale patches constrained by the single-channel input of most palmprint models. When deployed on skin, these patches contrast [...] Read more.
Physical-domain adversarial attacks have been extensively studied in face recognition and object detection, yet the field of palmprint recognition remains largely unexplored. Existing methods generate grayscale patches constrained by the single-channel input of most palmprint models. When deployed on skin, these patches contrast sharply with the surrounding tissue and are readily noticeable to human observers, undermining the covertness required in practical attacks. To address this limitation, we propose a Color Adversarial Patch (CAP) generation algorithm that leverages style transfer principles to produce visually natural color patches while maintaining high attack success rates. The method initiates the patch with a style prior using a pre-trained Contrastive Arbitrary Style Transfer (CAST) model and jointly optimizes adversarial loss, style loss, and smoothness loss within a unified framework. A three-channel averaging strategy is adopted to ensure compatibility with single-channel recognition models during gradient backpropagation. Experiments on the Tongji palmprint dataset show that the generated color patches achieve average cosine similarity values above the decision threshold in physical-domain tests, with peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM) values significantly higher than those for their grayscale counterparts. Ablation studies validate the indispensable role of each loss component. CAP offers a practical balance between attack effectiveness and visual camouflage, demonstrating the feasibility of concealed physical-domain attacks on palmprint recognition systems. Full article
►▼ Show Figures

Figure 1

19 pages, 2404 KB  
Article
Image Design Feature Enhancement Method Based on Transformer and Multimodal Neural Networks
by Qianning Xu and Jun Wang
Electronics 2026, 15(16), 3758; https://doi.org/10.3390/electronics15163758 - 21 Aug 2026
Viewed by 304
Abstract
With the rapid development of artificial intelligence technology, image design feature enhancement plays a crucial role in computer vision, image processing, and multimedia applications. Traditional feature enhancement methods often have limitations when dealing with complex and changeable image data. Therefore, the study proposes [...] Read more.
With the rapid development of artificial intelligence technology, image design feature enhancement plays a crucial role in computer vision, image processing, and multimedia applications. Traditional feature enhancement methods often have limitations when dealing with complex and changeable image data. Therefore, the study proposes an innovative MN-T (Multimodal Neural Network enhanced by Transformer) strategy to overcome these challenges. The MN-T strategy combines the attention mechanism of Transformer with the cross-modal learning capabilities of multimodal neural networks to more accurately capture and enhance key features in images. The Transformer attention mechanism enables MN-T to efficiently process global and local information in images, while the cross-modal learning capability of multimodal neural networks further enhances its ability to understand and express image features. The results show that MN-T strategy has excellent performance in processing different image data. Compared with the existing image design feature enhancement methods, MN-T strategy has achieved significant improvement in the root mean square error (RMSE), mean absolute error (MAE), peak signal-to-noise ratio (PSNR), and other key performance indicators. The research results of this paper not only provide a new idea and method for image design feature enhancement methods, but also provide a new theoretical support and experimental basis for the research of related fields. Full article
►▼ Show Figures

Figure 1

45 pages, 10907 KB  
Article
O-Mamba: Task-Driven Orthogonal Projection Spatial–Spectral Mamba for Few-Shot HSI Classification
by Dan Yang, Jiale Chen, Junsuo Qu, Yanli Feng, Linquan Li and Xiaobo Jia
Electronics 2026, 15(16), 3757; https://doi.org/10.3390/electronics15163757 - 21 Aug 2026
Viewed by 332
Abstract
Hyperspectral image classification relies heavily on the effective modeling of spatial–spectral representations. Recent deep learning architectures, including Transformers and state space models (SSMs), have shown promise for HSI classification. However, under few-shot scenarios, they may suffer from optimization instability in early-stage feature reduction, [...] Read more.
Hyperspectral image classification relies heavily on the effective modeling of spatial–spectral representations. Recent deep learning architectures, including Transformers and state space models (SSMs), have shown promise for HSI classification. However, under few-shot scenarios, they may suffer from optimization instability in early-stage feature reduction, weakened local spatial–spectral correlations after direct sequence flattening, and attenuation of center-pixel spectral information caused by deep spatial aggregation. To mitigate these issues, we propose orthogonal projection spatial–spectral Mamba (O-Mamba), a lightweight architecture for few-shot HSI classification. First, we introduce a task-driven orthogonal projection module (TOPM) for learnable end-to-end spectral dimensionality reduction. In this module, orthogonal parameterization, supervised initialization, and an auxiliary loss jointly improve the stability of the projection process, reducing feature redundancy and mitigating representation collapse. Second, we design a 3D spatial–spectral Mamba encoder that employs 3D Convolutional Neural Networks (CNN) as local tokenizers to preserve local spatial–spectral structures and then uses Mamba to capture long-range sequence dependencies with linear complexity with respect to sequence length. Finally, to alleviate over-smoothing in the target-pixel representation, we propose a decoupled target–context fusion strategy. This mechanism separately preserves the original spectral signature of the center pixel and fuses it with high-level contextual features, which may improve the separability of spectrally similar classes. Extensive experiments on four benchmark datasets show that O-Mamba achieves competitive classification performance under the evaluated few-shot settings, while maintaining a relatively small model size and low computational cost compared with representative CNN, Transformer, and Mamba-based methods. Full article
►▼ Show Figures

Figure 1

30 pages, 23294 KB  
Article
Structure-Aware Design of a Partially Overlapped Segmented Transmitter with a Position-Dependent Excitation Strategy for Automotive Power-Seat Wireless Power Transfer Under Wide Misalignment
by Chang-Su Shin, Dong-Hee Kim and Geun Wan Koo
Electronics 2026, 15(16), 3756; https://doi.org/10.3390/electronics15163756 - 21 Aug 2026
Viewed by 284
Abstract
Wireless power transfer (WPT) can eliminate moving power-supply harnesses in automotive power-seat systems, but seat travel and nearby metallic structures cause substantial variations in magnetic coupling and electromagnetic loss. This paper proposes a structure-aware, partially overlapped segmented transmitter and evaluates two predefined excitation [...] Read more.
Wireless power transfer (WPT) can eliminate moving power-supply harnesses in automotive power-seat systems, but seat travel and nearby metallic structures cause substantial variations in magnetic coupling and electromagnetic loss. This paper proposes a structure-aware, partially overlapped segmented transmitter and evaluates two predefined excitation states according to receiver position. In the single-segment state, only the reference segment CP1 is energized; in the simultaneous dual-segment state, CP1 and the adjacent segment CP2 are energized together. Three-dimensional finite element method (FEM) simulations compare candidate transmitter structures and evaluate the electromagnetic influence of the aluminum lower rail, steel upper rail, and steel seat frame. The transmitter geometry is determined by considering mutual inductance, winding loss, structural eddy-current loss, and partial-overlap characteristics. A three-coil equivalent circuit clarifies the branch-current distribution, and a two-state switched-capacitor network accommodates the different equivalent transmitter impedances. A 100 W, 110 kHz prototype separately evaluates representative states at x = 0 and 80 mm; automatic position-based state switching is not implemented. At x = 0 mm, CP1-only excitation achieves 78.79% efficiency. At x = 80 mm, CP1 + CP2 excitation produces 32.13 V and 72.15%, compared with 18.78 V and 67.84% under CP1-only excitation, thereby satisfying the 30 V minimum output requirement. Full article
(This article belongs to the Special Issue Advances in Wireless Power Transfer)
►▼ Show Figures

Figure 1

25 pages, 1324 KB  
Review
Physical Constraints on Comfort in Virtual Reality Headsets: A Review of Thermal, Mechanical, and Anthropometric Factors
by Daniela Zamora Alviarez, Emma Drew and Redwan Alqasemi
Electronics 2026, 15(16), 3755; https://doi.org/10.3390/electronics15163755 - 21 Aug 2026
Viewed by 428
Abstract
Head-mounted displays (HMDs) create a direct physical interface with the head and face that can constrain comfort during sustained and repeated virtual reality use. This focused narrative review examines three interacting domains of HMD physical comfort: thermal conditions at the headset–skin interface, mechanical [...] Read more.
Head-mounted displays (HMDs) create a direct physical interface with the head and face that can constrain comfort during sustained and repeated virtual reality use. This focused narrative review examines three interacting domains of HMD physical comfort: thermal conditions at the headset–skin interface, mechanical loading from head-supported mass and contact forces, and anthropometric compatibility between headset geometry and user anatomy. Exploratory searching was followed during revision by structured searches of Google Scholar, Scopus, and Web of Science, tracker-based screening, source classification, evidence extraction, and evidence-limitation appraisal. The tracker-based review retained 41 sources, including 28 domain-focused and 13 context or framing sources. Two additional contextual references supplied during peer review were incorporated outside the completed tracker-based process, resulting in 43 sources cited in the final manuscript. Thermal comfort was influenced by microclimate conditions, exposure, activity, sealing, and internally generated heat. Mechanical comfort depended on mass, center-of-mass position, pressure, movement, posture, and task demands. Anthropometric evidence demonstrated substantial variation relevant to headset fit and alignment. The heterogeneous evidence does not support universal limits for headset mass, temperature, pressure, cervical angle, or interface geometry. Sustained HMD comfort therefore requires integrated evaluation across devices, users, tasks, and exposure durations. Full article
(This article belongs to the Special Issue Shaping Human-Centered Virtual Worlds)
►▼ Show Figures

Figure 1

28 pages, 36017 KB  
Article
Process-Aware Feature Modulation for Fine-Grained Connector Detection
by Ziang Wang, Xitian Tian, Yolanda Bolea, Antoni Grau, Edmundo Guerra, Yuntong Chen, Fan Yang and Liping Ma
Electronics 2026, 15(16), 3754; https://doi.org/10.3390/electronics15163754 - 21 Aug 2026
Viewed by 189
Abstract
While deep learning has significantly improved the accuracy of visual detection systems, the integration of process semantics with visual perception for industrial assembly tasks remains largely unexplored. This study aims to develop a high-accuracy cable connector detection framework that incorporates process knowledge to [...] Read more.
While deep learning has significantly improved the accuracy of visual detection systems, the integration of process semantics with visual perception for industrial assembly tasks remains largely unexplored. This study aims to develop a high-accuracy cable connector detection framework that incorporates process knowledge to enhance feature discrimination under varying industrial imaging conditions. To achieve this goal, we build an enhanced Fully Convolutional One-Stage (FCOS) detector with a ConvNeXt V2 backbone and introduce a Process Feature Linear Modulation (PFNM) module. The proposed module adaptively modulates visual features using encoded process semantics, enabling the detector to align visual perception with assembly logic. Experiments conducted on an industrial connector dataset demonstrate that the proposed method achieves an mAP of 84.7%, outperforming representative state-of-the-art detectors including YOLOv11, RT-DETR, and DINO while maintaining an inference speed of 17.6 FPS. Ablation studies further show that each component contributes to progressive performance improvement, and the complete framework achieves a 5.5% AP gain over the ResNet-50 baseline. These results indicate that integrating process knowledge with visual feature learning effectively improves feature discrimination and provides a promising paradigm for process-aware perception in intelligent manufacturing. Full article
(This article belongs to the Special Issue Artificial Intelligence for Smart Mobility and Industrial Automation)
►▼ Show Figures

Figure 1

26 pages, 7899 KB  
Article
LTFANet: A Lightweight Time–Frequency Attention Network for Multi-Fault Diagnosis of Motor Bearings on an Edge Platform
by Maosen Chen and Xiaotian Zhang
Electronics 2026, 15(16), 3753; https://doi.org/10.3390/electronics15163753 - 21 Aug 2026
Viewed by 296
Abstract
Rolling bearings are critical components in rotating machinery, and their failures may cause unexpected downtime and safety risks. However, conventional deep diagnostic models are often difficult to deploy on resource-constrained edge devices because of their high computational cost and memory consumption. This paper [...] Read more.
Rolling bearings are critical components in rotating machinery, and their failures may cause unexpected downtime and safety risks. However, conventional deep diagnostic models are often difficult to deploy on resource-constrained edge devices because of their high computational cost and memory consumption. This paper proposes a lightweight time–frequency attention network (LTFANet) for multi-fault diagnosis of rolling bearings on an edge platform. The proposed model directly processes one-dimensional vibration signals and employs multi-scale depthwise separable convolutions to capture impact and periodic fault features with low computational complexity. A lightweight frequency branch is introduced to enhance fault-frequency representation, while an efficient channel attention module adaptively emphasizes fault-sensitive features. Moreover, a severity-aware multi-task extension is introduced to jointly identify the fault location and degradation level. To further improve edge inference efficiency, knowledge distillation, structured pruning, and TensorRT-based acceleration are integrated into the deployment pipeline. Experiments on CWRU-10 and Paderborn achieve 97.20% and 90.25% accuracy, respectively, while LTFANet contains only 0.020 M parameters and requires 0.610 M FLOPs. Knowledge distillation increases the CWRU-10 accuracy to 98.50%, and the severity-aware extension achieves 95.18% severity accuracy. On the NVIDIA Jetson Nano, the pruned TensorRT FP16 implementation achieves an average inference latency of 0.520 ms and a throughput of 1923.08 samples/s. The framework provides an effective solution for real-time and low-cost bearing condition monitoring at the edge. Full article
►▼ Show Figures

Figure 1

22 pages, 7271 KB  
Article
Resilience-Oriented Multi-Objective Optimal Placement of TCSC Based on Comprehensive Line Vulnerability Assessment
by Lixia Zhang, Ning Wang, Wei Kang, Bowen Zhu and Yunda Li
Electronics 2026, 15(16), 3752; https://doi.org/10.3390/electronics15163752 - 21 Aug 2026
Viewed by 174
Abstract
Modern power systems are increasingly exposed to uncertainties and face rising demands for operational resilience. To address this challenge, this paper investigates the optimal placement of thyristor-controlled series compensation (TCSC) devices within flexible AC transmission systems (FACTS). A comprehensive vulnerability evaluation index is [...] Read more.
Modern power systems are increasingly exposed to uncertainties and face rising demands for operational resilience. To address this challenge, this paper investigates the optimal placement of thyristor-controlled series compensation (TCSC) devices within flexible AC transmission systems (FACTS). A comprehensive vulnerability evaluation index is developed by integrating network structure, load impact, and branch disconnection factors, enabling a holistic identification of vulnerable transmission links. Subsequently, a multi-objective TCSC optimization model is formulated to simultaneously minimize the system-wide comprehensive vulnerability index and the total investment cost. To solve this model, an improved multi-objective particle swarm optimization (MOPSO) algorithm is devised, incorporating chaotic initialization and adaptive inertia weight adjustment to enhance both global exploration and local exploitation capabilities. The proposed method is validated using the IEEE 39-bus and IEEE 118-bus test systems. The results demonstrate that the optimized placement significantly reduces system vulnerability, maintains a favorable economic balance and improves the system security margin. Furthermore, uncertainty tests involving load variations, line parameter perturbations, and wind power fluctuations, as well as malicious attacks, confirm the robustness of the proposed placement strategy. This work provides a practical and effective framework for resilience-oriented TCSC planning, contributing to mitigating cascading failure risks and enhancing power system security. Full article
►▼ Show Figures

Figure 1

32 pages, 708 KB  
Article
Decoupled Decision-Stage Awareness for Conversational Recommendation with Large Language Model Agents in Information Analysis
by Chaoyang Li, Yiwei Lu, Bo Huang, Ruopeng Yang, Yongqi Shi, Zhaoyang Gu, Tianjin Ni and Yongqi Wen
Electronics 2026, 15(16), 3751; https://doi.org/10.3390/electronics15163751 - 21 Aug 2026
Viewed by 242
Abstract
Information analysis recommendation differs from conversational recommender systems (CRS) because relevance changes with the decision phase. The same event may support observation, interpretation, option selection, or action feedback, yet most large language model (LLM)-agent CRS represent dialogue state as intent and preference. This [...] Read more.
Information analysis recommendation differs from conversational recommender systems (CRS) because relevance changes with the decision phase. The same event may support observation, interpretation, option selection, or action feedback, yet most large language model (LLM)-agent CRS represent dialogue state as intent and preference. This study examines whether explicit decision-stage awareness improves recommendation and whether it can be added independently of the LLM backbone. We propose Stage-Aware Conversational Recommender System (SA-CRS), a plug-in layer guided by the Observe–Orient–Decide–Act cycle. It decouples stage detection from LLM reasoning and uses detected stages to guide dialogue strategy and candidate re-ranking. We evaluate SA-CRS on an information analysis recommendation dataset from event-structured reports, using multi-turn simulated dialogues and four LLM backbones. Oracle stage injection improves Hit@5 by 3.0 percentage points (pp), showing that decision stage provides a signal beyond topic matching. With a prompt-based detector, SA-CRS improves Hit@5 by 9.0 pp on a strong backbone; with an independent Bidirectional Encoder Representations from Transformers (BERT) detector and probabilistic re-ranking, gains range from 6.5 to 15.5 pp. Negative controls with uniform or random stage signals fail to reproduce the improvements and may reduce efficiency. The results suggest that, within the evaluated single-domain information-analysis setting, decoupled decision-stage awareness is practical for decision-intensive CRS. Full article
(This article belongs to the Section Artificial Intelligence)
►▼ Show Figures

Figure 1

19 pages, 8282 KB  
Article
Power Integrity Analysis and Evaluation of a Dual-Interposer HBM Structure
by Wenlong Li, Zhuangchao Zhan, Jingdong Li, Yiwei Wang, Yuxin Liang, Jingran Zhang and Daoguo Yang
Electronics 2026, 15(16), 3750; https://doi.org/10.3390/electronics15163750 - 21 Aug 2026
Viewed by 425
Abstract
High-bandwidth memory (HBM) faces critical power integrity challenges in high-stack configurations due to elongated power delivery paths and increased parasitic inductance. This paper proposes a dual-interposer HBM architecture with an interposer–HBM stack–interposer configuration, integrating an additional top interposer embedded with chip capacitors. This [...] Read more.
High-bandwidth memory (HBM) faces critical power integrity challenges in high-stack configurations due to elongated power delivery paths and increased parasitic inductance. This paper proposes a dual-interposer HBM architecture with an interposer–HBM stack–interposer configuration, integrating an additional top interposer embedded with chip capacitors. This topology redesigns the HBM’s power distribution network, reducing PDN impedance, and this technology enables bidirectional vertical power supply to DRAM chips during moments when they require current. The PDN impedance is systematically compared with a conventional trench-capacitance-enhanced structure (Structure A) and a deep-trench-capacitance-enhanced structure (Structure B). Results show that at 0.1–11.2 GHz, the proposed structure reduces peak PDN impedance by 66.41% and 65.7% versus Structures A and B, respectively, and decreases the loop inductance of the top-layer DRAM chip by 66.71%. The top interposer’s redistribution layer forms a parallel-plate capacitor complementing the embedded chip capacitors, achieving wideband impedance suppression. Without modifying existing protocols, this architecture provides a system-level PDN optimization strategy for high-stack HBM, offering quantitative insights for capacitor selection and layout design. Full article
►▼ Show Figures

Figure 1

30 pages, 2484 KB  
Article
AraCTI-NER: A Dataset and Benchmark for Arabic Cyber Threat Intelligence Named Entity Recognition
by Joud Alghamdi and Souham Meshoul
Electronics 2026, 15(16), 3749; https://doi.org/10.3390/electronics15163749 - 21 Aug 2026
Viewed by 431
Abstract
Automated extraction of structured threat information from unstructured cyber threat intelligence (CTI) underpins modern security operations, yet the supporting machine learning resources are almost exclusively English: no annotated Arabic CTI named entity recognition (NER) corpus has been published. We introduce AraCTI-NER, a dataset [...] Read more.
Automated extraction of structured threat information from unstructured cyber threat intelligence (CTI) underpins modern security operations, yet the supporting machine learning resources are almost exclusively English: no annotated Arabic CTI named entity recognition (NER) corpus has been published. We introduce AraCTI-NER, a dataset of 10,312 token-level annotated samples (275,530 tokens; 42,360 entity spans) over eight STIX-inspired entity types, built by an LLM-assisted pipeline seeded with authentic Arabic cybersecurity articles, structurally validated and rebalanced through targeted generation. We benchmark seven encoders from three families (Arabic-specialized, English cybersecurity-adapted, and multilingual) over three seeds under strict entity-level metrics, and release a 408-sentence expert-audited test subset (ATS-gold) whose reliability is quantified by a second independent expert validation (inter-annotator agreement 0.878 entity-level F1). XLM-RoBERTa Large attains the best mean F1 (0.7603; 0.7674 on ATS-gold), with AraBERTv2 close behind (0.7491), while both English-only cybersecurity encoders fall to ≈0.63, a separation that holds across every seed and survives expert correction, with the ≈3-point F1 decrease from ATS-silver to ATS-gold concentrated in Vulnerability and TTP. On 350 doubly annotated sentences from authentic Arabic cyber-incident news, a shift in both provenance and register, the strongest model reaches F1 = 0.5429 against an inter-annotator F1 of 0.616. AraCTI-NER establishes the first reproducible baseline for Arabic CTI NER and identifies domain-adaptive Arabic cybersecurity pre-training as the highest-value next step. Full article
(This article belongs to the Special Issue AI in Cybersecurity, 3rd Edition)
►▼ Show Figures

Figure 1

20 pages, 1416 KB  
Article
A Lightweight CNN–TCN–Attention Framework for Low-Latency Pre-Fall Transition and Fall Recognition on Resource-Constrained Edge Devices
by Woojin Cho, Seok-Oh Bang, Hyun-Seok Choi, Ki-Tae Kwon, Sang-Wuk Shin, Jin-Sung Roh, Jong-Min Lim and Hyun Mok Park
Electronics 2026, 15(16), 3748; https://doi.org/10.3390/electronics15163748 - 21 Aug 2026
Viewed by 315
Abstract
Falls are a major cause of severe injury and mortality among older adults, requiring rapid state recognition and alerts in healthcare and caregiving environments. However, many existing fall-recognition approaches primarily focus on post-fall detection or assume access to cloud/GPU computing, leaving limited evidence [...] Read more.
Falls are a major cause of severe injury and mortality among older adults, requiring rapid state recognition and alerts in healthcare and caregiving environments. However, many existing fall-recognition approaches primarily focus on post-fall detection or assume access to cloud/GPU computing, leaving limited evidence for short-term pre-fall recognition on low-end CPU-based edge devices. This study proposes a lightweight CNN–TCN–Attention framework for recognizing fall and pre-fall states on resource-constrained edge devices. Using MediaPipe, three-dimensional coordinates and visibility scores of 33 human body landmarks are extracted from input videos. Raw RGB frames are not directly used by the classifier, thereby reducing the processing of visually identifiable information. A total of 2300 fall-related videos from the AI-Hub dataset were reorganized into three classes: normal, pre-fall, and fall. The pre-fall onset was defined as the point at which the downward vertical velocity of the nose landmark exceeded a dataset-specific threshold. The model uses a CNN to extract frame-level skeletal patterns, a temporal convolutional network (TCN) to learn temporal changes in posture, and an attention module to aggregate temporal features. It was deployed on a Raspberry Pi Zero 2 W using ONNX Runtime. The proposed model achieved 94.71% accuracy and a macro F1-score of 93.63%, with an average state-decision latency of 325.88 ms/decision. It improved accuracy by 2.83 percentage points over the lightweight CNN–LSTM–Attention baseline while maintaining comparable latency and achieved approximately 1.75× faster processing than the MobileNetV3–TCN–Attention model. Full article
(This article belongs to the Special Issue Edge AI for Biomedical Applications: Innovations in Sensing, Computing and Security)
►▼ Show Figures

Figure 1

21 pages, 28322 KB  
Article
Design and Multi-Modal Locomotion Control of a Compound Leg-Wheel Robot
by Meng Gao, Changcheng Wang and Fuqun Zhao
Electronics 2026, 15(16), 3747; https://doi.org/10.3390/electronics15163747 - 21 Aug 2026
Viewed by 271
Abstract
To harness the terrain adaptability of legged systems and the high-speed efficiency of wheeled platforms, this paper presents the design of a novel leg-wheel hybrid mobile platform intended for enhanced obstacle negotiation. The proposed system comprises six identical leg-wheel modules, each integrating a [...] Read more.
To harness the terrain adaptability of legged systems and the high-speed efficiency of wheeled platforms, this paper presents the design of a novel leg-wheel hybrid mobile platform intended for enhanced obstacle negotiation. The proposed system comprises six identical leg-wheel modules, each integrating a closed-chain mechanical leg with an independently driven wheel mounted at the tip. The platform operates in two distinct modalities: a pure leg mode and a leg-wheel composite mode. In the leg mode, locomotion is driven by crank motors, providing exceptional mobility across uneven terrain. Conversely, the composite mode utilizes both crank and pitch link motors to facilitate obstacle surmounting, while hub motors ensure high-efficiency propulsion. Comprehensive gait planning for both modes is conducted, accompanied by a detailed analysis of the platform’s obstacle-negotiation capabilities. Kinematic analysis and gait simulations validate the platform’s superior mobility and efficient obstacle-crossing performance under the dual-mode strategy. Prototype experiments further confirm the feasibility of the mechanical design, demonstrating significant proficiency in traversing obstacles. This research contributes a novel design exploration by serially combining a closed-chain leg mechanism with an actuated wheel. The adopted closed-chain architecture offers the distinct advantages of high foot clearance and a single degree of freedom (DoF), which are critical for achieving reliable and effective obstacle-surmounting capabilities. Full article
►▼ Show Figures

Figure 1

20 pages, 2790 KB  
Article
Individual and Coordinated Mixed-Integer Linear Programming Dispatch of an Industrial Photovoltaic–Battery Prosumer Community on the Bulgarian Day-Ahead Market
by Antouan Hristov Anguelov, Roumen Trifonov and Galya Pavlova
Electronics 2026, 15(16), 3746; https://doi.org/10.3390/electronics15163746 - 21 Aug 2026
Viewed by 339
Abstract
Industrial consumers increasingly operate photovoltaic (PV) generation and battery energy storage systems (BESS) against volatile day-ahead electricity prices. This paper presents a simulation environment for a community of three heterogeneous industrial prosumers, anchored in twelve months of day-ahead prices from the Bulgarian Independent [...] Read more.
Industrial consumers increasingly operate photovoltaic (PV) generation and battery energy storage systems (BESS) against volatile day-ahead electricity prices. This paper presents a simulation environment for a community of three heterogeneous industrial prosumers, anchored in twelve months of day-ahead prices from the Bulgarian Independent Energy Exchange (IBEX), commercial hardware envelopes, and the terms of a market offtake contract that indexes remuneration to the day-ahead price, passes negative prices through to the producer, and mandates curtailment in strongly negative periods. A mixed-integer linear programming (MILP) dispatch model with binary charge/discharge modes and endogenous PV curtailment is solved daily in independent and coordinated regimes, under cost-only and capacity-aware objectives, for Sofia and Stara Zagora. On an energy-only basis, before capacity charges, storage turns the community from a net payer (39.1 kEUR/yr grid-only; 13.1 kEUR/yr with PV) into a net earner (8.64 kEUR/yr independent; 10.12 kEUR/yr coordinated), and the battery retrofit more than doubles the merchant plant’s net revenue. Cost-optimal dispatch synchronizes charging and raises the coincident grid peak to 241 kW; a capacity-aware objective cuts it to 133–140 kW (−42–45%) for under 0.9 kEUR/yr foregone income; under the two tested capacity-charge conventions, only the capacity-aware schedule remains net-positive after the modeled charge. The coordination gain grows with the import price adder and remains positive across 15–45 EUR/MWh. Full article
►▼ Show Figures

Figure 1

21 pages, 907 KB  
Article
Rule Graph-Based Low-Code Control for Renewable Energy and Storage Stations
by Jiacheng Li, Menghan Xiao, Chang Ye, Xun Xu and Yuwei Gui
Electronics 2026, 15(16), 3745; https://doi.org/10.3390/electronics15163745 - 21 Aug 2026
Viewed by 266
Abstract
Renewable energy and energy storage stations require frequent updates of monitoring and control logic across heterogeneous devices and changing operating strategies. This paper proposes a rule graph-based reference architecture that combines low-code logic configuration, graph–model semantic binding, and microservice-oriented functional decomposition. A component [...] Read more.
Renewable energy and energy storage stations require frequent updates of monitoring and control logic across heterogeneous devices and changing operating strategies. This paper proposes a rule graph-based reference architecture that combines low-code logic configuration, graph–model semantic binding, and microservice-oriented functional decomposition. A component status matrix separates the target architecture from the implemented subset. The runnable subset comprises a minimal FastAPI backend, REST/WebSocket telemetry interfaces, an in-process queue, and stateful rule evaluators; gateway, authentication, external message bus, time-series database, visual editor, and industrial protocol services remain design-level elements. Beyond the original single-rule example, a priority-ordered multi-device rule is implemented for cooperative BESS dispatch, communication/topology blocking, low-SOC protection, frequency-based load shedding, backup request, and five-sample recovery release. Existing local network benchmarks are complemented by a 600-step software-in-the-loop trace with scripted telemetry fluctuations and communication quality faults and by 500 in-process ASGI timing samples at each of the four point levels. The trace produced no safety dispatch or protected device violations. P99 application path latency ranged from 1.1962 to 5.5287 ms, but one 75.3065 ms outlier exceeded a 50 ms reference deadline, demonstrating that the Windows/FastAPI path is not deterministic. No industrial controller, hardware-in-the-loop facility, field data, or engineer usability study was used. Accordingly, the paper makes no claim of industrial real-time readiness or measured development effort reduction. Full article
►▼ Show Figures

Figure 1

30 pages, 3526 KB  
Article
Optimal Operation Strategy of Power Grids Integrated with High-Capacity Grid-Supporting Storage Devices Based on Trajectory Sensitivity Analysis and Improved Chaotic PSO Algorithm
by Yiqun Kang, Huizhen Huang, Bingyang Feng, Yuxuan Hu and Qiujie Wang
Electronics 2026, 15(16), 3744; https://doi.org/10.3390/electronics15163744 - 21 Aug 2026
Viewed by 302
Abstract
High penetration levels of renewable energy and power electronic apparatus create prominent obstacles for novel power grids, which mainly manifested as inadequate system inertia and a deteriorated stability margin. To overcome such drawbacks, this research develops an operational control method to maintain safe [...] Read more.
High penetration levels of renewable energy and power electronic apparatus create prominent obstacles for novel power grids, which mainly manifested as inadequate system inertia and a deteriorated stability margin. To overcome such drawbacks, this research develops an operational control method to maintain safe and steady grid operation with large-capacity grid-forming energy storage connected to the system. This paper first builds a dynamic voltage model covering grid-forming energy storage, distributed renewable generators, and distribution network frameworks. Then it explores how different control parameter settings of grid-forming storage affect dynamic voltage regulation capabilities under distinct R-L ratio scenarios. Since the correlation between energy storage control variables and voltage regulation features is highly nonlinear and complicated, trajectory sensitivity analysis is adopted to linearize these coupling constraints, which are further embedded into the power system security operation mathematical model. A chaotic particle swarm optimization (PSO) algorithm is used to solve the constructed optimization model. Simulation tests on a modified IEEE 33-bus test system ultimately prove that the proposed method is reliable and practically applicable. Simulation results on the modified IEEE 33-bus test system demonstrate that the proposed strategy restricts grid voltage fluctuation rate to only 2.41%, raises renewable energy accommodation rate up to 98.4%, and achieves a 30.6% reduction in overall system operation cost compared to traditional energy storage configuration schemes, which fully verifies the outstanding effectiveness and practical engineering feasibility of the proposed method. Full article
(This article belongs to the Special Issue Advanced Power Electronics and Sustainable Energy Systems: Recent Developments, Challenges and Future Perspectives)
►▼ Show Figures

Figure 1

18 pages, 1607 KB  
Article
Electromagnetic Inverse Scattering of Buried Conductors in Half-Space Using TE Waves and Deep Convolutional Neural Networks
by Po-Hsiang Chen, Chien-Ching Chiu, Hsin-Chien Wen and Hao Jiang
Electronics 2026, 15(16), 3743; https://doi.org/10.3390/electronics15163743 - 20 Aug 2026
Viewed by 377
Abstract
Electromagnetic Inverse Scattering (EMIS) is a key technology for reconstructing the geometry of buried conductors based on measured scattered fields. This research employs a Deep Convolutional Neural Network (DCNN) to address the intrinsic nonlinear characteristics in reconstructing perfect conductors in a half-space using [...] Read more.
Electromagnetic Inverse Scattering (EMIS) is a key technology for reconstructing the geometry of buried conductors based on measured scattered fields. This research employs a Deep Convolutional Neural Network (DCNN) to address the intrinsic nonlinear characteristics in reconstructing perfect conductors in a half-space using transverse electric (TE) waves. First, TE electromagnetic waves are transmitted to illuminate the conductor, and the corresponding scattered fields are collected. Subsequently, the scattered field is fed into the DCNN to reconstruct the precise shape of the conductor. Considering that real-world measurements may contain noise, we add 5%, 10%, and 20% noise levels in the simulation. Numerical results for various conductor geometries, including shield, oval, arrow, heart, and four-petal shapes, demonstrate the reconstruction capability of the proposed method under the investigated Gaussian noise levels. These results demonstrate the feasibility of applying a DCNN to the reconstruction of buried conductors under TE-wave illumination. Full article
(This article belongs to the Special Issue Trends and Perspectives in Microwave Imaging and Applications)
►▼ Show Figures

Figure 1

24 pages, 14861 KB  
Article
High-Precision Detection of Leather Creases via Dynamic, Cross-Calibrated, and Edge-Enhanced YOLOv8n-Pose
by Ran An, Gongchang Ren, Jiangong Sun, Yuan Huan, Jiaxuan Yang, Kaijie Zhang and Yuanbiao Wang
Electronics 2026, 15(16), 3742; https://doi.org/10.3390/electronics15163742 - 20 Aug 2026
Viewed by 309
Abstract
Residual creases generated during the leather spreading process exhibit highly variable morphologies and irregular feature distributions, causing significant challenges for feature extraction and leading to low localization accuracy. To address these issues, this paper proposes the Dynamic, Cross-Calibrated, and Edge-Enhanced YOLOv8n-Pose (DCE-YOLOv8n-Pose) algorithm. [...] Read more.
Residual creases generated during the leather spreading process exhibit highly variable morphologies and irregular feature distributions, causing significant challenges for feature extraction and leading to low localization accuracy. To address these issues, this paper proposes the Dynamic, Cross-Calibrated, and Edge-Enhanced YOLOv8n-Pose (DCE-YOLOv8n-Pose) algorithm. Instead of providing regional approximations, this framework outputs precise spatial coordinates for robotic grasping by integrating three synergistic components in a progressive network flow. First, dynamic snake convolution adaptively perceives the continuous geometric features of elongated creases; subsequently, an efficient multi-scale attention mechanism provides cross-dimensional weight calibration to suppress highly homochromatic background interference and correct spatial misalignments; finally, an edge-enhanced content-aware reassembly of features module preserves high-frequency gradients and prevents feature fracturing during multi-scale fusion. For comprehensive evaluation, a dataset comprising 700 original laboratory images was constructed. To prevent data leakage, the dataset was partitioned into training and validation sets based on individual leather specimens, ensuring that images of the same leather piece do not appear in both sets. Additionally, an independent test set of 500 images collected from an actual processing plant was designed for industrial validation. Experimental results indicate that, at an Intersection over Union (IoU) threshold of 0.5, the DCE-YOLOv8n-Pose model achieves a bounding box mean average precision (mAP@0.5) of 91.8% and a keypoint mAP@0.5 of 85.1%, with a keypoint precision of 87.9%. The computational load is maintained at 9.2 GFLOPs, alongside an inference speed of 114.3 FPS. Furthermore, consistent convergence across four independent training runs substantiates the model’s reliability in reducing missed detection rates and localization deviations. In conclusion, the proposed algorithm demonstrates practical applicability for the visual guidance of automated leather spreading equipment by balancing detection precision and inference speed, thereby offering an effective coordinate reference for subsequent robotic stretching operations. Full article
(This article belongs to the Section Artificial Intelligence)
►▼ Show Figures

Figure 1

41 pages, 5988 KB  
Article
Pump Noise Suppression in Continuous-Wave Mud Pulse Telemetry via Dual-Sensor Joint Delay and Amplitude Compensation
by Yang Zhao, Wanlu Jiang, Chengpeng Yu, Zhenbao Li and Yongyong Li
Electronics 2026, 15(16), 3741; https://doi.org/10.3390/electronics15163741 - 20 Aug 2026
Viewed by 261
Abstract
Continuous-wave mud pulse telemetry offers high spectral efficiency and transmission rates, making it an important technology for high-speed information transmission under complex well conditions. However, surface-received signals are highly susceptible to periodic pressure pulsations generated by mud pumps, which degrade phase extraction and [...] Read more.
Continuous-wave mud pulse telemetry offers high spectral efficiency and transmission rates, making it an important technology for high-speed information transmission under complex well conditions. However, surface-received signals are highly susceptible to periodic pressure pulsations generated by mud pumps, which degrade phase extraction and symbol decision performance. Dual-pressure-sensor delayed differential processing can exploit the correlated propagation characteristics of pump noise between two measurement locations to suppress its correlated components; however, its performance depends on accurately matching the propagation delay and amplitude compensation coefficient. To specifically address the dynamic variation in the pump noise propagation relationship between two measurement locations under actual operating conditions, a joint delay–amplitude compensation method is developed, in which pump noise suppression is formulated as the joint estimation of the signal propagation delay and amplitude compensation coefficient. Built upon LMS-based time delay estimation, the proposed method employs an enhanced time-varying step-size LMS time delay estimation algorithm (HTVSS-LMSTDE) to improve dynamic retracking capability following changes in propagation delay. A sliding-window weighted least-squares method (SWLS) is further introduced to estimate the amplitude compensation coefficient and correct differential mismatch caused by variations in the amplitude transfer ratio. With non-pump interference modeled as additive white Gaussian noise independent of the telemetry signal and pump noise, simulation results demonstrate that, when the propagation delay and amplitude transfer ratio vary simultaneously, the proposed method yields delay estimates and amplitude compensation coefficients close to their theoretically optimal values. Field wellbore tests further verify that the proposed method effectively attenuates low-frequency pump noise interference in continuous-wave mud pulse telemetry signals while preserving the BPSK-modulated information. Full article
►▼ Show Figures

Figure 1

36 pages, 10431 KB  
Article
A Simulation-Driven Hierarchical Stackelberg-DMPC Framework for UAV Swarm Interception
by Zhao Sun and Guangjun He
Electronics 2026, 15(16), 3740; https://doi.org/10.3390/electronics15163740 - 20 Aug 2026
Viewed by 265
Abstract
This paper proposes a simulation-driven hierarchical Stackelberg–distributed model predictive control framework (SHS-DMPC) for intercepting multi-wave UAV swarm attacks under limited defensive resources. The interaction between the defender and the attacker is modeled as a Stackelberg leader–follower game. At the strategy layer, a finite-response [...] Read more.
This paper proposes a simulation-driven hierarchical Stackelberg–distributed model predictive control framework (SHS-DMPC) for intercepting multi-wave UAV swarm attacks under limited defensive resources. The interaction between the defender and the attacker is modeled as a Stackelberg leader–follower game. At the strategy layer, a finite-response approximation of Stackelberg decision making is constructed under incomplete information: the attacker’s response type is inferred online from swarm-level motion features, and candidate defender strategies are subsequently evaluated through state-dependent short-horizon rollout simulations. This formulation avoids requiring explicit knowledge of the attacker’s utility function while retaining anticipatory leader–follower strategy evaluation. At the task-allocation layer, target value, threat level, spatial bias, and a reassignment penalty are incorporated into the allocation cost to translate the selected defense strategy into dynamic defender–attacker assignments. At the control layer, each defending UAV solves a local DMPC problem to generate continuous control inputs while satisfying kinematic, inter-UAV separation, and airspace-boundary constraints. Simulation results show that SHS-DMPC achieve a higher interception success rate, a lower value-weighted target loss rate, and fewer minimum-separation violations than the comparison methods under multi-wave heterogeneous attack scenarios, demonstrating the benefits of closed-loop coupling among response inference, strategy-conditioned allocation, and constraint-aware distributed trajectory optimization. Full article
►▼ Show Figures

Figure 1

22 pages, 628 KB  
Article
A Formal Framework of Architectural Intent Collapse for Tool-Level Attacks on LLM Agents
by Zhaowen Feng, Zhenhui Liu, Mingjun Ma, Dongran Zhuang and Jie Gao
Electronics 2026, 15(16), 3739; https://doi.org/10.3390/electronics15163739 - 20 Aug 2026
Viewed by 300
Abstract
Tool-level attacks on Large Language Model (LLM) agents—poisoned tool descriptions, prompt injection, and capability misrepresentation—are universally effective, yet no existing defense provides comprehensive protection. We propose Architectural Intent Collapse (AIC), a formal framework capturing the systematic loss of communicative intent when text from [...] Read more.
Tool-level attacks on Large Language Model (LLM) agents—poisoned tool descriptions, prompt injection, and capability misrepresentation—are universally effective, yet no existing defense provides comprehensive protection. We propose Architectural Intent Collapse (AIC), a formal framework capturing the systematic loss of communicative intent when text from heterogeneous sources is flattened into a single context window. Grounded as a novel instantiation of the Confused Deputy Problem, AIC reveals that the missing boundary is not permission but intent: the architecture cannot distinguish descriptive statements from prescriptive commands. We formalize AIC via an architectural collapse operator, introduce Intent Separation Degree (ISD) as a measurable metric, and develop a mechanism-based taxonomy of five intent-disguise attack types, including two previously undescribed (Conditional Latency and Inference Inducement). Experiments across 25 framework–model combinations (employing GPT-4o, Claude-4-Sonnet, Gemini-2.5-Pro, DeepSeek-V3, and Qwen3-32B as LLM backends) confirm that ISD degrades with description verbosity, strongly predicts defense effectiveness (r=−0.97), and is uniformly low across all current frameworks. Three root-cause defense principles are derived; one retains substantial protection against adaptive attackers. This research is useful for agent framework designers, security practitioners, and researchers seeking a principled understanding of why tool-level attacks succeed and how architectural defenses can address their root cause. Full article
(This article belongs to the Special Issue AI in Cybersecurity, 3rd Edition)
►▼ Show Figures

Figure 1

24 pages, 5633 KB  
Article
Design and Analysis of a CNN-Transformer-Based Differential Distinguisher for ARX Ciphers
by Lei Zhang, Yuxuan Wu, Jiao Lei, Quanrun Lv, Chaoen Xiao, Jianxin Wang, Ding Ding and Ruipeng Hong
Electronics 2026, 15(16), 3738; https://doi.org/10.3390/electronics15163738 - 20 Aug 2026
Viewed by 295
Abstract
Neural distinguishers are commonly developed and evaluated using cipher-specific data representations and model configurations. This paper presents a common CNN-Transformer differential-distinguisher architecture for the evaluated ARX cipher SPECK and the ARX-related addition–shift–XOR ciphers TEA and XTEA. The framework combines a supervised front-end purification [...] Read more.
Neural distinguishers are commonly developed and evaluated using cipher-specific data representations and model configurations. This paper presents a common CNN-Transformer differential-distinguisher architecture for the evaluated ARX cipher SPECK and the ARX-related addition–shift–XOR ciphers TEA and XTEA. The framework combines a supervised front-end purification gate, multi-scale convolutional feature extraction, multiple-ciphertext-pair representation, and self-attention-based aggregation. The purification gate is trained only on the training split and is treated as the first stage of an end-to-end classifier; samples rejected by the gate are not removed from the test-set evaluation. The same backbone architecture is trained separately for each evaluated cipher and round configuration. The resulting classifiers achieve accuracies of 98.64% for 7-round SPECK32/64 and 90.76% for 10-round TEA, and retain distinguishing capability for 5-cycle XTEA. These results demonstrate applicability across the evaluated word-oriented ciphers; they do not constitute an end-to-end key-recovery attack. Full article
(This article belongs to the Special Issue State of the Art in Cryptography Theory and Techniques)
►▼ Show Figures

Figure 1

21 pages, 2726 KB  
Article
A Privacy–Utility Balanced Trajectory Protection Scheme via Adaptive Perturbation of Markov Transition Matrices
by Zhihong Zhang, Yu Fu, Yaxuan Zhao, Taotao Liu and Yishuai An
Electronics 2026, 15(16), 3737; https://doi.org/10.3390/electronics15163737 - 20 Aug 2026
Viewed by 266
Abstract
The widespread adoption of Location-Based Services (LBSs) has significantly enhanced daily convenience, yet the upload and storage of user trajectory data pose substantial privacy leakage risks. Addressing the limitations of existing privacy protection methods in achieving personalized adaptation and balancing privacy with utility, [...] Read more.
The widespread adoption of Location-Based Services (LBSs) has significantly enhanced daily convenience, yet the upload and storage of user trajectory data pose substantial privacy leakage risks. Addressing the limitations of existing privacy protection methods in achieving personalized adaptation and balancing privacy with utility, this paper proposes a personalized privacy protection strategy for location trajectories based on weighted Kullback–Leibler (KL) divergence. The approach first employs a Markov transition matrix to model user movement patterns, utilizes quadtree-based dynamic grid partitioning for adaptive encoding of the state space, and introduces sensitivity scores weighted by dwell duration and visit frequency to identify critical privacy-sensitive points. It then develops an exponential decay perturbation mechanism combining regularization parameters and distortion thresholds to preserve trajectory spatial usability while protecting sensitive transitions. By quantifying privacy leakage through weighted KL divergence and measuring data utility via distortion metrics, a linearly weighted composite index is constructed, enabling personalized parameter optimization via grid search. Experimental results on the real-world Geolife dataset demonstrate that compared to three differential privacy baselines, this method reduces privacy leakage (measured by weighted KL divergence), improves POI Recall rates, and decreases average geographic errors. Paired t-tests confirm that all improvements are statistically significant (p < 0.001) with large effect sizes, validating its effectiveness and superiority in balancing privacy protection and data usability. Full article
(This article belongs to the Section Computer Science & Engineering)
►▼ Show Figures

Figure 1

42 pages, 1290 KB  
Systematic Review
CNN-Based Spatiotemporal Feature Extraction for Video Processing: A Systematic Review
by Adrian E. Lopez, Hugo Jimenez-Hernandez, Ana-Marcela Herrera-Navarro, Daniel Canton-Enriquez, Rodrigo Hernandez-Alvarado, Jorge-Luis Perez-Ramos, Arely-Guadalupe Morales-Hernandez and Julio-Cesar Mendez-Avila
Electronics 2026, 15(16), 3736; https://doi.org/10.3390/electronics15163736 - 20 Aug 2026
Viewed by 506
Abstract
The extraction of spatiotemporal features from video sequences allows for the recognition of actions and the analysis of behaviors in video, making it a key challenge in automated video processing. The literature shows widespread use of deep learning approaches, specifically convolutional neural networks [...] Read more.
The extraction of spatiotemporal features from video sequences allows for the recognition of actions and the analysis of behaviors in video, making it a key challenge in automated video processing. The literature shows widespread use of deep learning approaches, specifically convolutional neural networks (CNNs). In this context, researchers face the challenge of identifying the advantages, disadvantages, and emerging trends across different architectures, evaluation metrics, and even dataset selection. The objective of this study is to identify the most common CNN architectures, evaluation metrics, datasets, and trends in spatiotemporal feature extraction for video analysis. The selection of articles used the PRISMA methodology and the Joanna Briggs Institute (JBI) methodological framework. From the databases Scopus, Web of Science and the MDPI platform, and based on the inclusion/exclusion criteria, 31 articles that met the criteria were analyzed and synthesized. The search was conducted primarily using the keywords “Convolutional Neural Network,” “video processing,” and “feature extraction,” limiting the selected works to those published between 2020 and the end of 2025. The results mainly show the use of four neural network architectures: 2D CNNs, 3D CNNs, hybrid models (e.g., CNN–RNN, CNN–Transformer, and multi-stream models), and, to a lesser extent, lightweight architectures. Commonly used datasets were identified (e.g., UCF101 and HMDB51). Additionally, standardized evaluation metrics were identified, ranging from accuracy and F1-score to performance measures specific to each case study. The challenges identified center on the heterogeneity of the study datasets, the lack of standardized evaluation metrics, and maintaining a balance between accuracy and computational resource consumption. On the other hand, strong emerging trends toward the use of hybrid models and those integrating transformers have been identified. This systematic review emphasizes the need for clear and robust guidelines that allow for the appropriate selection of CNN architecture, test datasets, and evaluation metrics in applications for extracting spatiotemporal features from video sequences, as well as identifying trends and future lines of research. Full article
(This article belongs to the Special Issue AI-Driven Computer Vision: Foundation Models, Multimodal Fusion, and Real-World Applications)
►▼ Show Figures

Figure 1

27 pages, 3637 KB  
Article
A Spatial-Functional Two-Dimensional Hierarchical Group Decision-Making Architecture for Spectrum Management of Emergency Communication UAV Swarms
by Hengzhou Jin, Gang Wang, Yangqin Wei, Jin Zang, Yu Chen and Xinyu Zhao
Electronics 2026, 15(16), 3735; https://doi.org/10.3390/electronics15163735 - 20 Aug 2026
Viewed by 199
Abstract
This paper proposes a spatial-functional two-dimensional hierarchical group decision-making (HGDM) spectrum management architecture for emergency communication unmanned aerial systems (EC-UAS). The architecture handles the highly dynamic topology, large node population, and differentiated task priorities that characterize EC-UAS. Using the spectrum management properties of [...] Read more.
This paper proposes a spatial-functional two-dimensional hierarchical group decision-making (HGDM) spectrum management architecture for emergency communication unmanned aerial systems (EC-UAS). The architecture handles the highly dynamic topology, large node population, and differentiated task priorities that characterize EC-UAS. Using the spectrum management properties of EC-UAS, we develop a discrete-time closed-loop dynamic model of the architecture and design adaptive hierarchical iteration rules. We prove global stability of the model under the stated assumptions and analyze the convergence of the state error, deriving its theoretical upper bound and the relationship between convergence steps and accuracy. An input-to-state stability analysis further demonstrates that the system state error remains bounded under dynamic disturbances, with its magnitude scaling with the disturbance bound. Simulations verify the effectiveness of the architecture and the correctness of the theoretical analysis. Full article
(This article belongs to the Section Microwave and Wireless Communications)
►▼ Show Figures

Figure 1

23 pages, 14402 KB  
Article
Water Level Estimation by Means of Microwave Reflection Measurements and Machine Learning Processing in a Multimode Cavity
by José Gadea-Rodríguez, Alejandro Díaz-Morcillo and Juan Monzó-Cabrera
Electronics 2026, 15(16), 3734; https://doi.org/10.3390/electronics15163734 - 20 Aug 2026
Viewed by 336
Abstract
This paper presents a novel method based on microwave reflection measurements and machine-learning techniques to estimate the water level in a multimode microwave applicator. Accurate water level monitoring is essential to maximize heating efficiency and protect the microwave source from excessive reflected power. [...] Read more.
This paper presents a novel method based on microwave reflection measurements and machine-learning techniques to estimate the water level in a multimode microwave applicator. Accurate water level monitoring is essential to maximize heating efficiency and protect the microwave source from excessive reflected power. To supplement conventional physical sensors, four regression models were evaluated: a one-dimensional convolutional neural network (CNN-1D), a multilayer perceptron (MLP), a support vector regressor (SVR), and a random forest (RF) model. These models estimate the water level based on the reflection coefficient S11 measured under low-power conditions over the 2.2–2.8 GHz band using a waveguide-based measurement setup for embedded sensing. The models were trained and evaluated using either the magnitude, phase, or both of S11 over the full-band or two selected reduced sub-bands. Results demonstrated that several models can accurately estimate water level from low-power microwave measurements, especially when magnitude is used as input data. The results demonstrate the potential of microwave measurements combined with machine-learning as a non-invasive supplementary sensing approach for water level monitoring. Full article
(This article belongs to the Special Issue Feature Papers in "Microwave and Wireless Communications" Section, 2nd Edition)
►▼ Show Figures

Figure 1

53 pages, 775 KB  
Systematic Review
A Systematic Review of Machine Learning-Driven Software-Defined Wireless Sensor Networks: Architectures, Security, and Routing Trends
by Ahmed Nader Al-Dulaimy and Hannes Frey
Electronics 2026, 15(16), 3733; https://doi.org/10.3390/electronics15163733 - 20 Aug 2026
Viewed by 390
Abstract
Recent progress in intelligent, adaptive, and security-aware Software-Defined Wireless Sensor Networks (SDWSNs) is driven by the integration of Machine Learning (ML) with Software-Defined Networking (SDN) and Wireless Sensor Networks (WSNs). This systematic survey analyzes 46 peer-reviewed articles published between 2024 and 2025, providing [...] Read more.
Recent progress in intelligent, adaptive, and security-aware Software-Defined Wireless Sensor Networks (SDWSNs) is driven by the integration of Machine Learning (ML) with Software-Defined Networking (SDN) and Wireless Sensor Networks (WSNs). This systematic survey analyzes 46 peer-reviewed articles published between 2024 and 2025, providing a problem-oriented synthesis of ML-SDWSN research. Emphasizing security, routing, and performance optimization, with a particular focus on deployment architectures, the survey identifies three major trends: increased adoption of ensemble and Reinforcement Learning (RL) methods for security and adaptive control; broader implementation of edge-based ML to minimize inference latency; and greater emphasis on privacy-preserving techniques, especially Federated Learning (FL). The survey presents a structured taxonomy encompassing seven thematic areas: Distributed Denial-of-Service (DDoS) mitigation, Intrusion Detection Systems (IDSs), routing optimization, Quality of Service (QoS) management, privacy preservation, data integrity, and network-efficiency optimization. Findings are synthesized from over 120 experimental configurations reported in the literature. Due to substantial differences among the reviewed studies in terms of datasets, network topologies, hardware platforms, measurement definitions, and validation methodologies, the reported values are presented as descriptive cross-study aggregates rather than direct comparative benchmarks or formal effect-size estimates. Within these constraints, the survey identifies recurring trade-offs among accuracy, latency, scalability, and privacy. It provides evidence-based design considerations for researchers and practitioners. The survey also highlights eight critical research gaps, including limited multi-dataset validation, a lack of real-world deployments, insufficient scalability analysis, and the need for rigorous evaluation of RL-based SDWSN control. Full article
(This article belongs to the Special Issue Artificial Intelligence for Distributed Networks)
►▼ Show Figures

Figure 1

18 pages, 7470 KB  
Article
Contactless ECG Reconstruction from Millimeter-Wave Radar Signals Using a CNN-BiLSTM Network
by Mingda Liu, Xiaoyan Zhou, Bo Ni, Qida Yu and Xinnan Zhao
Electronics 2026, 15(16), 3732; https://doi.org/10.3390/electronics15163732 - 20 Aug 2026
Viewed by 366
Abstract
To investigate the feasibility of reconstructing electrocardiogram (ECG) waveforms from non-contact millimeter-wave radar measurements, a radar-based ECG reconstruction method using a CNN-BiLSTM network is presented. A synchronous acquisition platform integrating a millimeter-wave radar and a BIOPAC physiological signal acquisition system was established to [...] Read more.
To investigate the feasibility of reconstructing electrocardiogram (ECG) waveforms from non-contact millimeter-wave radar measurements, a radar-based ECG reconstruction method using a CNN-BiLSTM network is presented. A synchronous acquisition platform integrating a millimeter-wave radar and a BIOPAC physiological signal acquisition system was established to collect chest-wall vibration signals and reference ECG signals. A multi-channel cross-correlation-based channel selection and temporal alignment procedure was employed to construct paired radar–ECG samples. The radar chest-wall vibration signals were filtered using an 8–30 Hz band-pass filter and then fed into the CNN-BiLSTM model, while a joint time–frequency loss function was introduced to constrain ECG reconstruction. On the self-built vital sign dataset, the reconstructed ECG achieved a correlation coefficient of 0.5631 with the reference ECG, while the mean absolute errors of heart rate and R–R interval were 1.00 BPM and 10.02 ms, respectively. These results suggest that the reconstructed signals preserve basic heartbeat timing and overall rhythm-related information, although the waveform-level agreement varies among samples and does not yet demonstrate consistent recovery of fine-grained ECG morphology. Evaluation on a public dataset further showed condition-dependent reconstruction performance under Resting, Apnea, and Valsalva conditions. Published MultiRes-LinkNet values were included only as contextual numerical references because the baseline was not reimplemented within the same experimental pipeline. Overall, the results provide preliminary evidence for the feasibility of contactless ECG reconstruction from millimeter-wave radar signals and suggest its potential value for radar-based vital sign monitoring. Full article
(This article belongs to the Special Issue AI in Radar Signal Processing)
►▼ Show Figures

Figure 1

28 pages, 36444 KB  
Article
A Human-Centric Virtual World for Nuclear Power Plants: A Methodological Framework for Integrating BIM and Seismic Analysis Data
by Mathias Proboste Martínez, Javier Mora Serrano, Fernando Rastellini Canela, Cristhian Albert Padilla Leaños and Felipe Muñoz-La Rivera
Electronics 2026, 15(16), 3731; https://doi.org/10.3390/electronics15163731 - 20 Aug 2026
Viewed by 335
Abstract
Interpreting nonlinear seismic structural analysis results in nuclear power plants remains challenging when conventional post-processing tools are used, as these require analysts to reconstruct structural meaning from fragmented 2D or non-immersive 3D views. This limits spatial understanding, weakens traceability between global response and [...] Read more.
Interpreting nonlinear seismic structural analysis results in nuclear power plants remains challenging when conventional post-processing tools are used, as these require analysts to reconstruct structural meaning from fragmented 2D or non-immersive 3D views. This limits spatial understanding, weakens traceability between global response and local damage mechanisms, and constrains the communication of findings in critical infrastructure contexts. In response, this paper proposes a human-centered methodological framework for integrating BIM models and nonlinear seismic simulation results into an immersive virtual reality environment for structural interpretation and risk-free inspection in nuclear power plants. The proposed workflow connects structural seismic analysis, result post-processing, the reference BIM model, and its deployment in a VR environment developed in Unreal Engine. The framework was implemented through a case study based on a generic nuclear power plant, resulting in a functional demonstrator. A qualitative evaluation based on an expert walkthrough showed the potential of the proposed workflow to enhance spatial understanding, simplify comparison between structural states, enable risk-free inspection environments, and facilitate technical communication. The main contribution of the study is to demonstrate how an integrated virtual reality environment can act as a complementary, human-centered interpretive interface that reduces cognitive fragmentation in conventional structural analysis workflows. Full article
(This article belongs to the Special Issue Applications of Virtual, Augmented, and Mixed Reality in Creative Technology)
►▼ Show Figures

Figure 1

26 pages, 2394 KB  
Article
An Intrusion Detection Method Based on Dynamic Social Structure Gray Wolf Optimization and Multi-Scale Temporal Perception
by Yijian Weng, Zhiliang Zhu, Congjie Wen, Zekai Cai and Xinli Wang
Electronics 2026, 15(16), 3730; https://doi.org/10.3390/electronics15163730 - 20 Aug 2026
Viewed by 275
Abstract
The dispatching data network and information management system in smart grids constitute a critical communication infrastructure that requires continuous security monitoring. However, network attacks exhibit multi-scale temporal characteristics ranging from microsecond-level bursts to slow intrusions lasting minutes, making single-scale detection models insufficient. Moreover, [...] Read more.
The dispatching data network and information management system in smart grids constitute a critical communication infrastructure that requires continuous security monitoring. However, network attacks exhibit multi-scale temporal characteristics ranging from microsecond-level bursts to slow intrusions lasting minutes, making single-scale detection models insufficient. Moreover, the hyperparameter space of deep learning models is large and highly non-convex, rendering traditional manual tuning inefficient. To address these challenges, this paper proposes an intrusion detection method based on the Dynamic Social Gray Wolf Optimizer (DSGWO) and the Multi-Scale Temporal Convolutional Network (MSTCN). The DSGWO maintains population diversity via an underdog alliance and breaks elite monopoly through a rank promotion challenge mechanism, balancing exploration and exploitation to avoid premature convergence. The MSTCN employs multi-scale parallel branches whose key training hyperparameters are optimized by the DSGWO, with residual connections and feature fusion for robust temporal modeling. Experiments on UNSW-NB15 and CIC-IDS-2017 demonstrate that DSGWO-MSTCN achieves F1-scores of 0.9933 and 0.9892, respectively, outperforming GWO, PSO, and NGO-based optimization approaches. Full article
►▼ Show Figures

Figure 1

Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-268
Previous Issue
Next Issue
Back to TopTop