Electronics

18 pages, 12250 KB

Open AccessArticle

A Vision Transformer Model with Hyperparameter Optimization for Oral Cancer Image Classification

by Chun-Tai Huang, Ying-Lei Lin, Chung-Hui Lin and Ping-Feng Pai

Electronics 2026, 15(10), 2230; https://doi.org/10.3390/electronics15102230 - 21 May 2026

Viewed by 383

Oral cancer is a significant public health concern and is among the most common malignant tumors of the head and neck. Its incidence and mortality rates remain persistently high, especially in regions where smoking and betel nut chewing are prevalent. Due to its [...] Read more.

Oral cancer is a significant public health concern and is among the most common malignant tumors of the head and neck. Its incidence and mortality rates remain persistently high, especially in regions where smoking and betel nut chewing are prevalent. Due to its high mortality rate, early detection is crucial for improving patient outcomes. However, early symptoms of oral cancer often resemble benign oral lesions, leading to delayed diagnosis. In this study, a vision transformer (ViT) model with Optuna (ViTOPT) is employed to perform classification tasks of identifying oral cancer images. The Optuna is used to determine hyperparameters in ViT. Histological images are obtained from a publicly available dataset. Three classification tasks with histological images namely classifying oral squamous cell carcinoma (OSCC) and leukoplakia (LEUK), classifying the presence of dysplasia, and classifying OSCC and leukoplakia with or without dysplasia are performed in this study. Numerical results reveal that the proposed ViTOPT framework is able to provide satisfactory performance in oral cancer recognition. Thus, the proposed ViTOPT model is a feasible and effective alternative in identifying oral cancer. Full article

(This article belongs to the Special Issue Data-Centric Artificial Intelligence: New Methods for Data Processing, 2nd Edition)

► Show Figures

Figure 1

15 pages, 881 KB

Open AccessArticle

Iterative Learning Control for Linear Systems Subject to Random Disturbance: An Impulsive Approach

by Zijian Luo and Hongyu Yang

Electronics 2026, 15(10), 2229; https://doi.org/10.3390/electronics15102229 - 21 May 2026

Viewed by 277

Abstract

This paper investigates the tracking control problem of linear systems with random disturbances. Distinguishing itself from previous studies, this work models instantaneous external disturbances as impulse effects and employs a Bernoulli random variable to characterize their stochastic occurrence. After this, a random impulsive [...] Read more.

This paper investigates the tracking control problem of linear systems with random disturbances. Distinguishing itself from previous studies, this work models instantaneous external disturbances as impulse effects and employs a Bernoulli random variable to characterize their stochastic occurrence. After this, a random impulsive differential system is established and used for theoretical analysis. Based on this model, two iterative learning control strategies are applied to achieve the robust tracking. The theoretical results demonstrate that the proposed algorithms are successful in reducing the impact caused by random disturbances, enabling the system to perform robust tracking. Finally, a numerical simulation is provided to verify the effectiveness of the proposed approach. Full article

(This article belongs to the Section Systems & Control Engineering)

► Show Figures

Figure 1

29 pages, 3491 KB

Open AccessArticle

Generalized AUC Maximization Core Vector Machine: A Multi-Kernel Learning Approach for Fast Imbalanced Classification

by Yichen Sun, Min Wu, Erhao Zhou, Shitong Wang and Kai Zhu

Electronics 2026, 15(10), 2228; https://doi.org/10.3390/electronics15102228 - 21 May 2026

Viewed by 299

Abstract

Imbalanced classification remains a fundamental challenge in machine learning, where the Area Under the ROC Curve (AUC) is widely used for threshold-independent ranking evaluation, especially in AUC maximization studies. Existing AUC maximization methods suffer from two critical limitations: they rely on single fixed [...] Read more.

Imbalanced classification remains a fundamental challenge in machine learning, where the Area Under the ROC Curve (AUC) is widely used for threshold-independent ranking evaluation, especially in AUC maximization studies. Existing AUC maximization methods suffer from two critical limitations: they rely on single fixed kernels that fail to capture complex data structures, and they incur prohibitive computational costs due to pairwise constraint construction. To address these issues, we propose the Generalized AUC Maximization Core Vector Machine (GAM-CVM), a fast imbalanced classification framework integrating multi-kernel learning with core vector machine optimization. Multiple affinity graphs are constructed from complementary perspectives and fused via cross-diffusion into a unified kernel matrix that respects the intrinsic data manifold. This fused kernel is embedded into a generalized AUC objective with a flexible ranking margin. Given the fused kernel matrix, the optimization stage of GAM-CVM achieves asymptotic linear time complexity with respect to the number of sample pairs under a fixed approximation accuracy by reformulating the learning objective as a center-constrained minimum enclosing ball problem. Extensive experiments demonstrate that GAM-CVM achieves the best overall average ranking and significantly outperforms most competing methods while maintaining the lowest optimization-stage running time. Full article

(This article belongs to the Special Issue Multimodal Learning for Multimedia Content Analysis and Understanding)

► Show Figures

Figure 1

17 pages, 432 KB

Open AccessArticle

Reusing Wireless Power Transfer for Backscatter-Assisted Pairwise Cooperation in Multi-User WPCNs

by Yuan Zheng, Fengxian Tang, Weiqiang Wu and Yongxue Wang

Electronics 2026, 15(10), 2227; https://doi.org/10.3390/electronics15102227 - 21 May 2026

Viewed by 206

Abstract

This paper studies a backscatter-assisted pairwise cooperation scheme in a multi-user wireless powered communication network (WPCN), where pairs of wireless devices (WDs) first harvest wireless energy from an energy node (EN) and then transmit their information to an access point (AP). Under the [...] Read more.

This paper studies a backscatter-assisted pairwise cooperation scheme in a multi-user wireless powered communication network (WPCN), where pairs of wireless devices (WDs) first harvest wireless energy from an energy node (EN) and then transmit their information to an access point (AP). Under the proposed scheme, the two WDs in each pair first exchange their local messages and then cooperatively transmit to the AP in the uplink. To reduce the time and energy consumption of local information exchange, we exploit the short distance between paired users and realize message exchange through energy-conserving backscatter communication. Meanwhile, the proposed design effectively reuses the wireless power transfer (WPT) signal to enable simultaneous information exchange during the energy harvesting phase, thereby leaving more time and harvested energy for the subsequent cooperative uplink transmission. Based on this transmission protocol, we jointly optimize the time allocation, the user transmit power allocation, and the energy beamforming matrix at the EN to maximize the weighted sum rate. To tackle the resulting non-convex problem, we decompose it into two coupled subproblems and develop an alternating optimization algorithm to update the corresponding variables iteratively. Numerical results show that the proposed scheme achieves significant weighted sum rate improvement over representative benchmark methods. Full article

(This article belongs to the Special Issue Advances in Wireless Power Transfer)

► Show Figures

Figure 1

22 pages, 5705 KB

Open AccessArticle

A 20 Hz LTPS TFT-Only 8T1C AMOLED Pixel Circuit with over Tenfold Leakage Current Reduction by Source–Drain Voltage Control

by Kook Chul Moon and Jae-Hong Jeon

Electronics 2026, 15(10), 2226; https://doi.org/10.3390/electronics15102226 - 21 May 2026

Viewed by 280

Abstract

Low-refresh-rate driving is an effective way to reduce the power consumption of active-matrix organic light-emitting diode (AMOLED) displays. However, in conventional low-temperature polycrystalline silicon (LTPS) thin-film transistor (TFT) pixel circuits, leakage current through switching TFTs can disturb the stored gate voltage of the [...] Read more.

Low-refresh-rate driving is an effective way to reduce the power consumption of active-matrix organic light-emitting diode (AMOLED) displays. However, in conventional low-temperature polycrystalline silicon (LTPS) thin-film transistor (TFT) pixel circuits, leakage current through switching TFTs can disturb the stored gate voltage of the driving TFT during the long emission period. This causes the time-dependent variation in driving current and visible flicker. In this study, a novel pixel circuit for leakage suppression in low-refresh-rate driving is presented. Bias aging was first applied to reduce the leakage current of the LTPS TFT, and a device model was then built from the characteristics measured at 60 °C. Based on this model, the leakage-induced instability of a conventional 7T1C pixel circuit was analyzed. To suppress this effect, a new 8T1C pixel circuit was proposed. The key idea is to reduce the source–drain voltage of the leakage-sensitive switching TFT during the emission period by raising the initial line potential to a level close to the storage node potential. Simulation results show that the proposed circuit greatly reduces the time-dependent variation in both the driving TFT gate voltage and the driving current compared with the conventional 7T1C circuit. Perceptual evaluation based on human visual sensitivity also confirms stable low-refresh-rate operation down to 20 Hz over the practical gray range. These results show that the proposed circuit is an effective solution for moderate low-refresh-rate operation without relying on low-temperature polycrystalline silicon and oxide (LTPO) technology. Full article

► Show Figures

Figure 1

34 pages, 1405 KB

Open AccessArticle

CMTF-Net: A Complex-Valued Multi-Scale Time–Frequency Cross-Domain Attention Network for MIMO CSI Prediction

by Bin Ren and Chengqun Wang

Electronics 2026, 15(10), 2225; https://doi.org/10.3390/electronics15102225 - 21 May 2026

Viewed by 417

Abstract

With the widespread adoption of multiple-input–multiple-output (MIMO) technology, channel state information (CSI) prediction has become a crucial technique for enhancing the performance of wireless communication systems. Traditional channel prediction methods face performance bottlenecks under high-speed mobility and complex channel conditions, making it difficult [...] Read more.

With the widespread adoption of multiple-input–multiple-output (MIMO) technology, channel state information (CSI) prediction has become a crucial technique for enhancing the performance of wireless communication systems. Traditional channel prediction methods face performance bottlenecks under high-speed mobility and complex channel conditions, making it difficult to meet the requirements of modern communication systems. To address this issue, this paper proposes a fully complex-valued cross-domain modeling framework, termed a complex-valued multi-scale transformer with time–frequency cross-attention network (CMTF-Net), for MIMO CSI prediction. CMTF-Net integrates a learnable multi-scale short-time Fourier transform (LMS-STFT), complex-valued multi-head self-attention (C-MHSA), and bidirectional cross-domain attention for complex-valued sequences (BCDA-CVS). These modules are designed to preserve amplitude–phase consistency, adapt time–frequency representations to CSI evolution, and enable information interaction between temporal and spectral features. On the simulated Overall Test set, CMTF-Net achieves the lowest MAE of 0.000032 and the highest Corr. (

ρ

) of 0.8230 among the compared methods, while maintaining competitive SE and BER values of 0.4240 and 0.2411 at SNR = 10 dB. On the DICHASUS measured datasets, CMTF-Net also shows favorable Test-ID and Test-OOD performance. For example, on DICHASUS-2186, it obtains Corr. (

ρ

)/SE/BER values of 0.8367/0.4935/0.2243 on Test-ID and 0.8061/0.4697/0.2351 on Test-OOD. These results indicate that CMTF-Net provides a balanced performance profile across prediction accuracy, spatial alignment, and communication-oriented evaluation. Full article

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

► Show Figures

Figure 1

26 pages, 3868 KB

Open AccessArticle

Optimized Distributed Quasi-GRS-Coded Cooperation with Split Labeling Diversity

by Chen Chen, Fengfan Yang, Manman Yang and Pingxiang Zhou

Electronics 2026, 15(10), 2224; https://doi.org/10.3390/electronics15102224 - 21 May 2026

Viewed by 204

Abstract

In this paper, a distributed quasi-generalized Reed–Solomon (Q-GRS)-coded cooperative split labeling diversity (DQ-GRSCC-SLD) scheme is proposed to support reliable cooperative transmission of small-volume information in typical scenarios such as device-to-device (D2D) communication, vehicular ad hoc networks (VANETs) and wireless sensor networks. The system [...] Read more.

In this paper, a distributed quasi-generalized Reed–Solomon (Q-GRS)-coded cooperative split labeling diversity (DQ-GRSCC-SLD) scheme is proposed to support reliable cooperative transmission of small-volume information in typical scenarios such as device-to-device (D2D) communication, vehicular ad hoc networks (VANETs) and wireless sensor networks. The system employs distinct labeling mappers at the source and the relay, enabling single-antenna transmission while constructing equivalently a dual-antenna labeling diversity model at the destination, which enhances interference resistance and reduces transmission costs. In addition, an ingenious design is proposed to ensure that the destination obtains the joint Q-GRS code. To optimize the weight distribution of the joint code, a traversal search (TS) algorithm is developed. Furthermore, a low-complexity joint decoding algorithm for Q-GRS codes, namely bracketing decoding, is presented by leveraging the efficient decoding algorithm of generalized Reed–Solomon (GRS) codes. Compared to the conventional maximum likelihood (ML) decoding, its complexity has been reduced from comparing

q^{k}

codewords to evaluating q or

q + 1

promising codewords. A theoretical performance analysis of the DQ-GRSCC-SLD scheme is provided. Simulation results reveal that the proposed DQ-GRSCC-SLD scheme demonstrates its superior performance under practical scenarios. Full article

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

► Show Figures

Figure 1

31 pages, 456 KB

Open AccessTutorial

A Dual-Stage Ransomware Defense Framework Combining an Artificial Immune System and Honeyfile Traps

by Xiang Fang, Huseyn Huseynov and Tarek Saadawi

Electronics 2026, 15(10), 2223; https://doi.org/10.3390/electronics15102223 - 21 May 2026

Viewed by 350

Abstract

The escalating sophistication of ransomware requires defensive strategies that are both proactive against zero-day attacks and operationally efficient. Existing solutions often force a trade-off—sacrificing low false-positive rates for broad detection, or vice versa. This work introduces an integrated framework designed to transcend this [...] Read more.

The escalating sophistication of ransomware requires defensive strategies that are both proactive against zero-day attacks and operationally efficient. Existing solutions often force a trade-off—sacrificing low false-positive rates for broad detection, or vice versa. This work introduces an integrated framework designed to transcend this limitation. Our dual-stage approach synergizes pre-encryption behavioral analysis with definitive post-encryption confirmation. The first stage employs a specialized artificial immune system (AIS) that monitors a curated set of 47 features, including API-call n-grams and file entropy dynamics, to identify malicious activity before file encryption begins. This pre-emptive analysis is complemented by an enhanced, cross-platform R-Locker mechanism, which uses Windows named pipes and symbolic links to deploy honeyfiles that trap ransomware during I/O operations, providing a high-fidelity trigger for automated containment. We subjected this framework to a rigorous evaluation against 3500 real-world ransomware samples and 12,000 benign applications. The results demonstrate a 98.2% detection rate with a 0.8% false-positive rate, achieving a mean response time of 1.3 s. A key finding is the framework’s efficiency on both Windows and Linux (the only platforms tested), with the AIS and R-Locker modules consuming a combined 101 MB of memory. While the system excels in real-time detection, we note that its current memory forensics capability for key recovery is incompatible with certain ransomware families due to architectural obfuscations. Our findings suggest that the integrated approach performs well under laboratory conditions; further real-world validation is required to confirm robustness in diverse environments. Full article

(This article belongs to the Special Issue Cryptography and Computer Security, 2nd Edition)

► Show Figures

Figure 1

37 pages, 805 KB

Open AccessReview

Evaluating Large Language Models in Cybersecurity: A Systematic Taxonomy and Empirical Analysis

by Mantun Chen, Hua Cheng, Ting Su, Minghui Chen, Wenjun Cai and Hongcheng Zou

Electronics 2026, 15(10), 2222; https://doi.org/10.3390/electronics15102222 - 21 May 2026

Viewed by 357

Abstract

This paper presents a Systematization of Knowledge (SoK) on the evaluation methodologies and capability boundaries of Large Language Models (LLMs) in cybersecurity. We propose a Three-Dimensional Taxonomy Matrix to systematize existing metrics across offensive domains, defensive applications, and inherent architectural flaws. Beyond categorization, [...] Read more.

This paper presents a Systematization of Knowledge (SoK) on the evaluation methodologies and capability boundaries of Large Language Models (LLMs) in cybersecurity. We propose a Three-Dimensional Taxonomy Matrix to systematize existing metrics across offensive domains, defensive applications, and inherent architectural flaws. Beyond categorization, this matrix functions as a predictive framework to expose structural evaluation blind spots. Specifically, by intersecting target domains with failure attributions, it identifies a critical, unresolved frontier: measuring cross-architecture semantic equivalence in low-level reverse engineering. Empirically, synthesizing 39 frontier benchmarks reveals a systemic evaluation gap: static metric success rarely translates into end-to-end adversarial efficacy. In offensive domains, high penetration rates correlate strongly with pre-training data contamination. When subjected to semantics-preserving code obfuscation as a stress test, zero-shot, tool-free exploit success rates collapse to near 0%. In defensive contexts, cross-procedural code auditing struggles, yielding a peak F1-score of only 23.83%. Furthermore, models suffer from over-alignment-induced functional degradation, with joint-testing frameworks recording up to a 77% functional loss in automated program repair. Our analysis strongly suggests that purely autoregressive mechanisms drive severe technical hallucinations, evidenced by a 19.7% package dependency fabrication rate. Evaluations also expose significant attack surfaces and a significant safety-utility tradeoff: models succumb to prompt leakage attacks at rates up to 86.2%, while heavily aligned versions simultaneously exhibit excessively high False Refusal Rates (FRR) for benign, borderline security queries. Finally, we delineate a theoretical neuro-symbolic roadmap—integrating LLM heuristics with deterministic formal methods—to structurally mitigate the limitations of the autoregressive paradigm. Full article

► Show Figures

Figure 1

20 pages, 2293 KB

Open AccessArticle

Resonance Mechanism Analysis and Suppression of Grid-Connected Energy Storage Power Station Inverter

by Weiheng Kuang, Jinchuan Guo, Lianhui Ning, Junyuan Zhang, Xinmei Gu, Sisi Chen, Shihong Shi, Weihan Hao, Min Zhou, Tiantian He and Qingxin Wang

Electronics 2026, 15(10), 2221; https://doi.org/10.3390/electronics15102221 - 21 May 2026

Viewed by 297

Abstract

The increasingly prominent “double-high” characteristics (high penetration of renewable energy and high proportion of power electronic devices) in modern power systems pose severe challenges to secure and stable operation, especially due to wideband oscillations induced by grid-connected inverters. In view of the fact [...] Read more.

The increasingly prominent “double-high” characteristics (high penetration of renewable energy and high proportion of power electronic devices) in modern power systems pose severe challenges to secure and stable operation, especially due to wideband oscillations induced by grid-connected inverters. In view of the fact that existing impedance modeling for grid-forming control often neglects the decoupling effect of the LC filter capacitor and the dynamics of inner voltage/current loops, leading to inaccurate characterization of mid-to-high frequency impedance, this paper aims to establish more accurate impedance models for grid-connected inverters and to develop effective oscillation mitigation methods accordingly. First, the harmonic linearization method is adopted to derive refined positive- and negative-sequence impedance analytical models for NPC inverters under both grid-following and grid-forming control. Second, simulation-based frequency scanning is conducted to validate the accuracy of the proposed models, and the differences in system resonance characteristics under the two control modes are comparatively analyzed. Finally, oscillation suppression strategies based on active damping and virtual impedance are, respectively, designed. The results show that the proposed models can accurately characterize mid-to-high frequency impedance, reveal the distinct resonance mechanisms of different control modes, and the proposed suppression strategies can effectively attenuate wideband oscillations. These findings provide theoretical foundations and practical technical pathways for stability analysis and optimization design of inverter-grid systems in high-renewable-penetration scenarios. Full article

(This article belongs to the Special Issue Advanced Technologies for Future Electric Power Transmission Systems)

► Show Figures

Figure 1

22 pages, 3271 KB

Open AccessArticle

TextureCLIP: Cross-Dataset Zero-Shot Texture Anomaly Segmentation with Triadic Descriptive Prompting

by Xin Peng Ooi and Seong G. Kong

Electronics 2026, 15(10), 2220; https://doi.org/10.3390/electronics15102220 - 21 May 2026

Viewed by 288

Abstract

Texture anomaly segmentation aims to localize irregularities on textured surfaces, a task critical for industrial quality control. Supervised methods require extensive labeled data, while unsupervised approaches often struggle to generalize to unseen target domains. Recent zero-shot methods based on vision-language models such as [...] Read more.

Texture anomaly segmentation aims to localize irregularities on textured surfaces, a task critical for industrial quality control. Supervised methods require extensive labeled data, while unsupervised approaches often struggle to generalize to unseen target domains. Recent zero-shot methods based on vision-language models such as Contrastive Language-Image Pretraining (CLIP) enable anomaly detection through text prompts without target-domain training data. However, existing approaches typically rely on generic prompts and show limited sensitivity to fine-grained texture variations. To address these limitations, we propose TextureCLIP, a cross-dataset zero-shot framework with auxiliary training for texture anomaly segmentation. The framework is trained on source texture data from the MVTec AD texture subset using annotated source-domain samples and directly evaluated on six unseen target datasets without access to target-domain training images, annotations, or fine-tuning. The proposed Triadic Descriptive Prompting (TriDP) integrates normal prompts, generic anomaly prompts, and descriptive anomaly prompts to provide complementary semantic cues for improved cross-domain generalization. To enhance spatial sensitivity, Dual Attention Modules (DAMs) are incorporated into the CLIP image encoder to refine local feature representations. In addition, Softmax-Weighted Averaging (SMWA) aggregates multiple anomaly cues by emphasizing the prompt responses with higher similarity scores. Experimental results demonstrate that TextureCLIP achieves strong and consistent performance across diverse texture datasets, attaining 67.06% AP and 65.69% F1-max, with improvements of 5.17 and 2.66 percentage points over the competitive baselines, respectively. Full article

(This article belongs to the Section Artificial Intelligence)

► Show Figures

Figure 1

23 pages, 28053 KB

Open AccessArticle

Enhanced Composite Multi-Scale Slope Entropy and Its Application to Fault Diagnosis of Rolling Bearing

by Wei Li, Jiazhu Li, Shuyu Wang, Yan Chen and Jian Chen

Electronics 2026, 15(10), 2219; https://doi.org/10.3390/electronics15102219 - 21 May 2026

Viewed by 223

Abstract

The health status of rolling bearings is critical to the normal operation of rotating machinery. To effectively extract vibration signal features and accurately identify different fault types, a novel method based on enhanced composite multi-scale slope entropy (ECMSE) and a honey badger algorithm-optimized [...] Read more.

The health status of rolling bearings is critical to the normal operation of rotating machinery. To effectively extract vibration signal features and accurately identify different fault types, a novel method based on enhanced composite multi-scale slope entropy (ECMSE) and a honey badger algorithm-optimized kernel extreme learning machine (HBA–KELM) is proposed. Specifically, ECMSE integrates high-order differences into the composite multi-scale framework to capture high-frequency information while preserving low-frequency characteristics, thereby enhancing the discriminability of time-series representations. Meanwhile, an average coarse-graining strategy is incorporated to achieve a more comprehensive characterization of the signals. The extracted features are then input into the HBA–KELM classifier for fault identification. Experiments conducted on two public and private rolling bearing datasets demonstrate that our method achieves superior performance in distinguishing different fault types and damage levels compared with several existing approaches. Full article

(This article belongs to the Special Issue Intelligent Sensing Empowered by Artificial Intelligence)

► Show Figures

Figure 1

20 pages, 13558 KB

Open AccessArticle

Deep Hybrid Synesthesia Model for Audio-Image Transfer

by Zhaojie Luo, Jiayong Jiang and Ladóczki Bence

Electronics 2026, 15(10), 2218; https://doi.org/10.3390/electronics15102218 - 21 May 2026

Viewed by 326

Abstract

Most artistic expressions are conveyed through images (e.g., painting) and audio (e.g., music), and deep learning has been successfully applied to neural style transfer within each of these modalities. However, there is still a lack of deep models that explicitly learn to transfer [...] Read more.

Most artistic expressions are conveyed through images (e.g., painting) and audio (e.g., music), and deep learning has been successfully applied to neural style transfer within each of these modalities. However, there is still a lack of deep models that explicitly learn to transfer style between images and audio. Motivated by synesthesia, which reflects intrinsic connections between vision and hearing in the human brain, we propose a deep hybrid synesthesia model for audio–image style transfer. Our framework consists of two main components: (1) a component conversion module that learns cross-modal mappings between audio rhythm/spectrum and image color/shape in a continuous valence–arousal (VA) emotion space; and (2) a style conversion module that transfers high-level artistic styles between Eastern (ink-wash, shui-mo) and Western painting and their corresponding musical counterparts. We first learn emotion-aware feature networks that align low-level audio and visual components based on shared affective representations, and then model long-term stylistic structures for cross-modal style transfer. Experiments include “seeing the sound” (audio-to-image generation with controllable components) and full audio–image style transformations. Both objective analyses and subjective evaluations suggest that our model can produce cross-modal artworks whose perceived style and emotional content are consistent with human synesthetic impressions. Full article

(This article belongs to the Special Issue AI-Driven Image Generation: Algorithms, Architectures, and Applications)

► Show Figures

Figure 1

20 pages, 1010 KB

Open AccessArticle

Enhanced Discrete Multi-Objective Particle Swarm Optimization for Electromagnetic Spectrum Planning

by Liuyang Gao, Zhongfu Xu and Haili Li

Electronics 2026, 15(10), 2217; https://doi.org/10.3390/electronics15102217 - 21 May 2026

Viewed by 226

Abstract

Electromagnetic spectrum planning is a critical challenge in modern wireless communication systems, characterized by multiple conflicting objectives including spectrum utilization efficiency, interference minimization, and fairness among users. This paper proposes an Enhanced Discrete Multi-Objective Particle Swarm Optimization (EDMOPSO) algorithm specifically designed for spectrum [...] Read more.

Electromagnetic spectrum planning is a critical challenge in modern wireless communication systems, characterized by multiple conflicting objectives including spectrum utilization efficiency, interference minimization, and fairness among users. This paper proposes an Enhanced Discrete Multi-Objective Particle Swarm Optimization (EDMOPSO) algorithm specifically designed for spectrum assignment problems. The proposed method introduces a novel probabilistic discrete velocity update mechanism with adaptive dynamic bounds, an adaptive inertia weight strategy based on normalized population diversity, and an improved archiving technique with enhanced diversity preservation. To handle the discrete nature of spectrum allocation, we develop a binary encoding scheme combined with a problem-specific repair mechanism for constraint satisfaction. The algorithm is evaluated on both synthetic benchmark problems and real-world spectrum planning scenarios. Experimental results demonstrate that EDMOPSO achieves competitive performance advantages over seven established multi-objective evolutionary algorithms, with Hypervolume improvements of 18.7% and Inverted Generational Distance reductions of 23.4% compared to the second-best-performing algorithm. A comprehensive ablation study with 15 configurations validates the synergistic interaction between components. The proposed method provides an effective solution for macro-level periodic spectrum management in complex electromagnetic environments. Full article

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

► Show Figures

Figure 1

12 pages, 9106 KB

Open AccessArticle

A 20–43 GHz High-Dynamic-Range Amplifier with Current-Reused and Vertically Stacked Topology in GaAs Process

by Zhen Ye, Jiyu Zhang, Liulin Hu and Li Xu

Electronics 2026, 15(10), 2216; https://doi.org/10.3390/electronics15102216 - 21 May 2026

Viewed by 228

Abstract

This paper presents a current-reused vertically stacked (CRVS) topology for a high-dynamic-range amplifier (HDRA) implemented in a 0.1 μm GaAs pHEMT process, targeting wideband millimeter-wave (mm-wave) receiver front-ends. The proposed design breaks the inherent trade-off between noise figure (NF), linearity, and bandwidth, achieving [...] Read more.

This paper presents a current-reused vertically stacked (CRVS) topology for a high-dynamic-range amplifier (HDRA) implemented in a 0.1 μm GaAs pHEMT process, targeting wideband millimeter-wave (mm-wave) receiver front-ends. The proposed design breaks the inherent trade-off between noise figure (NF), linearity, and bandwidth, achieving simultaneous enhancement of transconductance efficiency, Miller effect suppression, and wideband matching. The fabricated prototype operates over a continuous 20–43 GHz bandwidth (covering K- and Ka-bands), demonstrating state-of-the-art performance: a flat gain of 24 ± 0.6 dB, a minimum NF of 2.2 dB, a maximum output 1 dB compression point (OP1dB) of 15.8 dBm and a low power consumption of 5 V/65 mA, with both input and output return losses better than −10 dB across the entire band. The results validate the effectiveness of the CRVS topology and highlight the competitiveness of GaAs pHEMT technology for high-performance wideband mm-wave front-ends, making the design suitable for applications including 5G/6G communication, satellite systems, and mm-wave test equipment. Full article

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

► Show Figures

Figure 1

15 pages, 8067 KB

Open AccessArticle

Large-Signal Equivalent Circuit Model for HighPower Laser Diode Mini-Array

by Lei Ling, Tao Duan, Shunhua Wu, Jiachen Liu, Junyue Zhang, Weizhou Huang, Qingkai Meng, Lang Chen, Jiachen Zhang, Te Li and Zhenfu Wang

Electronics 2026, 15(10), 2215; https://doi.org/10.3390/electronics15102215 - 21 May 2026

Viewed by 270

Abstract

High-power laser diodes are extensively utilized in advanced optoelectronic systems. These devices typically operate under high-current injection conditions, under which intrinsic parasitic parameters become non-negligible and exert a substantial influence on their electro-optical response characteristics. Furthermore, when multiple single emitters are monolithically integrated [...] Read more.

High-power laser diodes are extensively utilized in advanced optoelectronic systems. These devices typically operate under high-current injection conditions, under which intrinsic parasitic parameters become non-negligible and exert a substantial influence on their electro-optical response characteristics. Furthermore, when multiple single emitters are monolithically integrated into a linear array along the epitaxial-layer direction on a single substrate, additional parasitic elements are inevitably introduced. These parameters are critical for characterizing the output performance of high-power laser diodes. This paper presents the implementation of an equivalent circuit model for large-signal laser-diode operation within the Advanced Design System (ADS) computer-aided environment. The proposed model enables accurate simulation of the device’s operating-voltage waveform and optical-output-power response under both DC steady-state and pulsed-transient driving conditions, thereby achieving a coupled representation of electrical behavior and optical emission. Sensitivity analysis of various parasitic elements is performed to systematically evaluate their influence on output characteristics and device reliability. The results provide theoretical guidance for structural optimization and packaging design, offering new insights into future modeling and reliability assessment of high-power laser diodes. Full article

(This article belongs to the Section Optoelectronics)

► Show Figures

Figure 1

23 pages, 500 KB

Open AccessArticle

Beyond Tool Poisoning: Attack Surfaces of Malicious Remote MCP Servers Across LLM Platforms

by Jinwoo Park, Geonhee Kim, Hyeokjae Lee and Jeman Park

Electronics 2026, 15(10), 2214; https://doi.org/10.3390/electronics15102214 - 21 May 2026

Viewed by 480

Abstract

The Model Context Protocol (MCP) has become the de facto standard for connecting large language models (LLMs) to external tools, and its remote deployment mode lets users add third-party servers with a single URL—shifting a substantial portion of the host’s attack surface to [...] Read more.

The Model Context Protocol (MCP) has become the de facto standard for connecting large language models (LLMs) to external tools, and its remote deployment mode lets users add third-party servers with a single URL—shifting a substantial portion of the host’s attack surface to infrastructure operated by anonymous parties. Existing MCP security work has concentrated on tool-description poisoning and studied individual techniques in isolation, leaving it unclear what a malicious remote server can accomplish across its full surface. In this paper, we explore the malicious-server threat space along the axis of whether the host LLM participates in producing the harmful outcome, yielding two categories: LLM-passive attacks, which complete inside the server, and LLM-active attacks, which require the LLM to deliver the malicious content. We implement five scenarios spanning both categories—realizing each LLM-active scenario with both description-based and response-based variants against the same goal—and evaluate all configurations on ChatGPT, Claude Desktop, and Gemini CLI. We find that host-side filtering of MCP-bound data varies sharply across platforms (95% vs. 50% ASR on the same email request), that the description and response channels succeed on disjoint scenarios, and that successful attacks are almost never disclosed to the user. These findings suggest that defending remote MCP deployment requires a multi-layer approach combining host-side filtering, LLM-level response auditing, and user-visible output transparency. Full article

(This article belongs to the Special Issue Cryptography and Computer Security, 2nd Edition)

► Show Figures

Figure 1

21 pages, 9130 KB

Open AccessArticle

Semi-Supervised Facial Emotion Recognition via Valence-Arousal Pseudo-Label Refinement

by Seunghyun Kim, Hyunsoo Seo, Ill Hyung Jo and Eui Chul Lee

Electronics 2026, 15(10), 2213; https://doi.org/10.3390/electronics15102213 - 21 May 2026

Viewed by 303

Abstract

Facial expression recognition is a pivotal area in computer vision, traditionally focusing on categorical labels such as ‘Happy’, and ‘Sad’. Recent advancements have transitioned towards using the continuous indicators valence and arousal, reflecting the complicated nature of human emotions. This study introduces a [...] Read more.

Facial expression recognition is a pivotal area in computer vision, traditionally focusing on categorical labels such as ‘Happy’, and ‘Sad’. Recent advancements have transitioned towards using the continuous indicators valence and arousal, reflecting the complicated nature of human emotions. This study introduces a method using semi-supervised learning to generate valence and arousal labels for existing categorical datasets, addressing challenges like racial bias. We propose a pseudo-label refinement framework, VAP-Refine (Valence-Arousal Pseudo-label Refinement), which enhances facial expression recognition by combining teacher model predictions and category-level statistics. These predicted labels are adjusted using ground truth category information and combined with actual category labels to train a more robust facial expression recognition model. Our approach improved accuracy from 69.3% to 72.26% and from 62.94% to 67.24% across two datasets. Fine-tuning with a teacher model predicting valence and arousal achieved 75.82% and 91.58% accuracy, with an F1-score of 0.9147, despite data imbalances. These results highlight the potential of semi-supervised learning to enhance facial expression recognition by incorporating continuous emotional indicators, improving model performance and contributing to more accurate affective computing applications. Furthermore, the proposed framework consistently improved performance across various backbone architectures, including ResNet50, SHViT, and DDAMFN++, highlighting its generalizability and versatility. Full article

(This article belongs to the Special Issue Biometric Recognition: Latest Advances and Prospects, 2nd Edition)

► Show Figures

Figure 1

30 pages, 7666 KB

Open AccessArticle

NeSy-Drop: Interpretable Dropout Prediction and Personalized Intervention via Neuro-Symbolic Graph Learning in MOOCs

by Abdennour Redjaibia, Samia Drissi, Karima Boussaha, Yacine Lafifi and Sevinç Gülseçen

Electronics 2026, 15(10), 2212; https://doi.org/10.3390/electronics15102212 - 21 May 2026

Viewed by 344

Abstract

Dropout prediction in Massive Open Online Courses (MOOCs) has been extensively studied, yet existing systems share three fundamental limitations: Accurate models are black boxes, post-hoc explanations approximate rather than faithfully represent model decisions, and predictions are rarely translated into concrete instructor actions. This [...] Read more.

Dropout prediction in Massive Open Online Courses (MOOCs) has been extensively studied, yet existing systems share three fundamental limitations: Accurate models are black boxes, post-hoc explanations approximate rather than faithfully represent model decisions, and predictions are rarely translated into concrete instructor actions. This paper presents NeSy-Drop, a neuro-symbolic framework that simultaneously addresses prediction, explanation, and personalized intervention routing for MOOC dropout. NeSy-Drop constructs a heterogeneous graph per course cohort encoding student–resource–assessment interactions, processed through a heterogeneous graph transformer encoder, five behavioral atom predictor MLPs, and a differentiable symbolic rule layer producing guaranteed faithful ante-hoc explanations. A three-level explainability stack provides symbolic rule chains, SHAP embedding attribution, LIME raw-feature importance, and gradient-based counterfactual prescriptions. Each at-risk student is routed to one of five concrete interventions at one of three severity levels. Evaluated on OULAD covering 32,593 students across 22 cohorts, NeSy-Drop achieves AUC of 0.961 and macro F1 of 0.8983, within 2.2% AUC of the best non-interpretable baseline under a fair evaluation protocol, while being the only system that simultaneously predicts, explains, and prescribes actions at the individual student level. Full article

(This article belongs to the Section Artificial Intelligence)

► Show Figures

Figure 1

27 pages, 7724 KB

Open AccessArticle

AGCo-MATA: Air-Ground Collaborative Multi-Agent Task Allocation in Mobile Crowdsensing

by Lixin Yang, Kaixing Zhao, Tianhao Shao, Bohan Feng, Jian Di and Zuheng Ming

Electronics 2026, 15(10), 2211; https://doi.org/10.3390/electronics15102211 - 21 May 2026

Viewed by 288

Abstract

The rapid advancement of intelligent unmanned systems has brought new opportunities to mobile crowd sensing (MCS). Compared with traditional homogeneous frameworks, heterogeneous air-ground collaborative multi-agent frameworks consisting of unmanned aerial vehicles (UAVs) and unmanned ground vehicles (UGVs) exhibit superior flexibility and efficiency in [...] Read more.

The rapid advancement of intelligent unmanned systems has brought new opportunities to mobile crowd sensing (MCS). Compared with traditional homogeneous frameworks, heterogeneous air-ground collaborative multi-agent frameworks consisting of unmanned aerial vehicles (UAVs) and unmanned ground vehicles (UGVs) exhibit superior flexibility and efficiency in complex sensing tasks. Task allocation among agents is crucial for improving overall MCS quality. To achieve efficient task allocation for heterogeneous collaborative agents, this study investigated two typical complex multi-agent task allocation scenarios with dual optimization objectives: (1) For the Air-Ground Few-Agents-More-Tasks (AG-FAMT) scenario, the objectives are to maximize task completion and minimize total travel distance; (2) For the Air-Ground More-Agents-Few-Tasks (AG-MAFT) scenario (task allocation based on agent locations), the objectives are to minimize total travel distance and travel time cost. Overall, in this paper, we proposed two algorithms: a multi-task minimum cost maximum flow optimization algorithm called Multi-Task Minimum-Cost Maximum-Flow (MT-MCMF) tailored for AG-FAMT, and a multi-objective optimization algorithm called Weighted Integer Linear Programming (W-ILP) for AG-MAFT (with a focus on optimizing UAV charging path planning). Experiments on a large-scale real-world dataset demonstrated that both proposed algorithms outperform baseline methods under varying experimental settings (task quantity, difficulty, and distribution), providing a novel approach to enhance the overall quality of air-ground MCS tasks. Full article

(This article belongs to the Topic Advanced Technologies and Applications for Unmanned Systems)

► Show Figures

Figure 1

24 pages, 1305 KB

Open AccessArticle

FPCache: A Fingerprint-Rectified Learned Index Cache for Disaggregated Memory

by Chenyang Jia and Miao Cai

Electronics 2026, 15(10), 2210; https://doi.org/10.3390/electronics15102210 - 21 May 2026

Viewed by 213

Abstract

The rapid growth of data-intensive applications has increased the demand for efficient storage in large-scale key-value (KV) stores. Disaggregated memory architectures provide a scalable solution by separating compute and memory resources via RDMA. However, existing indexing schemes in these environments suffer from poor [...] Read more.

The rapid growth of data-intensive applications has increased the demand for efficient storage in large-scale key-value (KV) stores. Disaggregated memory architectures provide a scalable solution by separating compute and memory resources via RDMA. However, existing indexing schemes in these environments suffer from poor read efficiency, significantly degrading overall system throughput and scalability. Specifically, learned indexes often encounter substantial read amplification during remote data retrieval due to prediction errors. In addition, caching full keys incurs a high cache footprint, limiting the effective cache capacity on compute nodes and leading to additional remote memory accesses. This paper presents FPCache, a fingerprint-rectified learned index cache for disaggregated memory. We propose a fingerprint-assisted two-stage read approach to mitigate read amplification. FPCache first retrieves a compact fingerprint array for local matching. It then converts range reads into precise point accesses and directly reads the corresponding data item, thereby avoiding reading the entire range and reducing extra data transfers. Next, we design a fingerprint-offset compression strategy to maximize cache density. Leveraging fixed-length fingerprints and position offsets enables compute nodes to retain significantly more hotspot data within limited memory resources. Experimental evaluations using various YCSB workloads demonstrate that FPCache consistently outperforms state-of-the-art methods. Compared to systems like CHIME and ROLEX, FPCache improves system throughput by up to 62% and effectively maintains stable access efficiency under diverse data distributions. Full article

(This article belongs to the Special Issue New Challenges in High-Performance Computing and Computer Architecture)

► Show Figures

Figure 1

26 pages, 6746 KB

Open AccessArticle

Linear Parameter Varying Model Predictive Control with 3D Anomaly Perception for Autonomous Driving

by Zia Ur Rehman, Hongbin Ma and Ubaid Ur Rahman Qureshi

Electronics 2026, 15(10), 2209; https://doi.org/10.3390/electronics15102209 - 20 May 2026

Viewed by 253

Abstract

Accidents and vehicle damage caused by irregular road surfaces, such as potholes and cracks, remain a significant challenge in autonomous driving, particularly in terms of safety and trajectory reliability. Existing approaches often treat perception and control as separate processes, limiting their ability to [...] Read more.

Accidents and vehicle damage caused by irregular road surfaces, such as potholes and cracks, remain a significant challenge in autonomous driving, particularly in terms of safety and trajectory reliability. Existing approaches often treat perception and control as separate processes, limiting their ability to respond effectively to road-surface anomalies in real time. In the proposed work, a unified framework for road-surface anomaly-aware control that integrates 3D point cloud perception with a Linear Parameter-Varying Model Predictive Controller (LPV-MPC) is presented. The proposed approach utilizes onboard sensors to capture detailed geometric information of the road surface and detect anomalies relevant to vehicle motion. The detected anomalies are represented in a control-oriented form and incorporated into the LPV-MPC framework, enabling adaptive trajectory planning and speed regulation. This integration allows the controller to proactively adjust vehicle behavior in response to surface irregularities, improving both safety and tracking performance. Experimental results demonstrate that the proposed method enhances robustness against road disturbances and improves trajectory tracking compared to conventional control approaches without anomaly awareness. These results highlight the effectiveness of tightly coupling perception and control for reliable autonomous driving in real-world conditions. Full article

(This article belongs to the Special Issue AI-Powered Mobility Systems for Smart, Connected, and Autonomous Vehicles and Traffic Dynamics)

► Show Figures

Figure 1

26 pages, 5856 KB

Open AccessArticle

Adaptive SOC Estimation of Reconfigurable Battery Modules Based on a Hybrid Deep Learning Model

by Qiang Zhao, Fanqi Tang and Bing Zhang

Electronics 2026, 15(10), 2208; https://doi.org/10.3390/electronics15102208 - 20 May 2026

Viewed by 245

Abstract

Reconfigurable battery modules can dynamically adjust the connection topology among battery cells, significantly improving the energy utilization efficiency of battery energy storage systems. However, existing state estimation methods focus primarily on individual battery cells. Frequent topology changes cause traditional State of Charge (SOC) [...] Read more.

Reconfigurable battery modules can dynamically adjust the connection topology among battery cells, significantly improving the energy utilization efficiency of battery energy storage systems. However, existing state estimation methods focus primarily on individual battery cells. Frequent topology changes cause traditional State of Charge (SOC) estimation algorithms to accumulate large errors due to mismatches in equivalent capacity and internal resistance, making them ineffective for reconfigurable battery modules. To address this limitation, this paper proposes a Gated Recurrent Unit–Transformer architecture for precise SOC estimation in reconfigurable battery modules. The model uses a Gated Recurrent Unit to capture the temporal continuity of electrochemical evolution and employs the Transformer’s self-attention mechanism to analyze discrete topology changes. Experimental results show excellent estimation accuracy across different initial SOC levels, with a mean absolute error as low as 0.085% at full charge and 0.035% at 50% SOC. The architecture demonstrates strong topology self-identification capability and maintains high robustness even under abrupt voltage steps caused by reconfiguration. This method provides accurate and reliable state estimation for large-scale two-layer reconfigurable battery systems while reducing control complexity and improving operational efficiency. Full article

(This article belongs to the Topic Object Detection and Control of Networked Autonomous Systems: Theories, Analysis Tools and Applications)

► Show Figures

Figure 1

28 pages, 2945 KB

Open AccessArticle

Stability-Driven Feature Extraction–Kolmogorov–Arnold Network-Driven Ensemble Framework for Reliable Breast Cancer Detection

by Abdul Rahaman Wahab Sait and Yazeed Alkhurayyif

Electronics 2026, 15(10), 2207; https://doi.org/10.3390/electronics15102207 - 20 May 2026

Viewed by 208

Abstract

Breast cancer screening is a fundamentally probabilistic diagnostic task that requires precise identification of complex imaging characteristics from diverse patient cohorts. Despite improvements in deep learning techniques, current automatic tools are typically trained on well-curated datasets and do not generalize to heterogeneous data, [...] Read more.

Breast cancer screening is a fundamentally probabilistic diagnostic task that requires precise identification of complex imaging characteristics from diverse patient cohorts. Despite improvements in deep learning techniques, current automatic tools are typically trained on well-curated datasets and do not generalize to heterogeneous data, thereby limiting their application. This study aims to address these shortcomings by introducing a more effective and generalizable framework for breast cancer classification that focuses on the stability of features, the learning of complementary representations, and improved decision modeling. The proposed methodology incorporates stability-driven feature extraction (SDFE) with a multi-branch architecture that consists of EfficientNetV2 (Convolutional neural networks (CNNs)), EfficientFormer (Vision transformers (ViTs)), and multi-layer perceptron (MLP)-Mixer models to extract various feature representations. To improve non-linear decision boundaries, it uses a Kolmogorov–Arnold Network (KAN)-based classification head and selects the most credible prediction via an adaptive voting mechanism. This model is trained using patient-level splitting on the VinDr-Mammo dataset, evaluated using five-fold cross-validation, and subsequently externally validated on the CBIS-DDSM dataset. Experimental findings demonstrate the consistent performance of the proposed model, with accuracies of 94.5% in cross-validation, 93.3% on the VinDr-Mammo test set, and 94.6% on CBIS-DDSM, surpassing other recent state-of-the-art solutions. It demonstrates enhanced robustness and cross-dataset generalization, offering a scalable, consistent framework for breast cancer classification that supports the development of computer-aided diagnostic systems. Full article

(This article belongs to the Special Issue Next-Generation Machine Learning and Deep Learning Models for Complex Data, Vision, and Intelligent Applications)

► Show Figures

Figure 1

14 pages, 3086 KB

Open AccessArticle

Cyanate Ester–Lunar Regolith Composites for In Situ Fabrication of Structural Electronics on the Moon

by Guancheng Li, Batuhan Mirac Alasahin, Mark Mirotznik and Robert L. Opila

Electronics 2026, 15(10), 2206; https://doi.org/10.3390/electronics15102206 - 20 May 2026

Viewed by 222

Abstract

The development of electronic substrates from locally available materials is critical for sustainable lunar infrastructure. This work investigates the synthesis, processing, and characterization of cyanate ester–lunar regolith simulant (CE-LRS) composites designed specifically for the extreme lunar environment. LRS were evaluated as functional fillers [...] Read more.

The development of electronic substrates from locally available materials is critical for sustainable lunar infrastructure. This work investigates the synthesis, processing, and characterization of cyanate ester–lunar regolith simulant (CE-LRS) composites designed specifically for the extreme lunar environment. LRS were evaluated as functional fillers at loadings up to 55 wt.% with CE binder selected for its thermal stability (T_g > 230 °C), vacuum compatibility, and known radiation resistance from prior literature. A vacuum-assisted curing procedure was developed that utilizes the lunar environment as a processing advantage, reducing porosity from approximately 7% to less than 1% as quantified by X-ray micro-computed tomography. Dynamic mechanical analysis revealed that increased filler loading and vacuum processing enhanced the storage modulus and T_g through constraining polymer chain mobility at the filler-binder interface, confirming effective stress transfer and interfacial adhesion. Scanning electron microscopy also verified intimate polymer–filler wetting. Waveguide measurements in the microwave frequency range demonstrated that the composites remain non-magnetic while exhibiting moderately increased permittivity and low dielectric loss, meeting the requirements for radio-frequency substrate applications. Through material selection and process design that embraces, rather than ignores, lunar environmental constraints, this work establishes the CE-LRS composites that represent a viable pathway for the in situ fabrication of structural electronics on the Moon. Full article

► Show Figures

Figure 1

22 pages, 42854 KB

Open AccessArticle

The Study of UAV-Based Tea Shoots Detection with TSDet-UAV Method

by Kaihua Wei, Yulin Cai, Chengbo Lu, Jingcheng Zhang, Dong Ren, Shun Ren and Dongmei Chen

Electronics 2026, 15(10), 2205; https://doi.org/10.3390/electronics15102205 - 20 May 2026

Viewed by 230

Abstract

The picking of tea leaves in tea gardens requires multiple batches in the short and valuable tea harvest period. To realize timely and efficient tea plucking, it is feasible to use unmanned aerial vehicles (UAV) for tea shoot detection in large tea gardens. [...] Read more.

The picking of tea leaves in tea gardens requires multiple batches in the short and valuable tea harvest period. To realize timely and efficient tea plucking, it is feasible to use unmanned aerial vehicles (UAV) for tea shoot detection in large tea gardens. For the typical small targets of tea buds in unmanned aerial vehicle (UAV) aerial images, it is necessary to design an efficient tea buds detection model. In order to improve the accuracy and the speed of the tea buds detection in the UAV images, we designed the SH-CoordMapping hash space mapping algorithm to accelerate the remerging of the detection results into the original image. The C2PSA-BI module and the CARAFE upsampling module are applied to improve detail preservation during feature fusion. A lightweight detection head is further used to reduce redundant computation in the detection stage. By comparing with the traditional detection methods, it can be proved that the SWO sections are necessary for UAV-scale tea shoots detection. Based on the accuracy and the number of model parameters, the YOLO11n model with slice size as 640 and overlap rate as 0.1 performs the best. The TSDet-UAV was deployed on the NVIDIA Jetson Orin NX chip to construct an inspection system capable of real-time acquisition and detection. The experimental results demonstrate that the proposed TSDet-UAV achieves excellent performance, recording an mAP₅₀ of 52.9% on the constructed UAV-TS dataset while maintaining high efficiency. With a parameter size of 2.4 M and a total processing time of 1.32 s per high-resolution image under TensorRT FP16, the processing speed is highly suitable for real-time edge deployment on agricultural UAV platforms. The UAV image-based tea garden shoot inspection platform proposed in this paper can effectively confirm the growth status of tea shoots, assisting farm management in formulating precise picking plans. Full article

► Show Figures

Figure 1

33 pages, 11328 KB

Open AccessArticle

Artificial Intelligence for Autonomous Vehicles: Robustness Analysis in Complex Urban Traffic Scenarios

by Brandon Quezada-Godoy, Antonio Guerrero-González, Francisco García-Córdova, Francisco Lloret-Abrisqueta and Antonio Jesús Martínez-Espinosa

Electronics 2026, 15(10), 2204; https://doi.org/10.3390/electronics15102204 - 20 May 2026

Viewed by 343

Abstract

Autonomous driving in complex urban environments remains challenging due to perception uncertainty, dynamic multi-agent interactions, and control instability under adverse conditions. Despite advances in individual components, systematic evaluations of fully integrated modular pipelines under compounded urban disturbances remain scarce. This work presents a [...] Read more.

Autonomous driving in complex urban environments remains challenging due to perception uncertainty, dynamic multi-agent interactions, and control instability under adverse conditions. Despite advances in individual components, systematic evaluations of fully integrated modular pipelines under compounded urban disturbances remain scarce. This work presents a modular autonomous driving framework in CARLA Town10HD, integrating Convolutional Neural Network (CNN)-based perception using ResNet-18, global path planning via A* algorithm, and two control strategies: a classical Proportional–Integral–Derivative (PID) controller and a Deep Q-Network (DQN) agent with adaptive geometric steering assistance. A structured protocol assessed robustness across five scenarios: Heavy Rain, Dense Fog, Nighttime Driving, Dense Traffic, and Combined Extreme Conditions. The perception module achieved F1-scores close to 0.99 for traffic-sign, pedestrian, and lane classification; results reflect synthetic CARLA data and should not be interpreted as real-world generalization. The PID controller produced smoother trajectories with lower steering oscillations, while the DQN agent achieved faster traversal times at the cost of higher control variability. Route efficiency remained around 0.96 under isolated disturbances and decreased to 0.52 under compounded conditions, confirming sensitivity to multi-factor complexity. This study contributes a reproducible multi-scenario benchmark quantifying stability–adaptability trade-offs between classical and learning-based control, identifying scenario generalization and simulation-to-reality transfer as key future directions. Full article

(This article belongs to the Special Issue Electronic Architecture for Autonomous Vehicles)

► Show Figures

Figure 1

28 pages, 2401 KB

Open AccessArticle

Novel Positioning Scheme Based on Supervised Deep Reinforcement Learning for Indoor Wireless Localization

by Youngghyu Sun, Kyounghun Kim, Seongwoo Lee, Joonho Seon, Soohyun Kim and Jinyoung Kim

Electronics 2026, 15(10), 2203; https://doi.org/10.3390/electronics15102203 - 20 May 2026

Viewed by 290

Abstract

In this paper, a supervised deep reinforcement learning (SDRL)-based positioning scheme is proposed for indoor wireless localization. The proposed scheme formulates the positioning problem as a Markov decision process and introduces a target-aware reward design based on the artificial potential field (APF) to [...] Read more.

In this paper, a supervised deep reinforcement learning (SDRL)-based positioning scheme is proposed for indoor wireless localization. The proposed scheme formulates the positioning problem as a Markov decision process and introduces a target-aware reward design based on the artificial potential field (APF) to alleviate the sparse reward problem commonly encountered in search-based reinforcement learning. In the proposed scheme, supervision is provided at the reward level by incorporating the target position into the reward design, rather than at the action level via expert demonstrations. A multi-scale action set with 49 candidates is further adopted to provide a favorable trade-off between estimation accuracy and search efficiency. An anchor-based environment construction strategy is developed by selecting the four strongest reference points (RPs) and transforming their coordinates with respect to the strongest RP. Simulation results show that the proposed scheme achieves a mean absolute error (MAE) below 0.8 m and success rates above 99.1% within 1 m and 99.2% within 2 m under the default Bluetooth Low Energy setting, while the convex-valid rate of the anchor-based environment exceeds 99.5%. Compared with existing methods, the proposed scheme reduces the MAE by approximately 92.3%. Ablation studies confirm that multi-scale actions reduce the average search steps by approximately 69.5% compared with a single-scale baseline. The proposed scheme also retains stable performance across BLE, Wi-Fi, and Zigbee infrastructures when trained under a representative path-loss setting without retraining and maintains sub-meter accuracy under mild shadow fading. These results confirm that the proposed scheme can improve positioning accuracy and search efficiency for indoor wireless localization. Full article

(This article belongs to the Special Issue Advanced Indoor Localization Technologies: From Theory to Application)

► Show Figures

Figure 1

28 pages, 8600 KB

Open AccessArticle

A Reproducible FPGA-to-Silicon Verification Methodology for an Embedded SoC Platform in 28 nm CMOS

by Hyeseung Sun and Kwangki Ryoo

Electronics 2026, 15(10), 2202; https://doi.org/10.3390/electronics15102202 - 20 May 2026

Viewed by 411

Abstract

Many System-on-Chip (SoC) studies rely solely on simulation and tool-based results, encountering unexpected failures during post-silicon validation. In particular, silicon-level demonstrations of Hardware/Software (HW/SW) functional equivalence, which confirms that an FPGA-validated design operates identically on an ASIC with the same firmware, remain extremely [...] Read more.

Many System-on-Chip (SoC) studies rely solely on simulation and tool-based results, encountering unexpected failures during post-silicon validation. In particular, silicon-level demonstrations of Hardware/Software (HW/SW) functional equivalence, which confirms that an FPGA-validated design operates identically on an ASIC with the same firmware, remain extremely rare. This work proposes a reproducible FPGA-to-silicon verification methodology that establishes HW/SW functional equivalence at the silicon level by applying an identical firmware source code, device driver, and memory map to both platforms. The methodology is validated on an Arm Cortex-M0-based SoC platform fabricated in Samsung 28 nm Low Power Plus (LPP) CMOS technology with a dual Inter-Integrated Circuit (I2C) interface. The fabricated chip integrates two 64KB on-chip memories within a core area of 653

μ

m × 769

μ

m, operates at 125 MHz, and consumes 17.5 mW at the optimal operating point of 1.0 V. The primary contributions are: (1) a reproducible FPGA-to-silicon HW/SW functional equivalence verification methodology based on shared firmware source code, device driver, and memory map across both platforms, (2) silicon-measurement-based performance characterization with verified experimental data, (3) a reproducible design methodology documenting the complete flow from FPGA verification through ASIC fabrication, including static timing closure, place-and-route, and physical verification, and (4) an extensible SoC platform architecture enabling researchers to integrate and validate their own Intellectual Property (IP) via Advanced High-performance Bus (AHB) and I2C interfaces. Full article

(This article belongs to the Topic Advanced Integrated Circuit Design and Application)

► Show Figures

Figure 1

26 pages, 6226 KB

Open AccessArticle

Three-Stage Stochastic Optimal Operation and Game-Theoretic Benefit Allocation Strategy for a PV-Storage Virtual Power Plant Under Multi-Market Synergy

by Xiang Li, Gaoquan Ma, Bangcan Wang, Na Cai, Junwei Bao, Zishi Wang, Xuan Yang, Qian Ai and Chenyang Zhao

Electronics 2026, 15(10), 2201; https://doi.org/10.3390/electronics15102201 - 20 May 2026

Viewed by 267

Abstract

To address the output volatility of distributed photovoltaics, the low utilization efficiency of energy storage resources, and the challenge of optimal revenue for PV-storage virtual power plants (VPPs) in multi-market environments, this paper proposes a three-stage stochastic optimal operation strategy for PV-storage VPPs [...] Read more.

To address the output volatility of distributed photovoltaics, the low utilization efficiency of energy storage resources, and the challenge of optimal revenue for PV-storage virtual power plants (VPPs) in multi-market environments, this paper proposes a three-stage stochastic optimal operation strategy for PV-storage VPPs under multi-market synergy and develops a benefit allocation model based on the Nash–Harsanyi bargaining game. A Monte Carlo simulation was adopted to capture the uncertainties of market electricity prices and PV power output, and the stochastic dual-dynamic-programming (SDDP) algorithm was employed to solve the three-stage optimization framework consisting of day-ahead bidding, real-time optimization, and real-time frequency regulation. Bargaining power was quantified from four dimensions—the marginal contribution rate, PV prediction accuracy, energy storage capacity, and utilization rate—to establish a fair and reasonable internal benefit allocation mechanism. Case studies verified that the proposed method improved the single-day market revenue by up to 20.79% compared with traditional operation modes, achieved a near-zero curtailment rate for distributed PV, and maintained frequency regulation performance scores above 0.4 at all times. The benefits of all investment entities in the alliance increased by 3.36–99.43%, significantly enhancing the multi-market profitability of PV-storage VPPs and the stability of alliance cooperation. Full article

► Show Figures

Figure 1

Journal Menu

Journal Browser

Electronics, Volume 15, Issue 10 (May-2 2026) – 260 articles

Further Information

Guidelines

MDPI Initiatives

Follow MDPI