Computers & Electrical Engineering最新文献

{"title":"Stress detection through wearable EEG technology: A signal-based approach","authors":"Rakesh Kumar Rai,&nbsp;Dushyant Kumar Singh","doi":"10.1016/j.compeleceng.2025.110478","DOIUrl":"10.1016/j.compeleceng.2025.110478","url":null,"abstract":"<div><div>Accurate and non-invasive stress detection is critical for mental health monitoring and early intervention. Among various physiological signals, electroencephalography (EEG) offers a unique advantage as it captures direct neural activity, making it more responsive to cognitive and emotional stress compared to peripheral markers like heart rate or skin conductance. In this work, we propose a novel EEG-based stress detection framework that combines Fuzzy Attention-based Fully Convolutional Network (FA-FCN) for dynamic signal segmentation, Hybrid Wavelet-Short-Time Fourier Transform (HW-STFT) for rich time–frequency feature extraction, Recursive LASSO (RLASSO) for optimal feature selection, and an Adam-Optimized Sequential Generative Adversarial Network (AOS-GAN) for robust classification. Each component is specifically designed to address the limitations of existing methods FA-FCN enhances signal relevance by focusing on stress-related EEG regions, while HW-STFT captures transient frequency shifts with high resolution. RLASSO improves computational efficiency by reducing dimensionality, and AOS-GAN enhances classification in imbalanced conditions using adversarial learning. Our model achieves an accuracy of 94%, significantly outperforming other state-of-the-art methods, which typically report accuracies around 85%–89%. This demonstrates the model’s strong potential for real-world deployment in high-stress environments such as emergency response, cognitive workload monitoring, or workplace mental health systems. Future work will focus on validating this approach across larger and more diverse EEG datasets, enabling real-time deployment on edge devices, and exploring multimodal integration with other physiological signals for holistic stress assessment.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"126 ","pages":"Article 110478"},"PeriodicalIF":4.0,"publicationDate":"2025-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144298696","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A neural cantonese speech converter using QCA for nanocomputing","authors":"Angshuman Khan ,&nbsp;Rohit Kumar Shaw ,&nbsp;Ali Newaz Bahar","doi":"10.1016/j.compeleceng.2025.110536","DOIUrl":"10.1016/j.compeleceng.2025.110536","url":null,"abstract":"<div><div>This research explores the pioneering integration of Quantum-dot Cellular Automata (QCA) for designing a combinational neuro-fuzzy logic circuit within a feedforward neural network-based Cantonese speech converter. Cantonese, a tonal language with intricate phonetic structures, presents substantial speech recognition and synthesis challenges. The proposed QCA-based speech conversion circuit leverages the quantum mechanical tunnelling properties of quantum dots to achieve ultra-fast processing, minimal power dissipation, and enhanced energy efficiency, making it a highly suitable alternative to conventional speech recognition systems. The architectural design ensures precise phonetic recognition, tone preservation, and high intelligibility, optimizing real-time speech processing. Simulation results confirm that the circuit consumes only 2.338 nanowatts of power, demonstrating a 45 % enhancement in energy-delay cost compared to conventional speech recognition systems. Additionally, the proposed system achieves excellent recognition accuracy for frequently used Cantonese keywords in eBook reading applications. The study underscores QCA’s transformative potential in low-power nanocomputing, positioning it as a breakthrough technology for efficient, high-speed, and sustainable speech processing in next-generation natural language interfaces.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"126 ","pages":"Article 110536"},"PeriodicalIF":4.0,"publicationDate":"2025-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144306961","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"3DSPECSN: Adaptive 3D spatial patch based siamese network for robust hyperspectral image analysis","authors":"Ravikant Kumar Nirala ,&nbsp;Gautam Kumar ,&nbsp;Rishav Singh ,&nbsp;Chandra Prakash","doi":"10.1016/j.compeleceng.2025.110520","DOIUrl":"10.1016/j.compeleceng.2025.110520","url":null,"abstract":"<div><div>The classification of hyperspectral image (HSI) remains a challenging task due to high dimensionality, spatial correlation of the features and variability of the HSI data sources. This research introduces 3D Spatial Patch Extraction (3DSPE) technology which uses a Siamese Network framework to successfully capture multi-dimensional spectral-spatial data patterns for enhanced unmixing outcomes. The 3DSPE technique generates superior multiple spectral mixture signature detection due to its fusion of spectral and spatial features. The performance increases for endmember recovery and spatial distribution analysis when spectral patterns combine with spatial dependencies through the implementation of a Siamese Network architecture. A trainable image stratification approach for hyperspectral data increases speed of convergence and limits overfitting and builds generalized performance through adaptive optimization techniques at lower processing times for large datasets. The proposed framework shows strong performance in terms of Overall Accuracy (OA), Average Accuracy (AA), and Kappa Coefficient (κ) measurements from Indian Pines, Pavia University, and Salinas tests which attains nearly 99 % accuracies. The methodology not only achieved high accuracy, also enabling stable development of improved hyperspectral unmixing models that deliver better remote sensing results through improved precision along with enhanced flexibility and scalability. The approach provides a scalable efficient solution which can be applied to multiple remote sensing tasks, land-cover analyses and resource monitoring operations with potential seamless integration of other spectral analysis tools.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"126 ","pages":"Article 110520"},"PeriodicalIF":4.0,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144298695","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Image encryption architecture exploiting finite-precision error for System-on-Chip and programmable logic devices","authors":"Arthur M. Lima ,&nbsp;Lucas G. Nardo ,&nbsp;Erivelton Nepomuceno ,&nbsp;Janier Arias-Garcia ,&nbsp;Jones Yudi","doi":"10.1016/j.compeleceng.2025.110482","DOIUrl":"10.1016/j.compeleceng.2025.110482","url":null,"abstract":"<div><div>Security in interconnected System-on-Chip (SoC) devices is increasingly critical as their adoption expands, driving demand for cryptographic solutions that are both robust and cost-effective. Hence, this work addresses the challenges of implementing chaos-based encryption schemes on digital hardware by introducing a novel image encryption approach based on the mathematical complexity of Chua’s circuit. It exploits finite-precision computational errors as a noise-like entropy source within the encryption process. Using High-Level Synthesis (HLS) with C/C<span><math><mrow><mo>+</mo><mo>+</mo></mrow></math></span>, we implement two parallelized Digital Chua’s Circuits (DCCs), each numerically resolved via the fourth-order Runge–Kutta method and using distinct natural interval extensions. The lower-bound error is then extracted from the chaotic circuits and used to generate the final keystream. Optimized for IEEE-754 floating-point arithmetic, our stream cipher enables efficient hardware synthesis and achieves high cryptographic performance, reaching the target error threshold within a few clock cycles. The proposed image encryption design passes the NIST SP 800-22 test suite and achieves strong NPCR and UACI scores (over 99.6 % and 33.4 %, respectively), demonstrating efficiency and robustness against common attacks. Moreover, implemented on a ZCU104 evaluation board, the architecture can process one pixel in just 70 clock cycles with a low energy cost of 0.42 <span><math><mrow><mi>μ</mi><mi>J</mi></mrow></math></span> using only 48 DSPs (2.48 % of available resources). By bridging chaotic dynamics and practical SoC architectures, this work provides an encryption solution that balances hardware consumption and energy efficiency.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"126 ","pages":"Article 110482"},"PeriodicalIF":4.0,"publicationDate":"2025-06-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144291477","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Stealthy Backdoor Attacks on Graph Neural Networks via Relational Constraint Modeling","authors":"Deshan Yang ,&nbsp;Limin Pan ,&nbsp;Jinjie Zhou ,&nbsp;Senlin Luo ,&nbsp;Peng Luan","doi":"10.1016/j.compeleceng.2025.110483","DOIUrl":"10.1016/j.compeleceng.2025.110483","url":null,"abstract":"<div><div>Graph Neural Networks (GNNs) have achieved significant success in tasks like node and graph classification, yet they remain vulnerable to backdoor attacks. Traditional attack methods often rely on node feature manipulation, neglecting the structural relationships within graphs, which makes their triggers more detectable. To address this limitation, we propose a <u>R</u>elationally <u>C</u>onstrained <u>C</u>onditional GAN <u>B</u>ackdoor <u>A</u>ttack (RCCBA). Our method employs a hybrid expert model that combines diverse graph neural network architectures to capture both feature and structural information, thus enabling more robust trigger node selection via a centrality-based rule that identifies nodes with minimal impact on neighboring nodes. Additionally, relationship constraints ensure that the triggers generated by the conditional GAN closely mimic the original graph, enhancing imperceptible and attack success. Experimental results demonstrate that RCCBA achieves an average attack success rate exceeding 90% with a minimal poisoning rate of less than 0.1%, and successfully circumvents pruning-based and outlier detection defenses. The real-world implications of this work are significant for domains such as social networks, recommendation systems, and fraud detection, and our findings highlight the need for future defense strategies that address both feature and structural vulnerabilities in GNN security.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"126 ","pages":"Article 110483"},"PeriodicalIF":4.0,"publicationDate":"2025-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144279906","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Secure and optimized drone swarm operations with decentralized Adaptive Differential Evolution","authors":"Usama Arshad ,&nbsp;Zahid Halim","doi":"10.1016/j.compeleceng.2025.110487","DOIUrl":"10.1016/j.compeleceng.2025.110487","url":null,"abstract":"<div><div>Efficient drone swarm management requires real-time adaptive optimization and secure decentralized communication to ensure robust performance in dynamic environments. Traditional optimization methods such as Particle Swarm Optimization (PSO) and Genetic Algorithms (GA) suffer from premature convergence and lack the adaptability required for large-scale swarm coordination. Similarly, centralized communication frameworks introduce security vulnerabilities, including single points of failure and susceptibility to cyberattacks. This study presents a novel integration of Adaptive Differential Evolution (ADE) and blockchain technology, leveraging ADE’s dynamic parameter tuning to improve swarm intelligence while utilizing blockchain’s decentralized ledger to secure inter-drone communication. The proposed framework was evaluated through extensive simulations on drone swarms ranging from 20 to 200 drones, demonstrating a 27% improvement in convergence speed and a 35% increase in task efficiency compared to PSO-based methods. Blockchain integration ensured 99.3% data integrity, preventing unauthorized modifications and cyber threats such as man-in-the-middle attacks and data corruption attempts. Energy consumption analysis indicated that ADE reduced power usage by 18% compared to traditional heuristic approaches. Additionally, adversarial testing revealed that denial-of-service (DoS) resilience improved by 42% due to the blockchain’s consensus validation mechanisms. These results highlight the feasibility of secure and adaptive drone swarm management, making it suitable for real-world applications in disaster response, autonomous surveillance, and smart logistics.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"126 ","pages":"Article 110487"},"PeriodicalIF":4.0,"publicationDate":"2025-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144279909","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Robust electric load forecasting through ensemble learning: A stacking approach with empirical mode decomposition and transfer learning","authors":"Mohit Choubey,&nbsp;Rahul Kumar Chaurasiya,&nbsp;J.S. Yadav","doi":"10.1016/j.compeleceng.2025.110511","DOIUrl":"10.1016/j.compeleceng.2025.110511","url":null,"abstract":"<div><div>Recent advancements in artificial intelligence (AI) have significantly influenced various disciplines, including electricity demand forecasting within power systems. This study introduces a methodology that emphasizes on predicting total energy consumption rather than limiting the scope to specific sectors. By integrating Empirical Mode Decomposition (EMD) with Transfer Learning (TL), the proposed model enhances the accuracy and generalization capability of ensemble models. The methodology achieves this by decomposing input data features into linear and nonlinear components, that optimized the resource allocation, encourages to use simpler models, and mitigating overfitting risks. TL further strengthens the model's adaptability, allowing it to accommodate diverse load patterns from multiple sectors. This adaptability facilitates the integration of sector-specific load into a comprehensive framework, leading to more accurate predictions of net load demand for power station operations. Experimental evaluations validated the model’s superior performance, achieving a Mean Absolute Error (MAE) of 82.58, a Root Mean Square Error (RMSE) of 95.375, and a Mean Absolute Percentage Error (MAPE) of 1.06 %, this contributes 2.85 % improvement over conventional methods. The proposed model further validated on the widely utilized New South Wales (NSW) dataset revealed an MAE of 96.23, an RMSE of 102.16, and a MAPE of 1.02 %. A predictive accuracy of 98.98 % was achieved using the proposed model, which outperforms state-of-the-art models like N-BEATS and DLinear and other advanced ensemble techniques. Statistical tests, such as the Friedman test and Nemenyi post-hoc analysis, confirm the strength of the proposed model, regularly placing it as the top performer among the other methods. This enables the proposed model to be applicable in the real world by predicting the energy consumption at a broader level rather than at a sector level. Moreover, the proposed model’s outcomes illustrate, that the framework is reliable and generalizable in nature which leads to better resource optimization and promotion of energy efficiency practices in load forecasting can be achieved.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"126 ","pages":"Article 110511"},"PeriodicalIF":4.0,"publicationDate":"2025-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144279923","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Smart residential electric vehicle charging and discharging scheduling via multi-agent asynchronous-updating deep reinforcement learning","authors":"Qiang Zhao ,&nbsp;Chengwei Xu ,&nbsp;Chuan Sun ,&nbsp;Yinghua Han","doi":"10.1016/j.compeleceng.2025.110473","DOIUrl":"10.1016/j.compeleceng.2025.110473","url":null,"abstract":"<div><div>Despite the increasing penetration of electric vehicle (EV), significant efforts are still required to transition towards a low-carbon future while balancing economy and power system stability. Coordinating EV charging and discharging scheduling in residential areas faces challenges due to uncertainties in EV owners’ commuting behaviors, complex energy demands, and unpredictable power information. This paper formulates the <strong>R</strong>esidential <strong>EV C</strong>harging and <strong>D</strong>ischarging <strong>S</strong>cheduling (REV-CDS) as a Markov Decision Process with an unknown transition function. <strong>M</strong>ulti-<strong>A</strong>gent-<strong>A</strong>synchronous-<strong>S</strong>oft-<strong>A</strong>ctor-<strong>C</strong>ritic (MAASAC) algorithm is proposed to solve the Markov Decision Process. The proposed method employs the asynchronous updating process instead of synchronous updating, which allows agents to maintain consistent policy update direction, enabling better learning of a charging and discharging strategy to improve coordination among EVs and highly align with the overall scheduling goals in the REV-CDS environment. Finally, several numerical studies were conducted to compare the proposed approach with classical multi-agent reinforcement learning methods. The studies demonstrate the effectiveness of improvement in minimizing charging costs, reducing carbon emissions, alleviating charging anxiety, and preventing transformer overload.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"126 ","pages":"Article 110473"},"PeriodicalIF":4.0,"publicationDate":"2025-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144271026","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A decision framework for privacy-preserving synthetic data generation","authors":"Pablo Sanchez-Serrano,&nbsp;Ruben Rios,&nbsp;Isaac Agudo","doi":"10.1016/j.compeleceng.2025.110468","DOIUrl":"10.1016/j.compeleceng.2025.110468","url":null,"abstract":"<div><div>Access to realistic data is essential for various purposes, including training machine learning models, conducting simulations, and supporting data-driven decision making across diverse domains. However, the use of real data often raises significant privacy concerns, as it may contain sensitive or personal information. Generative models have emerged as a promising solution to this problem by generating synthetic datasets that closely resemble real data. Nevertheless, these models are typically trained on original datasets, which carries the risk of leaking sensitive information. To mitigate this issue, privacy-preserving generative models have been developed to balance data utility and privacy guarantees. This paper examines existing generative models for synthetic tabular data generation, proposing a taxonomy of solutions based on the privacy guarantees they provide. Additionally, we present a decision framework to aid in selecting the most suitable privacy-preserving generative model for specific scenarios, using privacy and utility metrics as key selection criteria.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"126 ","pages":"Article 110468"},"PeriodicalIF":4.0,"publicationDate":"2025-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144271028","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Fault diagnosis of multimodal feature fusion convolutional neural network based on differential evolution optimization","authors":"Min Ji ,&nbsp;Shaofeng Zhang ,&nbsp;Jinghui Yang","doi":"10.1016/j.compeleceng.2025.110518","DOIUrl":"10.1016/j.compeleceng.2025.110518","url":null,"abstract":"<div><div>To overcome the performance degradation caused by data scarcity in bearing fault diagnosis -where acquiring sufficient fault samples proves particularly challenging, an effective fault detection approach has been developed. To utilize fault feature information more effectively, we have presented the vibration signal of bearings in a multimodal manner. Two forms of data, time series data in one dimension and gray images in two dimensions, are used as sample data. According to the characteristics of these two kinds of data, two modules are used for targeted feature extraction, and obtain the local frequency, vibration mode characteristics and global dependence of the original vibration signal. A multimodal feature fusion convolutional neural network (AMFCNN) is constructed. Simultaneously, the differential evolution algorithm undergoes optimization to ascertain the optimal combination of key hyper-parameters for the model. Three datasets are used for comparative and variable operating condition experiments to validate this method. The experimental results show that AMFCNN achieves an average accuracy of 91.56 % with only 21.43 % of the data used, which is &gt;13 % improvement over the unimodal approach. AMFCNN is constructed from the two starting points of optimizing the vibration signal data and improving the feature extraction ability of the model, effectively avoiding the problems of over fitting and insufficient feature extraction ability of most fault diagnosis models.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"126 ","pages":"Article 110518"},"PeriodicalIF":4.0,"publicationDate":"2025-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144271029","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}