Jingbin Hao , Jianhua Hu , Ci Liang , Xinhua Liu , Xiaokai Sun , Dezheng Hua
{"title":"A lightweight network for driver gesture recognition","authors":"Jingbin Hao , Jianhua Hu , Ci Liang , Xinhua Liu , Xiaokai Sun , Dezheng Hua","doi":"10.1016/j.compeleceng.2025.110577","DOIUrl":"10.1016/j.compeleceng.2025.110577","url":null,"abstract":"<div><div>In the domain of smart car technology, driver gesture recognition models often face the challenge of balancing efficiency with accuracy, typically requiring substantial computational resources and memory. To address these challenges, this paper introduces a lightweight network structure, Intelligent Cockpit Gesture recognition-You Only Look Once version 7 (ICG-YOLOv7), based on YOLOv7. The contributions of this study include proposing an improved Convolutional Block Attention Module (CBAM) to enhance feature extraction and devising an unstructured pruning method to compress the model. Specifically, to achieve continuous feature recalibration, a residual connection structure is designed using the concept of residual learning. Moreover, to prevent the loss of feature information, a Four Conv BN SiLU Module (FCBSM) structure is designed and integrated into YOLOv7, which retains important original feature information. Furthermore, a self-made cockpit environment gesture dataset is developed, on which comparative experiments and ablation experiments are conducted. An unstructured pruning method based on layer adaptive sparsification is then designed to compress the improved network, thereby enhancing the model's generalization ability and significantly reducing its computational effort, number of parameters, and size. Experimental results demonstrate that the proposed approach effectively integrates neural networks and deep learning into vehicle development. Finally, a gradient-weighted class activation map (Grad-CAM) method is utilized to visualize and analyze the pruned model, thereby enhancing the interpretability of the gesture recognition model and facilitating its deployment in vehicle systems to accomplish the task of driver gesture recognition in cockpit environments.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"127 ","pages":"Article 110577"},"PeriodicalIF":4.9,"publicationDate":"2025-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144861256","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":"Protection of high voltage transmission line connected large scale solar photovoltaic plant using green anaconda optimized machine learning method","authors":"Sikander Singh, Paresh Kumar Nayak","doi":"10.1016/j.compeleceng.2025.110618","DOIUrl":"10.1016/j.compeleceng.2025.110618","url":null,"abstract":"<div><div>Transmission lines are widely used engineering systems designed to transport large amounts of power across a country from one location to the furthest points in the other direction. Transmission line protection is a significant concern in power system engineering since transmission lines account for the vast majority of power system faults (85–87%). This paper presents a hybrid artificial neural network and support vector machine technique to detect and classify faults on a transmission line. MATLAB/Simulink software is utilized to simulate various fault and operating conditions on high-voltage transmission lines. The empirical wavelet transform is used to decompose fault transients due to its capability to extract information from the transient signal. This method's optimal hyper-parameter selection is obtained by using the green Anaconda optimization algorithm. The results showed that the proposed technique acquired a high accuracy of 99.86%, precision of 99.23%, sensitivity of 99.23%, specificity of 99.92%, recall of 99.23%, F1-score of 99.23%, Mean Square Error of 0.187, Root Mean Square Error of 0.433 and Mean Absolute Error of 0.031. The proposed technique has been shown to be highly efficient and accurate, making it a reliable classifier for fault identification.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"127 ","pages":"Article 110618"},"PeriodicalIF":4.9,"publicationDate":"2025-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144861253","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}
Jayasri Kotti , M. Belsam Jeba Ananth , Rajeshkannan Regunathan
{"title":"Hybrid EfficientQNet for brain tumor detection using MRI images","authors":"Jayasri Kotti , M. Belsam Jeba Ananth , Rajeshkannan Regunathan","doi":"10.1016/j.compeleceng.2025.110601","DOIUrl":"10.1016/j.compeleceng.2025.110601","url":null,"abstract":"<div><div>Brain tumors (BT) result from the abnormal growth of brain cells and are associated with high morbidity and mortality rates. Malignant tumors spread quickly, while early-stage tumors grow slowly. Detection is challenging due to their varied sizes and shapes. To address this, EfficientQNet is proposed for effective BT detection using MRI. The process starts with preprocessing using the Fuzzy Local Information C-Means clustering model (FLICM) for Region of Interest (ROI) extraction and skull stripping, followed by SegNet for segmentation and image augmentation. Subsequently, texture features such as Correlation, Angular Second Moment, Inverse Difference Moment, Contrast, and Discrete Cosine Transform (DCT) with Fuzzy Local Binary Pattern (FLBP) are extracted. Finally, EfficientQNet is used for detection. Here, EfficientQNet combines the existing technologies, such as EfficientNet-B3-attn-2 with Deep Q-Learning to optimize layer configurations, achieving superior performance in brain tumor detection. Furthermore, EfficientQNet achieved an accuracy of 90.3 %, sensitivity of 93.2 %, specificity of 91.2 %, precision of 92.4 %, and F1-score of 92.8 %, with a loss of 9.7 %. The accuracy improvement over Fine-tuned Visual Geometry Group 16 (Fine-tuned VGG16), EfficientNet B0, Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM), Ultra-Light Brain Tumor Detection (UL-BTD), Deep Learning-based Brain Tumor Detection and Classification using Magnetic Resonance Imaging (DLBTDC-MRI), and Parallel Deep Convolutional Neural Network (PDCNN) methods is 12.2 %, 9.63 %, 6.31 %, 5.42 %, 2.65 %, and 2.54 %, respectively.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"127 ","pages":"Article 110601"},"PeriodicalIF":4.9,"publicationDate":"2025-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144861254","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}
Gunavathie M A , A. Arivarasi , Puneet Kumar Aggarwal , Krishna Prakash Arunachalam
{"title":"Blockchain-enabled 6G wireless network securities using multi-relational graph attention disentangled cascaded graph convolution networks","authors":"Gunavathie M A , A. Arivarasi , Puneet Kumar Aggarwal , Krishna Prakash Arunachalam","doi":"10.1016/j.compeleceng.2025.110617","DOIUrl":"10.1016/j.compeleceng.2025.110617","url":null,"abstract":"<div><div>The emergence of sixth-generation (6 G) wireless networks introduces transformative capabilities such as ultra-low latency, massive device connectivity, and real-time intelligent services. However, the inherently distributed and heterogeneous architecture of 6 G significantly escalates vulnerabilities to cyber threats, unauthorized access, and data breaches across large-scale decentralized infrastructures. Existing security mechanisms are inadequate in modeling complex multi-relational interactions among network entities and often fail to ensure trust, transparency, and tamper-resistance at scale. To address these challenges, this research proposes a novel Blockchain-Enabled 6 G Wireless Network Security framework integrating Multi-Relational Graph Attention and Disentangled Cascaded Graph Convolution Network (Multi-RACG) model. This hybrid graph-based model captures high-order relational dependencies while disentangling semantic features across graph channels to enable precise, context-aware intrusion detection. A Dandelion Optimization Algorithm (DOA) is employed to fine-tune model parameters and optimize network architecture, ensuring rapid convergence and reduced computational overhead. Additionally, a Proof-of-Work-Based Weighted Mining (PoWBWM) consensus protocol strengthens blockchain operations by incorporating dynamic trust metrics, enhancing data integrity and resilience against malicious manipulation. Experimental results demonstrate the framework's superiority, achieving 99.9 % detection accuracy with minimal false positives and computational loss, positioning it as a highly scalable and intelligent security solution for future 6 G ecosystems.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"127 ","pages":"Article 110617"},"PeriodicalIF":4.9,"publicationDate":"2025-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144852940","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":"SeAF-PrS: Effective and adaptive priority scheduling scheme for improving the quality of service in wireless body area networks","authors":"Shaik Afzal Ahammed M S , Manjaiah D H","doi":"10.1016/j.compeleceng.2025.110616","DOIUrl":"10.1016/j.compeleceng.2025.110616","url":null,"abstract":"<div><div>The Wireless Body Area Networks (WBAN) hold immense capability in medical services with the significant benefits of remote patient health monitoring. Since data transmission is the major step for real-time diagnosis, conventional healthcare applications exhibited lower adaptability and continuous transmission of data resulting in higher energy losses. Therefore, the research focuses on improving the quality of services through an efficient technique Known as Sensitive and Adaptive Flitter optimized priority scheduler (SeAF-PrS). The data transmission is performed when deviations from the original signal levels is noticed. The application of the Sensitive and Adaptive Flitter Optimization (SeAFO) algorithm improves data scheduling by assisting in the selection of high-priority signals for patients that require immediate diagnosis. In addition, the modified shift rows (MSR) encryption scheme provides higher security through the encrypting of data packets as well as the authentication check allows only the verified users to access the patient data. The proposed approach is compared against the traditional healthcare monitoring systems which reveal improved efficiency with a throughput of 0.74 kbps, minimum delay of 0.100 ms, and energy loss of 7.37 J for 10 nodes with 250 users.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"127 ","pages":"Article 110616"},"PeriodicalIF":4.9,"publicationDate":"2025-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144858292","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}
Karam Khairullah Mohammed , Saad Mekhilef , Marizan Mubin , Ibrahim Ismael Alnaib , Houssem Jerbi , Mehdi Seyedmahmoudian
{"title":"A novel hybrid global maximum power point tracking method based on partial shading mitigation for grid connected photovoltaic systems","authors":"Karam Khairullah Mohammed , Saad Mekhilef , Marizan Mubin , Ibrahim Ismael Alnaib , Houssem Jerbi , Mehdi Seyedmahmoudian","doi":"10.1016/j.compeleceng.2025.110619","DOIUrl":"10.1016/j.compeleceng.2025.110619","url":null,"abstract":"<div><div>Maximizing the power output from photovoltaic (PV) panels is a critical concern in PV system operation, particularly in partial shading conditions (PSCs). While the global maximum power point (GMPP) might not be captured by conventional methods. Consequently, several optimization techniques have been presented for following the GMPP. On the other hand, partial shading cannot be distinguished from uniform shading using optimization techniques. A new hybrid MPPT approach is presented in this paper to overcome this limitation and prevent an unneeded search area of the entire P-V curve via the USCs. If the system is subjected to uniform shading conditions (USCs), an ANFIS technique is suggested for finding the best power point while the irradiance sensor is eliminated in order to reduce costs. Moreover, a modified hybrid levy rat swarm optimization technique (LRSO) has been applied to mitigate the partial shading in order to enhance the convergence speed. Using dSPACE and Micro Lab Box with a sampling time of 0.05 s, the proposed strategy was implemented experimentally. The results illustrated that the suggested strategy had been successfully applied, with uniform and PSCs having a high efficiency of 99.7 % and an average tracking time of 0.34 s. The effectiveness of the suggested approach has been evaluated by contrasting it with the most significant approaches in this domain. Additionally, the suggested approach has been verified with a grid connected to demonstrate the tracking ability under different shading patterns with lower oscillation and fast tracking efficiency.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"127 ","pages":"Article 110619"},"PeriodicalIF":4.9,"publicationDate":"2025-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144842280","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}
Riccardo La Cesa, Simone Acciarito, Gian Carlo Cardarilli, Luca Di Nunzio, Marco Re, Sergio Spanò, Cristian Valenti
{"title":"High efficiency hardware implementation of a ZP-OTFS modulator for next generation high-mobility wireless systems","authors":"Riccardo La Cesa, Simone Acciarito, Gian Carlo Cardarilli, Luca Di Nunzio, Marco Re, Sergio Spanò, Cristian Valenti","doi":"10.1016/j.compeleceng.2025.110614","DOIUrl":"10.1016/j.compeleceng.2025.110614","url":null,"abstract":"<div><div>Orthogonal Time Frequency Space (OTFS) is an emerging transmission technology poised to become the dominant paradigm in High-mobility wireless communications and a potential successor to the current Orthogonal Frequency Division Multiplexing (OFDM) technology. In this article, we propose a Zero Padded OTFS (ZP-OTFS) modulator that incorporates the logic required for pilot insertion, zero padding, cyclic prefix, and payload data management. Additionally, the Register Transfer Level (RTL) architecture on Field-Programmable Gate Arrays (FPGAs) and Application-Specific Integrated Circuits (ASIC) will be discussed, with particular focus on the first one. A comparison with state-of-the-art architectures is also conducted to evaluate its performance. The implementation results demonstrate that the proposed architecture offers significant advantages in terms of resource utilization, achieving an improvement percentage up to about 90%, being capable of attaining clock frequencies up to over 600 MHz while maintaining low power consumption. Furthermore, our study explores the relationship between latency, dynamic power consumption, and resource utilization as functions of frame size, proposing design parameters that account for both transmission channel characteristics and hardware constraints, allowing the development of a real OTFS-based transceiver system. Finally, a complete transmission system is presented by integrating the proposed modulator into a complete transmission chain. The transmitted signal is analyzed to validate the effectiveness of the proposed approach.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"127 ","pages":"Article 110614"},"PeriodicalIF":4.9,"publicationDate":"2025-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144809330","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":"Explainable artificial intelligence-based identification of the localized events in imagined speech electroencephalogram","authors":"Arun Balasubramanian, Kartik Pandey, Gautam Veer, Debasis Samanta","doi":"10.1016/j.compeleceng.2025.110608","DOIUrl":"10.1016/j.compeleceng.2025.110608","url":null,"abstract":"<div><div>Localizing events in Imagined Speech Electroencephalogram (IS-EEG) signals is considered vital for analyzing significant neural activity that may otherwise be obscured by non-event segments. In this study, a Deep Learning framework is introduced, utilizing a hybrid Long Short-Term Memory-Convolutional Neural Network (LSTM-CNN) for effective IS-EEG signal recognition and event localization. A key component of this framework is the implementation of Gradient-weighted Class Activation Mapping (Grad-CAM) as an explainability technique, which is used to generate heatmaps that highlight critical regions in the IS-EEG signals and validate the model’s classification decisions. The IS-EEG signals were used to train and validate the proposed LSTM-CNN model, which achieved an accuracy of 97.24%. Subsequently, the IS-EEG signals were analyzed to estimate the theoretical duration of localized events, which was found to lie between 0.9 and 1.8 s. The trained model and the derived duration estimates were then utilized to determine an optimal threshold of 0.28 based on the average performance with masking. Furthermore, masking of non-critical segments led to an accuracy improvement to 99.17%, while masking of essential regions resulted in poor performance. The robustness of the model was also evaluated by introducing controlled levels of noise, with the Grad-CAM heatmaps demonstrating reasonable consistency in the presence of noise.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"127 ","pages":"Article 110608"},"PeriodicalIF":4.9,"publicationDate":"2025-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144780786","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":"Enhancing blockchain security against quantum threats through integration of post-quantum cryptographic algorithms","authors":"Revathi K , Suganthi K","doi":"10.1016/j.compeleceng.2025.110610","DOIUrl":"10.1016/j.compeleceng.2025.110610","url":null,"abstract":"<div><div>The fast progress in quantum computing presents a substantial menace to existing cryptographic systems, requiring post-quantum cryptographic techniques that can withstand quantum attacks. This research aims to increase the security of blockchain ecosystems against quantum-related threats by incorporating robust post-quantum cryptographic methods. We provide a new encryption technique that relies on the Key Encryption process, namely, Crystals-Kyber, in conjunction with an effective digital signature methodology centered on Lattice-based techniques: Falcon and Crystal-Dilithium. The techniques are meticulously included in the Hyperledger Fabric 4.0 platform, showcasing improved security and durability in cryptocurrency exchanges built around blockchain technology. We have effectively achieved a cryptographic resistance above 90 % against quantum assaults, greatly enhancing the protection of blockchain operations. The proposed strategy makes cryptographic design more robust to quantum attacks and establishes a basis for secure blockchain technology in the quantum era.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"127 ","pages":"Article 110610"},"PeriodicalIF":4.9,"publicationDate":"2025-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144780787","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}
Chin-Ling Chen , Yu-Jung Lin , Hsing-Chung Chen , Yong-Yuan Deng , Der-Chen Huang , Ling-Chun Liu
{"title":"Application of Blockchain in Marriage Registration System","authors":"Chin-Ling Chen , Yu-Jung Lin , Hsing-Chung Chen , Yong-Yuan Deng , Der-Chen Huang , Ling-Chun Liu","doi":"10.1016/j.compeleceng.2025.110526","DOIUrl":"10.1016/j.compeleceng.2025.110526","url":null,"abstract":"<div><div>The decline in marriage rates in the international community in 2020 as a result of Covid-19 has led to significant losses in the marriage industry in many countries. In addition to the problem of marriage rates, when a couple divorces, there may also be legal disputes over marital property, maintenance, or child custody. We propose a blockchain marriage registration system that combines Hyperledger Fabric and IPFS (Interplanetary File System) to address these issues. The contributions of our system include the use of smart contracts to avoid disputes, the use of multiple signatures to ensure the validity of prenuptial agreements, and the traceability and verifiability of data through the features of Hyperledger Fabric. Our system can provide a secure, convenient, and cost-effective method of registering marriages. The security and performance analysis shows the feasibility of this study in enabling online marriage registration agreements.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"127 ","pages":"Article 110526"},"PeriodicalIF":4.9,"publicationDate":"2025-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144780785","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}