Computers & Electrical Engineering最新文献

{"title":"Development of MD-LSTM for malicious detection and hybrid coati and dolphin swarm optimization for QoS-aware multicast routing in MANET environment","authors":"Sanjaya Kumar Sarangi ,&nbsp;Rasmita Lenka ,&nbsp;Arabinda Nanda ,&nbsp;Bhavani Sankar Panda","doi":"10.1016/j.compeleceng.2025.110086","DOIUrl":"10.1016/j.compeleceng.2025.110086","url":null,"abstract":"<div><div>Mobile Ad-hoc Networks (MANETs) also known as infrastructure-less network consist of mobile nodes that create a form of internet connectivity. Routing protocols are essential in MANETs, playing a crucial role in their functionality and extensively studied for enhancing the efficiency. Most of the routing protocols select unstable and congested intermediate nodes thereby it frequently causes the network path failures and packet loss. The recent studies in wireless MANET have focused on developing the new strategies for multicast routing. Quality of Service (QoS) enhances predictability of network performance in MANET. However, the application of QoS in ad hoc networks faces challenges due to the limited battery life and dynamic capacity of mobile devices. To boost the performance of MANETs, a novel hybrid heuristic approach for QoS-aware multicast transmission is proposed. The inherent openness of the network also increases its vulnerability to malicious attacks. Therefore, the malicious node detection is handled by Multiscale Dilated Long Short-Term Memory (MD-LSTM), where the attacked nodes are identified. Subsequently, the multicast routing is accomplished by proposing the Hybrid Coati Optimization Algorithm and Dolphin Swarm Optimization (HCDSO), which is used to find the optimal paths for mitigating the attacked nodes. This algorithm is also used to derive the multi-objective function along with different constraints. Finally, the analysis is effectively done by utilizing different performance measures. The accuracy of the suggested framework attains 97.75 in 1st dataset, 96.6 in 2nd dataset and 97.75 in 3rd dataset. On the contrary, the proposed method outperforms better results for malicious detection and multicast routing in MANET while enhancing the throughput performance than the existing techniques.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110086"},"PeriodicalIF":4.0,"publicationDate":"2025-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143148994","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":"Advancements in medical image segmentation: A review of transformer models","authors":"S.S. Kumar","doi":"10.1016/j.compeleceng.2025.110099","DOIUrl":"10.1016/j.compeleceng.2025.110099","url":null,"abstract":"<div><div>Medical image segmentation is crucial for precise diagnosis, treatment planning, and disease monitoring in healthcare. Traditional methods often struggle with the complexity and variability inherent in medical images. However, recent advancements in deep learning, particularly Transformer models, have revolutionized the field. This comprehensive review explores the transformative impact of Transformer models on medical image segmentation. Beginning with an overview of the limitations of traditional approaches, the review introduces foundational Transformer architectures such as the Vision Transformer, Swin Transformer, and Pyramid Vision Transformer. Systematically categorizing Transformer-based segmentation techniques, it delves into their applications across diverse medical imaging tasks, including brain tumor segmentation, polyp detection, cardiac segmentation, and more. Additionally, the review examines the challenges and considerations in benchmarking Transformer models using evaluation metrics and benchmark datasets. By analyzing current research trends and insights, this review provides valuable guidance for researchers and practitioners seeking to harness the power of Transformer models in medical image segmentation.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110099"},"PeriodicalIF":4.0,"publicationDate":"2025-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143144557","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":"AWE-DPFL: Adaptive weighting and dynamic privacy budget federated learning for heterogeneous data in IoT","authors":"Guiping Zheng,&nbsp;Bei Gong,&nbsp;Chong Guo,&nbsp;Tianqi Peng,&nbsp;Mowei Gong","doi":"10.1016/j.compeleceng.2025.110070","DOIUrl":"10.1016/j.compeleceng.2025.110070","url":null,"abstract":"<div><div>In the era of data-driven artificial intelligence, the widespread deployment of IoT devices has amplified concerns around privacy and data security. Federated learning (FL) offers a promising solution by enabling local model training without exposing raw data, effectively mitigating privacy risks. However, the inherent heterogeneity of IoT data leads to significant disparities in data distributions across different clients, negatively impacting the global model’s performance. Furthermore, conventional fixed differential privacy mechanisms lack the adaptability needed to dynamically adjust to the evolving requirements of different training phases, limiting their effectiveness in privacy-preserving federated learning. To address these challenges, we propose a federated learning framework called AWE-DPFL, which integrates adaptive weight fusion and dynamic privacy budget adjustment mechanisms. AWE-DPFL employs a dynamic privacy budget adjustment strategy to allocate privacy budgets based on the variance in client model updates, thereby improving model performance while ensuring robust privacy protection. Additionally, the adaptive weight fusion mechanism assigns different weights to each client’s model, taking into account data heterogeneity and quality, which leads to an enhanced global model that better reflects individual client contributions. Moreover, AWE-DPFL incorporates meta-learning alongside differential privacy techniques during local model training, resulting in an effective balance between data privacy and model performance. This approach not only improves model adaptability and generalization across diverse data distributions but also ensures that privacy requirements are met throughout the training process. Experimental evaluations demonstrate that AWE-DPFL significantly outperforms existing approaches on the MNIST, FashionMNIST, HAR, and Edge-IIoTset datasets, showcasing its effectiveness as a federated learning solution for real-world IoT applications.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110070"},"PeriodicalIF":4.0,"publicationDate":"2025-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143144616","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":"Improving the traffic prediction process efficiency using novel cohesive model","authors":"G. Balamurugan ,&nbsp;K.B. Gurumoorthy ,&nbsp;Suganyadevi S ,&nbsp;Balasamy K","doi":"10.1016/j.compeleceng.2025.110082","DOIUrl":"10.1016/j.compeleceng.2025.110082","url":null,"abstract":"<div><div>Internet of Vehicles (IoVs) systems is useful in handling hefty good across a city in a reliable and time-efficient manner. In a smart city environment, transportation is assisted through guiding systems that provide seamless support for the drivers. Guiding systems are automated and process dependent to perform traffic data analysis for reliable transportation assistance. In this article, a cohesive prediction model (CPM) is introduced with process efficiency (PE) for handling traffic data analysis. This model performs both traffic prediction and effective data analysis for achieving better prediction analysis and PE. The prediction process is performed for validating necessary traffic data preventing unnecessary analysis. This helps to refine the complex analysis in traffic prediction with efficient big data management. The process is eased using predictive learning through the classification of time and request depending on data analyses. The performance of the proposed model is verified using the metrics of resource utilization, prediction accuracy, and response delay for the varying vehicle density and responses, respectively. The proposed CPM-PE improves the prediction accuracy by 15.04 %, and reduces the response delay by 11.96 % and overhead by 9.92 % for the maximum requests considered.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110082"},"PeriodicalIF":4.0,"publicationDate":"2025-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143144561","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":"An ensemble model with attention based mechanism for image captioning","authors":"Israa Al Badarneh ,&nbsp;Bassam H. Hammo ,&nbsp;Omar Al-Kadi","doi":"10.1016/j.compeleceng.2025.110077","DOIUrl":"10.1016/j.compeleceng.2025.110077","url":null,"abstract":"<div><div>Image captioning creates informative text from an input image by creating a relationship between the words and the actual content of an image. Recently, deep learning models that utilize transformers have been the most successful in automatically generating image captions. The capabilities of transformer networks have led to notable progress in several activities related to vision. In this paper, we thoroughly examine transformer models, emphasizing the critical role that attention mechanisms play. The proposed model uses a transformer encoder–decoder architecture to create textual captions and a deep learning convolutional neural network to extract features from the images. To create the captions, we present a novel ensemble learning framework that improves the richness of the generated captions by utilizing several deep neural network architectures based on a voting mechanism that chooses the caption with the highest bilingual evaluation understudy (BLEU) score. The proposed model was evaluated using publicly available datasets. Using the Flickr8K dataset, the proposed model achieved the highest BLEU-[1-3] scores with rates of 0.728, 0.495, and 0.323, respectively. The suggested model outperformed the latest methods in Flickr30k datasets, determined by BLEU-[1-4] scores with rates of 0.798, 0.561, 0.387, and 0.269, respectively. The model efficacy was also obtained by the Semantic propositional image caption evaluation (SPICE) metric with a scoring rate of 0.164 for the Flicker8k dataset and 0.387 for the Flicker30k. Finally, ensemble learning significantly advances the process of image captioning and, hence, can be leveraged in various applications across different domains.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110077"},"PeriodicalIF":4.0,"publicationDate":"2025-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143144614","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":"Unmasking hidden threats: Enhanced detection of embedded malicious domains in pirate streaming videos","authors":"Yingshuo Wang,&nbsp;Changyong Guo,&nbsp;Jianen Yan,&nbsp;Zhaoxin Zhang,&nbsp;Yanan Cheng","doi":"10.1016/j.compeleceng.2025.110087","DOIUrl":"10.1016/j.compeleceng.2025.110087","url":null,"abstract":"<div><div>A significant number of malicious domains are hidden within public service websites in cyberspace. Pirated video streaming sites cleverly embed links to illicit content such as pornography and gambling within videos to conduct illegal activities. These contents pose significant threats to the physical and mental health of children and adolescents, yet effective detection and extraction methods are lacking. This paper investigates the embedding of malicious domains in pirated video streaming websites and proposes the Enhanced Detection of Embedded Malicious Domains (EDEMD) framework, which combines webpage text, URL features, and visual information. The study first develops an efficient framework to acquire URLs from public service websites using a dynamic keyword expansion algorithm and search engine APIs. Next, random forest and convolutional neural networks are employed to filter and classify embedded URLs on pirated video streaming pages, achieving accuracy rates exceeding 96% and 98%, respectively. Detection of most pages requires only 0.1 s, with nearly a 100% improvement in detection efficiency for playback pages. Finally, automation web testing tools are used to extract and analyze suspected malicious domains. An analysis of 1,347 pirated video streaming sites uncovers their underlying operational methods. This study provides robust technical support for curbing the spread of malicious domains in pirate videos.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110087"},"PeriodicalIF":4.0,"publicationDate":"2025-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143144559","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":"Multifactor interpretability method for offshore wind power output prediction based on TPE-CatBoost-SHAP","authors":"Shen Ruan ,&nbsp;Yun Chen ,&nbsp;Gengyang Lu ,&nbsp;Zhi Li ,&nbsp;Shu Chen ,&nbsp;Chenghao Wang ,&nbsp;Ting Li","doi":"10.1016/j.compeleceng.2025.110081","DOIUrl":"10.1016/j.compeleceng.2025.110081","url":null,"abstract":"<div><div>Accurate forecasting of offshore wind power output is crucial to the operation and scheduling of wind farms. However, predicting offshore wind power output is challenging because numerous factors influence wind energy. Current research focuses on achieving precise predictions, but the interpretability of models remains poor. In this context, this paper proposes a method that integrates Tree-structured Parzen estimator (TPE), categorical boosting (CatBoost), and SHapley Additive exPlanations (SHAP) to predict and interpret the offshore wind power output. PCA (principal component analysis) is used for meteorological feature preprocessing. The TPE algorithm iteratively searches for the optimal hyperparameters for the CatBoost prediction model. SHAP values are used to analyze the importance of various factors. By examining interactions, we identify the meteorological conditions conducive to efficient wind farm operation. The results indicate that the TPE-CatBoost model outperforms various other models in terms of evaluation metrics. SHAP explains the dependency between features and the seasonal variations in wind farms.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110081"},"PeriodicalIF":4.0,"publicationDate":"2025-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143144562","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":"An energy-aware link fault detection and recovery scheme for QoS enhancement in Internet of Things-enabled wireless sensor network","authors":"Prasanth Aruchamy ,&nbsp;Lavina Balraj ,&nbsp;K.K. Devi Sowndarya","doi":"10.1016/j.compeleceng.2025.110092","DOIUrl":"10.1016/j.compeleceng.2025.110092","url":null,"abstract":"<div><div>In the era of digitalization, the proliferation of the Internet of Things (IoT) has assisted in transforming the entire world into a smarter globe. Wireless Sensor Networks (WSNs) play an important role in developing many IoT-based real-time applications. Due to energy resource constraints, there are possibilities for hardware, communication links, and software faults occurring in the IoT-enabled WSN (IWSN) environment. These faults affect the data reading and cause serious impairment in the IWSN. To alleviate this scenario, it is imperative to identify a reliable fault diagnosis technique to accurately detect and respond to faults promptly. This work proposes a novel Energy-Aware Hardware and Link Fault Diagnosis (EAHLFD) scheme to enhance the quality of service in IWSN. Initially, the suggested EAHLFD scheme applies the Hardware Fault Detection and Classification method to detect various hardware faults. The Adaptive Communication Link Fault Detection and Recovery method is then invoked in the second stage for detecting the link faults. In this method, the optimized recovery link is determined based on three factors namely: probability value, link cost, and battery energy. The efficacy of the proposed scheme has been tested under two different simulation environments (IWSN#1 and IWSN#2). From the simulation results, it is manifest that the suggested EAHLFD scheme attains a superior fault diagnosis accuracy of 99.04% as compared to existing schemes.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110092"},"PeriodicalIF":4.0,"publicationDate":"2025-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143144560","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":"EcoWatch: Region of interest-based multi-quantization resource-efficient framework for migratory bird surveillance using wireless sensor networks and environmental context awareness","authors":"Oussama Hadji ,&nbsp;Moufida Maimour ,&nbsp;Abderezzak Benyahia ,&nbsp;Ouahab Kadri ,&nbsp;Eric Rondeau","doi":"10.1016/j.compeleceng.2025.110076","DOIUrl":"10.1016/j.compeleceng.2025.110076","url":null,"abstract":"<div><div>Global sustainability initiatives increasingly rely on innovative technologies to safeguard biodiversity and mitigate environmental impacts. In this paper, we present EcoWatch, a novel framework that leverages Wireless Multimedia Sensor Networks (WMSNs) using LoRaWAN technology for efficient data transmission to enable real-time bird species detection and counting in their natural habitat. EcoWatch combines YOLOv8 <em>You Only Look Once</em> for object detection and <em>Learning to Count Everything</em> (LTCE) for precise object counting at the base station. To address the inherent limitations of WSNs in terms of energy and bandwidth, EcoWatch incorporates a multi-level ROI-based video compression technique. Extensive evaluation demonstrates that EcoWatch significantly reduces energy consumption (up to 58.7%) and data transmission load (by 69.8%) compared to other methods while maintaining acceptable image quality, detection and counting accuracy. Notably, EcoWatch exhibits robust performance across seasons and adapts well to varying environmental conditions, making it a promising solution for real-world ecological monitoring applications.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110076"},"PeriodicalIF":4.0,"publicationDate":"2025-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143144615","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A comprehensive survey on RPL routing-based attacks, defences and future directions in Internet of Things","authors":"Anil Kumar Prajapati ,&nbsp;Emmanuel S. Pilli ,&nbsp;Ramesh Babu Battula ,&nbsp;Vijay Varadharajan ,&nbsp;Abhishek Verma ,&nbsp;R.C. Joshi","doi":"10.1016/j.compeleceng.2025.110071","DOIUrl":"10.1016/j.compeleceng.2025.110071","url":null,"abstract":"<div><div>The Internet of Things (IoT) is a network of digital devices like sensors, processors, embedded and communication devices that can connect to and exchange data with other devices and systems over the internet. IoT devices have limitations on power, memory, and computational resources. Researchers have developed the IPv6 Over Low-power Wireless Personal Area Network (6LoWPAN) protocols to provide wireless connectivity among these devices while overcoming the constraints on resources. 6LoWPAN has been approved subsequently by the Internet Engineering Task Force (IETF). The IETF Routing Over Low-power and Lossy Networks (ROLL) standardized the Routing Protocol for LLNs known as RPL (IETF RFC 6550), which is part of the 6LoWPAN stack. However, IoT devices are vulnerable to various attacks on RPL-based routing. This survey provides an in depth study of existing RPL-based attacks and defense published from year 2011 to 2024 from highly reputed journals and conferences. By thematic analysis of existing routing attacks on RPL, we developed a novel attack taxonomy which focuses on the nature of routing attacks and classifies them into 12 major categories. Subsequently, the impact of each attack on the network is analyzed and discussed real life scenarios of these attacks. Another contribution of this survey proposed a novel taxonomy for classification of defense mechanisms into 8 major categories against routing attacks based on type of defense strategy. The detailed analysis of each defense mechanism with real life applicability is explained. Furthermore, evaluation tools such as testbeds and simulators for RPL-based attack and defense are discussed and critically analyzed in terms of real world applicability. Finally, open research challenges are presented on the basis of research gaps of existing literature along with research directions for practitioners and researchers. We believe our study will give actionable insights and solid foundation for researchers to expand effective defense solutions against emerging RPL routing attacks in IoT networks.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110071"},"PeriodicalIF":4.0,"publicationDate":"2025-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143144563","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}