IEEE Access最新文献_第8页

A Transmission Axle Fault Diagnosis System for Massive Rapid Transit System With Enhanced Attention-Based Shuffle Networks 基于增强注意力洗牌网络的大型快速交通系统传动轴故障诊断系统

IF 3.6 3区计算机科学

IEEE Access Pub Date : 2025-09-01 DOI: 10.1109/ACCESS.2025.3604605

Leehter Yao;Hsuan Su;Matthew Cheng

{"title":"A Transmission Axle Fault Diagnosis System for Massive Rapid Transit System With Enhanced Attention-Based Shuffle Networks","authors":"Leehter Yao;Hsuan Su;Matthew Cheng","doi":"10.1109/ACCESS.2025.3604605","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3604605","url":null,"abstract":"The transmission axle faults can cause severe damage to gearbox components and jeopardize motor power transmission. A transmission axle fault (TAF) diagnosis method for train propulsion systems in mass rapid transit (MRT) networks is proposed, utilizing an Enhanced Attention-based Shuffle Network (EASN). Time-domain vibration signals, directly acquired from sensors mounted on the transmission axles, are employed as input to the EASN model. The proposed model is specifically designed for deployment on lightweight AI edge devices, with an emphasis on computational efficiency and real-time performance for onboard diagnostic applications. The EASN architecture is composed of multiple building blocks, referred to as Split-Attention Shuffle Units (SASUs). Each SASU integrates a Shuffling Processing Module (SPM) and a Split Attention Module (SAM) in a cascaded configuration. While the SPM is based on ShuffleNet, which is known for its computational efficiency but relatively lower classification accuracy, the proposed SASU mitigates this limitation through the introduction of an even channel shuffling mechanism combined with a hybrid attention strategy. The hybrid attention scheme leverages both Spatial Excitation (SPE) and Squeeze-and-Excitation (SnE) mechanisms, significantly enhancing the network’s diagnostic accuracy without compromising its lightweight design. Experimental results demonstrate that the proposed EASN achieves a Top-1 classification accuracy of 93.9%, representing a 5.6% improvement over ResNet-50 while reducing the model size by 98.8%. Compared with lightweight models such as MobileNet V2, EASN improves accuracy by 19.4% with only a 28.6% increase in parameter size. These findings indicate that EASN offers an effective balance between diagnostic accuracy and model compactness, making it well-suited for real-time, edge-based fault detection in mass rapid transit systems.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"152253-152265"},"PeriodicalIF":3.6,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11145795","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145005502","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}

引用次数: 0

Enhancing Student Retention in Introductory Programming Courses: Leveraging Advanced Learning Validation Tools and Educational Data Mining 提高编程入门课程的学生留存率：利用高级学习验证工具和教育数据挖掘

IF 3.6 3区计算机科学

IEEE Access Pub Date : 2025-09-01 DOI: 10.1109/ACCESS.2025.3604563

Alan Mutka;Fatima Živković Mutka;Martin Žagar;Domagoj Tolić

{"title":"Enhancing Student Retention in Introductory Programming Courses: Leveraging Advanced Learning Validation Tools and Educational Data Mining","authors":"Alan Mutka;Fatima Živković Mutka;Martin Žagar;Domagoj Tolić","doi":"10.1109/ACCESS.2025.3604563","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3604563","url":null,"abstract":"Student retention in introductory programming courses remains a persistent challenge in higher education, with high failure and dropout rates impacting both learners and institutions. This article presents a novel, behavior-based approach to addressing this issue through Educational Data Mining (EDM) and a custom-built learning validation framework called AssessMe. Developed by the authors in collaboration with SmoothSoft Ltd., AssessMe is an advanced software tool that monitors the real-time development of programming assignments. It captures detailed behavioral data–including active coding time, code changes (added, modified, removed lines), and submission timelines–to generate learning indicators reflecting how students approach problem-solving tasks. Unlike traditional assessment methods that focus solely on final code correctness, AssessMe emphasizes the coding process, offering deeper insights into student engagement, effort, and learning strategies. This study focuses on Programming I and II courses within the Web & Mobile Computing program at RIT Croatia. The dataset includes 3,537 student submissions from the 2024/2025 academic year, covering homework, practicals, and in-class activities, all enriched with AssessMe indicators. We apply traditional machine learning models combined with the TSFRESH library to extract meaningful time-series features from students’ coding activity. This enables the identification of temporal learning patterns and supports early prediction of academic outcomes. Our models achieve over 93% accuracy in forecasting pass/fail status by the fifth week of a 15-week semester, demonstrating a strong correlation between AssessMe indicators and final grades. This behavior-based assessment approach enhances early intervention strategies and provides actionable insights for improving student retention and learning outcomes.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"153614-153626"},"PeriodicalIF":3.6,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11145819","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145027897","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}

引用次数: 0

Deep Learning-Based Drowsiness Detection System for Driver’s Safety 基于深度学习的驾驶员睡意检测系统

IF 3.6 3区计算机科学

IEEE Access Pub Date : 2025-09-01 DOI: 10.1109/ACCESS.2025.3604728

Sindhu Vidyanathan Dixith;Shrikant Jadhav;Youngsoo Kim;Naveenkumar Jayakumar

{"title":"Deep Learning-Based Drowsiness Detection System for Driver’s Safety","authors":"Sindhu Vidyanathan Dixith;Shrikant Jadhav;Youngsoo Kim;Naveenkumar Jayakumar","doi":"10.1109/ACCESS.2025.3604728","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3604728","url":null,"abstract":"Driver drowsiness is a leading contributor to road accidents, accounting for over 100,000 crashes and approximately ~1000 fatalities each year in the United States alone, as per National Safety Council (NSC). To mitigate this urgent public-safety risk, we propose a real-time Driver Drowsiness Detection System that achieves both high accuracy and fast inference on standard hardware. Our key idea is to combine two complementary deep-learning strategies: 1) a custom Convolutional Neural Network (CNN) paired with a Support Vector Machine (SVM) classifier, and 2) a lightweight transfer-learning model built upon a pre-trained convolutional backbone. We evaluate these approaches on two datasets: a four-class Kaggle collection of open/closed eyes and yawn/no-yawn images, and the 37-subject MRL eye dataset. For the custom CNN+SVM pipeline, we optimized the split ratios, dropout rates, and L2 regularization to achieve 100% training accuracy and 99.7% validation accuracy. For the transfer-learning model, we leveraged an existing network to accelerate training, achieving 99.4% training accuracy and 99.1% validation accuracy. Finally, we compare both models using metrics including loss curves, confusion matrices, precision, recall, and F1-score. Our system demonstrates that real-time, highly accurate drowsiness detection is achievable without specialized hardware, paving the way for broader deployment in road-safety applications.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"154080-154102"},"PeriodicalIF":3.6,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11145796","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145021167","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}

引用次数: 0

Vision-Based Time Series Crowd Forecasting Using Semi-Supervised Learning 基于视觉的半监督学习时间序列人群预测

IF 3.6 3区计算机科学

IEEE Access Pub Date : 2025-09-01 DOI: 10.1109/ACCESS.2025.3604713

Salma Saud Alghamdi;Lama Al Khuzayem;Ohoud Alzamzami

{"title":"Vision-Based Time Series Crowd Forecasting Using Semi-Supervised Learning","authors":"Salma Saud Alghamdi;Lama Al Khuzayem;Ohoud Alzamzami","doi":"10.1109/ACCESS.2025.3604713","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3604713","url":null,"abstract":"Crowd forecasting is a crucial component of public safety, urban planning, and event management, enabling proactive decision-making based on anticipated crowd dynamics. Traditional sensor-based approaches, such as WiFi-based methods, suffer from accuracy issues due to device penetration limitations. On the other hand, vision-based approaches, while more precise, typically require fully extensive labeled data and high computational resources. These demands restrict their application to forecasting often limited to predicting the next frame or a few seconds ahead. To overcome these challenges, this research presents a vision-based time series forecasting framework that exploits a semi-supervised deep learning approach. A semi-supervised crowd counting model, trained on just 5% of labeled images from a single day, is used to extract time series crowd counts from images captured over 16 days at 5-minute intervals. These extracted time series data are then used for training multiple Long Short-Term Memory (LSTM) variants to analyze the dynamics of crowd forecasting. Experimental results demonstrate that the proposed framework enables accurate crowd forecasting while reducing annotation costs. Unlike existing vision-based approaches, which are constrained to forecasting seconds ahead, our approach can forecast a horizon of one hour ahead. Notably, the CNN Autoencoder LSTM and ConvLSTM models achieved an RMSE of 61.93 and a MAPE of 26.13%. These findings highlight the effectiveness of semi-supervised learning with minimal labeled data in vision-based crowd forecasting. Future work will focus on improving generalizability and robustness across different urban environments.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"153523-153541"},"PeriodicalIF":3.6,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11145448","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145027959","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}

引用次数: 0

High-Fidelity Virtual Model for Industrial Robot Control Under Uncertain and Disturbed Scenarios: A Comparative Study on the UR5e 不确定与扰动环境下工业机器人控制的高保真虚拟模型——基于UR5e的比较研究

IF 3.6 3区计算机科学

IEEE Access Pub Date : 2025-09-01 DOI: 10.1109/ACCESS.2025.3604653

Heni Belgacem;Mohammad Abuabiah;Inés Chihi

{"title":"High-Fidelity Virtual Model for Industrial Robot Control Under Uncertain and Disturbed Scenarios: A Comparative Study on the UR5e","authors":"Heni Belgacem;Mohammad Abuabiah;Inés Chihi","doi":"10.1109/ACCESS.2025.3604653","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3604653","url":null,"abstract":"Robust control of industrial manipulators under real-world uncertainties is critical for reliable automation. This work presents a comprehensive framework for modeling, control, and performance evaluation of the UR5e robotic manipulator. High-fidelity kinematic and dynamic models are developed and validated against experimental data to create a realistic virtual environment. Four control strategies, including Computed Torque Control, Proportional Integral Derivative, Sliding Mode Control, and Nonlinear Model Predictive Control are implemented and systematically compared. The comparison considers tracking accuracy, robustness, energy efficiency, and computational demand under nominal conditions as well as in the presence of external disturbances, sensor noise, and model uncertainties. Sliding Mode Control demonstrates consistent tracking under disturbances, Nonlinear Model Predictive Control achieves reduced energy consumption with smooth motion profiles, Computed Torque Control provides balanced accuracy and response, and Proportional Integral Derivative performs effectively under low-disturbance conditions. The methodology provides a validated simulation platform for benchmarking robotic control strategies and supports data-driven selection of controllers for industrial applications.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"152063-152089"},"PeriodicalIF":3.6,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11145441","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144997873","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}

引用次数: 0

High-Isolation Twin-Cluster Fan-Blade Metasurface-Embedded Wideband CP MIMO Antenna With Shorting-Pin Vias for IoT Applications 用于物联网应用的高隔离双集群风扇叶片超表面嵌入式宽带CP MIMO天线

IF 3.6 3区计算机科学

IEEE Access Pub Date : 2025-08-29 DOI: 10.1109/ACCESS.2025.3604009

Nathapat Supreeyatitikul;Pongsathorn Chomtong;Jessada Konpang;Prayoot Akkaraekthalin

{"title":"High-Isolation Twin-Cluster Fan-Blade Metasurface-Embedded Wideband CP MIMO Antenna With Shorting-Pin Vias for IoT Applications","authors":"Nathapat Supreeyatitikul;Pongsathorn Chomtong;Jessada Konpang;Prayoot Akkaraekthalin","doi":"10.1109/ACCESS.2025.3604009","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3604009","url":null,"abstract":"This study proposes a twin-cluster fan-blade metasurface (MTS)-embedded wideband circularly polarized (CP) multiple-input multiple-output (MIMO) antenna incorporating shorting-pin vias for Internet of Things (IoT) applications. The antenna is implemented on a double-layered substrate. The upper layer comprises two identical clusters of fan-blade MTS elemental arrays, while the lower layer integrates a dual-element coplanar waveguide feeding structure. Centrally placed shorting-pin vias are introduced to mitigate inter-cluster mutual coupling. Characteristic mode analysis is employed to investigate the CP behavior of each cluster of the twin-cluster antenna, with circular polarization achieved through two orthogonal modes (Modes 3 and 4). For coupling suppression, the optimal placement of the shorting-pin vias is determined based on minimizing the normalized electric field magnitude between clusters. The measured return loss bandwidth, axial ratio bandwidth, and maximum gain at the center frequency of 5.5 GHz are 46.72% (4.2–6.77 GHz), 11% (5.29–5.9 GHz), and 5.82 dBic, respectively. The antenna scheme achieves high isolation exceeding 25 dB, an envelope correlation coefficient below 0.002, a diversity gain above 9.99 dB, a mean effective gain below -3 dB, a total active reflection coefficient below -10 dB, and a channel capacity loss below 0.2 bits/s/Hz. This study is the first to utilize a fan-blade MTS-embedded structure for CP generation and to integrate shorting-pin vias within a CP MIMO antenna for mutual coupling suppression. Essentially, the proposed twin-cluster antenna scheme is well-suited for advanced IoT applications.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"152392-152411"},"PeriodicalIF":3.6,"publicationDate":"2025-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11145025","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145003519","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}

引用次数: 0

Perceptual Quality Assessment for NeRF-Generated Scenes: A Training Reference Metric nerf生成场景的感知质量评估：一个训练参考度量

IF 3.6 3区计算机科学

IEEE Access Pub Date : 2025-08-29 DOI: 10.1109/ACCESS.2025.3603970

Shihao Luo;Nguyen Tien Phong;Chibuike Onuoha;Truong Cong Thang

{"title":"Perceptual Quality Assessment for NeRF-Generated Scenes: A Training Reference Metric","authors":"Shihao Luo;Nguyen Tien Phong;Chibuike Onuoha;Truong Cong Thang","doi":"10.1109/ACCESS.2025.3603970","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3603970","url":null,"abstract":"Perceptual quality assessment is a key challenge in traditional image processing as well as emerging AI-based novel view synthesis methods such as neural radiation fields (NeRF). NeRF has revolutionized 3D scene reconstruction by leveraging neural networks for volumetric rendering, achieving unprecedented photorealistic results. Currently, the perceptual quality assessment of NeRF still relies heavily on full-reference (FR) metrics, such as PSNR and SSIM, which require external reference images from predefined camera positions and suffer from significant limitations. In this paper, we propose a new quality metric that directly leverages training views to quantify the perceptual quality of NeRF-generated scenes, eliminating the need for external predefined reference images and camera position metadata. In the proposed approach, we first extracted hierarchical and abstract features from training views using pretrained deep convolutional neural networks (CNNs) and then constructed a reference representation for perceptual quality evaluation through feature space interpolation. We evaluated the effectiveness of the proposed approach with two options: one with pretrained CNNs only (without calibration) and another with calibration applied to learn the importance of hierarchical feature stages. The experimental results demonstrate the effectiveness of the proposed method, which outperforms traditional no-reference (NR) metrics while being comparable to popular FR metrics. We found that deep features trained for high-level classification tasks have a strong potential to quantify perceptual quality across different viewpoints of the same object in NeRF. The code is released at <uri>https://github.com/WemoLuo/NVQS</uri>","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"152277-152292"},"PeriodicalIF":3.6,"publicationDate":"2025-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11145061","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145005394","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}

引用次数: 0

Development of a Hybrid Control Algorithm Based on a Combination of Robust Nonlinear Techniques for Electric Power Steering 基于鲁棒非线性技术的电动助力转向混合控制算法研究

IF 3.6 3区计算机科学

IEEE Access Pub Date : 2025-08-29 DOI: 10.1109/ACCESS.2025.3603623

Tuan Anh Nguyen

{"title":"Development of a Hybrid Control Algorithm Based on a Combination of Robust Nonlinear Techniques for Electric Power Steering","authors":"Tuan Anh Nguyen","doi":"10.1109/ACCESS.2025.3603623","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3603623","url":null,"abstract":"This work proposes a hybrid nonlinear control algorithm for electric power steering (EPS) systems, designed by integrating backstepping control (BSC), sliding mode control (SMC), and proportional-integral (PI) control. The robustness of SMC against external disturbances and the flexible trajectory handling of BSC are combined to achieve high-performance control. Meanwhile, PI controllers compensate for phase differences and tracking errors. Their gains are optimally tuned using a dual-loop optimization framework, which is later extended with fuzzy logic to enhance adaptability under varying operating conditions. System stability is rigorously analyzed and proven via Lyapunov theory. A dynamic EPS model is developed to account for disturbances (road reaction torques and external disturbances). Numerical simulations validate the proposed method, showing that the maximum tracking error remains below 0.01 rad, with root mean square error (RMSE) approaching zero. The proposed controller effectively suppresses chattering and eliminates phase lag, ensuring robust and precise steering assistance under various driving conditions.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"152452-152468"},"PeriodicalIF":3.6,"publicationDate":"2025-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11143188","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145005446","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}

引用次数: 0

Benchmark Algorithm for Asynchronous Detection of M-ASPM Packets Combined With Measuring Carrier Frequency Offset 结合载波频偏测量的M-ASPM报文异步检测基准算法

IF 3.6 3区计算机科学

IEEE Access Pub Date : 2025-08-29 DOI: 10.1109/ACCESS.2025.3604244

Alexei V. Nikitin;Ruslan L. Davidchack

{"title":"Benchmark Algorithm for Asynchronous Detection of M-ASPM Packets Combined With Measuring Carrier Frequency Offset","authors":"Alexei V. Nikitin;Ruslan L. Davidchack","doi":"10.1109/ACCESS.2025.3604244","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3604244","url":null,"abstract":"This paper demonstrates that in M-ary Aggregate Spread Pulse Modulation (M-ASPM) a relatively short, low-gain front portion of the transmitted packet can be used for robust asynchronous detection of the packet at low computational cost. With low processing gain, this detection is insensitive to large carrier frequency offsets (CFO) between the transmitter and the receiver, and it can be combined with measuring the CFO within the desired range and with desired precision. The subsequent costly processing of the high-gain, CFO-sensitive segments of the packet (e.g., those allocated to the synchronization and the payload) is implemented only after the detection, and the measured CFO is used for modifying the processing to prevent signal deterioration. In the receiver, only low-order matched filtering is continuously performed, and a single detection channel can be shared by multiple quasi-orthogonal channels used for payloads. This greatly economizes M-ASPM networks, especially their multi-channel, high-throughput configurations. However, since the contrast in the processing gains between different portions of the packet can exceed two orders of magnitude, matching the detection sensitivity with that of the payload, while maintaining the portion of the packet dedicated to the detection relatively small, poses a significant challenge. This paper presents the detection algorithm that overcomes this challenge, together with detailed explanation of the procedures and the tools employed in implementation of its steps. In particular, the properties, scope, and limitations of such essential algorithm components as the modulo exponential averaging (MEA) and adaptive quantile tracking filters (QTFs) for troughs are analyzed.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"152914-152933"},"PeriodicalIF":3.6,"publicationDate":"2025-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11145042","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144998345","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}

引用次数: 0

Design and Computational Modeling of an AI-Based Automated Cybersecurity Incident Response System 基于人工智能的自动网络安全事件响应系统设计与计算建模

IF 3.6 3区计算机科学

IEEE Access Pub Date : 2025-08-29 DOI: 10.1109/ACCESS.2025.3603975

Jiehao Zhang;Simin Li;Weiwei Huang;Haoxin Jing;Qin Zhang;Xing Xia

{"title":"Design and Computational Modeling of an AI-Based Automated Cybersecurity Incident Response System","authors":"Jiehao Zhang;Simin Li;Weiwei Huang;Haoxin Jing;Qin Zhang;Xing Xia","doi":"10.1109/ACCESS.2025.3603975","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3603975","url":null,"abstract":"Modern cybersecurity operations face unsustainable alert volumes, averaging 22000 weekly alerts with 68% false positives, overwhelming defenses and delaying incident response due to limitations in conventional SOAR platforms. To address this, an AI-driven Automated Incident Response (AIR) system is proposed, integrating STIX/TAXII multimodal fusion for unified data ingestion, attention-LSTM networks for adaptive threat recognition across temporal sequences, Bayesian game-theoretic decision layers for strategic response planning, and DRL validation for real-time optimization. This architecture reduces false negatives by 42% in C2 tunneling detection and achieves Nash equilibrium in 97.3% of adversarial engagements. Rigorous testing on hybrid infrastructure datasets (100 K normal events, 20K DDoS, 5K C2 attacks) demonstrates a 93% mean F1-score across attack scenarios, end-to-end latency of 58.3 ms, and <inline-formula> <tex-math>$12.5times $ </tex-math></inline-formula> higher strategy updates/sec versus baselines. Compared to existing models, the system improves detection F1 by 10.7%, reduces false positives by 39%, and enhances energy efficiency to 1850 events/Joule (<inline-formula> <tex-math>$2.98times $ </tex-math></inline-formula> Snort). The framework establishes a new paradigm for agile, auditable incident response validated by STIX action chains.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"154383-154394"},"PeriodicalIF":3.6,"publicationDate":"2025-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=11145017","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145021218","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}

引用次数: 0