检索结果-内蒙古大学图书馆

AI-enabled driver assistance: monitoring head and gaze movements for enhanced safety

Complex & intelligent Systems 2025年第7期11卷 1-28页

作者： Mudassar Shah, Sayyed Zengkang, Gan Sun, Zhaoyun Hussain, Tariq Zaman, Khalid Alwabli, Abdullah Jaffar, Amar Y. Ali, Farman School of Civil and Transportation Engineering Shenzhen University Guangdong China Institute of Intelligent Manufacturing Technology Shenzhen Polytechnic University Shenzhen China School of Computer Science and Technology Zhejiang Gongshang University Hangzhou China Department of Electrical Engineering College of Engineering and Computing in Alqunfudah Umm Al-Qura University Mecca Saudi Arabia Computer and Network Engineering Department College of Computing Umm Al-Qura University Makkah Saudi Arabia Department of Computer Science and Engineering School of Convergence College of Computing and Informatics Sungkyunkwan University Seoul South Korea

This paper introduces a real-time head-pose detection and eye-gaze estimation system for Automatic Driver Assistance Technology (ADAT) aimed at enhancing driver safety by accurately collecting and transmitting data on the driver’s head position and eye gaze to mitigate potential risks. Existing methods are constrained by significant limitations, including reduced accuracy under challenging conditions such as varying head orientations and lighting, higher latency in real-time applications (e.g., Faster-RCNN and TPH-YOLOv5), and computational inefficiency, which hinders their deployment in resource-constrained environments. To address these challenges, we propose a novel framework using the Transformer Detection of Gaze Head - YOLOv7 (TDGH-YOLOv7) object detector. The key contributions of this work include the development of a reference image dataset encompassing diverse vertical and horizontal gaze positions alongside the implementation of an optimized detection system that achieves state-of-the-art performance in terms of accuracy and latency. The proposed system achieves superior precision, with a weighted accuracy of 95.02% and Root Mean Square Errors of 2.23 and 1.68 for vertical and horizontal gaze estimation, respectively, validated on the MPII-Gaze and DG-Unicamp datasets. A comprehensive comparative analysis with existing models, such as CNN, SSD, Faster-RCNN, and YOLOv8, underscores the robustness and efficiency of the proposed approach. Finally, the implications of these findings are discussed, and potential avenues for future research are outlined.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Enhancing Low-Latency and High-Reliability Data Forwarding for In-Vehicle Time-Sensitive Networks

引用

IEEE Transactions on Vehicular Technology 2025年

作者： Sun, Haitong Zhang, Haijun Guan, Wanqing Liu, Xiangnan Ma, Hui Leung, Victor C.M. Hebei Key Laboratory of Space-Air-Ground Intelligent Communication China Beijing Engineering and Technology Research Center for Convergence Networks and Ubiquitous Services University of Science and Technology Beijing Beijing100083 China School of Electrical Engineering and Computer Science KTH Royal Institute of Technology Stockholm11428 Sweden Department of Computer Science and Software Engineering Shenzhen University Shenzhen518061 China

To satisfy the low delay, low jitter, and high success rate requirements for in-vehicle networks, IEEE 802.1 Task Group proposed Time-Sensitive Networking (TSN), which has aroused increasing attention in managing time-critical services in networks. This paper resolves the following issue: how to minimize the worst-case response time (WCRT) of frame forwarding in TSN under no-wait scenarios to guarantee low-latency transmission, stability, and high success rate for in-vehicle TSN networks? Currently, some work resolves this problem by leveraging integer linear programming, and others apply heuristic methods to avoid the time complexity problem, but the success rate is affected since they are prone to get trapped in local optima. To reach a balance between time-complexity and success rate, a scheme named Frame Dynamic Priority Scheduling Mechanism (FDPSM) is proposed, where the frame dynamic priority levels are set up and frames are scheduled accordingly. The target is to maximize the success rate and minimize the WCRT for in-vehicle networks. We evaluate the proposed mechanism from aspects such as traffic forwarding time, size of windows, number of hops, and the network load, together with success rate and the simulation results demonstrate the advantages of the FDPSM for in-vehicle TSN networks. © 2025 IEEE. All rights reserved,

关键词： Integer programming

来源：评论

学校读者我要写书评

暂无评论

Differentially Private Synthetic Adversarial Network Traffic Generation Based on Tabular Diffusion Processes

Differentially Private Synthetic Adversarial Network Traffic...

引用

International Conference on Artificial Intelligence in Information and Communication (ICAIIC)

作者： Minjae Kang Gunhee Cho Sungju Yun Yeonjoon Lee Department of Computer Science and Engineering Major in Bio-Artificial Intelligence Hanyang University Ansan Republic of Korea Department of Intelligent Information Convergence Engineering Hanyang University Ansan Republic of Korea College of Computing Hanyang University Ansan Republic of Korea

ISBN: (数字)9798331507022

ISBN: (纸本)9798331507039

IDS is becoming increasingly important as networks become more vulnerable to a variety of attacks. However, traditional ML and DL-based IDS have limited accuracy and false positive rates due to the lack of open adversarial network datasets and vendors' concerns regarding potential data leakage during the training process. To address these challenges, VAE and GAN-based network packet generation models have emerged, but they still face issues with low data fidelity. In this context, TabDDPM is a suitable option for generating network packets, as it surpasses VAE and GAN in accurately capturing data distribution characteristics and effectively learning the inherent structure of network packet formats. Despite its potential, research applying TabDDPM to network packet generation has not yet been explored. In this paper, we propose and evaluate DP-NetDDPM, a novel framework that enables adversarial network traffic generation while preserving both fidelity and privacy guarantees. Specifically, we examine the use of diffusion to generate adversarial network packet datasets and compare its performance with baseline models. Our framework focuses on three key criteria: data fidelity, adaptability for machine learning applications, and data privacy. The results show that DP-NetDDPM surpasses traditional models in both fidelity and adaptability, achieving a notable 72% improvement in fidelity and a 45% enhancement in adaptability over baselines.

关键词： Training Adaptation models Privacy Telecommunication traffic Machine learning Network security Generative adversarial networks Data models Faces Payloads

来源：评论

学校读者我要写书评

暂无评论

Adhesive bioelectronics for closed-loop therapy

引用

Med-X 2025年第1期3卷 1-32页

作者： Jung, Hyunjin Lee, Hyelim Shin, Mikyung Son, Donghee Department of Electrical and Computer Engineering Sungkyunkwan University (SKKU) Suwon Republic of Korea Center for Neuroscience Imaging Research Institute for Basic Science Suwon Republic of Korea Department of Intelligent Precision Healthcare Convergence Sungkyunkwan University (SKKU) Suwon Republic of Korea Department of Biomedical Engineering Sungkyunkwan University (SKKU) Suwon Republic of Korea Department of Artificial Intelligence System Engineering Sungkyunkwan University (SKKU) Suwon Republic of Korea

Wearable and implantable devices mark a major advancement in biomedical engineering, enabling continuous bio-signal monitoring and treatment based on precise feedback. However, traditional rigid bioelectronic devices face challenges in seamlessly interfacing with soft organs, often leading to tissue damage and inflammation. To address these issues, soft bioelectronics that better conform to the body while maintaining performance accuracy have been developed. Despite these advancements, the dynamic nature of body movements and the complex chemical environment in the body can compromise the stability of devices over time. Therefore, recent innovations have focused on ultrathin and stretchable designs that enhance conformal contact through van der Waals forces, as well as the incorporation of adhesive hydrogels, to improve adhesion, even in the presence of biofluids. This review provides a comprehensive overview of the advancements in wearable and implantable bioelectronics aimed at achieving conformal tissue adhesion and long-term stability, covering structural modifications to bilayer designs. Finally, the integration of bioelectronics with closed-loop therapies to significantly enhance treatment efficacy is explored. • Materials strategies and device fabrication/integration processes optimally tailored to various organs are addressed. • Mechanical and electrical characterizations of wearable and implantable adhesive bioelectronics are described, while emphasizing challenges and opportunities in long-term stable tissue interfacing. • Various demonstrations of closed-loop bio-integrated systems using soft stretchable adhesive materials, and their futuristic perspectives are introduced.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Improved MIMO Signal Detection Based on DNN in MIMO-OFDM System

引用

computers, Materials & Continua 2022年第2期70卷 3625-3636页

作者： Jae-Hyun Ro Jong-Gyu Ha Woon-Sang Lee Young-Hwan You Hyoung-Kyu Song Department of Convergence Engineering for Intelligent Drone Sejong UniversitySeoul05006Korea Department of Information and Communication Engineering Convergence Engineering for Intelligent DroneSejong UniversitySeoul05006Korea Department of Computer Engineering Convergence Engineering for Intelligent DroneSejong UniversitySeoul05006Korea

This paper proposes the multiple-input multiple-output(MIMO)detection scheme by using the deep neural network(DNN)based ensemble machine learning for higher error performance in wireless communication *** the MIMO detection based on the ensemble machine learning,all learning models for the DNN are generated in offline and the detection is performed in online by using already learned *** the offline learning,the received signals and channel coefficients are set to input data,and the labels which correspond to transmit symbols are set to output *** the online learning,the perfectly learned models are used for signal detection where the models have fixed bias and *** performance improvement,the proposed scheme uses the majority vote and the maximum probability as the methods of the model combinations for obtaining diversity gains at the MIMO *** simulation results show that the proposed scheme has improved symbol error rate(SER)performance without additional receive antennas.

关键词： MIMO DNN ensemble machine learning ML

来源：评论

学校读者我要写书评

暂无评论

Efficient Gauss-Seidel Precoding with Parallel Calculation in Massive MIMO Systems

引用

computers, Materials & Continua 2022年第1期70卷 491-504页

作者： Hyun-Sun Hwang Jae-Hyun Ro Chan-Yeob Park Young-Hwan You Hyoung-Kyu Song Department of Information and Communication Engineering Convergence Engineering for Intelligent DroneSejong UniversitySeoul05006Korea Department of Convergence Engineering for Intelligent Drone Sejong UniversitySeoul05006Korea Department of Computer Engineering Convergence Engineering for Intelligent DroneSejong UniversitySeoul05006Korea

A number of requirements for 5G mobile communication are satisfied by adopting multiple input multiple output(MIMO)*** inter user interference(IUI)which is an inevitable problem in MIMO systems becomes controllable when the precoding scheme is *** this paper,the horizontal Gauss-Seidel(HGS)method is proposed as precoding scheme in massive MIMO *** massive MIMO systems,the exact inversion of channel matrix is impractical due to the severe computational ***,the conventionalGauss-Seidel(GS)method is used to approximate the inversion of channel *** GS has good performance by using previous calculation results as ***,the required time for obtaining the precoding symbols is too long due to the sequential process of ***,the HGS with parallel calculation is proposed in this paper to reduce the required *** rows of channel matrix are eliminated for parallel calculation *** addition,HGSuses the ordered channelmatrix to prevent performance degradation which is occurred by parallel *** HGS with proper number of parallelly computed symbols has better performance and reduced required time compared to the traditional GS.

关键词： Massive MIMO GS matrix inversion linear precoding

来源：评论

学校读者我要写书评

暂无评论

Adaptive Relay Selection Scheme for Minimization of the Transmission Time

引用

computers, Materials & Continua 2021年第10期69卷 1361-1373页

作者： Yu-Jin Na Ji-Sung Jung Young-Hwan You Hyoung-Kyu Song Department of Information and Communication Engineering and Convergence for Intelligent DroneSejong UniversitySeoul05006Korea Department of Computer Engineering and Convergence for Intelligent DroneSejong UniversitySeoul05006Korea

As the installation of small cells increases,the use of relay also *** relay operates as a base station as well as just an *** the roles and types of relays become more diverse,appropriate relay selection technology is an effective way to improve communication *** researches for relay selection have been studied to secure the reliability of relay *** this paper,the relay selection scheme is proposed for a cooperative system using decode-and-forward(DF)relaying scheme in the mobile communication *** maintain the transmission rate,the proposed scheme classifies a candidate group considering the outage probability of multiple *** the applicable candidate group,the proposed scheme selects the relay considering the amount of data allocated to each ***,the proposed scheme defines the unit transmission time through each user’s data and relay ***,the proposed scheme selects a relay that minimizes the total transmission time through the relay transmission time that calculates the unit transmission time for all *** this adaptive relay selection scheme,an optimal relay can be assigned for each *** the same transmission rate and the amount of data,the proposed scheme improves the performance of transmission time and *** results show that the proposed scheme reduces the total transmission time for the same amount of data and signal to noise ratio(SNR).

关键词： Relay selection cooperative relay MIMO outage probability transmission time

来源：评论

学校读者我要写书评

暂无评论

Impact of en-APTAS membrane on the performance of HfO₂-based RRAM devices

引用

Surfaces and Interfaces 2025年 64卷

作者： Chae, Myoungsu Ahmad, Ibtisam Jang, Yuseong Lee, Doowon Kim, Hee-Dong Department of Semiconductor Systems Engineering and Convergence Engineering for Intelligent Drone Institute of Semiconductor and System IC Sejong University 209 Neungdong-ro Gwangjin-gu Seoul 05006 South Korea Institute of Industrial Science Department of Electrical Engineering University of Tokyo 4-6-1 Komaba Tokyo Meguro City 153-8505 Japan Division of Electrical Electronic and Control Engineering Kongju National University Cheonan 31080 South Korea

We propose the use of an N-[3-trimethoxysilyl)propyl]ethylenediamine (en-APTAS) to enhance the resistive switching characteristics of hafnium oxide (HfO2)-based resistive random-access memory (RRAM) with a porous carb... 详细信息

We propose the use of an N-[3-trimethoxysilyl)propyl]ethylenediamine (en-APTAS) to enhance the resistive switching characteristics of hafnium oxide (HfO₂)-based resistive random-access memory (RRAM) with a porous carbon nanotube (CNT)-top electrode (CNT-TE). First, we confirm that immersion in en-APTAS significantly increased the on/off ratio compared to conventional HfO₂-based RRAM with CNT-TE. The en-APTAS immersion for 15 min in particular resulted in a 24-fold increase in the on/off ratio and a stable on/off behavior for over 100 cycles and 10,000 s of the retention evaluation. Furthermore, we demonstrate that the oxygen migration into the porous structure of CNT-TE is enhanced by the surface passivation with en-APTAS, which leads to improved endurance and retention characteristics. Finally, the reliability evaluations reveal that the en-APTAS-treated HfO₂-based RRAM exhibited reduced operating voltage and resistance state variability. These results suggest that the reaction between the amine groups and HfO₂ induced surface passivation and homogenization of oxygen vacancies, which thereby enhance the proposed device's stability and performance. © 2025 Elsevier B.V.

关键词： en-APTAS Hafnium oxide Passivation, Reliability RRAM

来源：评论

学校读者我要写书评

暂无评论

Ransomware Classification Framework Using the Behavioral Performance Visualization of Execution Objects

引用

computers, Materials & Continua 2022年第8期72卷 3401-3424页

作者： Jun-Seob Kim Ki-Woong Park Department of Computer and Information Security and Convergence Engineering for Intelligent DroneSejong UniversitySeoul05006Korea

A ransomware attack that interrupted the operation of Colonial Pipeline(a large *** pipeline company),showed that security threats by malware have become serious enough to affect industries and social infrastructure rather than individuals *** agents and characteristics of attacks should be identified,and appropriate strategies should be established accordingly in order to respond to such *** this purpose,the first task that must be performed is malware *** creators are well aware of this and apply various concealment and avoidance techniques,making it difficult to classify *** study focuses on new features and classification techniques to overcome these *** propose a behavioral performance visualization method using utilization patterns of system resources,such as the central processing unit,memory,and input/output,that are commonly used in performance analysis or tuning of *** extracted the usage patterns of the system resources for ransomware to performbehavioral performance *** results of the classification performance evaluation using the visualization results indicate an accuracy of at least 98.94%with a 3.69%loss ***,we designed and implemented a framework to perform the entire process—from data extraction to behavioral performance visualization and classification performance measurement—that is expected to contribute to related studies in the future.

关键词： Behavioral performance visualization ransomware malware classification

来源：评论

学校读者我要写书评

暂无评论

AI-Driven Prioritization and Filtering of Windows Artifacts for Enhanced Digital Forensics

引用

computers, Materials & Continua 2024年第11期81卷 3371-3393页

作者： Juhwan Kim Baehoon Son Jihyeon Yu Joobeom Yun Department of Computer and Information Security and Convergence Engineering for Intelligent DroneSejong UniversitySeoul05006Republic of Korea

Digital forensics aims to uncover evidence of cybercrimes within compromised *** cybercrimes are often perpetrated through the deployment of malware,which inevitably leaves discernible traces within the compromised *** analysts are tasked with extracting and subsequently analyzing data,termed as artifacts,from these systems to gather ***,forensic analysts must sift through extensive datasets to isolate pertinent ***,manually identifying suspicious traces among numerous artifacts is time-consuming and *** studies addressed such inefficiencies by integrating artificial intelligence(AI)technologies into digital *** the efforts in previous studies,artifacts were analyzed without considering the nature of the data within them and failed to prove their efficiency through specific *** this study,we propose a system to prioritize suspicious artifacts from compromised systems infected with malware to facilitate efficient digital *** system introduces a double-checking method that recognizes the nature of data within target artifacts and employs algorithms ideal for anomaly *** key ideas of this method are:(1)prioritize suspicious artifacts and filter remaining artifacts using autoencoder and(2)further prioritize suspicious artifacts and filter remaining artifacts using logarithmic *** evaluation demonstrates that our system can identify malicious artifacts with high accuracy and that its double-checking method is more efficient than alternative *** system can significantly reduce the time required for forensic analysis and serve as a reference for future studies.

关键词： Digital forensics autoencoder logarithmic entropy prioritization anomaly detection windows artifacts artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：