检索结果-内蒙古大学图书馆

2023 ieee 66th International Midwest symposium on Circuits and Systems, MWSCAS 2023

作者： El-Zeiny, Ahmed N. Hassan, Adham Mostafa, Hassan Khalil, Ahmed H. Cairo University Electronics and Communications Engineering Department Egypt University of Science and Technology Nanotechnology and Nanoelectronics Program Zewail City of Science and Technology Giza12578 Egypt

ISBN: (纸本)9798350302103

In this paper, we present an efficient real-time object detector designed for embedded system applications. Although increased memory and power consumption pose significant challenges, our network adopts techniques to overcome them. Depthwise and pointwise separable convolutions are used across the backbone, neck, and head of the network to reduce the number of parameters. To compensate for accuracy loss, we employ several techniques including starting with a larger baseline, using depthwise separable convolution mixed with blueprint separable convolution in the backbone, replacing the SiLU activation function with the Mish activation function, and switching from Adam to SGD optimization algorithms. These techniques help to achieve a state-of-the-art object detection network called Mini-YOLOX, which optimizes YOLOX-S from the YOLOX series and reaches only ∽57% of its parameter size and ∽52% of its floating-point operations with better accuracy. Mini-YOLOX achieves impressive results with only 5.08M parameters and 13.95 GFLOPs, resulting in a 40.9 AP score in just 5.71 ms. © 2023 ieee.

关键词： convolutional neural network (CNN) embedded applications real-time object detector YOLOX

来源：评论

学校读者我要写书评

暂无评论

The Role of Artificial Intelligence in Boosting Cybersecurity and Trusted embedded Systems Performance: A Systematic Review on Current and Future Trends

引用

ieee ACCESS 2025年 13卷 55258-55276页

作者： Oun, Ahmed Wince, Kaden Cheng, Xiangyi Ohio Univ Russ Coll Engn & Technol Sch Elect Engn & Comp Sci Athens OH 45701 USA Ohio Northern Univ TJ Smull Coll Engn Dept Elect & Comp Engn & Comp Sci Ada OH 45810 USA Loyola Marymount Univ Seaver Coll Sci & Engn Dept Mech Engn Los Angeles CA 90045 USA

As technology becomes increasingly interconnected, ensuring the security of cyber and embedded systems is critical due to escalating vulnerabilities and sophisticated cyber threats. Researchers are exploring artificial intelligence (AI) to improve security mechanisms, yet there is a lack of a comprehensive technical, AI-focused analysis detailing the integration of AI into existing security hardware and frameworks. To address this gap, this article systematically reviews 63 articles on AI in cybersecurity and trusted embedded systems. The reviewed articles are categorized into four application domains: 1) Intrusion Detection and Prevention (IDPS), 2) Malware Detection, 3) Industrial Control and Cyber-Physical Systems (CPS) and 4) Distributed Denial-of-Service (DDoS) Detection and Prevention. We investigated current trends in integrating AI into security domains by summarizing the hardware used, the AI methodologies adopted, and the statistical distribution by publication year and region. The key findings of our review indicate that AI significantly enhances security measures by enabling capabilities such as detection, classification, feature selection, data privacy preservation, model combination, data generation, output interpretation, optimization, and adaptation. In addition, the benefits and challenges identified in these studies provide insight into the future potential of AI integration in security. Suggested directions for future work include improving generalization and scalability, exploring continuous or real-time monitoring, and improving AI model performance. This analysis serves as a foundation for advancing AI applications in the effective securing of cyber and embedded systems effectively.

关键词： Security cybersecurity embedded system artificial intelligence machine learning deep learning review systematic review cybersecurity embedded system artificial intelligence machine learning deep learning review systematic review

来源：评论

学校读者我要写书评

暂无评论

real-time Pedestrian Detection and Tracking System Using Deep Learning and Kalman filter: applications on embedded Systems in Advanced Driver Assistance Systems 20

Real-Time Pedestrian Detection and Tracking System Using Dee...

引用

20th Latin American Robotics symposium (LARS) / 15th Brazilian symposium on Robotics (SBR) / 14th Workshop on Robotics in Education (WRE)

作者： Bruno, Diego Renan Osorio, Fernando Santos Univ Sao Paulo Inst Math & Comp Sci Sao Carlos Brazil

ISBN: (纸本)9798350315387

In this paper, we present a perception system for assisting robotic vehicles in smart cities, facilitating interaction with pedestrians, cyclists, and other motor vehicles while adhering to local traffic rules, all with the aim of enhancing traffic safety. Multiple Object Tracking (MOT) is a complex and fundamental problem in computer vision for robotic vehicles, requiring individual evaluation of various detected mobile agents to make informed decisions. To address this challenge, we utilize embedded and dedicated hardware systems, along with Deep Learning algorithms, as powerful tools for real-time processing of computer vision. In this work, we developed an Advanced Driver Assistance System (ADAS) with 91.85% (mAP) and 78.2% (IoU) accuracy for MOT using Nvidia's Jetson-Nano and optimized the Deep-SORT YOLOv7 model in conjunction with the Kalman filter algorithm to achieve this capability, and a rate equal to or greater than 50% is already considered relevant for the task of detecting dynamic obstacles.

关键词： Smart Cities Autonomous Vehicles Advanced Driver Assistance Systems Computer Vision Deep Learning

来源：评论

学校读者我要写书评

暂无评论

Rate-Distortion and Complexity Analysis of Fast Video Encoders 15

Rate-Distortion and Complexity Analysis of Fast Video Encode...

引用

15th Latin American symposium on Circuits and Systems (LASCAS)

作者： Lodi, Eduardo Ferraz Scarpatto, Arthur Seidel, Ismael Correa, Guilherme Grellert, Mateus Fed Univ Santa Catarina UFSC Embedded Comp Lab ECL Florianopolis Brazil Fed Univ Pelotas UFPel Pelotas Brazil Fed Univ Rio Grande do Sul UFRGS Porto Alegre Brazil

ISBN: (纸本)9798350381221;9798350381238

Hardware accelerators for video processing are always on demand due to the tight real-time restrictions of video applications, so it is important for designers to know which encoding tasks must be prioritized. In this paper, we compare three state-of-the-art fast video encoding implementations by evaluating encoding time, compression efficiency and the video quality of the end product. The fast implementations are called VVEnc, XEVE-EVC and SVT-AV1, and they were selected for their emerging adoption as fast video encoders by the community. Systematic experiments are performed to analyze these codecs in terms of quality, time, and compression efficiency. Our results show that VVEnc is the most efficient fast encoder in terms of compression, but it also takes more time to encode compared to the others. To simulate a realistic scenario, we evaluate how these implementations perform in a multi-core platform. Our experiments suggest that only SVT-AV1 was capable of achieving real-time encoding for HD (> 24 frames per second), and higher frame rates can only be achieved with multiple threads, showing that dedicated hardware solutions are still required for real-time encoding of high-resolution media. We hope that with this work, future researchers will have an easier time identifying the more daunting parts of the video encoding process.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

HRMP3+TECS: Component Framework for Multiprocessor real-time Operating System with Memory Protection 26

HRMP3+TECS: Component Framework for Multiprocessor Real-time...

引用

26th ieee International symposium on real-time Distributed Computing (ISORC)

作者： Takaso, Yoshitada Oyama, Hiroshi Takada, Hiroaki Azumi, Takuya Saitama Univ Grad Sch Sci & Engn Saitama Japan OKUMA Corp Oguchi Japan Nagoya Univ Inst Innovat Future Soc Nagoya Aichi Japan

ISBN: (纸本)9798350339024

The scale and demand for protection functionalities in embedded systems continue to grow as Internet of things technology develops. Simultaneously, multiprocessor real-time operating systems (RTOSs) with memory protection functionalities are broadening in use. On the other hand, the large development effort and poor reusability of multiprocessor RTOS-based development remain hindrances to their use. To solve this problem, this paper proposes a component framework for multiprocessor RTOSs that includes memory protection. In the proposed framework, OS functionalities are reframed as components. The allocation of objects to processors and protection settings, which are supported by the OS, can then be configured based on the component description. Plugins are implemented here to generate files for object generation from component descriptions of OS functionalities. In addition, access to the protection domain can be defined based on the component description. Finally, test programs are used in a performance evaluation. The proposed framework enables extensions to the component-based OS while maintaining most of the functionality and performance of the target OS.

关键词： embedded Systems Component-based Development Multiprocessors real-time Operating Systems Memory Protection

来源：评论

学校读者我要写书评

暂无评论

real-time Human Posture Recognition Using embedded Optical Fiber Sensors and Machine Learning

引用

ieee TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2025年 74卷

作者： Zhou, Xue Wang, Weijia Hui, Yaping Li, Xuegang Zhang, Xuenan Yan, Xin Cheng, Tonglei Northeastern Univ Coll Informat Sci & Engn State Key Lab Synthet Automat Proc Ind Shenyang 110819 Peoples R China Northeastern Univ Foshan Grad Sch Innovat Foshan 528311 Guangdong Peoples R China Northeastern Univ Natl Frontiers Sci Ctr Ind Intelligence & Syst Opt Shenyang 110819 Peoples R China

Smart wearable technology is highly popular in biomedical and kinesiology fields due to its lightweight, flexible, comfortable, and adaptive nature. A smart wearable knee pad was designed using a fiber Bragg grating (FBG) flexible sensor for human posture recognition, combined with machine learning algorithms to enhance recognition accuracy. By placing three FBGs with different central wavelengths in key areas of the knee, posture recognition was realized by observing shifts in the FBG central wavelengths. Although FBG sensors are highly sensitive to small deformations, manual analysis of wavelength shifts introduces errors. Therefore, machine learning algorithms were integrated with the sensor to improve automatic posture recognition accuracy. A voting model combining eXtreme gradient boosting (XGBoost) and random forest (RF) was applied to the sensor. Experimental results showed an accuracy of up to 0.9908, with precision, recall, and F1 score reaching 0.9883, 0.9897, and 0.9886, respectively. The area under the receiver operating characteristic (ROC) curve reached 0.9966, demonstrating significant improvement over traditional machine learning models. These results highlight the superior performance of machine learning techniques in human posture recognition combined with FBG sensing technology, providing an effective solution for human posture analysis and virtual reality applications.

关键词： Fiber Bragg grating (FBG) human posture recognition machine learning algorithms real-time monitoring spectrum Fiber Bragg grating (FBG) human posture recognition machine learning algorithms real-time monitoring spectrum

来源：评论

学校读者我要写书评

暂无评论

Improved Frequency Estimation for Range-Finding applications using Phase Unwrapping

Improved Frequency Estimation for Range-Finding Applications...

引用

ieee International Instrumentation and Measurement technology Conference (I2MTC)

作者： Sivanathan, Sivagunalan Roula, Mohammed Ali Li, Kang Univ South Wales Fac Comp Engn & Sci Pontypridd M Glam Wales

ISBN: (纸本)9798350380903;9798350380910

The Fast Fourier Transform (FFT) is a common signal processing operation used in frequency estimation, a technique that has many applications in a wide range of fields including radar, LiDAR, sonar, medical sciences, telecommunications, and metrology. However, in real-time applications, computational intensity often imposes restrictions on frequency resolution, and consequently introducing spectral errors. To address this limitation, the frequency resolution of FFT can be refined for enhanced accuracy by unwrapping the phase information of the frequency component, while maintaining the required FFT points and sampling rate. This paper presents an analysis of diverse windowing techniques, showcasing the potential enhancement of FFT spectral accuracy, particularly for precision-centric measurements. The proposed method is adaptable for deployment in both programmable and non-programmable real-time systems, including embedded Systems and Field-Programmable Gate Arrays (FPGAs).

关键词： Frequency Estimation Phase unwrapping Fast Fourier Transform LiDAR measurement radar Spectral Leakage

来源：评论

学校读者我要写书评

暂无评论

RD-Gen: Random DAG Generator Considering Multi-rate applications for Reproducible Scheduling Evaluation 26

RD-Gen: Random DAG Generator Considering Multi-rate Applicat...

引用

26th ieee International symposium on real-time Distributed Computing (ISORC)

作者： Yano, Atsushi Azumi, Takuya Tier IV Inc Nagoya Aichi Japan Saitama Univ Grad Sch Sci & Engn Saitama Japan

ISBN: (纸本)9798350339024

real-time systems have various requirements such as the deadline and resource constraints. In addition, real-time systems are becoming larger and more complex, and studies on performance analysis and efficient scheduling algorithms are becoming increasingly important. Directed acyclic graph (DAG) models, which can express task dependencies and parallelism, are used for such studies. Random DAG sets are used to demonstrate the effectiveness and objectivity of methods proposed for real-time systems. However, there is no random DAG generation tool available that can generate a DAG set that considers the latest multi-rate applications. Therefore, researchers need to generate random DAG sets on their own, leading to additional effort and reduced reliability and reproducibility. To solve this problem, we propose a random DAG generator considering multi-rate applications for reproducible scheduling evaluation (RD-Gen). RD-Gen also enables batch generation of random DAG sets with different parameters. Case studies are used to demonstrate that RD-Gen can manage various problem settings and DAG study requirements.

关键词： DAG Random generation tool Multi-rate applications

来源：评论

学校读者我要写书评

暂无评论

Managing and Optimizing 5G & Beyond Network Resources for Multi-Task Digital Twin applications in Industry 4.0 26

Managing and Optimizing 5G & Beyond Network Resources for Mu...

引用

26th ieee International symposium on real-time Distributed Computing (ISORC)

作者： Min, Ziran Zhou, Shuang Kang, Zhuangwei Shekhar, Shashank Mahmoudi, Charif Gokhale, Swapna Gokhale, Aniruddha Vanderbilt Univ Dept CS Nashville TN 37235 USA Siemens Technol Princeton NJ 08540 USA Univ Connecticut Dept CSE Storrs CT USA

ISBN: (纸本)9798350339024

Industry 4.0 is leading factories to undergo a significant transformation, where automation is achieved through the use of modern smart technologies, such as 5G & beyond (5G+) network and digital twins. Yet, many Industrial Internet of Things (IIoT) applications, including smart factories and robotic repair, present challenges in delivering dedicated and real-time network services between the physical world entities and their digital twins due to the different network requirements of each sub tasks of the applications. Although 5G+ networks can provide high-speed, low-latency, and reliable network services, managing and optimizing the network resources in real-time remains complex and time-consuming. To address these challenges, this paper proposes solutions to manage and optimize 5G+ network resources in real-time, and deliver dynamic and real-time network requirements of multi-task digital twin applications.

关键词： Digital Twins 5G and beyond Industry 4.0 Industrial Internet of Things (IIoT) Software-defined Networking (SDN)

来源：评论

学校读者我要写书评

暂无评论

Design and Implementation of FPGA based System for Object Detection and Range Estimation used in ADAS applications utilizing FMCW Radar

Design and Implementation of FPGA based System for Object De...

引用

ieee International symposium on Circuits and Systems (ISCAS)

作者： Khan, Mujeev Mahajan, Pranjal Khan, Gani Nawaz Chaudhary, Devansh Benny, Jewel Wajid, Mohd Srivastava, Abhishek IHT Hyderabad Ctr VLSI & Embedded Syst Technol CVEST Hyderabad Telangana India ZHCET Aligarh Muslim Univ Dept Elect Engn Aligarh Uttar Pradesh India

ISBN: (纸本)9798350330991;9798350331004

This paper presents the design and implementation of a hardware system for real-time object detection and range estimation utilizing Frequency-Modulated Continuous Wave (FMCW) millimeter wave radar signals, which are commonly used in Advance Driver Assistance Systems (ADAS) and robotic applications. The proposed system utilizes the Fast Fourier Transform (FFT) algorithm to process the FMCW radar signal for estimating the range of an object. For developing a resource-efficient and low latency range-estimation hardware, this paper also presents a comparative analysis for the implementation of three popular FFT architectures (iterative Radix-2, iterative Radix-2(2), and pipelined Radix-2) on FPGA. Based on the presented analysis the lowest latency range-estimation architecture has been chosen for FFT implementation and a system prototype is developed as a proof-of-concept by integrating the commercially available Texas Instruments' 77 GHz AWR1642BOOST FMCW radar module with a FPGA to host the chosen FFT architecture. Measurement results of the proposed system hardware are also presented in this paper, which shows > 99.21% range-estimation accuracy.

关键词： ADAS FFT FMCW Radar FPGA range-estimation Object-detection

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：