检索结果-内蒙古大学图书馆

Enhanced UAV Pursuit-Evasion Using Boids Modelling:A Synergistic Integration of Bird Swarm Intelligence and DRL

computers, Materials & Continua 2024年第9期80卷 3523-3553页

作者： Weiqiang Jin Xingwu Tian Bohang Shi Biao Zhao Haibin Duan Hao Wu School of Information and Communications Engineering Faculty of Electronic and Information EngineeringXi’an Jiaotong UniversityXi’an710049China China Academy of Electronics and Information Technology China Electronics Technology Group Corporation(CETC)Beijing100041China Department ofAutomatic Control School of Automation Science and Electrical EngineeringBeihangUniversityBeijing100191China

TheUAV pursuit-evasion problem focuses on the efficient tracking and capture of evading targets using unmanned aerial vehicles(UAVs),which is pivotal in public safety applications,particularly in scenarios involving intrusion monitoring and *** address the challenges of data acquisition,real-world deployment,and the limited intelligence of existing algorithms in UAV pursuit-evasion tasks,we propose an innovative swarm intelligencebased UAV pursuit-evasion control framework,namely“Boids Model-based DRL Approach for Pursuit and Escape”(Boids-PE),which synergizes the strengths of swarm intelligence from bio-inspired algorithms and deep reinforcement learning(DRL).The Boids model,which simulates collective behavior through three fundamental rules,separation,alignment,and cohesion,is adopted in our *** integrating Boids model with the Apollonian Circles algorithm,significant improvements are achieved in capturing UAVs against simple evasion *** further enhance decision-making precision,we incorporate a DRL algorithm to facilitate more accurate strategic *** also leverage self-play training to continuously optimize the performance of pursuit *** experimental evaluation,we meticulously designed both one-on-one and multi-to-one pursuit-evasion scenarios,customizing the state space,action space,and reward function models for each *** simulations,supported by the PyBullet physics engine,validate the effectiveness of our proposed *** overall results demonstrate that Boids-PE significantly enhance the efficiency and reliability of UAV pursuit-evasion tasks,providing a practical and robust solution for the real-world application of UAV pursuit-evasion missions.

关键词： UAV pursuit-evasion swarm intelligence algorithm Boids model deep reinforcement learning self-play training

来源：评论

学校读者我要写书评

暂无评论

Efficient Adaptation of Pre-trained Vision Transformer via Householder Transformation 38

Efficient Adaptation of Pre-trained Vision Transformer via H...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Dong, Wei Sun, Yuan Yang, Yiting Zhang, Xing Lin, Zhijun Yan, Qingsen Zhang, Haokui Wang, Peng Yang, Yang Shen, Hengtao College of Information and Control Engineering Xi'an University of Architecture and Technology China School of Computer Science Northwestern Polytechnical University China School of Computer Science and Engineering University of Electronic Science and Technology of China China School of Computer Science and Technology Tongji University China

A common strategy for Parameter-Efficient Fine-Tuning (PEFT) of pre-trained Vision Transformers (ViTs) involves adapting the model to downstream tasks by learning a low-rank adaptation matrix. This matrix is decomposed into a product of down-projection and up-projection matrices, with the bottleneck dimensionality being crucial for reducing the number of learnable parameters, as exemplified by prevalent methods like LoRA and Adapter. However, these low-rank strategies typically employ a fixed bottleneck dimensionality, which limits their flexibility in handling layer-wise variations. To address this limitation, we propose a novel PEFT approach inspired by Singular Value Decomposition (SVD) for representing the adaptation matrix. SVD decomposes a matrix into the product of a left unitary matrix, a diagonal matrix of scaling values, and a right unitary matrix. We utilize Householder transformations to construct orthogonal matrices that efficiently mimic the unitary matrices, requiring only a vector. The diagonal values are learned in a layer-wise manner, allowing them to flexibly capture the unique properties of each layer. This approach enables the generation of adaptation matrices with varying ranks across different layers, providing greater flexibility in adapting pre-trained models. Experiments on standard downstream vision tasks demonstrate that our method achieves promising fine-tuning performance. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Comprehensive Survey on Machine and Deep Learning for Optical Communications

引用

IEEE Access 2025年 13卷 88794-88846页

作者： Amirabadi, Mohammad Ali Nezamalhosseini, S. Alireza Kahaei, Mohammad Hossein Chen, Lawrence R. Ferdowsi University of Mashhad Department of Electrical Engineering Mashhad*** Iran Iran University of Science and Technology School of Electrical Engineering Tehran*** Iran McGill University Department of Electrical and Computer Engineering MontrealQCH3A 0G4 Canada

The increasing complexity of optical communication systems and networks necessitates advanced methodologies for extracting valuable insights from vast and heterogeneous datasets. Machine learning (ML) and deep learning (DL) have emerged as pivotal tools in this domain, revolutionizing data analysis and enabling automated self-configuration in optical communication systems. Their adoption is fueled by the growing intricacy of systems and links, driven by numerous adjustable and interdependent parameters. This complexity is particularly evident in areas such as coherent transceivers, advanced digital signal processing, optical performance monitoring, cross-layer network optimizations, and nonlinearity compensation. While ML and DL offer immense potential, their application in optical communications is still in its early stages, with significant opportunities remaining unexplored. Many algorithms have yet to be fully deployed in practical settings, underscoring the emerging nature of this research area. This paper presents a comprehensive survey of ML and DL applications across optical fiber communication (OFC), optical wireless communication (OWC), and optical communication networking (OCN), highlighting the challenges, current advancements, and future potential of these approaches. To address the identified gaps, this survey evaluates and compares ML and DL algorithms in terms of their performance, complexity, objectives, input data, metrics, and applications in optical communication. Specific emphasis is placed on identifying how these algorithms enhance system performance and optimization. Furthermore, the advantages and limitations of existing methods are analyzed, offering a clear perspective on the role of ML and DL in this domain. The survey also includes updated visual representations and domain-specific examples to elucidate the practical applications of ML and DL in OFC, OWC, and OCN. It concludes by discussing the open challenges, proposing potential soluti

关键词： Fiber to the x

来源：评论

学校读者我要写书评

暂无评论

Simulation and Modeling of Emerging Heterojunction Transistors: Toward Rational Design and Optimization

IEEE Journal on Flexible Electronics

引用

IEEE Journal on Flexible Electronics 2024年第11期3卷 492-501页

作者： Yoo, Hocheon Hayakawa, Ryoma Wakayama, Yutaka Kim, Chang-Hyun Gachon University School of Electronic Engineering Seongnam13120 Korea Republic of Tsukuba305-0044 Japan University of Ottawa School of Electrical Engineering and Computer Science OttawaONK1N 6N5 Canada

This review summarizes the recent progress in the simulation and modeling of two important types of thin-film heterojunction transistors. Both structures functionally rely on the formation of a lateral p-n junction inside a single source-to-drain channel. However, different extents of vertical overlap between the p- and n-type semiconductor films produce unique switching characteristics of the respective systems. Experimental evidences behind recent theoretical models and the predictive capabilities of these models are illustrated with published examples. Simulation results of multivalued logic (MVL) circuits built with these heterojunction transistors are also discussed to emphasize practical aspects of device modeling for circuit- and system-level optimization. © 2024 IEEE.

关键词： Heterojunctions

来源：评论

学校读者我要写书评

暂无评论

Research on Grayscale Images Flame Detections Based on Colorazion Algorithm and YOLOv8

Research on Grayscale Images Flame Detections Based on Color...

引用

2024 IEEE International Conference on Imaging Systems and Techniques, IST 2024

作者： Liu, Peng Sun, Zhenfu Zhou, Biao Wang, Hongyu Qu, Haozhen Institute of Energy and Electrical Engineering Changchun University of Science and Technology Changchun China School of Control and Computer Engineering North China Electric Power University Beijing China

ISBN: (纸本)9798350378214

Flame detection algorithms are crucial for real-time fire monitoring using surveillance cameras. Current flame detection algorithms perform excellently on color cameras;however, many night vision cameras can only capture grayscale images, making flame detection less effective in such scenarios. Traditional flame detection algorithms rely heavily on color features, especially the red and yellow hues of flames. Conversely, grayscale images lack color information, making these algorithms ineffective. To address this limitation, we propose a novel flame detection algorithm that integrates the DeOldify model with the you only look once(YOLOv8) model. DeOldify can effectively add color to grayscale images. YOLOv8 is a powerful object detection model that excels in identifying various objects within an image. This algorithm leverages DeOldify's colorization capabilities and YOLOv8's powerful object detection abilities to enhance flame detection accuracy in grayscale images. This algorithm has been tested multiple times on public datasets, converting original images to grayscale and then colorizing them to construct the dataset. The results show that, compared to grayscale images, this algorithm improves the mean average precision (mAP@0.5) by 19.7% on different validation sets, by 4% on different training sets, and by 12.9% on grayscale images captured by night vision cameras. By comparing the training results of models such as YOLOv8n and YOLOv8s, the colorized images improve the model's performance. This demonstrates the high generalizability of this algorithm. © 2024 IEEE.

关键词： Premixed flames

来源：评论

学校读者我要写书评

暂无评论

The propagation of Quasi-static region during granular impact

引用

Particuology 2023年第12期83卷 1-7页

作者： Moyu Yang Ran Li Yingnan Xiu Qiang Zeng Xiaoyan Ye Hui Yang School of Optical-Electrical and Computer Engineering University of Shanghai for Science and TechnologyShanghai200093China Center for Particle-Laden Turbulence Key Laboratory of Mechanics on Disaster and Environment in Western China attached to the Ministry of Education of ChinaDepartment of Mechanics and Engineering ScienceSchool of Civil Engineering and MechanicsLanzhou UniversityLanzhou730000China

In this paper,through experiments and DEM simulations,it is found that there is a ring-shaped region named Quasi-static Region between the particles expanding outward and the particles collapsing inward during the impact ***-static Region is always generated from the impact point at the same time,and then spreads out at a uniform *** the propagation of Quasi-static Region,the velocity of the particles in the particle extend region varies linearly in space and *** the change rate is only related to the properties of the particle *** simulation results show that the particle flow in Quasi-static Region is elastic-inertial flow,while the particle flow expanding outward and collapsing inward is inertial flow.

关键词： Impact Quasi-staticregion Propagation Shearflow

来源：评论

学校读者我要写书评

暂无评论

Joint trust-based detection and signature-based authentication technique for secure localization in underwater wireless sensor network

引用

Multimedia Tools and Applications 2025年 1-20页

作者： Kumar, Manni Goyal, Nitin Qaisi, Ramy Mohammed Aiesh Najim, Mohd Gupta, Sachin Kumar Department of Computer Science & Engineering Chandigarh University Gharuan Punjab India Department of Computer Science Engineering School of Engineering and Technology Central University of Haryana Haryana Mahendergarh123031 India Department of Electrical and Electronic Engineering College of Engineering University of Jeddah Jeddah21589 Saudi Arabia Department of Electronics and Communication Engineering Central University of Jammu UT of J&K Jammu181143 India

Ensuring secure and accurate node localization in Underwater Wireless Sensor Networks (UWSN) is a significant challenge, as conventional methods tend to neglect the security risks associated with malicious node interference. This research addresses this issue by proposing a Joint Trust-based Detection and Signature-based Authentication (JTDSA) technique, aimed at improving both the security and accuracy of node localization in UWSNs. The technique utilizes Intrusion Detection System (IDS) methods to calculate the trust values of secondary-anchor nodes (SNs), ensuring only the most trusted SNs are selected. Furthermore, a digital signature algorithm secures the Received Signal Strength (RSS) data from tampering. The method incorporates a hybrid approach combining Angle of Arrival (AoA) and RSS ranging techniques to achieve robust localization. Simulation results demonstrate that JTDSA improves localization accuracy by 4% compared to existing methods, achieving an overall accuracy of 99.2%. The technique also reduces packet drop by 35% and enhances attack detection accuracy by 3%. These results indicate that the proposed JTDSA technique offers an effective solution for secure and reliable localization in UWSNs. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Sensor nodes

来源：评论

学校读者我要写书评

暂无评论

Suitability of SDN and MEC to facilitate digital twin communication over LTE-A

引用

Digital Communications and Networks 2024年第2期10卷 347-354页

作者： Hikmat Adhami Mohammad Alja’afreh Mohamed Hoda Jiaqi Zhao Yong Zhou Abdulmotaleb El Saddik Mohamed bin Zayed University of Artificial Intelligence Masdar CityAbu DhabiUnited Arab Emirates School of Computer Science and Technology China University of Mining and TechnologyXuzhou221116China School of Electrical Engineering and Computer Science University of OttawaOttawaONK1N 6N5Canada

Haptic is the modality that complements traditional multimedia,i.e.,audiovisual,to evolve the next wave of innovation at which the Internet data stream can be exchanged to enable remote skills and control *** will require ultra-low latency and ultra-high reliability to evolve the mobile experience into the era of Digital Twin and Tactile *** the 5th generation of mobile networks is not yet widely deployed,Long-Term Evolution(LTE-A)latency remains much higher than the 1 ms requirement for the Tactile Internet and therefore the Digital *** work investigates an interesting solution based on the incorporation of Software-defined networking(SDN)and Multi-access Mobile Edge Computing(MEC)technologies in an LTE-A network,to deliver future multimedia applications over the Tactile Internet while overcoming the QoS *** network scenarios were designed and simulated using Riverbed modeler and the performance was evaluated using several time-related Key Performance Indicators(KPIs)such as throughput,End-2-End(E2E)delay,and *** best scenario possible is clearly the one integrating MEC and SDN approaches,where the overall delay,jitter,and throughput for haptics-attained 2 ms,0.01 ms,and 1000 packets per *** results obtained give clear evidence that the integration of,both SDN and MEC,in LTE-A indicates performance improvement,and fulfills the standard requirements in terms of the above KPIs,for realizing a Digital Twin/Tactile Internet-based system.

关键词： LTE-A 5G Digital twin Tactile Internet KPIs SDN MEC NFV Haptic

来源：评论

学校读者我要写书评

暂无评论

Frequency Attacks Based on Invertible Neural Networks

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年第2期6卷 292-300页

作者： Shao, Ming-Wen Yang, Jian-Xin Meng, Ling-Zhuang Hu, Zhi-Yong China University of Petroleum School of Computer Science and Technology East China Qingdao266580 China China University of Petroleum College of Control Science and Engineering East China Qingdao266580 China

Adversarial attacks reveal the vulnerability of classifiers based on deep neural networks to well-designed perturbations. Most existing attack methods focus on adding perturbations directly to the pixel space. However, the perturbations generated by these methods may be easily perceived by humans. To alleviate the aforementioned problem, we propose a novel high-frequency attack based on invertible neural networks (HA-INN) that relies on INNs to adds perturbations to the high-frequency space of the image instead of the pixel space. In this way, we can fool the classifier while the perturbations are not easily detected by humans. Specifically, we introduce INNs to separate the high-frequency and low-frequency components of the image. And the low-frequency components are guaranteed to be reconstructed as the original image, while the high-frequency components are replaced with resampled high-frequency latent variables with additional adversarial information. Then, the low-frequency components and the high-frequency components with adversarial information are inversely fed into the INN to generate effective adversarial examples. Extensive experiments on two datasets (CIFAR-10 and CIFAR-100) show that our method can generate misleading and transferable cross-architectural adversarial examples with greatly reduced computational resource requirements. Under the white-box setting, the attack success rate of CIFAR10 and CIFAR100 is 99.8% and 99.74%, respectively. Further, under the black-box setting, the adversarial examples generated by our method are more effective than the other methods. © 2020 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Fu-Rec:Multi-Task Learning Recommendation Model Fusing Neighbor-Discrimination and Self-Discrimination

引用

Wuhan University Journal of Natural sciences 2024年第2期29卷 134-144页

作者： ZHENG Sirui HUANG Bo LIU Jin ZENG Guohui YIN Ling LI Zhi SUN Tie School of Electronic and Electrical Engineering Shanghai University of Engineering ScienceShanghai 201600China School of Computer Wuhan UniversityWuhan 430072HubeiChina School of Computer Science and Engineering Guangxi Normal UniversityGuilin 541004GuangxiChina AIoT Manufacturing Solutions Technology Co. Ltd.Hefei 230000AnhuiChina

In recent years,self-supervised learning has achieved great success in areas such as computer vision and natural language processing because it can mine supervised signals from unlabeled data and reduce the reliance on manual ***,the currently generated self-supervised signals are either neighbor discrimination or self-discrimination,and there is no model to integrate neighbor discrimination and *** on this,this paper proposes Fu-Rec that integrates neighbor-discrimination contrastive learning and self-discrimination contrastive learning,which consists of three modules:(1)neighbor-discrimination contrastive learning,(2)selfdiscrimination contrastive learning,and(3)recommendation *** neighbor-discrimination contrastive learning and selfdiscrimination contrastive learning tasks are used as auxiliary tasks to assist the recommendation *** Fu-Rec model effectively utilizes the respective advantages of neighbor-discrimination and self-discrimination to consider the information of the user’s neighbors as well as the user and the item itself for the recommendation,which results in better performance of the recommendation *** results on several public datasets demonstrate the effectiveness of the Fu-Rec proposed in this paper.

关键词： self-supervised learning recommendation system contrastive learning multi-task learning

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：