检索结果-内蒙古大学图书馆

Multi modal data fusion defense strategy for campus network security: research on Kolmogorov Arnold Networks combined with B-spline function

引用

Discover Computing 2025年第1期28卷 1-25页

作者： Hu, Zhiying Wang, Lifeng Ding, Xiaomei Zhao, Lihong Xue, Minghe School of Computer Engineering Anhui Wenda University of Information Engineering Anhui Hefei 231201 China Institute of Plasma Physics Institutes of Physical Science Chinese Academy of Sciences Anhui Hefei 230031 China

This article proposes a Kolmogorov Arnold Networks (KANs) model that combines B-spline functions and is applied to a multimodal data fusion defense strategy for campus network security. Simulation experiments analyze the optimization effect of KANs models in various scenarios, including factors like network layers, adaptive learning rate, early stopping, L2 regularization, weighted loss function, and batch normalization. Experimental results show that, after optimizing the network structure and adopting adaptive learning rates, the KANs model converges more quickly and avoids overfitting. Especially with the introduction of early stopping strategy, L2 regularization, and Batch Normalization, the generalization ability of the model has been significantly improved. The strategy based on multimodal data fusion effectively improves the robustness and accuracy of the model when processing different types of data, such as images, text, time series data, etc. Experiments have shown that through weighted loss functions and multi task learning strategies, the model exhibits higher accuracy and lower false alarm rates when dealing with imbalanced samples and complex attack patterns. Overall, the proposed KANs model, combined with B-spline functions, not only optimizes campus network security defense but also improves the accuracy and robustness of multimodal data fusion tasks. © The Author(s) 2025.

关键词： B-spline function Campus network security Kolmogorov Arnold Networks (KANs) Multimodal data fusion Network optimization Overfitting Robustness

来源：评论

学校读者我要写书评

暂无评论

When liquid crystals encounter structured light 20

When liquid crystals encounter structured light

引用

Emerging Liquid Crystal Technologies XX 2025

作者： Zhou, Le Liu, Yuanfeng Neyts, Kristiaan Shen, Yang Sun, Jingbo Zhou, Ji Physics Department Colorado School of Mines 1523 Illinois St. GoldenCO80401 United States School of Materials Science and Engineering State Key Laboratory of New Ceramics and Fine Processing Tsinghua University Beijing100084 China Department of Electronic and Computer Engineering State Key Laboratory of Advanced Displays and Optoelectronics Technologies The Hong Kong University of Science and Technology Clear Water Bay Kowloon Hong Kong

ISBN: (纸本)9781510685222

Structured light refers to the light field tailored by various degrees of freedom including intensity, phase, and polarization states in both spatial and temporal domains, which may greatly vitalize the technologies in both optics, such as the next-generation optical communication as well as imaging and the materials science in both fabrications and characterizations. The structured characteristics of the structure light need materials also with structured optical properties that can generate or manipulate structured light in a straightforward way, which can be well satisfied by liquid crystals or glassy liquid crystals such as special spherulites. © 2025 SPIE.

关键词： Degrees of freedom (mechanics)

来源：评论

学校读者我要写书评

暂无评论

Study of heat transfer and flow within atherosclerotic plaques in a focused ultrasound field

引用

Ultrasonics 2025年 154卷 107699页

作者： Lei, Weirui Hu, Jiwen Zhai, Jintao Gong, Jinru Tian, Feng Chang, Shuai Zou, Xiao Ju, Fangfang Qian, Shengyou School of Physics and Electronics Hunan Normal University Changsha410081 China School of Mathematics and Physics University of South China Hengyang421001 China College of Computer Science and Electronic Engineering Hunan University Changsha410012 China

Focused ultrasound has been widely used for the thermotherapy of soft tissue lesions. In this process, non-Fourier heat conduction and porous medium theory has to be considered because of non-homogeneous media. The study estimates the effects of the temperature lag and porous medium on the plaque ablation and drug treatment by focused ultrasound (FU). This study integrated TWMBT with the porous media heat transfer equation to characterize the internal temperature distribution within atherosclerotic plaque (AP) during FU application. The coupling equations are solved with finite element method. This paper focuses on the effects of porosity, permeability, and attenuation coefficient on the temperature and flow rate within the AP. The results consider artery wall thickness on heating of AP by FU. In addition, this study qualitatively analyzed the differences among the Pennes, TWMBT, and porous media heat conduction equations. The results show that the temperature responses of biological tissues exhibits lagging behaviors, which are inherently related to the physical time scale. Because of the disparities in the physical characteristics of the target and surrounding tissues, fluid flow within AP can have an impact on the distribution of tissue temperature, the direction of flow between solid tissues is determined by the permeability coefficient and ultrasonic intensity. The permeability coefficient, frequency and attenuation coefficient have a significant effect on the fluid flow within AP. Both heat dissipation and heat convergence are characteristics of fluid flow within the tissue, the focal location and the physical property parameters may affect the fluid heat dissipation and heat collection properties within the tissue. Furthermore, the temperature peak may not occur at the focus. The model can provide an analytical template for different types of precise thermal ablation AP, including radiofrequency ablation, microwave therapy, and laser ablation besides FU ablat

关键词： Electric heating

来源：评论

学校读者我要写书评

暂无评论

Intelligent 5G Evolution: AI-Augmented Hybrid Model with Network Slicing and Autonomous Self-Healing for Future-Ready Connectivity 3

Intelligent 5G Evolution: AI-Augmented Hybrid Model with Net...

引用

3rd International Conference on Electronics and Renewable Systems, ICEARS 2025

作者： Endla, Purushotham Pure, Ramya S. Soni, Anita Andal, C. Jagadeeswari, D. Devi, K. Nirmala Sr University School of Sciences and Humanities Department of Physics Telangana Warangal506371 India Mlr Institute of Technology Department of Computer Science and Engineering Hyderabad India Ies University Cse Department Bhopal India Dr. Mgr Educational and Research Institute Chennai India Sri muthukumaran Institute of Technology Department of Ece India Madanapalle Institute of Technology & Science Deartment of Computer Applications India

ISBN: (纸本)9798331509675

The increasing demand for dynamic, efficient, and resilient network management in 5G systems has driven advancements in AI-enhanced architectures. This research proposes an AI-Augmented Hybrid 5G Model that integrates dynamic network slicing, self-healing capabilities, and AI-driven predictive analytics to optimize network performance and reliability. The hybrid model employs supervised and unsupervised machine learning algorithms, achieving a fault detection accuracy of 95.2% and traffic demand prediction accuracy of 92.8%. Key innovations include dynamic resource allocation for enhanced mobile broadband (eMBB), ultra-reliable low-latency communications (URLLC), and massive machine-type communication (mMTC). Self-healing capabilities reduced fault recovery times to an average of 300 ms, significantly improving system uptime to 99.8%. Compared to traditional 5G systems, the model demonstrates a 50% reduction in latency, a 28.6% improvement in resource efficiency, and an 80% faster slice reconfiguration. These results highlight the model's potential to revolutionize 5G network management, ensuring reliability and adaptability for diverse applications. The study concludes by exploring the applicability of the hybrid model in future 6G networks and cross-domain orchestration. © 2025 IEEE.

关键词： Predictive analytics

来源：评论

学校读者我要写书评

暂无评论

A blockchain-oriented covert communication technology with controlled security level based on addressing confusion ciphertext

引用

Frontiers of computer science 2025年第2期19卷 133-135页

作者： Lejun ZHANG Bo ZHANG Ran GUO Zhujun WANG Guopeng WANG Jing QIU Shen SU Yuan LIU Guangxia XU Zhihong TIAN Sergey GATAULLIN Cyberspace Institute Advanced Technology Guangzhou UniversityGuangzhou 510006China Research and Development Center for E-Learning Ministry of EducationBeijing 100039China College of Information Engineering Yangzhou UniversityYangzhou 225009China School Math and Computer Science Quanzhou Normal UniversityQuanzhou 362000China Pazhou Lab Guangzhou 510330China School of Physics and Materials Science Guangzhou UniversityGuangzhou 510006China Central Economic and Mathematics Institute Russian Academy of SciencesMoscow 117418Russia Industrial Programming Department MIREA-Russian Technological UniversityMoscow 119454Russia

1 Introduction Traditional covert communication[1]usually relies on centralized channels,which are efficient in transmission but susceptible to eavesdropping and attacks,leading to information ***,distributed methods also exist,such as peer-to-peer networks and distributed hash *** technology,which is decentralized,anonymous,non-tamperable and anti-attack,can be applied to covert communication to solve the problems of traditional methods and improve the quality of communication.

关键词： communication tables networks

来源：评论

学校读者我要写书评

暂无评论

Rapid Single-Flux-Quantum Circuit Routing Algorithm Considering Length Matching

引用

Jisuanji Yanjiu yu Fazhan/computer Research and Development 2025年第5期62卷 1151-1163页

作者： Liu, Genggeng Yu, Yantao Zhou, Ruping Wei, Rongshan Xu, Ning College of Computer and Data Science Fuzhou University Fuzhou350116 China Fuzhou350116 China College of Physics and Information Engineering Fuzhou University Fuzhou350108 China School of Information Engineering Wuhan University of Technology Wuhan430070 China

The high frequency characteristics of rapid single-flux-quantum (RSFQ) circuits poses a great challenge to circuit layout design. In order to solve the circuit delay problem caused by the high frequency characteristics of RSFQ, delay elements such as passive transmission line can be used in the routing stage. The delay of a passive transmission line is roughly proportional to its length, and the power consumption of the passive transmission line does not increase with the increase of the wirelength, so length matching routing is a crucial problem for RSFQ circuits. Therefore, we propose an efficient RSFQ circuit routing algorithm considering length matching, including the following key strategies: 1) when generating the initial path, a method of detour routing is presented to meet the partial length matching of passive transmission lines without changing the initial routing space;2) an iterative resource insertion algorithm based on region-awareness is utilized to reduce the area of additional resources needed to be added;3) a length-matching driven routing algorithm considering blocking cost is designed, which improves the resource utilization of routing space. Experimental results show that, compared with existing multi-terminal routing algorithms, the proposed algorithm reduces the area required for routing by 8% and the running time by 36%, thus achieving fast and high-quality routing results. © 2025 science Press. All rights reserved.

关键词： Electric delay lines

来源：评论

学校读者我要写书评

暂无评论

Swarm Characteristic Classification using Robust Neural Networks with Optimized Controllable Inputs

arXiv

引用

arXiv 2025年

作者： Peltier, Donald W. Kaminer, Isaac Clark, Abram Orescanin, Marko Naval Postgraduate School MontereyCA93943 United States Department of Mechanical & Aerospace Engineering Department of Physics Department of Computer Science

Having the ability to infer characteristics of autonomous agents would profoundly revolutionize defense, security, and civil applications. Our previous work was the first to demonstrate that supervised neural network time series classification (NN TSC) could rapidly predict the tactics of swarming autonomous agents in military contexts, providing intelligence to inform counter-maneuvers. However, most autonomous interactions, especially military engagements, are fraught with uncertainty, raising questions about the practicality of using a pretrained classifier. This article addresses that challenge by leveraging expected operational variations to construct a richer dataset, resulting in a more robust NN with improved inference performance in scenarios characterized by significant uncertainties. Specifically, diverse datasets are created by simulating variations in defender numbers, defender motions, and measurement noise levels. Key findings indicate that robust NNs trained on an enriched dataset exhibit enhanced classification accuracy and offer operational flexibility, such as reducing resources required and offering adherence to trajectory constraints. Furthermore, we present a new framework for optimally deploying a trained NN by the defenders. The framework involves optimizing defender trajectories that elicit adversary responses that maximize the probability of correct NN tactic classification while also satisfying operational constraints imposed on the defenders. © 2025, CC BY.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

High-brightness betatron X-ray source driven by SULF-1 PW laser

引用

High Power Laser science and engineering 2025年

作者： Zhang, Hong Deng, Zhigang Jiang, Hai Wang, Shaoyi Wei, Jianmeng Ge, Yanjie Chu, Genbai Chen, Xizhuan Wang, Hao Yan, Yonghong Feng, Ke Jiang, Kangnan Hu, Runshu Tan, Fang Zeng, Gaojie Guo, Hang Yang, Xintao Qian, Jiayi Zhu, Jiacheng Zhang, Zongxin Xu, Yi Leng, Yuxin Zhou, Weimin Li, Song Wang, Wentao Li, Ruxin Shanghai China University of Chinese Academy of Sciences Beijing China Science and Technology on Plasma Physics Laboratory Laser Fusion Research Center China Academy of Engineering Physics Mianyang China School of Optical-Electrical and Computer Engineering University of Shanghai for Science and Technology Shanghai China

The betatron radiation source, features a micrometer-scale source size, femtosecond-scale pulse duration, milliradian-level divergence angles, and a broad spectrum exceeding tens of keV. It is conducive to the high-contrast imaging of minute structures and for investigating interdisciplinary ultrafast processes. In this study, we present a betatron X-ray source derived from a high-charge, high-energy electron beam through a laser wakefield accelerator driven by the 1 PW/0.1 Hz laser system at the Shanghai Superintense Ultrafast Laser Facility (SULF). The critical energy of betatron X-ray source is 22±5 keV. The maximum X-ray flux reaches up to 4×109 photons for each shot in the spectral range of 5-30 keV. Correspondingly, the experiment demonstrates a peak brightness of 1.0 × 1023 photons·s-1·mm-2·mrad-2·0.1%BW-1, comparable to those demonstrated by third-generation synchrotron light sources. Additionally, the imaging capability of the betatron X-ray source is validated. This study lays the foundation for future imaging applications. © 2025 Cambridge University Press. All rights reserved.

关键词： Betatrons

来源：评论

学校读者我要写书评

暂无评论

MadVoro: Parallel Construction of Voronoi Diagrams in Distributed Memory Systems

arXiv

引用

arXiv 2025年

作者： Mizrachi, Maor Raveh, Barak Steinberg, Elad School of Computer Science and Engineering The Hebrew University Jerusalem9190401 Israel Racah Institute of Physics The Hebrew University Jerusalem9190401 Israel

Voronoi diagrams are essential geometrical structures with numerous applications, particularly astrophysics-driven finite volume methods. While serial algorithms for constructing these entities are well-established, parallel construction remains challenging. This is especially true in distributed memory systems, where each host manages only a subset of the input points. This process requires redistributing points across hosts and accurately computing the corresponding Voronoi cells. In this paper, we introduce a new distributed construction algorithm, which is implemented in our open-source C++ 3-dimensional Voronoi construction framework. Our approach leverages Delaunay triangulation as an intermediate step, which is then transformed into a Voronoi diagram. We introduce the algorithms we implemented for the precise construction and our load-balancing approach and compare the running time with other state-of-the-art frameworks. MadVoro is a versatile tool that can be applied in various scientific domains, such as mesh decomposition, computational physics, chemistry, and machine learning. © 2025, CC BY.

关键词： Finite volume method

来源：评论

学校读者我要写书评

暂无评论

Enhancing Healthcare Recommendations for Special-Need Patients Using Deep Learning and Blockchain Security 15th

Enhancing Healthcare Recommendations for Special-Need Patien...

引用

15th International Conference on Intelligent Computing and Networking, IC-ICN 2024

作者： Mudhafar, Mustafa Raparthi, Mohan Thota, Vijayakumar Sivakumar, B.N. Degadwala, Sheshang Department of Medical Physics College of Applied Medical Sciences University of Kerbala Karbala56001 Iraq Alphabet Life Science DallasTX75063 United States NSB Business School Bangalore University Bangalore India Department of Computer Engineering Sigma University Gujarat Vadodara India

ISBN: (纸本)9789819786305

Special needs patients deserve increasingly specialised and secure healthcare supervision as the healthcare environment evolves. Our ground-breaking approach combines deep learning and blockchain security to enhance healthcare recommendations for people with special needs. This novel approach might assist with the difficulties of providing tailored healthcare advice. To get the best results, we mix two efficient technologies. Deep learning allows you to detect specific patterns in large patient datasets. Modern neural networks and algorithms are what power it. In the recommended approach, deep learning insights resulted in 92% accuracy, 89% precision, 94% recall, and a 91% F1 score. Deep learning allows for extremely accurate and customised healthcare suggestions. Special needs individuals need customised treatment regimens due to their specific medical challenges. Blockchain security is critical for ensuring patient data validity and confidentiality. To secure patient data, blockchain technology generates a distributed, immutable ledger and an open, auditable database. Encryption protects data's privacy, prohibiting unauthorised access to patient data. These developments in healthcare advice may increase treatment quality and patient trust in the system. Our analysis showed positive and game-changing findings, with the suggested technique outperforming traditional methods. These results illustrate the system's effectiveness and potential to revolutionise special-needs healthcare. This study reveals that the proposed approach is durable and has the potential to significantly alter healthcare recommendations. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Blockchain Deep learning Healthcare Integration Patients Recommendations Security Special needs Technology Treatment

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：