检索结果-内蒙古大学图书馆

Lightweight Task-Oriented Semantic Communication Empowered by Large-Scale AI Models

IEEE Transactions on Vehicular technology 2025年

作者： Liu, Chuanhong Guo, Caili Yang, Yang Chen, Mingzhe Quek, Tony Q. S. Beijing University of Posts and Telecommunications Beijing Key Laboratory of Network System Architecture and Convergence School of Information and Communication Engineering Beijing100876 China Beijing University of Posts and Telecommunications Beijing Laboratory of Advanced Information Networks School of Information and Communication Engineering Beijing100876 China University of Miami Department of Electrical and Computer Engineering Institute for Data Science and Computing Coral GablesFL United States Singapore University of Technology and Design Dept. of Information Systems Technology and Design 487372 Singapore

Recent studies have focused on leveraging large-scale artificial intelligence (LAI) models to improve semantic representation and compression capabilities. However, the substantial computational demands of LAI models pose significant challenges for real-time communication scenarios. To address this, this paper proposes utilizing knowledge distillation (KD) techniques to extract and condense knowledge from LAI models, effectively reducing model complexity and computation latency. Nevertheless, the inherent complexity of LAI models leads to prolonged inference times during distillation, while their lack of channel awareness compromises the distillation performance. These limitations make standard KD methods unsuitable for task-oriented semantic communication scenarios. To address these issues, we propose a fast distillation method featuring a pre-stored compression mechanism that eliminates the need for repetitive inference, significantly improving efficiency. Furthermore, a channel adaptive module is incorporated to dynamically adjust the transmitted semantic information based on varying channel conditions, enhancing communication reliability and adaptability. In addition, an information bottleneck-based loss function is derived to guide the fast distillation process. Simulation results verify that the proposed scheme outperform baselines in term of task accuracy, model size, computation latency, and training data requirements. © 1967-2012 IEEE.

关键词： NP-hard

来源：评论

学校读者我要写书评

暂无评论

Improved siamese network for video-based person re-identification

Improved siamese network for video-based person re-identific...

引用

2019 International Conference on networking and network Applications, NaNA 2019

作者： Wang, Zhongmin Cai, Lanlan Duan, Na Fan, Lin Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xi'An University of Posts and Telecommunications Xi'an shaanxi710121 China School of Computer Science and Technology Xi'An University of Posts and Telecommunications Xi'an shaanxi710121 China

ISBN: (纸本)9781728126296

In the field of human-computer interaction and visual surveillance, person re-identification is undoubtedly a key task. In this paper, we propose improved Siamese network for video-based person re-identification, which combines the verification model and the classification model, in a way that all the supervision information of the samples is fully utilized. The key frame extraction is able to select informative frames over the sequence, which removes occluded frames according to optical flow vector and measures the similarity of two adjacent frames' contents in terms of the block mutual information entropy. A convolutional network and long short-term memory joint training is used as a feature extractor to re-identify using the improved Siamese network framework. Experimental data on the iLIDS-VID and PRID-2011 datasets show that this approach is superior to existing video-based re-identification methods. © 2019 IEEE.

关键词： Optical flows

来源：评论

学校读者我要写书评

暂无评论

Emotional analysis based on dynamic functional brain network

Emotional analysis based on dynamic functional brain network

引用

2019 International Conference on networking and network Applications, NaNA 2019

作者： Wang, Zhongmin Tong, Yue Heng, Xia Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing Xi'An University of Posts and Telecommunications Xi'an Shaanxi710121 China School of Computer Science and Technology Xi'An University of Posts and Telecommunications Xi'an Shaanxi710121 China

ISBN: (纸本)9781728126296

The human brain dynamically changes in the second and millisecond levels, so constructing a functional brain network using static connections can result in the absence of some time-related effective features. In this paper, a dynamic network model construction method is proposed. The basic idea of the proposed dynamic emotion analysis method is to construct different network modes based on the channel correlation mode and band power distribution method, and use the sliding window method to estimate the dynamic correlation mode and dynamic power distribution changes, Then the multivariate features of network-level brain dynamics are extracted and classified. We conducted experiments on the SEED database to study network-level dynamic transformations in the process of positive, neutral, and negative emotions. The results verify the feasibility of the emotional assessment method based on dynamic functional connection, and provide a new way to establishing brain dynamic model under different emotional states. The results show that emotion analysis based on dynamic networks is more accurate than static networks. © 2019 IEEE.

关键词： Band power Brain function network Channel correlation mode Dynamic emotion analysis Sliding window

来源：评论

学校读者我要写书评

暂无评论

Image recognition and detection based on fast area convolutional neural network

引用

Journal of Physics: Conference Series 2021年第1期1976卷

作者： Hanwen Zhang Zhen Qin Hua Xie School of Automation Engineering UESTC Chengdu China Network and Data Security Key Laboratory of Sichuan Province University of Electronic Science and Technology of China

Nowadays, image recognition and detection technologies based on traditional artificial neural networks and convolutional neural networks are slightly inadequate in terms of training and recognition time and accuracy, and are difficult to deploy on devices with limited hardware resources. Therefore, this article proposes a recognition and detection technology based on fast regional convolutional neural networks. We use RPN (Region Proposal network) instead of Selective Search method to rebuild the network, and add a new ROI pooling layer before the fully connected layer of CNN. Determine the category. The average detection accuracy on our data set can reach 83.8%, and the training time is only 0.34 hours.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Authentication of metropolitan quantum key distribution network with post-quantum cryptography

arXiv

引用

arXiv 2021年

作者： Yang, Yong-Hua Li, Pei-Yuan Ma, Shi-Zhao Qian, Xiao-Cong Zhang, Kai-Yi Wang, Liu-Jun Zhang, Wan-Li Zhou, Fei Tang, Shi-Biao Wang, Jia-Yong Yu, Yu Zhang, Qiang Pan, Jian-Wei CAS Quantum Network Co. Ltd Shanghai201315 China QuantumCTek Co. Ltd Hefei230088 China Jinan Institute of Quantum Technology Jinan250101 China Department of Computer Science and Engineering Shanghai Jiao Tong University Shanghai200240 China Shanghai Qizhi Institute Shanghai200232 China School of Physics and Astronomy and Yunnan Key Laboratory for Quantum Information Yunnan University Kunming650500 China Hefei National Laboratory for Physical Sciences Microscale and Department of Modern Physics University of Science and Technology of China Hefei230026 China

Quantum key distribution (QKD) provides information theoretically secures key exchange requiring authentication of the classic data processing channel via pre-sharing of symmetric private keys. In previous studies, the lattice-based post-quantum digital signature algorithm Aigis-Sig, combined with public-key infrastructure (PKI) was used to achieve highefficiency quantum security authentication of QKD, and its advantages in simplifying the MAN network structure and new user entry were demonstrated. This experiment further integrates the PQC algorithm into the commercial QKD system, the Jinan field metropolitan QKD network comprised of 14 user nodes and 5 optical switching nodes. The feasibility, effectiveness and stability of the post-quantum cryptography (PQC) algorithm and advantages of replacing trusted relays with optical switching brought by PQC authentication large-scale metropolitan area QKD network were verified. QKD with PQC authentication has potential in quantum-secure communications, specifically in metropolitan QKD networks. Copyright © 2021, The Authors. All rights reserved.

关键词： Quantum cryptography

来源：评论

学校读者我要写书评

暂无评论

Querying a matrix through matrix-vector products 46

Querying a matrix through matrix-vector products

引用

46th International Colloquium on Automata, Languages, and Programming, ICALP 2019

作者： Sun, Xiaoming Woodruff, David P. Yang, Guang Zhang, Jialin CAS Key Lab of Network Data Science and Technology Institute of Computing Technology Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing China Carnegie Mellon University PittsburghPA United States Institute of Computing Technology Chinese Academy of Sciences Beijing China Conflux Beijing China

ISBN: (纸本)9783959771092

We consider algorithms with access to an unknown matrix M ∈ Fn×d via matrix-vector products, namely, the algorithm chooses vectors v1,..., vq, and observes Mv1,..., Mvq. Here the vi can be randomized as well as chosen adaptively as a function of Mv1,..., Mvi−1. Motivated by applications of sketching in distributed computation, linear algebra, and streaming models, as well as connections to areas such as communication complexity and property testing, we initiate the study of the number q of queries needed to solve various fundamental problems. We study problems in three broad categories, including linear algebra, statistics problems, and graph problems. For example, we consider the number of queries required to approximate the rank, trace, maximum eigenvalue, and norms of a matrix M;to compute the AND/OR/Parity of each column or row of M, to decide whether there are identical columns or rows in M or whether M is symmetric, diagonal, or unitary;or to compute whether a graph defined by M is connected or triangle-free. We also show separations for algorithms that are allowed to obtain matrix-vector products only by querying vectors on the right, versus algorithms that can query vectors on both the left and the right. We also show separations depending on the underlying field the matrix-vector product occurs in. For graph problems, we show separations depending on the form of the matrix (bipartite adjacency versus signed edge-vertex incidence matrix) to represent the graph. Surprisingly, this fundamental model does not appear to have been studied on its own, and we believe a thorough investigation of problems in this model would be beneficial to a number of different application areas. © Xiaoming Sun, David P. Woodruff, Guang Yang, and Jialin Zhang;licensed under Creative Commons License CC-BY

关键词： Matrix algebra

来源：评论

学校读者我要写书评

暂无评论

In situ monitoring of natural labile copper during the growth process of a marine diatom

引用

Environmental pollution (Barking, Essex : 1987) 2025年 382卷 126662页

作者： Haitao Han Ying Li Yinchu Wang Dawei Pan CAS Key Laboratory of Coastal Environmental Processes and Ecological Remediation Shandong Key Laboratory of Coastal Environmental Processes Research Center for Coastal Environment Engineering Technology of Shandong Province Yantai Institute of Coastal Zone Research Chinese Academy of Sciences Yantai 264003 People's Republic of China. School of Information and Electronic Engineering Shandong Technology and Business University Yantai 264003 People's Republic of China. Key Laboratory of Coastal Biology and Biological Resource Utilization Yantai Institute of Coastal Zone Research Chinese Academy of Sciences Yantai 264003 People's Republic of China National Basic Science Data Center Beijing 100190 People's Republic of China. CAS Key Laboratory of Coastal Environmental Processes and Ecological Remediation Shandong Key Laboratory of Coastal Environmental Processes Research Center for Coastal Environment Engineering Technology of Shandong Province Yantai Institute of Coastal Zone Research Chinese Academy of Sciences Yantai 264003 People's Republic of China University of Chinese Academy of Sciences Beijing 100049 People's Republic of China. Electronic address: dwpan@***.

A novel iridium-needle electrode (Ir-NE) functionalized with agarose gel (AG-gel) and gold nanoparticles (AuNPs) was fabricated and adopted for the real-time in situ monitoring of natural labile copper (Cu´) in the culture medium of marine diatom. AuNPs possessed excellent electro-catalytic capacity towards the voltammetric determination of Cu, and AG-gel acted as the protective layer which enhanced the stability and lifespan of the electrode. The separation-catalysis detection mechanism based on the AG-gel and AuNPs made the electrode both high sensitivity and excellent anti-biofouling capability. In addition, the AG-gel/AuNPs/Ir-NE successfully achieved the real-time in situ monitoring of Cu´ in culture medium with Phaeodactylum tricornutum (P. tricornutum) as the model organism for marine diatoms. The correlation analysis between the concentrations of different Cu species and P. tricornutum cell density indicated that Cu´ or the Cu´ to total dissolved Cu (TdCu) ratio (Cu´/TdCu) might be a more accurate indicator of the bioavailability of Cu to marine diatom than TdCu. Furthermore, this electrode might also be extended to the real-time in situ determination of other labile metals in the culture medium of different microalgae, which provides an effective technique for studying feedback interactions between metals and algae.

关键词： Phaeodactylum tricornutum Real-time in situ monitoring copper speciation iridium-needle electrode natural labile copper

来源：评论

学校读者我要写书评

暂无评论

Improving preparation and luminescence properties of Eu and dy doped Sr₂MgSi₂O₇ phosphors by adding excess silica

引用

Scientific Reports 2025年第1期15卷 1-14页

作者： Zhang, Chi Wei, Liuming Liu, Qin Cui, Ruirui Deng, Chaoyong Li, Jingyu College of Electrical Engineering Henan University of Technology Zhengzhou 450001 China Department of Network Security Henan Police College Zhengzhou 450046 China Spallation Neutron Source Science Center Dongguan 523803 China Key Laboratory of Functional Composite Materials of Guizhou Province College of Big Data and Information Engineering Guizhou University Guiyang 550025 China Anqiu Qingyun Bilingual School Weifang China

Solid state reaction method is most commonly used to synthesize Sr2MgSi2O7: Eu2+, Dy3+ phosphors. In order to avoid the energy waste caused by high sintering temperature, we systematically studied and optimized the pr... 详细信息

Solid state reaction method is most commonly used to synthesize Sr₂MgSi₂O₇: Eu²⁺, Dy³⁺ phosphors. In order to avoid the energy waste caused by high sintering temperature, we systematically studied and optimized the preparation process of Sr₂MgSi₂O₇: Eu²⁺, Dy³⁺ long afterglow phosphors by increasing the Si/Sr ratio in the raw materials. The appropriate addition of silica not only facilitates the preparation of single phase Sr₂MgSi₂O₇: Eu²⁺, Dy³⁺ phosphor but also enhances the rate of solid state reaction, so as to achieve the purpose of reducing the sintering temperature. The photoluminescence performance of the sample with a Si/Sr ratio of 1.1 sintered at 1350 °C is the highest, followed by that of the sample sintered at 1400 °C with a Si/Sr ratio of 1.0. Notably, the photoluminescence properties of the Si/Sr ratio of 1.3 sample sintered at 1300 °C are comparable to those of the Si/Sr ratio of 1.0 sample sintered at 1400 °C. However, in terms of afterglow properties, the former exhibits an enhancement by a factor of 3.5 compared to the latter. This finding suggests that strategically increasing the Si/Sr ratio can reduce the sintering temperature by 100 °C while maintaining or even enhancing the luminescence and afterglow properties of the samples. Additionally, the electronic structures of the Sr₂MgSi₂O₇ host, the doped luminescence center Eu, the oxygen vacancies and the co-doped activator Dy have been methodically investigated through first-principles calculations. By integrating the findings on the distribution of the dopants and defects levels, we predict and propose the photoluminescence and afterglow mechanisms of Sr₂MgSi₂O₇: Eu²⁺, Dy³⁺. These studies suggest an efficient, cheap and energy-saving strategy for t

关键词： Electron traps First-principles calculation Optimizing synthesis Persistent luminescence Sr2MgSi2O7

来源：评论

学校读者我要写书评

暂无评论

Next-generation sequencing revealed influenza and Chlamydia infection in recurrent pneumonia in a patient who had recovered from COVID-19

引用

Precision Clinical Medicine 2020年第4期3卷 294-296页

作者： Wen Wen Guorong Zhang Sihui Luo Zhaoqing Bai Wanyin Tao Meng Guo Kaixin He Shuafang Jia Wei Liu Kaiguang Zhang Jianping Weng Shu Zhu The First Affiliated Hospital of USTC Division of Life Sciences and MedicineUniversity of Science and Technology of ChinaHefei 230001China Hefei National Laboratory for Physical Sciences at Microscale the CAS Key Laboratory of Innate Immunity and Chronic DiseaseSchool of Basic Medical SciencesDivision of Life Sciences and MedicineUniversity of Science and Technology of ChinaHefei 230026China Anqing Hospital Affiliated to Anhui Medical University(Anqing Municipal Hospital) Anqing 246003China School of Data Science University of Science and Technology of ChinaHefei 230026China CAS Centre for Excellence in Cell and Molecular Biology University of Science and Technology of ChinaHefei 230026China

Dear Editor,The severe acute respiratory syndrome coronavirus 2(SARS-CoV-2)was identified as the cause of an outbreak of a global pandemic of coronavirus disease 2019(COVID-19)in December 2019.1 COVID-19 often starts with nonspecific upper respiratory tract symptoms,making it difficult to distinguish from other diseases that present similar symptoms,such as fever,lung infection,and asthenia.

关键词： COVID-19 diagnosis next-generation sequencing influenza Chlamydia

来源：评论

学校读者我要写书评

暂无评论

Prior-Informed AGN-Host Spectral Decomposition Using PyQSOFit

arXiv

引用

arXiv 2024年

作者： Ren, Wenke Guo, Hengxiao Shen, Yue Silverman, John D. Burke, Colin J. Wang, Shu Wang, Junxian CAS Key Laboratory for Research in Galaxies and Cosmology Department of Astronomy University of Science and Technology of China Anhui Hefei230026 China School of Astronomy and Space Science University of Science and Technology of China Hefei230026 China Kavli Institute for the Physics and Mathematics of the Universe The University of Tokyo WPI Kashiwa277-8583 Japan Shanghai Astronomical Observatory Chinese Academy of Sciences 80 Nandan Road Shanghai200030 China Department of Astronomy University of Illinois at Urbana-Champaign 1002 West Green Street UrbanaIL61801 United States National Center for Supercomputing Applications 1205 West Clark Street UrbanaIL61801 United States Department of Astronomy School of Science The University of Tokyo 7-3-1 Hongo Bunkyo Tokyo113-0033 Japan Center for Data-Driven Discovery Kavli IPMU WPI UTIAS The University of Tokyo Chiba Kashiwa277-8583 Japan Center for Astrophysical Sciences Department of Physics & Astronomy Johns Hopkins University BaltimoreMD21218 United States Department of Astronomy Yale University 266 Whitney Avenue New HavenCT06511 United States Department of Physics & Astronomy Seoul National University Seoul08826 Korea Republic of

We introduce an improved method for decomposing the emission of active galactic nuclei (AGN) and their host galaxies using templates from principal component analysis (PCA). This approach integrates prior information from PCA with a penalized pixel fitting mechanism which improves the precision and effectiveness of the decomposition process. Specifically, we have reduced the degeneracy and overfitting in AGN-host decomposition, particularly for those with low signal-to-noise ratios (SNR), where traditional methods tend to fail. By applying our method to 76,565 SDSS data Release 16 quasars with z BH). Furthermore, we obtained stellar velocity dispersion (σ★) measurements for 4,137 quasars. The slope of the MBH − σ★ relation in this subsample is generally consistent with previous quasar studies beyond the local universe. Our method provides a robust and efficient approach to disentangle the AGN and host galaxy components across a wide range of SNRs and redshifts. © 2024, CC BY.

关键词： Active Galactic Nuclei

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：