检索结果-内蒙古大学图书馆

SAR in Big Data Era: Models, Methods and Applications (BIGSARDATA)

作者： Fangzhou Han Lamei Zhang Hongwei Dong Department of Information Engineering Harbin Institute of Technology Harbin China Science & Technology on Integrated Information System Laboratory Institute of Software Chinese Academy of Sciences Beijing China

Benefiting from the rapid advancements in Deep Learning (DL), data-driven features exhibit remarkable separability in PolSAR image classification. However, data-driven features are challenging to interpret and susceptible to resolution and noise. On the contrary, polarimetric features based on the target decomposition not only have high robustness, but also has physical interpretability. Consequently, researchers have begun to explore ways to integrate both types of features to achieve enhanced performance, which typically concat the data-driven features and polarimetric features directly. However, polarimetric features are not stable in separability, which is because the inherent defects of the target decomposition algorithm. To mitigate the adverse effects of low-quality features, this paper proposes a trusted polarimetric feature fusion (TPFF). The new paradigm calculates the confidence of features and employs a novel feature fusion method. The proposed method has been tested on the EMISAR Foulum dataset, with experimental results demonstrating improved classification performance and robustness.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Evaluation and Quantitative Attribution Analysis of Water Yield Services in the Peak-cluster Depression Basins in Southwest of Guangxi,China

引用

Chinese Geographical science 2023年第1期33卷 116-130页

作者： WANG Donghua TIAN Yichao ZHANG Yali HUANG Liangliang TAO Jin YANG Yongwei LIN Junliang ZHANG Qiang School of Resources and Environment Beibu Gulf UniversityQinzhou 535011China College of Environmental Science and Engineering Guilin University of TechnologyGuilin 541004China Key Laboratory of Marine Geographic Information Resources Development and Utilization in the Beibu Gulf Beibu Gulf UniversityQinzhou 535011China

Karst environmental issues have become one of the hot spots in contemporary international geological research. The same problem of water shortage is one of the hot spots of global concern. The peak-cluster depression basins in southwest of Guangxi is an important water connotation and ecological barrier areas in the Pearl River Basin of China. Thus, studying the spatial and temporal variations and the influencing factors of its water yield services is critical to achieve the sustainable development of water resources and ecological environmental protection in this region. As such, this paper uses the Integrated Valuation of Ecosystem Services and Tradeoffs(InVEST) model to assess the spatial and temporal variabilities of water yield services and its trends in the peak-cluster depression basins in southwest of Guangxi from 2000 to 2020. This work also integrates precipitation(Pre), reference evapotranspiration(ET), temperature(Tem), digital elevation model(DEM), slope, normalized difference vegetation index(NDVI), land use/land cover(LULC) and soil type to reveal the main factors that influence water yield services with the help of Geodetector. Results show that: 1) in time scale,the total annual water yield in the study area show a fluctuating and increasing trend from 2000 to 2020, with a growth rate of 7.3753 × 10^(8)m^(3)/yr, and its multi-year average water yield was 538.07 mm;2) in spatial pattern, with high yield areas mainly distributed in the south of the study area(mainly including Shangsi County, Pingxiang City, Ningming County, Longzhou County and Jingxi County), and low yield areas mainly distributed in Baise City and Nanning City;3) the dominant factor of water yield within karst and non-karst landforms is not necessarily controlled by precipitation, and the explanation degree of DEM factors in karst areas is significantly higher than that in non-karst areas;4) amongst the climatic factors, Pre, ET and Tem are dominant in the spatial pattern of region w

关键词： water yield Integrated Valuation of Ecosystem Services and Trade-offs(InVEST) Geodetector peak-cluster depression basins in southwest of Guangxi China

来源：评论

学校读者我要写书评

暂无评论

A Digitally assisted nonlinearity suppression architecture using over-the-air beam cancelation

A Digitally assisted nonlinearity suppression architecture u...

引用

作者： Mu, Qin Li, Nanxi Chen, Cong Wei, Xiaozheng Wen, Rong Wu, Muqing Li, Xiang Liu, Ying School of Information and Communication Engineering Beijing University of Posts and Telecommunications Beijing China National Key Lab of Science and Technology on Communications University of Electronic Science and Technology of China Chengdu China School of Network and Communication Engineering Chengdu Technological University Chengdu China Laboratory of Electromagnetic Space Cognition and Intelligent Control Beijing China

Using a large number of power amplifiers brings serious nonlinear distortion to 5G communication systems, which becomes a challenging problem for conventional digital predistortion (DPD) to linearize signals. To address this challenge, we propose a digitally assisted nonlinearity suppression architecture for applications in multiple antenna arrays. In this architecture, we use an auxiliary array to generate a nonlinear cancelation signal to cancel with the main beam at the receiving end. We explore a two-step method to model the nonlinear signal and combine neural network to complete the modeling. In addition, we introduce a novel method for estimating and tracking the channel to update the model coefficients. To verify the proposed methods, we design a scene and compare the performance of our architecture with that of conventional DPD. Our experiments demonstrate that our proposed methods can significantly enhance the adjacent channel power leakage ratio performance by 12 dB while also optimizing in-band distortion. Moreover, our architecture outperforms conventional DPD approaches even with enhanced distortion. © 2023 John Wiley & Sons Ltd.

关键词： Power amplifiers

来源：评论

学校读者我要写书评

暂无评论

A New Predefined Time Stability Theorem and its Application in Multi-Modal Memristive Neural Network Synchronization

A New Predefined Time Stability Theorem and its Application ...

引用

2023 IEEE International Conference on Memristive Computing and Applications, ICMCA 2023

作者： Zhao, Hui Zhou, Lei Wang, Qingjie Liu, Aidi Peng, Jingliang School of Information Science and Engineering University of Jinan Shandong Provincial Key Laboratory of Network Based Intelligent Computing Jinan China College of Information Science and Engineering Northeastern University Shenyang China Collaborative Innovation Center of Memristive Computing Application Qilu Institute of Technology Jinan China

ISBN: (纸本)9798350316230

This article proposes a new predefined-time (PDT) stability theorem, which offers more evident advantages compared to previous predefined-time stability. Firstly, based on the limitations of existing PDT stability, this theorem relaxed the limitations of sufficient conditions, which fill the gap in the research on PDT stability. Secondly, the PDT synchronization of multi-modal memristive neural networks (MM-MNN) is analyzed by the usability of the new judgment theorem. A matching PDT controller is designed and sufficient synchronization control conditions is obtained by utilizing generalized Lyapunov functional method, set-valued mapping, inequality analysis technique and differential inclusion theory. Meanwhile, the topological characteristics of multi-modal neuron and its impact on network synchronization behavior are presented. Finally, we provide some confirmatory numerical simulations. © 2023 IEEE.

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

Parallelism or Fairness? How to Be Friendly for SSDs in Cloud Environments

Parallelism or Fairness? How to Be Friendly for SSDs in Clou...

引用

IEEE International Conference on Cluster Computing

作者： Yang Zhou Fang Wang Zhan Shi Dan Feng Wuhan National Laboratory for Optoelectronics Key Laboratory of Information Storage System Engineering Research Center of data storage systems and Technology Ministry of Education of China Wuhan China School of Computer Science and Technology Huazhong University of Science and Technology Wuhan China

ISBN: (数字)9798350358711

ISBN: (纸本)9798350358728

Modern SSDs achieve low latency and high throughput by utilizing multiple levels of SSD parallelism. Fairness is also a critical design consideration in cloud environments and has inspired great interest in recent years to study it. However, we find that excessive striving for SSD parallelism can inadvertently harm the SSD fairness due to the striped data layout. We observe from our experimental analysis that some common access patterns in cloud loads (e.g., high-intensity I/O) are not addressed by SSD parallelism, which can affect both SSD parallelism and fairness. In this paper, we analyze two address mapping policies with orthogonal characteristics from SSD parallelism and fairness perspectives. Based on extensive experiments and observations, we schedule a hybrid address mapping policy to achieve fairness of SSDs in cloud environments by distinguishing different I/O access modes and cold/hot data, while ensuring SSD performance. Experimental results on the latest block-level I/O traces show that our approach significantly improves performance and fairness by up to 77.3% and 62.8%, respectively, over the previous schemes with negligible cost. Meanwhile, it can also reduce the write amplification by 23.5% to 33.5%.

关键词： Schedules Layout Interference Parallel processing Predictive models Throughput Data models Resource management Decision trees Low latency communication

来源：评论

学校读者我要写书评

暂无评论

Vulnerability Detection Based on Adapter Tuning and Enhanced Feature Learning 36

Vulnerability Detection Based on Adapter Tuning and Enhanced...

引用

36th International Conference on Software engineering and Knowledge engineering, SEKE 2024

作者： Luo, Hui Lu, Lu Liang, Zhihong Suo, Siliang School of Computer Science and Engineering South China University of Technology Guangzhou China Pazhou Laboratory Guangzhou China PengCheng Laboratory Shenzhen China Electric Power Research Institute CSG Guangdong Guangzhou China Guangdong Provincial Key Laboratory of Power System Network Security Guangdong Guangzhou China

ISBN: (纸本)1891706594

Pre-trained code models have achieved promising results in the vulnerability detection field. The prevailing approach is to adapt these models with vulnerability datasets using inefficient full-model fine-tuning. These pre-trained models primarily capture the semantic features of code, while neglecting its structural characteristics. To address these issues, this paper proposes a vulnerability detection method with adapter tuning and enhanced feature learning. First, adapter modules are introduced to UniXcoder and tune only the parameters in the adapters to extract semantic features. This significantly reduces the number of training parameters and better adapts the model to downstream tasks. Then, structural features, including control and data flow information, are extracted from the Program Dependence Graph (PDG) to compensate for the limitations of pre-trained models that rely solely on semantic features. Finally, the semantic and structural features are integrated to train the detection model. The experimental results demonstrate that our proposed method outperforms state-of-the-art approaches and is highly efficient in terms of both training parameters and training data. © 2024 Knowledge systems Institute Graduate School. All rights reserved.

关键词： Data flow graphs

来源：评论

学校读者我要写书评

暂无评论

Dual-Frequency High-Power Rectifying Metasurface for Microwave Power Transfer 6

Dual-Frequency High-Power Rectifying Metasurface for Microwa...

引用

6th IEEE International Conference on Electronic information and Communication technology, ICEICT 2023

作者： Yang, Jinlong Deng, Li Feng, Botao Xiao, Long Chen, Lijie Yao, Gengjiang School of Information and Communication Engineering Beijing University of Posts and Telecommunications Beijing China College of Electronics and Information Engineering Shenzhen University Shenzhen China Science and Technology on Electromagnetic Compatibility Laboratory China Ship Development and Design Center Wuhan China

ISBN: (纸本)9798350399059

This paper presents a design for a dual-frequency high-power rectifying metasurface. It comprises a dual-frequency receiving metasurface and a rectifier. The metasurface utilizes an ELC structure and achieves nearly 90% power absorption at both 2.4GHz and 3.5GHz, while consuming minimal energy in the medium and on the surface. The rectifying circuit incorporates GaN HEMT transistors and is designed based on a power amplifier. At an input power of approximately 40dBm, the rectifier demonstrates a high rectification efficiency, reaching 70% at 2.4GHz and 60% at 3.5GHz. © 2023 IEEE.

关键词： Power amplifiers

来源：评论

学校读者我要写书评

暂无评论

Fed-AugMix: Balancing Privacy and Utility via Data Augmentation

arXiv

引用

arXiv 2024年

作者： Li, Haoyang Chen, Wei Zhang, Xiaojin National Engineering Research Center for Big Data Technology and System Services Computing Technology and System Laboratory Cluster and Grid Computing Laboratory School of Computer Science and Technology Huazhong University of Science and Technology Wuhan430074 China School of Software Engineering Huazhong University of Science and Technology Wuhan430074 China

Gradient leakage attacks pose a significant threat to the privacy guarantees of federated learning. While distortion-based protection mechanisms are commonly employed to mitigate this issue, they often lead to notable performance degradation. Existing methods struggle to preserve model performance while ensuring privacy. To address this challenge, we propose a novel data augmentation-based framework designed to achieve a favorable privacy-utility trade-off, with the potential to enhance model performance in certain cases. Our framework incorporates the AugMix algorithm at the client level, enabling data augmentation with controllable severity. By integrating the Jensen-Shannon divergence into the loss function, we embed the distortion introduced by AugMix into the model gradients, effectively safeguarding privacy against deep leakage attacks. Moreover, the JS divergence promotes model consistency across different augmentations of the same image, enhancing both robustness and performance. Extensive experiments on benchmark datasets demonstrate the effectiveness and stability of our method in protecting privacy. Furthermore, our approach maintains, and in some cases improves, model performance, showcasing its ability to achieve a robust privacy-utility trade-off. Copyright © 2024, The Authors. All rights reserved.

关键词： Differential privacy

来源：评论

学校读者我要写书评

暂无评论

Single-molecule MIR absorption detection and nanoscale imaging

Single-molecule MIR absorption detection and nanoscale imagi...

引用

CLEO: science and Innovations in CLEO 2024, CLEO: S and I 2024 - Part of Conference on Lasers and Electro-Optics

作者： Meng, Zhao-Dong You, En-Ming Yi, Jun Tian, Zhong-Qun State Key Laboratory of Physical Chemistry of Solid Surfaces College of Chemistry and Chemical Engineering School of Electronic Science and Engineering Fujian Key Laboratory of Ultrafast Laser Technology and Applications Xiamen University Xiamen361005 China Xiamen361005 China School of Ocean Information Engineering Jimei University Xiamen361021 China

来源：评论

学校读者我要写书评

暂无评论

Using Hash Learning For a Multi-Model Fashion Recommendation system 5

Using Hash Learning For a Multi-Model Fashion Recommendation...

引用

5th International Conference on Computer Vision and information technology, CVIT 2024

作者： Shaohua, Du Wei, Zhang Zeyu, He School of Computer Science Beijing Information Science and Technology University Beijing China Beijing Laboratory of National Economic Security Early-warning Engineering Beijing China Beijing Advanced Innovation Center for Future Blockchain and Privacy Computing Beijing China

ISBN: (纸本)9781510686755

In the fashion recommendation system, with the increase in the number of clothing, the combination of clothing matching is growing exponentially, which leads to the slowdown of training speed and excessive memory usage, and other problems are becoming more and more prominent. Hashing technology can effectively reduce memory usage and improve the recommendation speed. With more and more product images accompanied by text descriptions, multi-modal modeling has become a research hotspot in recommendation systems. Most of the past research adopts the method of processing different modal information separately and then weighting and averaging, but the correlation between the two modalities is not fully utilized. We believe that the visual and textual features of the same product are semantically consistent and have the same aesthetic characteristics. Therefore, we try new modeling approaches to mine the higher-order connections between the two modalities. Another issue is that visual information is much more important than textual information in fashion recommendation systems, so it is not reasonable to equate the two modalities. We model the hash-processed binary representation, optimize the network structure to include visual features in the ranking to compute the loss, and also use textual features to assist the image features for modeling. Tests on two larger datasets from Polyvore show that our model exhibits significant results compared to state-of-the-art models on key evaluation metrics such as user-personalized recommendation and compatibility modeling. © 2024 SPIE.

关键词： Recommender systems

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：