检索结果-内蒙古大学图书馆

2024 IEEE Global Communications Conference, GLOBECOM 2024

作者： Wang, Wei Li, Fan Yin, Mingzhu Ni, Weihao Sun Yat-Sen University School of Electronics and Information Technology Guangdong Provincial Key Laboratory of Optoelectronic Information Processing Chips and Systems Guangzhou510275 China

ISBN: (纸本)9798350351255

Digital-to-analog converter (DAC) plays an important role in a communication system, which determines the quality of the generated signal. A high-resolution DAC is necessary to guarantee the system performance. However, the application of a high-resolution DAC results in high system cost and power consumption, which is incompatible with the future low-cost data centers. To be cost-effective and energy-efficient, a system with low-resolution DAC and negligible quantization noise induced performance degradation is highly desirable. Noise shaping (NS) technique is proposed to effectively suppress the quantization noise induced by low-resolution DAC. In this paper, we review the principle of the NS technique, and its effectiveness is investigated experimentally in 100 Gb/s intensity modulation and direct detection (IM/DD) PAM-4 system with low-resolution DAC. The experimental results show that the system with 4-bit DAC coupled with NS technique shows almost the same receiver sensitivity as the system with 8-bit DAC without NS technique over 2-km and 10-km single mode fiber (SMF) transmission in O-band. It indicates that significant performance improvement can be achieved by using NS technique in low-resolution DAC systems, and it is a promising solution for future low-cost data centers. © 2024 IEEE.

关键词： Single mode fibers

来源：评论

学校读者我要写书评

暂无评论

Specific Emitter Identification based on CNN and Transformer 5

Specific Emitter Identification based on CNN and Transformer

引用

5th International Academic Exchange Conference on Science and technology Innovation, IAECST 2023

作者： Wang, Bin Gao, Ning Wang, Feng The 38th Research Institute of China Electronics Technology Group Corporation Key Laboratory of Aperture Array and Space Application Key Laboratory of Intelligent Information Processing Anhui Hefei230088 China Hefei University of Technology School of Computer and Information Hefei230601 China

ISBN: (纸本)9798350357738

Specific emitter identification(SEI) plays an increasingly crucial and potential role in both military and civilian scenarios. It refers to a process to discriminate individual emitters from each other by analyzing extracted characteristics from given radio signals. In this paper, we present a novel radiation source identification method named CNNFormer based on a hybrid architecture of Convolutional Neural Network and Transformer, harnessing the strengths of both architectures to enhance feature extraction and sequence modeling. We demonstrate the effectiveness of our approach on simulation datasets, achieving state-of-the-art accuracy in emitter identification. We showcase outstanding performance of our method on simulation datasets, achieving 98.29% three-class identification accuracy the emitter dataset with three devices and 97.60% ten-class identification accuracy on an RF fingerprint dataset with ten XSPR transmitters. Furthermore, our model allows for insight into the key features contributing to identification. This work contributes to enhancing the security and reliability of wireless communication systems. © 2023 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

SERGCN: Graph Convolutional Neural Networks with Skip-Edge Residual Connection for Node Classification Tasks 6

SERGCN: Graph Convolutional Neural Networks with Skip-Edge R...

引用

6th International Conference on Electronic Engineering and Informatics, EEI 2024

作者： Qin, Aqian He, Guoqing Wang, Huiwei College of Electronic and Information Engineering Southwest University Chongqing400715 China Chongqing Three Gorges University Key Laboratory of Intelligent Information Processing Chongqing404100 China Beijing Institute of Technology Chongqing Innovation Center Chongqing401120 China

ISBN: (纸本)9798350353594

Graph Convolutional Networks (GCNs) have been widely used for node categorization, graph embedding, and link prediction, etc. Two main challenges regarding developing deep GCNs for node classifying are overfitting and oversmoothing. The main causes of oversmoothing are overfitting and training barriers, which is directly caused by gradient disappearance. To resolve this issue, we propose the Skip-Edge Residual Graph Con-volution Network (SERGCN), which extracts a certain proportion of edge information at each layer in the novel way and integrates the initial information and the information of the previous layer in the convolution process, so as to prevent and slow down the gradient disappearance and further oversmoothing. Based on extensive experiments and real-world data sets used in several benchmarks, our proposed solution incorporates the benefits of a deep architecture while equaling or even exceeding the accuracy of existing approaches. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Design of Welding Robot Control System based on Visual Recognition technology 23

Design of Welding Robot Control System based on Visual Recog...

引用

3rd International Conference on Robotics and Control Engineering, RobCE 2023

作者： Han, Yu Han, Xu Guo, Xiaoqian Yang, Li Guo, Tinghang Qin, Zhuanping Zhao, Zhuangzhuang Tianjin Key Laboratory of Information Sensing and Intelligent Control Tianjin University of Technology and Education China

ISBN: (纸本)9781450398107

The level of industrial automation has gradually improved in recent years. Various industries gradually realize automatic welding through the combination of robot system and field equipment. To overcome the defects of traditional welding process, such as high labor costs, low safety, low welding efficiency, and unstable welding quality, a welding robot control system based on visual recognition technology is designed. The system includes three modules, namely visual module, control module and execution module. By Combining visual recognition technology, through the weld image feature measurement and processing method, the system can achieve functions such as weld image acquisition and recognition, weld feature point extraction, welding robot motion trajectory planning, and welding motion control. The system can make the welding quality and efficiency higher and is an automatic innovation of industrial welding technology. © 2023 ACM.

关键词： Quality control

来源：评论

学校读者我要写书评

暂无评论

ToolEyes: Fine-Grained Evaluation for Tool Learning Capabilities of Large Language Models in Real-world Scenarios 31

ToolEyes: Fine-Grained Evaluation for Tool Learning Capabili...

引用

31st International Conference on Computational Linguistics, COLING 2025

作者： Ye, Junjie Li, Guanyu Gao, Songyang Huang, Caishuang Wu, Yilong Li, Sixian Fan, Xiaoran Dou, Shihan Ji, Tao Zhang, Qi Gui, Tao Huang, Xuanjing School of Computer Science Fudan University China Institute of Modern Languages and Linguistics Fudan University China Research Institute of Intelligent Complex Systems Fudan University China Shanghai Key Laboratory of Intelligent Information Processing China Pengcheng Laboratory China

ISBN: (纸本)9798891761964

Existing evaluations of tool learning primarily focus on validating the alignment of selected tools (e.g., various APIs) for large language models (LLMs) with expected outcomes. However, these approaches rely on a limited set of scenarios where answers can be pre-determined. Furthermore, a sole emphasis on outcomes disregards the complex capabilities required for LLMs to effectively use tools. To tackle this issue, we propose ToolEyes, a fine-grained system tailored for the evaluation of the LLMs' tool learning capabilities in authentic scenarios. The system meticulously examines seven real-world scenarios, analyzing five dimensions crucial to LLMs in tool learning: format alignment, intent comprehension, behavior planning, tool selection, and answer organization. Additionally, ToolEyes incorporates a tool library boasting approximately 600 tools, serving as an intermediary between LLMs and the physical world. Evaluations involving ten LLMs across three categories reveal a preference for specific scenarios and limited cognitive abilities in tool learning. Intriguingly, expanding the model size even exacerbates the hindrance to tool learning. The code and data are available at https://***/Junjie-Ye/ToolEyes. © 2025 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Privacy-Preserving Fingerprint Recognition via Federated Adaptive Domain Generalization

引用

Computers, Materials & Continua 2025年第3期82卷 5035-5055页

作者： Yonghang Yan Xin Xie Hengyi Ren Ying Cao Hongwei Chang Henan Key Laboratory of Big Data Analysis and Processing Computer and Information EngineeringHenan UniversityKaifeng475004China College of Information Science and Technology Nanjing Forestry UniversityNanjing210037China Henan Branch China Life Insurance Co.Ltd.Zhengzhou450000China

Fingerprint features,as unique and stable biometric identifiers,are crucial for identity ***,traditional centralized methods of processing these sensitive data linked to personal identity pose significant privacy risks,potentially leading to user data *** Learning allows multiple clients to collaboratively train and optimize models without sharing raw data,effectively addressing privacy and security ***,variations in fingerprint data due to factors such as region,ethnicity,sensor quality,and environmental conditions result in significant heterogeneity across *** heterogeneity adversely impacts the generalization ability of the global model,limiting its performance across diverse *** address these challenges,we propose an Adaptive Federated Fingerprint Recognition algorithm(AFFR)based on Federated *** algorithm incorporates a generalization adjustment mechanism that evaluates the generalization gap between the local models and the global model,adaptively adjusting aggregation weights to mitigate the impact of heterogeneity caused by differences in data quality and feature ***,a noise mechanism is embedded in client-side training to reduce the risk of fingerprint data leakage arising from weight disclosures during model *** conducted on three public datasets demonstrate that AFFR significantly enhances model accuracy while ensuring robust privacy protection,showcasing its strong application potential and competitiveness in heterogeneous data environments.

关键词： Fingerprint recognition privacy protection federated learning adaptive weight adjustment

来源：评论

学校读者我要写书评

暂无评论

Improved R3D for Human Action Recognition Based on Transfer Learning 6

Improved R3D for Human Action Recognition Based on Transfer ...

引用

6th International Conference on Robotics and Computer Vision, ICRCV 2024

作者： Han, Xiaowei Guo, Qiuyang Lu, Yunjing Liu, Jiwu Fei, Chaolong Harbin University of Commerce Postdoctoral research workstation of Northeast Asia Service Outsourcing Research Center Station of Applied Economics Heilongjiang Provincial Key Laboratory of Electronic Commerce and Information Processing Harbi China Harbin University of Commerce Heilongjiang Provincial Key Laboratory of Electronic Commerce and Information Processing Harbin China Guangdong Technology College Heilongjiang Provincial Key Laboratory of Electronic Commerce and Information Processing Zhaoqing China

ISBN: (纸本)9798331527426

Human action recognition is an important research direction in the field of computer vision and machine learning, which is widely used in intelligent surveillance, human-computer interaction, sports analysis and other fields. Traditional action recognition methods usually rely on large-scale labeled datasets and complex feature extraction processes, which have problems such as high data acquisition cost and long model training time, as an effective solution, transfer learning can significantly improve the performance of the model and reduce the dependence on labeled data by transferring the features learned by the pre-trained model on large-scale datasets to the target task. In this paper, the R3D model is used as the backbone network, and the model is improved by strengthening the channel and temporal feature extraction capabilities. The model achieves good performance on Shopping Action_ v l by transfer learning based on HMDB51 and UCF101 datasets, and obtains 89.53% and 94.14% of the optimal Top-1 and Top-5 indicators after pre-training on UCF101 dataset. In summary, the C-TSAB ResNets model is proposed in this paper performs well in accuracy and robustness on human action recognition tasks through transfer learning, which verifies the actual effect of the innovation in the research work of this topic. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Multimodal Sentiment Analysis Based on Contrastive Learning and Cross-Modal Guided Fusion

引用

The Journal of China Universities of Posts and Telecommunications 2025年

作者： Liu Huanqi Sun Jingtao Zhang Fengling Hou Wenyan School of Computer Science and Technology， Xi'an University of Posts and Telecommunications Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing， Xi'an University of Posts and Telecommunications

Multimodal sentiment analysis， which integrates text， speech， and image modalities， has emerged as a prominent research direction in artificial intelligence for precise emotion assessment. However， current techniques experience difficulties in efficiently managing redundancy and inconsistency across features from different modalities， compromising sentiment analysis accuracy. Additionally， while the analysis of intraclass emotional features has garnered substantial attention， studies of interclass relationships have been neglected. To address these challenges， a multimodal sentiment analysis method based on contrastive learning and cross-modal guided fusion (CLCGF) is proposed. This method encodes text and images to derive latent representations and employs a cross-modal guided module with sparse attention mechanisms to effectively integrate textual and visual features， thereby mitigating redundancy issues within each modality's features. In addition to the sentiment classification task， a supervised contrastive learning task is incorporated to aid the model in learning effective features from multimodal data related to emotions. To assess the efficacy of the CLCGF method， experiments were conducted on three public datasets: MVSA-Single， MVSA-Multiple and HFM. The experimental results indicate that CLCGF significantly improves sentiment analysis accuracy compared with traditional methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Joint Estimation of SOH and RUL for Lithium-Ion Batteries Based on Improved Twin Support Vector Machineh

引用

Energy Engineering 2025年第1期122卷 243-264页

作者： Liyao Yang Hongyan Ma Yingda Zhang Wei He School of Electrical and Information Engineering Beijing University of Civil Engineering and ArchitectureBeijing100044China Institute of Distributed Energy Storage Safety Big Data Beijing100044China Beijing Key Laboratory of Intelligent Processing for Building Big Data Beijing100044China

Accurately estimating the State of Health(SOH)and Remaining Useful Life(RUL)of lithium-ion batteries(LIBs)is crucial for the continuous and stable operation of battery management ***,due to the complex internal chemical systems of LIBs and the nonlinear degradation of their performance,direct measurement of SOH and RUL is *** address these issues,the Twin Support Vector Machine(TWSVM)method is proposed to predict SOH and ***,the constant current charging time of the lithium battery is extracted as a health indicator(HI),decomposed using Variational Modal Decomposition(VMD),and feature correlations are computed using Importance of Random Forest Features(RF)to maximize the extraction of critical factors influencing battery performance ***,to enhance the global search capability of the Convolution Optimization Algorithm(COA),improvements are made using Good Point Set theory and the Differential Evolution *** Improved Convolution Optimization Algorithm(ICOA)is employed to optimize TWSVM parameters for constructing SOH and RUL prediction ***,the proposed models are validated using NASA and CALCE lithium-ion battery *** results demonstrate that the proposed models achieve an RMSE not exceeding 0.007 and an MAPE not exceeding 0.0082 for SOH and RUL prediction,with a relative error in RUL prediction within the range of[-1.8%,2%].Compared to other models,the proposed model not only exhibits superior fitting capability but also demonstrates robust performance.

关键词： State of health remaining useful life variational modal decomposition random forest twin support vector machine convolutional optimization algorithm

来源：评论

学校读者我要写书评

暂无评论

A Novel Two-Stage Method for Texture Defect Detection 4

A Novel Two-Stage Method for Texture Defect Detection

引用

4th International Conference on Consumer Electronics and Computer Engineering, ICCECE 2024

作者： Liu, Lu Jiang, Min Xu, Lifa School of Computer Science and Technology Wuhan University of Science and Technology Hubei WuHan430065 China Wuhan University of Science and Technology Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Hubei Wuhan430065 China

ISBN: (纸本)9798350384062

Texture defect detection has wide applications in both textiles and industry, where traditional defect detection methods focus on costly manual labeling and defect samples are complex to collect in real-world industrial scenarios. Furthermore, labeling errors are common in datasets from industrial scenarios, and the problem still needs to be explored. Therefore, we propose a novel texture defect detection method. It has two stages;the defective samples are synthesized in the first stage. In the second stage, we represent normal samples using a self-supervised learning method. Then, we construct a single classifier based on this learned representation. Subsequently, the model is trained to distinguish between normal and defective samples within the embedding space. To increase the robustness of our method, we adopt a novel strategy to construct defective samples and compute the feature similarity through a similarity-weighted aggregation method to achieve defect detection, which is more robust to noise. Experimental results on the DAGM2007 dataset show that our method performs comparably to the leading texture defect detection methods without label noise. In contrast, in the presence of noise, our method can effectively minimize the impact of noise on model performance. © 2024 IEEE.

关键词： Defects

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：