检索结果-内蒙古大学图书馆

12th IEEE International Conference on Information and Communication Networks, ICICN 2024

作者： Yang, Yuhang Sun, Hongrui Yang, Hongyan He, Quanlin Yu, Fangzhou Li, Qiming Li, Jianqing Zhi, Wenlong School of Optoelectronic Engineering Guilin University of Electronic Technology Guilin China School of Computer Science and Engineering Macau University of Science and Echnology China

ISBN: (纸本)9798350355802

Metasurfaces represent an innovative class of artificial composite materials, consisting of sub-wavelength structural units that enable the flexible manipulation of electromagnetic waves through structural design. This capability results in an exceptionally high information capacity and pixel resolution. To enhance the practicality and security of encryption and storage devices, we have designed and fabricated an all-dielectric metasurface based on the transmission phase theory. This metasurface facilitates dual-channel polarization multiplexing holographic display within the visible spectrum. The impact of metasurface parameters on phase is examined using the finite-difference-time-domain (FDTD) method, and the electromagnetic wave's polarization is controlled by adjusting the metasurface's parameters under 700nm illumination. The Gerchberg-Saxton (GS) algorithm was employed to encode the metasurface with 2-bit information. By controlling the polarization of the incident light, the holographic reconstruction of different images is achieved at the same position in the far field. The realization of polarization multiplexing dual-channel holography can provide a new idea for metasurface in the field of multifunctional multiplexing. © 2024 IEEE.

关键词： Holographic displays

来源：评论

学校读者我要写书评

暂无评论

Remember This Event That Year? Assessing Temporal Information and Understanding in Large Language Models

Remember This Event That Year? Assessing Temporal Informatio...

引用

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Beniwal, Himanshu Patel, Dishant Kowsik Nandagopan, D. Ladia, Hritik Yadav, Ankit Singh, Mayank Department of Computer Science and Engineering Indian Institute of Technology Gandhinagar India

ISBN: (纸本)9798891761681

Large Language Models (LLMs) are increasingly ubiquitous, yet their ability to retain and reason about temporal information remains limited, hindering their application in real-world scenarios where understanding the sequential nature of events is crucial. Our study experiments with 12 state-of-the-art models (ranging from 2B to 70B+ parameters) on a novel numerical-temporal dataset, TempUN, spanning from 10,000 BCE to 2100 CE, to uncover significant temporal retention and comprehension limitations. We propose six metrics to assess three learning paradigms to enhance temporal knowledge acquisition. Our findings reveal that open-source models exhibit knowledge gaps more frequently, suggesting a trade-off between limited knowledge and incorrect responses. Additionally, various fine-tuning approaches significantly improved performance, reducing incorrect outputs and impacting the identification of 'information not available' in the generations. The associated dataset and code are available at https://***/lingoiitgn/TempUN. © 2024 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

Design and Analysis of a 4-element Wideband MIMO Antenna for THz Applications 2

Design and Analysis of a 4-element Wideband MIMO Antenna for...

引用

2nd International Conference on Power, Electrical, Electronics and Industrial Applications, PEEIACON 2024

作者： Jafor, Md. Shakiul Paul, Liton Chandra Mimi, Afsana Hossain, Md. Sadik Al Mamun, Md. Shajol Rahman, Md. Hasibur Department of Electrical and Electronic Engineering Pabna University of Science & Technology Electronic and Communication Engineering Department of Electrical Pabna Bangladesh Pundra University of Science & Technology Department of Computer Science & Engineering Bogura Bangladesh

ISBN: (纸本)9798331517984

In this work, initially a rectangular microstrip patch antenna measuring (94.8 × 110 × 10) μm3 with a polyamide substrate has been analyzed and developed. The antenna has a bandwidth of 170 GHz (1.98 - 2.15 THz) and is tuned to resonate at a frequency of 2.08 THz. Its return loss is -26.733 dB, radiation efficiency is 92%, and maximum gain is 6.35 dB. In addition, the MIMO technology has been implemented to improve performance. Its bandwidth is 210 GHz (1.98-2.19 THz), and it offers a reflection coefficient of -37.75 dB, a radiation efficiency of 95%, and a peak gain of 6.01 dB. The substrate of the four-element MIMO antenna measures (154.8x170x10) μm3. The HFSS platform has been used to perform the modelling and simulation. © 2024 IEEE.

关键词： Slot antennas

来源：评论

学校读者我要写书评

暂无评论

Retinal Disease Detection for Low Resource Devices Through Pruned and Quantized CNNs

Retinal Disease Detection for Low Resource Devices Through P...

引用

2025 IEEE International Students' Conference on Electrical, Electronics and computer science, SCEECS 2025

作者： Gavel, Harish Azad, Chandrashekhar National Institute of Technology Jamshedpur Department of Computer Science and Engineering Jamshedpur India

ISBN: (纸本)9798331529833

Retinal disease diagnosis is one of the crucial step for preventing vision loss, especially in conditions such as Diabetic Retinopathy(DR), Glaucoma, and Age-related Macular Degeneration(AMD). In this paper, we propose an efficient approach for retinal disease classification using lightweight convolutional neural networks (CNNs) deployed on a low-resource device, the Raspberry Pi 4. To address the limitations of resource-constrained hardware, we employ model optimization techniques, including pruning and quantization, to reduce computational requirements while maintaining classification accuracy. The proposed system is evaluated on the Messidor datasets. Our optimized MobileNet and EfficientNetLite models decreased the accuracy scores by 1.5% to 2%, while reducing inference time by over 50% and memory usage by 45%. These results show the feasibility of deploying real-time retinal disease models on edge devices for telemedicine and mobile health in resource-limited settings. © 2025 IEEE.

关键词： Constrained optimization

来源：评论

学校读者我要写书评

暂无评论

Cyber intrusion detection using dual interactive Wasserstein generative adversarial network with war strategy optimization in wireless sensor networks

引用

Multimedia Tools and Applications 2024年第18期84卷 1-31页

作者： Anusha, N. Tapas Bapu, B.R. Selvakumaran, S. Vijayaraj, A. Ramesh Kumar, C. Raji, P. Department of IoT School of Computer science and Engineering VIT Tamil Nadu Vellore India Department of Electronics and Communication Engineering S.A.Engineering College Tamil Nadu Chennai India Department of Artificial Intelligence and Data Science Rajalakshmi Institute of Technology Tamil Nadu Chennai India Department of Information Technology R.M.K Engineering College R.S.M Nagar Tamil Nadu Kavaraipettai India Department of Electronics and Communication Engineering Panimalar Engineering College Tamil Nadu Chennai India Department of Electronics and Communication Engineering Saveetha Engineering College Tamil Nadu Chennai India

Wireless sensor network (WSN) is one of the essential components of a multi-hop cyber-physical system comprising many fixed or moving sensors. There are many common attacks in WSN, which can quickly harm a WSN system. In this manuscript, Dual Interactive Wasserstein Generative Adversarial Network (DIWGAN) with war strategy optimization (WSO) algorithm is introduced for detecting intrusion in WSN. Initially, the input data is attained from the NSL-KDD and CICIDS2017 datasets. The data is then passed onto Morphological filtering and Extended Empirical wavelet transformation (MFEEWT) based pre-processing. After that, the filtered data is sent to the DIWGAN, which detects the intrusion and classifies the fraud data in the network. Additionally, the WSO algorithm is introduced to enhance the classification parameters of DIWGAN. By then the performance of the proposed DIWGAN-WSO-IDS methodology is evaluated using Python Platform. Also, the performance of the proposed methodology is calculated using various metrics like accuracy, precision, recall, f-measure, mean square error, and attack detection rate. Thus, the proposed technique has attained 29.87%, 31.7%, and 26.8% higher accuracy, 23.54%, 20.09%, and 26.98% higher recall, 12.76%, 10.96%, and 6.7% higher precision. And 6.85%, 4.56%, and 8.56% lower MSE than the existing techniques like BFSGSRF, MHICA-SSA, and CNN-LSTM methods. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Wireless sensor networks

来源：评论

学校读者我要写书评

暂无评论

Battery pack capacity estimation for electric vehicles based on enhanced machine learning and field data

引用

Journal of Energy Chemistry 2024年第5期92卷 605-618页

作者： Qingguang Qi Wenxue Liu Zhongwei Deng Jinwen Li Ziyou Song Xiaosong Hu College of Mechanical and Vehicle Engineering Chongqing UniversityChongqing 400044China School of Mechanical and Electrical Engineering University of Electronic Science and Technology of ChinaChengdu 611731SichuanChina Department of Mechanical Engineering National University of SingaporeSingapore 117575Singapore

Accurate capacity estimation is of great importance for the reliable state monitoring,timely maintenance,and second-life utilization of lithium-ion *** numerous works on battery capacity estimation using laboratory datasets,most of them are applied to battery cells and lack satisfactory fidelity when extended to real-world electric vehicle(EV)battery *** challenges intensify for large-sized EV battery packs,where unpredictable operating profiles and low-quality data acquisition hinder precise capacity *** fill the gap,this study introduces a novel data-driven battery pack capacity estimation method grounded in field *** proposed approach begins by determining labeled capacity through an innovative combination of the inverse ampere-hour integral,open circuit voltage-based,and resistance-based correction ***,multiple health features are extracted from incremental capacity curves,voltage curves,equivalent circuit model parameters,and operating temperature to thoroughly characterize battery aging behavior.A feature selection procedure is performed to determine the optimal feature set based on the Pearson correlation ***,a convolutional neural network and bidirectional gated recurrent unit,enhanced by an attention mechanism,are employed to estimate the battery pack capacity in real-world EV ***,the proposed method is validated with a field dataset from two EVs,covering approximately 35,000 *** results demonstrate that the proposed method exhibits better estimation performance with an error of less than 1.1%compared to existing *** work shows great potential for accurate large-sized EV battery pack capacity estimation based on field data,which provides significant insights into reliable labeled capacity calculation,effective features extraction,and machine learning-enabled health diagnosis.

关键词： Electricvehicle Lithium-ion battery pack Capacity estimation Machine learning Field data

来源：评论

学校读者我要写书评

暂无评论

A Survey on Security and Network management of SDWSN with ML Technique

A Survey on Security and Network management of SDWSN with ML...

引用

2023 International Conference on Intelligent Systems for Communication, IoT and Security, ICISCoIS 2023

作者： Jaya Lakshmi, A. Joe Prathap, P.M. Gojan School of Business and Technology Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350335835

Wireless Sensor Networks (WSN), is the foundation of today's technology of the sensor world(IoT). A traditional WSN tends to gather datafrom dynamic environmental conditions to perform a specific task but it finds difficulty in responding always to the frequently changing environment apart from the assigned task. To address this problem, Software Defined Wireless Sensor Network (SDWSN) has been proposed which is the association of a Software Defined Network and WSN. This paper focuses on the present scenario and issues of SDWSN in terms of security and network management. From this survey, reader can witness the state of art of the evolving investigation that integratesMachine Learning (ML) with SDWSN together called, ML-SDWSN. Many implementation problems of SDWSN can be overridden with a centralized approach and intelligent algorithms of ML-SDWSN. This survey gives the perception of educational and industrial researchers on the present state of SDWSN with ML techniques to contribute to the flourishing technology. © 2023 IEEE.

关键词： Wireless sensor networks

来源：评论

学校读者我要写书评

暂无评论

Power Metric Based BBR Congestion Control Using XGBoost 2

Power Metric Based BBR Congestion Control Using XGBoost

引用

2nd International Conference on Machine Learning and Autonomous Systems, ICMLAS 2025

作者： Midhula, K.S. Arun Raj Kumar, P. National Institute of Technology Calicut Computer Science and Engineering Department Kerala India

ISBN: (数字)9798331505745

ISBN: (纸本)9798331505745

Congestion is a prevalent challenge to every networks today due to the higher network access and inefficient utilization of available bandwidth. Bottleneck Bandwidth and Round-trip propagation time (BBR) is a popular congestion control algorithm introduced by Google. BBR estimates maximum bottleneck bandwidth and minimum propagation delay at end devices periodically to regulate the sending rate. However, there are still limitations exist due to the control parameter (called pacing gain value) used in the sending rate calculation. Since the pacing gain values used in different phases of BBR are fixed, there is an overestimation of the sending rate as well as inflight data. Thus, there is a persistent bottleneck queue formation and experiences higher network delay. In order to resolve the problem, we proposed a pacing gain prediction algorithm using XGBoost regression algorithm controlled by the power metric. The proposed algorithm adaptively predicts the pacing gain values with respect to the network parameters. The research focuses on multimedia applications, emphasizing the importance of reducing network delay to enhance Quality of Service (QoS). The proposed model achieves performance improvement than the existing regression models. © 2025 IEEE.

关键词： Quality of service

来源：评论

学校读者我要写书评

暂无评论

A CNN and HBA Based Approach for Grape Disease Identification 4

A CNN and HBA Based Approach for Grape Disease Identificatio...

引用

4th International Conference on Sentiment Analysis and Deep Learning, ICSADL 2025

作者： Vishnuvarthan, K. Renugadevi, R. Santhi, S. Kit - Kalaignarkarunanidhi Institute of Technology Department of Computer Science and Engineering Coimbatore India

ISBN: (纸本)9798331523923

Grapes are among the most extensively cultivated fruit crops globally, with over 10,000 varieties. They serve as a significant economic resource for many nations due to their diverse applications, including wine production, fresh juice consumption, raisin production, and juice manufacturing. However, grape diseases pose a severe threat to the viticulture industry, leading to substantial financial losses and a decline in wine quality. Manual evaluation is the foundation of traditional disease detection techniques, which are error-prone in addition to saving time. This highlights the need for an automated system capable of analyzing plant health, identifying diseases, and addressing undernourishment efficiently and reliably. Accurate disease diagnosis and timely control measures are crucial for mitigating these challenges. Many scholars have studied various algorithms such as CNN, SVM, Neural Networks, Random Forests, etc., yet each has disadvantages. So we have proposed an automated deep learning-based approach that integrates the Honey Badger Algorithm (HBA) with Convolutional Neural Networks (CNN) for the detection of grape diseases. A phenomenal 95% accuracy rate was attained by the prototype after it was trained on a dataset of 10,000 photos. Its performance was evaluated across various grape varieties and environmental conditions, demonstrating its robustness and adaptability. © 2025 IEEE.

关键词： Wine

来源：评论

学校读者我要写书评

暂无评论

Outlier Selection with Local Intrinsic Dimensionality and Visualization Using t-SNE 17th

Outlier Selection with Local Intrinsic Dimensionality and Vi...

引用

17th International Conference on Multi-disciplinary Trends in Artificial Intelligence, MIWAI 2024

作者： Bheekya, Dharamsotu Moiz, Salman Abdul Rao, C. Raghavendra Department of Computer Science and Engineering Geethanjali College of Engineering and Technology Telangana Hyderabad501301 India School of Computer and Information Sciences University of Hyderabad Telangana Hyderabad500046 India

ISBN: (纸本)9789819606917

The k-Nearest Neighbor (k-NN) graph is an essential technique in data mining, machine learning, and computer vision for identifying local data patterns;however, its efficacy is significantly hindered in high-dimensional datasets due to the challenges of the curse of dimensionality, which includes issues like distance concentration and heightened computational requirements. To mitigate these challenges, various strategies, including dimensionality reduction and sophisticated graph construction methods, have been proposed. Outlier detection, vital in domains such as fraud detection and healthcare, similarly faces obstacles in high-dimensional contexts, leading to the development of diverse statistical and learning-based methodologies. This research aims to enhance k-NN graph construction and outlier detection by utilizing sparsification and subspace selection techniques, which effectively capture both local and global outliers through precise estimation of intrinsic dimensionality in localized areas. Furthermore, the study underscores the significance of visualizing high-dimensional data in reduced dimensions to enhance human interpretability, proposing an algorithm that leverages sparsified data representations to improve the accuracy of outlier detection and provide a more efficient framework for high-dimensional data analysis. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Dimensionality reduction

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：