检索结果-内蒙古大学图书馆

IEEE International Symposium on Electromagnetic Compatibility (EMC)

作者： P.G. Bremner J.C. West R. Afra C.F. Bunting G. Vazquez D.H. Trout RobustPhysics Del Mar CA School of Electrical & Computer Engineering Oklahoma State University Stillwater Ok USA Launch Services Program NASA Kennedy Space Center Cape Canaveral FL

In the space community, there is increasing interest in augmentation of launch fairing thermal-acoustic blankets, to also control electromagnetic environment threats. This second paper by the authors reports on the development of simulation methods to predict the maximum expected electric field levels when RF absorbing materials are used to reduce the mean field. To account for frequency variance in a “frequency-stirred” ensemble, the paper reports the experimental validation of a new unconditional probability density function model – an enhancement to Rayleigh statistics – for the reverberant electric field magnitude at any location and any frequency.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Sensing Beyond the Exceptional Point for High Detectivity

Sensing Beyond the Exceptional Point for High Detectivity

引用

Conference on Lasers and Electro-Optics (CLEO)

作者： Zhichao Li Ciril S. Prasad Xielin Wang Ding Zhang Gururaj V. Naik Applied Physics Graduate Program Smalley-Curl Institute Rice University Houston TX USA Department of Electrical and Computer Engineering Rice University Houston TX USA Department of Physics and Astronomy Rice University Houston TX USA

Exceptional point (EP)-based optical sensors exhibit exceptional sensitivity but poor detectivity. Slightly off EP operation boosts detectivity without much loss in sensitivity. We experimentally demonstrate a high-de... 详细信息

ISBN: (纸本)9798350369311

关键词： Proteins Optical losses Sensitivity Optical device fabrication Measurement by laser beam Lasers and electrooptics Optical sensors Laser applications Electro-optical waveguides

来源：评论

学校读者我要写书评

暂无评论

Systolic Array for Parallel Solution of the Robust Kalman Filter Used for Attitude and Position Estimations in UAVs

Systolic Array for Parallel Solution of the Robust Kalman Fi...

引用

International Conference on Unmanned Aircraft Systems (ICUAS)

作者： Leandro José Evilásio Campos Marco Henrique Terra Ricardo Menotti Roberto Santos Inoue Graduate Program in Electrical Engineering Federal University of São Carlos São Carlos SP Brazil Department of Electrical and Computer Engineering São Carlos School of Engineering University of São Paulo São Carlos SP Brazil Department of Computing Federal University of São Carlos

The efficient Kalman filter has been widely used in recent decades to obtain air navigation information in UAVs. However, for a good performance of the Kalman filter, the model that describes the system dynamics must not contain uncertainties. This paper presents the implementation of a robust Kalman filter to estimate the attitude, velocity, and position of UAVs. The robust filter considers uncertainties in the sensor models. A mathematical structure based on the solution of linear systems synthesizes the predictor-corrector robust estimation algorithm. The main contribution of this study is the proposed QR decomposition based on Givens rotation to solve the linear system. The simulated experiments used sensory data collected in Zürich-Switzerland and ground truth referencing attitude, velocity, and position. The offline simulation results express the effectiveness of the robust Kalman filter for this application, with a reduction of up to 18.9% in the estimation error, in relation to the standard Kalman filter. The proposal to use systolic arrays for numerical solutions has shown promise for implementation in parallel processing platforms, such as FPGAs.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Superpixel Tokenization for Vision Transformers: Preserving Semantic Integrity in Visual Tokens

arXiv

引用

arXiv 2024年

作者： Lew, Jaihyun Jang, Soohyuk Lee, Jaehoon Yoo, Seungryong Kim, Eunji Lee, Saehyung Mok, Jisoo Kim, Siwon Yoon, Sungroh Interdisciplinary Program in AI Seoul National University Korea Republic of Department of Electrical and Computer Engineering Seoul National University Korea Republic of Amazon United States AIIS ASRI INMC ISRC Seoul National University Korea Republic of

Transformers, a groundbreaking architecture proposed for natural language processing (NLP), have also achieved remarkable success in computer vision. A cornerstone of their success lies in the attention mechanism, which models relationships among tokens. While the tokenization process in NLP inherently ensures that each token maintains semantic integrity without containing multiple meanings, the grid-based tokenization of Vision Transformer (ViT) relies on uniformly partitioned square image patches, which may result in an arbitrary mixing of visual concepts within a token. In this work, we propose a novel tokenization pipeline that replaces the grid-based tokenization with superpixels, encouraging each token to capture a distinct visual concept. Unlike square image patches, superpixels are formed in varying shapes, sizes, and locations, making direct substitution challenging. To address this, our pipeline first generates pixel-level embeddings and efficiently aggregates them within superpixel clusters, producing superpixel tokens that seamlessly replace patch tokens in ViT. Extensive experiments demonstrate that our approach enhances the performance of ViT on various downstream tasks and introduces intriguing properties such as adaptive inference and semantic integrity in tokens. © 2024, CC BY.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Exploring Shifted Domain Problem within MNIST Dataset

Exploring Shifted Domain Problem within MNIST Dataset

引用

computer Science, Information Technology and engineering (ICCoSITE), International Conference on

作者： Hidayaturrahman Yaya Heryadi Lukas Wayan Suparta Yulyani Arifin Computer Science Department BINUS Graduate Program – Computer Science Bina Nusatara University Jakarta Indonesia Cognitive Engineering Research Group (CERG) Faculty of Engineering Universitas Katolik Indonesia Atma Jaya Jakarta Indonesia Department of Electrical Engineering Faculty of Industrial Technology Institut Teknologi Nasional Yogyakarta Yogyakarta Indonesia

Domain data can be shifted in any direction so it will be shared in different distributions to its original domain. This could be a problem since the model was trained with different distributions. It is found that adversarial domain adaptation using domain adversarial neural networks (DANN) can help to solve this problem on some scale. DANN can minimize the discrepancy between source and target data so the model can work well in both domains. The experiment is done by utilizing MNIST dataset that shifted into some conditions. In a condition when the shifting of distribution is too far, DANN is struggling to maintain the knowledge extracted from source data which leads to underperformance in the source and target domain. In contrast, when the shifting is closer, DANN can easily fit the model so it can perform well in both domains. It proves DANN is one of the good approaches to performing domain adaptation in small discrepancies.

关键词： computer science Adaptation models Neural networks Feature extraction Data models Data mining Information technology

来源：评论

学校读者我要写书评

暂无评论

On the Evaluation of Access-point Handovers for UAVs in Long-distance Missions

On the Evaluation of Access-point Handovers for UAVs in Long...

引用

2021 International Conference on Unmanned Aircraft Systems, ICUAS 2021

作者： Fagundes, Leonardo A. Souza, Vitor Barbosa Brandao, Alexandre S. Núcleo de Especialização em Robótica Federal University of Viçosa Department of Electrical Engineering and Graduate Program on Computer Science Viçosa Brazil Federal University of Viçosa Graduate Program on Computer Science Department of Informatics Viçosa Brazil

ISBN: (纸本)9780738131153

Unmanned Aerial Vehicle (UAV) systems are being increasingly used in a broad range of applications requiring extensive communications either to interconnect the UAVs with each other or to connect them with Ground Control Stations (GCS). Due to the complexity and difficulty of performing long distance experiments with these robots, simulation becomes a viable and necessary alternative. However, focusing either on the modeling of UAV operations or communication dynamics, available simulation tools are not able to capture the complex interdependencies between these two aspects, especially in applications where extensive communications are required. The main contribution of this work is the modeling and implementation of strategies for enhancing connectivity of a UAV performing long-distance navigation tasks, taking into account applications' requirements. The proposed strategies aim at assessing the impact of frequent handovers between wireless access points in terms of packet loss and available bandwidth. The first strategy prioritizes maintaining a connection until the signal-to-noise ratio (SNR) observed by the robot reaches a predefined threshold. For this, it is assumed that the data to be transmitted is small enough so that the robot may change its modulation to obtain smaller bandwidths while improving connection time. The second one prioritizes bandwidth availability, thus, it always searches for a network that is able to offer data rates higher than a minimum threshold. A hybrid model is validated by considering the characteristics of each individual strategy. The capabilities of the simulation tool designed for this work are illustrated through some case-study scenarios. © 2021 IEEE.

关键词： Unmanned aerial vehicles (UAV)

来源：评论

学校读者我要写书评

暂无评论

An Explainable Deep Learning Model for Prediction of Severity of Alzheimer’s Disease

An Explainable Deep Learning Model for Prediction of Severit...

引用

IEEE Symposium on Computational Intelligence and Bioinformatics and Computational Biology (CIBCB)

作者： Godwin Ekuma Daniel B. Hier Tayo Obafemi-Ajayi Computer Science Dept. Missouri State University Springfield MO USA Electrical & Computer Eng. Dept Missouri University of Science & Technology Rolla MO USA Engineering Program Missouri State University Springfield MO USA

Deep Convolutional Neural Networks (CNNs) have become the go-to method for medical imaging classification on various imaging modalities for binary and multiclass problems. Deep CNNs extract spatial features from image data hierarchically, with deeper layers learning more relevant features for the classification application. Despite the high predictive accuracy, usability lags in practical applications due to the black-box model perception. Model explainability and interpretability are essential for successfully integrating artificial intelligence into healthcare practice. This work addresses the challenge of an explainable deep learning model for the prediction of the severity of Alzheimer’s disease (AD). AD diagnosis and prognosis heavily rely on neuroimaging information, particularly magnetic resonance imaging (MRI). We present a deep learning model framework that integrates a local data-driven interpretation method that explains the relationship between the predicted AD severity from the CNN and the input MR brain image. The deep explainer uses SHapley Additive exPlanation values to quantity the contribution of different brain regions utilized by the CNN to predict outcomes. We conduct a comparative analysis of three high-performing CNN models: DenseNet121, DenseNet169, and Inception-ResNet-v2. The framework shows high sensitivity and specificity in the test sample of subjects with varying levels of AD severity. We also correlated five key AD neurocognitive assessment outcome measures and the APOE genotype biomarker with model misclassifications to facilitate a better understanding of model performance.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Impact of High-k Materials and Tube Numbers on CNFET Gate’s Performances

Impact of High-k Materials and Tube Numbers on CNFET Gate’s...

引用

International Conference on Smart Systems and Inventive Technology (ICSSIT)

作者： Christofer Yalung Doldet Tantraviwat Sermsak Uatrongjit Wittawat Yamwong Philosophy Program in Nanoscience and Nanotechnology (International Program/Interdisciplinary) Chiang Mai University Chiang Mai Thailand Department of Electrical Engineering Faculty of Engineering Chiang Mai University Chiang Mai Thailand Thai Microelectronics Center (TMEC) National Electronics and Computer Technology Center (NECTEC) Chachoengsao Thailand

Carbon nanotube field effect transistor (CNFET) performance is primarily influenced by a variety of characteristics, such as nanotube diameter, operating voltage, pitch, the number of tubes, dielectric constant, and contact materials. In this study, the dielectric constant and the number of carbon nanotubes (CNTs) were emphasized to analyze the propagation delay and average power dissipation of basic logic gates. Zirconium dioxide (ZrO 2 ), hafnium oxide (HfO 2 ), and silicon dioxide (SiO 2 ) were used as dielectric materials. In this experiment, the Stanford CNFET model was utilized, and variable factors such as the number of CNTs and the dielectric constant $(K_{ox})$ accumulated on the nanotube gate's surface were diversified. The correlation coefficients of the delay, the average power, and the CNT numbers were also presented. Furthermore, the multi-objective genetic algorithm and the Rmethod were utilized to find the optimal K ox and the number of CNTs for minimizing the average dissipation power and the propagation delay of some logic gates designed with CNFET. For all the gates considered, the optimal number of CNTs is 4 while the dielectric constant is within the ranges of 19 to 24.

关键词： Silicon compounds Dielectric constant Correlation Logic gates CNTFETs Power dissipation Electron tubes

来源：评论

学校读者我要写书评

暂无评论

Design and Optimization of the Porous Metamaterial EM Absorbers in X-band

Design and Optimization of the Porous Metamaterial EM Absorb...

引用

IEEE International Symposium on Electromagnetic Compatibility (EMC)

作者： Kanat Anurakparadorn Alan Taub Eric Michielssen Macromolecular Science and Engineering Program Department of Material Science and Engineering University of Michigan Ann Arbor Michigan USA Department of Mechanical Engineering University of Michigan Ann Arbor Michigan USA Department of Electrical Engineering and Computer Science University of Michigan Ann Arbor Michigan USA

A strategy that combines experiment and simulation to design and optimize electromagnetic (EM) metamaterial absorbers containing a periodic porous structure is described. The approach provides the ability to produce absorbers that meet multiple user-specified objectives. Using the measured intrinsic properties of the baseline materials as an input to EM-field based computational modelling and optimization, absorption by the studied metamaterials measured by their reflection loss (RL) increases significantly. The resulting metamaterials have the potential for lower cost and lighter weight while providing greater protection than traditional metal gaskets and foams.

关键词：

来源：评论

学校读者我要写书评

暂无评论

GPU Performance Tuning and Power Efficiency on the DGX A100 Cluster

GPU Performance Tuning and Power Efficiency on the DGX A100 ...

引用

IEEE International Conference on Cloud Computing Technology and Science (CloudCom)

作者： Khanin Udomchoksakul Orathai Sangpetch Akkarit Sangpetch Electrical and Computer Engineering Carnegie Mellon University Pittsburgh USA Carnegie Mellon-KMITL Thailand Program CMKL University Bangkok Thailand

ISBN: (纸本)9781665463683

The complexity of current Deep learning has been growing rapidly nowadays. Such advancement allows various organizations such as private sectors and government to leverage intelligent systems on their use cases. High Performance Computing (HPC) infrastructure nowadays has pivoted to GPU-oriented systems, enabling developers and researchers to train complex models with large datasets unlike conventional clusters equipped only with CPU cores. However, focus on power efficiency on the HPC system has not been prevalent especially on the new system such as DGX A100 that does not have datapoints on how GPUs consumed power. Even though such HPC cluster can be powerful, always allowing it to run at the maximum capacity results to financial cost to the HPC provider at the end. Therefore, for any organization providing the system, it is crucial for them to balance the cluster capabilities while maintaining overall power consumption which can potentially be costly in the long term. This paper reveals A100 GPU metrics that are relevant to Power usage and explains GPU profiling applied to Deep learning workload on the cluster, saving up to 32% of the power usage while compromising only 11.5% of training time compared to a default profile. Then, the paper investigates literature review that could be learned further adopted to the current system at CMKL university as the next milestone.

关键词： Deep learning Training Measurement Cloud computing Power demand High performance computing Government

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：