检索结果-内蒙古大学图书馆

Solving Lur'e equations through zeroing neural networks

Information Sciences 2025年 718卷

作者： Yu He Xiaojuan Wang Yafei Tie Huanqi Yang Theodore E. Simos Spyridon D. Mourtas Vasilios N. Katsikis School of Computer Science and Artificial Intelligence Huanghuai University Zhumadian 463000 China Henan International Joint Laboratory of Behavior Optimization Control for Smart Robots Henan 463000 China Henan Key Laboratory of Smart Lighting Zhumadian 46300 China Hangzhou Dianzi University School of Mechanical Engineering Er Hao Da Jie 1158 Xiasha Hangzhou 310018 China Center for Applied Mathematics and Bioinformatics Gulf University for Science and Technology West Mishref 32093 Kuwait Section of Mathematics Dept. of Civil Engineering Democritus Univ. of Thrace Xanthi 67100 Greece Department of Economics Division of Mathematics-Informatics and Statistics-Econometrics National and Kapodistrian University of Athens Sofokleous 1 Street 10559 Athens Greece Laboratory “Hybrid Methods of Modelling and Optimization in Complex Systems” Siberian Federal University Prosp. Svobodny 79 660041 Krasnoyarsk Russia

Solving Lur'e equations plays a critical role in addressing linear-quadratic optimal control (LQOC) problems, especially in cases where the control cost matrices are singular. This paper introduces, for the first time, two novel zeroing neural network (ZNN) models—ZNNLE and ZNNLE-LQOC—specifically designed to solve the Lur'e equation system and the LQOC problem, respectively. The proposed models extend the applicability of the ZNN methodology to these challenging scenarios by offering robust and efficient solutions to time-varying matrix equations. Theoretical analyses confirm the validity of both models, while numerical simulations and practical applications demonstrate their effectiveness. Moreover, a comparative study with an enhanced alternating-direction implicit (ADI) method highlights the superior performance of the ZNNLE-LQOC model in solving LQOC problems.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Integrated sidelobe cancellation and linear prediction kalman filter for joint multi-microphone speech dereverberation, interfering speech cancellation, and noise reduction

arXiv

引用

arXiv 2019年

作者： Dietzen, Thomas Doclo, Simon Moonen, Marc van Waterschoot, Toon STADIUS Center for Dynamical Systems Signal Processing and Data Analytics and ETC Technology Cluster Electrical Engineering Leuven Belgium University of Oldenburg Dept. of Medical Physics and Acoustics Cluster of Excellence Hearing4all Oldenburg Germany KU Leuven ESAT STADIUS

In multi-microphone speech enhancement, reverberation as well as additive noise and/or interfering speech are commonly suppressed by deconvolution and spatial filtering, e.g., using multi-channel linear prediction (MCLP) on the one hand and beamforming, e.g., a generalized sidelobe canceler (GSC), on the other hand. In this paper, we consider several reverberant speech components, whereof some are to be dereverberated and others to be canceled, as well as a diffuse (e.g., babble) noise component to be suppressed. In order to perform both deconvolution and spatial filtering, we integrate MCLP and the GSC into a novel architecture referred to as integrated sidelobe cancellation and linear prediction (ISCLP), where the sidelobe-cancellation (SC) filter and the linear prediction (LP) filter operate in parallel, but on different microphone signal frames. Within ISCLP, we estimate both filters jointly by means of a single Kalman filter. We further propose a spectral Wiener gain post-processor, which is shown to relate to the Kalman filter's posterior state estimate. The presented ISCLP Kalman filter is benchmarked against two state-of-the-art approaches, namely first a pair of alternating Kalman filters respectively performing dereverberation and noise reduction, and second an MCLP+GSC Kalman filter cascade. While the ISCLP Kalman filter is roughly M2 times less expensive than both reference algorithms, where M denotes the number of microphones, it is shown to perform similarly as compared to the former, and to outperform the latter. Copyright © 2019, The Authors. All rights reserved.

关键词： Beamforming

来源：评论

学校读者我要写书评

暂无评论

Square root-based multi-source early PSD estimation and recursive RETF update in reverberant environments by means of the orthogonal procrustes problem

arXiv

引用

arXiv 2019年

Multi-channel short-time Fourier transform (STFT) domain-based processing of reverberant microphone signals commonly relies on power-spectral-density (PSD) estimates of early source images, where early refers to reflections contained within the same STFT frame. State-of-the-art approaches to multi-source early PSD estimation, given an estimate of the associated relative early transfer functions (RETFs), conventionally minimize the approximation error defined with respect to the early correlation matrix, requiring non-negative inequality constraints on the PSDs. Instead, we here propose to factorize the early correlation matrix and minimize the approximation error defined with respect to the early-correlation-matrix square root. The proposed minimization problem - constituting a generalization of the so-called orthogonal Procrustes problem - seeks a unitary matrix and the square roots of the early PSDs up to an arbitrary complex argument, making non-negative inequality constraints redundant. A solution is obtained iteratively, requiring one singular value decomposition (SVD) per iteration. The estimated unitary matrix and early PSD square roots further allow to recursively update the RETF estimate, which is not inherently possible in the conventional approach. An estimate of the said early-correlation-matrix square root itself is obtained by means of the generalized eigenvalue decomposition (GEVD), where we further propose to restore non-stationarities by desmoothing the generalized eigenvalues in order to compensate for inevitable recursive averaging. Simulation results indicate fast convergence of the proposed multi-source early PSD estimation approach in only one iteration if initialized appropriately, and better performance as compared to the conventional approach. Copyright © 2019, The Authors. All rights reserved.

关键词： Singular value decomposition

来源：评论

学校读者我要写书评

暂无评论

Domain-Adversarial Graph Neural Networks for Text Classification

Domain-Adversarial Graph Neural Networks for Text Classifica...

引用

IEEE International Conference on Data Mining (ICDM)

作者： Man Wu Shirui Pan Xingquan Zhu Chuan Zhou Lei Pan Dept. of Computer & Electrical Engineering and Computer Science Florida Atlantic University Boca Raton USA Faculty of Information Technology Monash University Melbourne Australia Academy of Mathematics and Systems Science Chinese Academy of Sciences Beijing China School of Cyber Security University of Chinese Academy of Sciences Beijing China School of Information Technology Deakin University Geelong VIC Australia

Text classification, in cross-domain setting, is a challenging task. On the one hand, data from other domains are often useful to improve the learning on the target domain; on the other hand, domain variance and hierarchical structure of documents from words, key phrases, sentences, paragraphs, etc. make it difficult to align domains for effective learning. To date, existing cross-domain text classification methods mainly strive to minimize feature distribution differences between domains, and they typically suffer from three major limitations - (1) difficult to capture semantics in non-consecutive phrases and long-distance word dependency because of treating texts as word sequences, (2) neglect of hierarchical coarse-grained structures of document for feature learning, and (3) narrow focus of the domains at instance levels, without using domains as supervisions to improve text classification. This paper proposes an end-to-end, domain-adversarial graph neural networks (DAGNN), for cross-domain text classification. Our motivation is to model documents as graphs and use a domain-adversarial training principle to lean features from each graph (as well as learning the separation of domains) for effective text classification. At the instance level, DAGNN uses a graph to model each document, so that it can capture non-consecutive and long-distance semantics. At the feature level, DAGNN uses graphs from different domains to jointly train hierarchical graph neural networks in order to learn good features. At the learning level, DAGNN proposes a domain-adversarial principle such that the learned features not only optimally classify documents but also separates domains. Experiments on benchmark datasets demonstrate the effectiveness of our method in cross-domain classification tasks.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Broadband Elliptical Slotted Patch Antenna for 5G Communications

Broadband Elliptical Slotted Patch Antenna for 5G Communicat...

引用

International Conference on computer, Communication, Chemical, Material and Electronic engineering (IC4ME2)

作者： S. Das M. Samsuzzaman M. N. Rahman M. T. Islam M. Tarikul Islam Dept. of Computer and Communication Engineering Patuakhali Science and Technology University Patuakhali Bangladesh Dept. of Electrical Electronic and Systems Engineering Universiti Kebangsaan Malaysia Malaysia

ISBN: (纸本)9781538647769

In this paper, a broadband elliptical slotted patch antenna has been proposed for 5G wireless communications. The proposed antenna apparatus has a minimal size of 0.308λ 0 × 0.308λ 0 × 0.009λ 0 at 22 GHz. The antenna bandwidth is enhanced for an elliptical ground plane. Microstrip feed line has been connected by a stub, to achieve better impedance. Array design has been designed for better gain, therefore array design has been made. The array has a high gain of about 18 dBi at 28 GHz and 21dBi at 38 GHz. The proposed antenna covers a wide band area from 22 GHz to 48 GHz. The proposed antenna covers 28/38 GHz. The antenna additionally gives omnidirectional to required 5G bands.

关键词： 5G mobile communication Antenna radiation patterns Broadband antennas Broadband communication Patch antennas Wireless communication

来源：评论

学校读者我要写书评

暂无评论

A Periodic Cusped Magnetic - Quad Magnetic Focusing System for Low Voltage Ultra-Compact W-Band Klystron

A Periodic Cusped Magnetic - Quad Magnetic Focusing System f...

引用

International Conference on Vacuum Electronics

作者： Yuan Zheng Neville C. Luhmann Diana Gamzina Joe Olszewski Ann Sy Brandon R. Weatherford Dept. of Electrical and Computer Engineering University of California Davis Davis CA U.S.A. Stanford Linear Accelerator Center Menlo Park CA US Technology Innovation Directorate SLAC National Acceleration Laboratory Menlo Park CA U.S.A.

A Periodic Cusped Magnet (PCM) - Tunable Quadrupolar Magnet (TQM) is proposed to focus the sheet electron beam of a low voltage W-Band, ultra-compact continuous wave sheet beam klystron (LUWK). The design utilizes a PCM to prevent the vertical beam spreading, and a TQM to control transverse beam spreading the force. Employing the PCM-TQM not only benefits the compact size of the LUWK, but it also mitigates the Diocotron instability of the sheet beam. Using the PCM-QM focusing system, the MAGIC-3D simulations show that an elliptical sheet beam with transverse size 6.0 mm x 0.5 mm can achieve a 99% cold beam transmission, and a 97% hot beam transmission through a 140 mm beam tunnel.

关键词：

来源：评论

学校读者我要写书评

暂无评论

An information theoretic interpretation to deep neural networks

arXiv

引用

arXiv 2019年

作者： Huang, Shao-Lun Xu, Xiangxiang Zheng, Lizhong Wornell, Gregory W. DSIT Research Center Tsinghua-Berkeley Shenzhen Institute Shenzhen518055 China Dept. of Electronic Engineering Tsinghua University Beijing100084 China Dep. of Electrical & Computer Eng. Massachusetts Institute of Technology CambridgeMA02139-4307 United States

It is commonly believed that the hidden layers of deep neural networks (DNNs) attempt to extract informative features for learning tasks. In this paper, we formalize this intuition by showing that the features extracted by DNN coincide with the result of an optimization problem, which we call the "universal feature selection" problem, in a local analysis regime. We interpret the weights training in DNN as the projection of feature functions between feature spaces, specified by the network structure. Our formulation has direct operational meaning in terms of the performance for inference tasks, and gives interpretations to the internal computation results of DNNs. Results of numerical experiments are provided to support the analysis. Copyright © 2019, The Authors. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

AutoRARE: An Automated Tool For Generating FPGA-Based Multi-Memory Hardware Accelerators For Compute-Intensive Applications

AutoRARE: An Automated Tool For Generating FPGA-Based Multi-...

引用

IEEE International Performance Computing and Communications Conference

作者： Shrikant S. Jadhav Clay Gloster Christopher Doss Youngsoo Kim Jannatun Naher Dept. of Electrical and Computer Engineering North Carolina A&T State University Greensboro NC Dept. of Computer Systems Technology North Carolina A&T State University Greensboro NC Dept. of Electrical and Computer Engineering Bradley University Peoria IL

ISBN: (纸本)9781538668092;9781538668085

In this paper, we present AutoRARE, a Java-based automated design tool for generating Field Programmable Gate Array (FPGA)-based hardware accelerators. AutoRARE automatically generates all VHDL models needed to build/synthesize a processor specifically tailored for each application. The user needs only provide the VHDL description of a special-purpose floating point Arithmetic Logic Unit (ALU) or function core. The tool generates the VHDL description for the memory interface, memory controller, host processor interface, and the application specific processor. We also present details of the FPGA-based multi-memory hardware accelerator for accelerating computationally intensive applications, generated using AutoRARE. The multi-memory hardware accelerator is highly pipelined and able to simultaneously read and write multiple floating point values from multiple memories. The multi-memory architecture is the key to providing hardware accelerators that execute 10X-100X faster than typical multi-core processors. The Taylor Series expansion of the sine/cosine function is used as an application to demonstrate the merits of the multi-memory hardware accelerator. In our experiments, we executed the Taylor Series in software and compared execution times with an FPGA-based hardware implementation. Our experiments show that the FPGA-based multi-memory Taylor Series hardware accelerator is 481X faster than software executing the Taylor Series on a typical server.

关键词： Reconfigurable computing Field-Programmable gate arrays Field-Programmable custom computing machines High performance computing Multi-Memory architecture Multiple memory banks

来源：评论

学校读者我要写书评

暂无评论

Characterization of Sub-Bandgap Energy States in CuInxGa(i-x)Se2 and Transparent Conducting Oxides with Electron Energy-Loss Spectroscopy

引用

Microscopy and Microanalysis 2018年第S1期24卷 456-457页

作者： Julia I. Deitz Pran K. Paul Shankar Karki Sylvain Marsillac Aaron R. Arehart John M. Walls Jake W. Bowers Tyler J. Grassman David W. McComb Dept. of Materials Science & Engineering The Ohio State University Columbus OH USA. Dept. of Electrical & Computer Engineering The Ohio State University Columbus OH USA. Dept. of Electrical & Computer Engineering Old Dominion University Norfolk VA USA. Centre for Renewable Energy Systems Technology Dept. of Mechanical Electrical and Manufacturing Engineering Loughborough University Loughborough Leicestershire UK.

来源：评论

学校读者我要写书评

暂无评论

FCLAB: An EEGLAB module for performing functional connectivity analysis on single-subject EEG data

FCLAB: An EEGLAB module for performing functional connectivi...

引用

2018 IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2018

作者： Pezoulas, Vasileios C. Athanasiou, Alkinoos Nolte, Guido Zervakis, Michalis Fratini, Antonio Fotiadis, Dimitrios I. Klados, Manousos A. Unit of Medical Technology and Intelligent Information Systems Dept. of Materials Science and Engineering University of Ioannina Ioannina45112 Greece Lab of Medical Physics Faculty of Medicine School of Health Sciences Aristotle University of Thessaloniki Thessaloniki54124 Greece Dept. of Neurophysiology and Pathophysiology University Medical Center Hamburg-Eppendorf Hamburg Germany School of Electrical and Computer Engineering Technical University of Crete Chania73100 Greece Dept. of Biomedical Engineering School of Life and Health Sciences Aston University Aston United Kingdom

ISBN: (纸本)9781538624050

Functional connectivity (FC) analysis constitutes a fundamental neuroscientific approach that has been extensively used for the investigation of brain's connectivity and activation patterns. To that end, several software tools have been developed. This paper presents FCLAB, the only EEGLAB-based plugin, which is able to work with EEG signals in order to estimate and visualize brain functional connectivity networks based on a variety of similarity measures as well as run a complete graph analysis procedure followed by a detailed visualization of the ensuing local and global measures distribution. FCLAB entails optimization procedures for the implementation of the connectivity structures and is the result of long-term research in EEG functional connectivity. The computed functional connectivity measures have been carefully selected to reflect the state-of-art in the field. Future work focuses on extending the platform for multi-subject analysis in order to enable the implementation of statistical analysis tools. © 2018 IEEE.

关键词： Neurophysiology

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：