检索结果-内蒙古大学图书馆

Edge Learning for B5G Networks With Distributed signal processing: Semantic Communication, Edge Computing, and Wireless Sensing

引用

IEEE JOURNAL OF SELECTED TOPICS IN signal processing 2023年第1期17卷 9-39页

作者： Xu, Wei Yang, Zhaohui Ng, Derrick Wing Kwan Levorato, Marco Eldar, Yonina C. Debbah, Merouane Southeast Univ Natl Mobile Commun Res Lab Nanjing 210096 Peoples R China Purple Mt Labs Nanjing 211111 Peoples R China Zhejiang Lab Hangzhou 311121 Peoples R China Zhejiang Univ Coll Informat Sci & Elect Engn Hangzhou 310027 Zhejiang Peoples R China Univ New South Wales Sch Elect Engn & Telecommun Sydney NSW 2052 Australia Univ Calif Irvine Dept Comp Sci Irvine CA 92697 USA Weizmann Inst Sci Fac Math & CS IL-7610001 Rehovot Israel Technol Innovat Inst Masdar U Arab Emirates Mohamed Bin Zayed Univ Artificial Intelligence Masdar 9639 Abu Dhabi U Arab Emirates

To process and transfer large amounts of data in emerging wireless services, it has become increasingly appealing to exploit distributed data communication and learning. Specifically, edge learning (EL) enables local model training on geographically disperse edge nodes and minimizes the need for frequent data exchange. However, the current design of separating EL deployment and communication optimization does not yet reap the promised benefits of distributed signal processing, and sometimes suffers from excessive signalling overhead, long processing delay, and unstable learning convergence. In this paper, we provide an overview on practical distributed EL techniques and their interplay with advanced communication optimization designs. In particular, typical performance metrics for dual-functional learning and communication networks are discussed. Also, recent achievements of enabling techniques for the dual-functional design are surveyed with exemplifications from the mutual perspectives of "communications for learning " and "learning for communications. " The application of EL techniques within a variety of future communication systems are also envisioned for beyond 5G (B5G) wireless networks. For the application in goal-oriented semantic communication, we present a first mathematical model of the goal-oriented source entropy as an optimization problem. In addition, from the viewpoint of information theory, we identify fundamental open problems of characterizing rate regions for communication networks supporting distributed learning-and-computing tasks. We also present technical challenges as well as emerging application opportunities in this field, with the aim of inspiring future research and promoting widespread developments of EL in B5G.

关键词： Wireless communication Optimization Training Reinforcement learning Measurement Deep learning signal processing algorithms Artificial intelligence (AI) deep learning (DL) edge learning (EL) federated learning (FL) multi-agent reinforcement learning (MARL) communication optimization Internet-of-Everything (IoE) beyond 5G (B5G)

来源：评论

学校读者我要写书评

暂无评论

COMPLEX DIGITAL signal-processing USING QUADRATIC RESIDUE NUMBER-SYSTEMS

引用

IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND signal processing 1986年第1期34卷 166-177页

作者： KRISHNAN, R JULLIEN, GA MILLER, WC UNIV WINDSOR DEPT ELECT ENGNWINDSOR N9B 3P4ONTARIOCANADA

Recently, the quadratic residue number system (QRNS) has been introduced [6], [7] which allows the multiplication of complex integers with two real multiplications. The restriction is that the number system has either all prime moduli of the form 4K + 1, or composite numbers with prime factors of that form. If an increase in real multiplications from two to three can be tolerated, then the restriction can be lifted to allow moduli of any form: the resulting number system is termed the modified quadratic residue number system (MQRNS). In this paper, the MQRNS is defined, and residue-to-binary conversion techniques in both the QRNS and MQRNS are presented. Hard-ware implementations of multiplication intensive, complex nonrecursive, and recursive digital filters are also presented in this paper where the QRNS and MQRNS structures are realized using a bit-slice architecture.

关键词： Digital signal processing Dynamic range Hardware Digital filters Digital arithmetic Throughput signal processing Decoding Cathode ray tubes signal processing algorithms

来源：评论

学校读者我要写书评

暂无评论

Kernel methods and their potential use in signal processing -: An overview and guidelines for future development

引用

IEEE signal processing MAGAZINE 2004年第3期21卷 57-65页

作者： Pérez-Cruz, F Bousquet, O Univ Carlos III Madrid Dept Signal Theory & Commun Madrid Spain Max Planck Inst Biol Cybernet D-72076 Tubingen Germany

The notion of kernels, recently introduced, has drawn much interest as it allows one to obtain nonlinear algorithms from linear ones in a simple and elegant manner. This, in conjunction with the introduction of new linear classification methods such as the support vector machines (SVMs), has produced significant progress in machine learning and related research topics. The success of such algorithms is now spreading as they are applied to more and more domains. signal processing procedures can benefit from a kernel perspective, making them more powerful and applicable to nonlinear processing in a simpler and nicer way. We present an overview of kernel methods and provide some guidelines for future development in kernel methods, as well as, some perspectives to the actual signal processing problems in which kernel methods are being applied.

关键词： Kernel signal processing Vectors signal processing algorithms Data analysis Hilbert space Symmetric matrices Machine learning Principal component analysis Least squares methods

来源：评论

学校读者我要写书评

暂无评论

Forensic Applications of signal processing Bridging the gap between scientific and legal requirements

引用

IEEE signal processing MAGAZINE 2009年第2期26卷 104-111页

作者： Tibbitts, Jo Lu, YiBin University of Sydney Australia University of New South Wales Australia Shanǵhai Jiao Tonǵ University China

This article highlights some problems encountered by forensic signal processing experts in the area of speech and video processing. We have demonstrated that there is a need for speech, video, and other signal processing experts within the I II community to work together to provide standardized guidelines to court systems around the world to assist them in dealing with this complex form of evidence. We also need to develop and provide law enforcement agencies with supplemental tools to assist them to improve the quality of speech and video evidence gathered for the court. In particular, quantitative measures need to be developed to allow law enforcement agents to determine during the investigation whether speech and video material is above or below a standard expected of court evidence.

关键词： Forensics signal processing Law Legal factors signal processing algorithms Data engineering Appropriate technology Documentation Qualifications Ethics

来源：评论

学校读者我要写书评

暂无评论

Advanced signal processing Methods for Characterization of Schizophrenia

引用

IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING 2021年第4期68卷 1123-1130页

作者： Masychev, Kirill Ciprian, Claudio Ravan, Maryam Reilly, James P. MacCrimmon, Duncan New York Inst Technol Dept Comp Sci New York NY USA New York Inst Technol Dept Elect & Comp Engn New York NY 10023 USA McMaster Univ Dept Elect & Comp Engn Hamilton ON L8S 4L8 Canada McMaster Univ Dept Psychiat Hamilton ON Canada

Objective: Schizophrenia is a severe mental disorder associated with nerobiological deficits. Auditory oddball P300 have been found to be one of the most consistent markers of schizophrenia. The goal of this study is to find quantitative features that can objectively distinguish patients with schizophrenia (SCZs) from healthy controls (HCs) based on their recorded auditory odd-ball P300 electroencephalogram (EEG) data. Methods: Using EEG dataset, we develop a machine learning (ML) algorithm to distinguish 57 SCZs from 66 HCs. The proposed ML algorithm has three steps. In the first step, a brain source localization (BSL) procedure using the linearly constrained minimum variance (LCMV) beamforming approach is employed on EEG signals to extract source waveforms from 30 specified brain regions. In the second step, a method for estimating effective connectivity, referred to as symbolic transfer entropy (STE), is applied to the source waveforms. In the third step the ML algorithm is applied to the STE connectivity matrix to determine whether a set of features can be found that successfully discriminate SCZ from HC. Results: The findings revealed that the SCZs have significantly higher effective connectivity compared to HCs and the selected STE features could achieve an accuracy of 92.68%, with a sensitivity of 92.98% and specificity of 92.42%. Conclusion: The findings imply that the extracted features are from the regions that are mainly affected by SCZ and can be used to distinguish SCZs from HCs. Significance: The proposed ML algorithm may prove to be a promising tool for the clinical diagnosis of schizophrenia.

关键词： Entropy Electroencephalography Feature extraction Brain Standards signal processing algorithms Location awareness Effective connectivity electroencephalography (EEG) EEG beamforming machine learning odd-ball paradigm schizophrenia symbolic transfer entropy

来源：评论

学校读者我要写书评

暂无评论

THE INTEGRATED signal-processing SYSTEM ISP

引用

IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND signal processing 1984年第4期32卷 842-851页

作者： KOPEC, GE Fairchild Laboratory for Artificial Intelligence Research Palo Alto CA USA

The Integrated signal processing System (ISP) is a Lisp machine-based workstation which provides a unified environment for signal data processing and the development of signal processing algorithms. ISP is based on a model of signal processing computation in which the fundamental activities are creating and manipulating abstract signal objects. ISP consists of three main subsystems. The signal representation language (SRL) formalizes the semantic foundation of ISP and provides a set of facilities for defining signal classes and creating instances of signal objects. The ISP environment provides a signal stack, display windows, and signal pictures which are used to create and view selected signals from the universe defined by SRL. Finally, the user interface consists of a number of interactive mechanisms for manipulating components of the environment.

关键词： signal processing signal processing algorithms Data processing Programming profession Speech processing Speech analysis Libraries Command languages Workstations Displays

来源：评论

学校读者我要写书评

暂无评论

Sidelobe Mitigation in Noise Radar Using Sparse signal processing

引用

IEEE AEROSPACE AND ELECTRONIC SYSTEMS MAGAZINE 2020年第9期35卷 32-40页

作者： de Martin, Lilian van Rossum, Wim Ribeiro, Diogo Anitori, Laura TNO Radar Technol The Hague Netherlands

It is well known that noise-like waveforms have the inherent property of suppressing ambiguities due to their nonrepetitive nature, as opposed to the more classical linear frequency modulated (LFM) pulsed waveforms. This aperiodicity also improves the robustness against jammers, making these waveforms harder to detect and classify by an enemy. However, noise waveforms are Doppler sensitive, requiring the use of Doppler-dependent matched filters for proper range compression. Also, when processed with classical matched filters, noise waveforms have sidelobes that spread out across the whole range-Doppler plane, potentially masking weak targets in the presence of strong ones. A possible solution to reduce the sidelobe interference is to use sparse signal processing (SSP). In this article, we demonstrate with simulated and experimental data how noise waveforms can be used in combination with SSP for unambiguous detection of strong as well as weak targets, improving upon the performance of more conventional LFM pulsed waveforms.

关键词： Doppler effect Radar Matching pursuit algorithms Radar signal processing signal processing algorithms Computational modeling Matched filters Frequency modulation Jamming

来源：评论

学校读者我要写书评

暂无评论

Algorithm Unrolling: Interpretable, Efficient Deep Learning for signal and Image processing

引用

IEEE signal processing MAGAZINE 2021年第2期38卷 18-44页

作者： Monga, Vishal Li, Yuelong Eldar, Yonina C. Penn State Univ Sch Elect Engn University Pk PA 16802 USA Penn State Univ Elect Engn & Comp Sci Fac University Pk PA 16802 USA Amazon Lab 126 San Jose CA 94089 USA Weizmann Inst Sci Dept Math & Comp Sci IL-7610001 Rehovot Israel Weizmann Inst Sci Ctr Biomed Engn & Signal Proc IL-7610001 Rehovot Israel MIT Cambridge MA 02139 USA Broad Inst Cambridge MA USA Duke Univ Durham NC 27706 USA Israel Acad Sci & Humanities Jerusalem Israel

Deep neural networks provide unprecedented performance gains in many real-world problems in signal and image processing. Despite these gains, the future development and practical deployment of deep networks are hindered by their black-box nature, i.e., a lack of interpretability and the need for very large training sets. An emerging technique called algorithm unrolling, or unfolding, offers promise in eliminating these issues by providing a concrete and systematic connection between iterative algorithms that are widely used in signal processing and deep neural networks. Unrolling methods were first proposed to develop fast neural network approximations for sparse coding. More recently, this direction has attracted enormous attention, and it is rapidly growing in both theoretic investigations and practical applications. The increasing popularity of unrolled deep networks is due, in part, to their potential in developing efficient, high-performance (yet interpretable) network architectures from reasonably sized training sets.

关键词： Training data Systematics Neural networks signal processing algorithms Performance gain Network architecture Deep learning Machine learning

来源：评论

学校读者我要写书评

暂无评论

Convex Optimization in signal processing

引用

IEEE signal processing MAGAZINE 2010年第3期27卷 19-+页

作者： Eldar, Yonina C. Luo, Zhi-Quan Ma, Wing-Kin Palomar, Daniel P. Sidiropoulos, Nicholas D. Spanish University for Distance Education Ciudad Universitaria Madrid Spain Technical University of Valencia Alcoi Spain Burgos University Burgos Spain DMR Consulting Madrid Spain Spanish University for Distance Education Ciudad Úniversitaria Madrid Spain

In recent years, we have witnessed technical breakthroughs in a wide variety of topics where the key to success is the use of convex optimization. In fact, convex optimization has now emerged as a major signal processing tool that has made a significant impact on numerous problems previously considered intractable. Considering the foundational nature and potential impact of convex optimization in signal processing, we have put together this special issue that aims to provide tutorials of convex optimization techniques (including available software) and various successful signal processing applications. Our goal is not only to contribute to the diffusion of recent developments in this research area within the signal processing community, but also to spur further advances in and applications of convex optimization for signal processing.

关键词： signal processing Finite impulse response filter signal processing algorithms Array signal processing Optimization methods Game theory Application software Design optimization Iterative algorithms MIMO

来源：评论

学校读者我要写书评

暂无评论

Kernel-Based Learning for Statistical signal processing in Cognitive Radio Networks

引用

IEEE signal processing MAGAZINE 2013年第4期30卷 126-136页

作者： Ding, Guoru Wu, Qihui Yao, Yu-Dong Wang, Jinlong Chen, Yingying PLA Univ Sci & Technol Coll Commun Engn Nanjing Jiangsu Peoples R China Southeast Univ Nanjing Jiangsu Peoples R China PLA Univ Sci & Technol Nanjing Jiangsu Peoples R China Stevens Inst Technol Hoboken NJ 07030 USA Carleton Univ Ottawa ON K1S 5B6 Canada Spar Aerosp Ltd Montreal PQ Canada Qualcomm Inc San Diego CA USA IEEE Nanjing Sect Nanjing Jiangsu Peoples R China Stevens Inst Technol Dept Elect & Comp Engn Hoboken NJ USA Alcatel Lucent Holmdel NJ USA

Kernel-based learning (KBL) methods have recently become prevalent in many engineering applications, notably in signal processing and communications. The increased interest is mainly driven by the practical need of being able to develop efficient nonlinear algorithms, which can obtain significant performance improvements over their linear counterparts at the price of generally higher computational complexity. In this article, an overview of applying various KBL methods to statistical signal processing-related open issues in cognitive radio networks (CRNs) is presented. It is demonstrated that KBL methods provide a powerful set of tools for CRNs and enable rigorous formulation and effective solutions to both long-standing and emerging design problems.

关键词： Machine learning Learning systems Kernel Cognitive radio signal processing algorithms Nonlinear algorithms

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

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

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：