检索结果-内蒙古大学图书馆

Wide and Deep Graph Neural Network With Distributed Online Learning

IEEE TRANSACTIONS ON signal processing 2022年 70卷 3862-3877页

作者： Gao, Zhan Gama, Fernando Ribeiro, Alejandro Univ Cambridge Dept Comp Sci Technol Cambridge CB3 0FD England Univ Calif Berkeley Dept Elect Engn & Comp Sci Berkeley CA 94709 USA Univ Penn Dept Elect & Syst Engn Philadelphia PA 19104 USA

Graph neural networks (GNNs) are naturally distributed architectures for learning representations from network data. This renders them suitable candidates for decentralized tasks. In these scenarios, the underlying graph often changes with time due to link failures or topology variations, creating a mismatch between the graphs on which GNNs were trained and the ones on which they are tested. Online learning can be leveraged to retrain GNNs at testing time to overcome this issue. However, most online algorithms are centralized and usually offer guarantees only on convex problems, which GNNs rarely lead to. This paper develops the Wide and Deep GNN (WD-GNN), a novel architecture that can be updated with distributed online learning mechanisms. The WD-GNN consists of two components: the wide part is a linear graph filter and the deep part is a nonlinear GNN. At training time, the joint wide and deep architecture learns nonlinear representations from data. At testing time, the wide, linear part is retrained, while the deep, nonlinear one remains fixed. This often leads to a convex formulation. We further propose a distributed online learning algorithm that can be implemented in a decentralized setting. We also show the stability of the WD-GNN to changes of the underlying graph and analyze the convergence of the proposed online learning procedure. Experiments on movie recommendation, source localization and robot swarm control corroborate theoretical findings and show the potential of the WD-GNN for distributed online learning.

关键词： Robot kinematics signal processing algorithms Convergence Heuristic algorithms Graph neural networks Testing Training Graph neural networks distributed learning online learning stability analysis convergence analysis

来源：评论

学校读者我要写书评

暂无评论

Accelerating LMS-Based Equalization With Correlated Training Sequence in Bandlimited IM/DD Systems

引用

JOURNAL OF LIGHTWAVE TECHNOLOGY 2022年第13期40卷 4268-4275页

作者： Yao, Shuang Tang, Xizi Zhang, Rui Zhou, Qi Su, Shang-Jen Hsu, Chin-Wei Alfadhli, Yahya Ma, Xiaoli Chang, Gee-Kung Georgia Inst Technol Sch Elect & Comp Engn Atlanta GA 30308 USA

As higher symbol rates are utilized in the intensity modulation and direct detection (IM/DD) scheme to meet the unrelenting growth of data traffic, overcoming the inter-symbol interference (ISI) induced by the limited bandwidth has become increasingly crucial. Channel equalization based on digital signal processing (DSP) is an effective solution, where least-mean squares (LMS) algorithm is adopted to adjust tap coefficients. However, the LMS algorithm usually has slow rate of convergence and requires lots of training symbols. This work proposes a novel training sequence to accelerate the LMS-based equalization. A first-order Markov chain (MC) is employed for sequence generation, which introduces correlation between samples and shapes signal spectrum. Compared with the conventional training sequence that consists of independent and identically distributed (i.i.d.) samples and has a white spectrum, the MC sequence enables faster convergence of tap coefficients and mean-squared error (MSE). Moreover, an experimental demonstration of a 43 Gbaud PAM-4 signal shows that the proposed sequence can achieve a lower pre-forward-error-correction (pre-FEC) bit error rate (BER) than that of the i.i.d. sequence with the same length. When the PAM-4 signal is transmitted over a 5-km standard single mode fiber (SSMF) with 6-dB system bandwidth of 10 GHz, more than 70% training sequence length reduction can be attained. When the fiber length is increased to 10 km and the signal suffers from severe power fading, more than 48% reduction can be achieved.

关键词： Training Bandwidth Convergence Equalizers signal processing algorithms Eigenvalues and eigenfunctions Symmetric matrices Channel equalization digital signal processing intensity modulation and direct detection least-mean squares

来源：评论

学校读者我要写书评

暂无评论

Generic Reversible Visible Watermarking via Regularized Graph Fourier Transform Coding

引用

IEEE TRANSACTIONS ON IMAGE processing 2022年 31卷 691-705页

作者： Qi, Wenfa Guo, Sirui Hu, Wei Peking Univ Wangxuan Inst Comp Technol Beijing 100871 Peoples R China Xidian Univ State Key Lab Integrated Serv Networks Xian 710071 Peoples R China

Reversible visible watermarking (RVW) is an active copyright protection mechanism. It not only transparently superimposes copyright patterns on specific positions of digital images or video frames to declare the copyright ownership information, but also completely erases the visible watermark image and thus enables restoring the original host image without any distortion. However, existing RVW algorithms mostly construct the reversible mapping mechanism for a specific visible watermarking scheme, which is not versatile. Hence, we propose a generic RVW framework to accommodate various visible watermarking schemes. In particular, we obtain a reconstruction data packet-the compressed difference image between the watermarked image and the original host image, which is embedded into the watermarked image via any conventional reversible data hiding method to facilitate the blind recovery of the host image. The key is to achieve compact compression of the difference image for efficient embedding of the reconstruction data packet. To this end, we propose regularized Graph Fourier Transform (GFT) coding, where the difference image is smoothed via the graph Laplacian regularizer for more efficient compression and then encoded by multi-resolution GFTs in an approximately optimal manner. Experimental results show that the proposed framework has much better versatility than state-of-the-art methods. Due to the small amount of auxiliary information to be embedded, the visual quality of the watermarked image is also higher.

关键词： Watermarking Image coding Visualization Image restoration Image reconstruction signal processing algorithms Approximation algorithms Reversible visible watermarking graph Fourier transform graph Laplacian regularizer versatility

来源：评论

学校读者我要写书评

暂无评论

Lamb Wave Wireless Communication Through Healthy and Damaged Channels With Symmetrical and Asymmetrical Steps and Notches

引用

IEEE TRANSACTIONS ON ULTRASONICS FERROELECTRICS AND FREQUENCY CONTROL 2022年第7期69卷 2390-2399页

作者： Bahouth, Rudy Benmeddour, Farouk Moulin, Emmanuel Assaad, Jamal Univ Lille Univ Polytech Hauts France UMR IEMN 8520 Cent LilleCNRS F-59000 Lille France Inst Natl Sci Appl INSA F-59313 Le Mont Houy Valenciennes France

Data transmission through solid metallic channels is recommended in certain industries where no other options are proposed, such as nuclear, aerospace, and smart vehicles. In addition to the Faraday shielding effect of electromagnetic waves, another issue related to damage presence due to mechanical loads exists. Severe damage in the transmission channel leads to signal loss at the receiver. For this sake, ultrasonic guided waves, such as Lamb waves, maybe a good substitute since they can propagate through long distances in solid metallic structures. The scope of this work is to build a reliable, reproducible, and high data-rate wireless communication experimental platform, using ultrasonic guided waves, through healthy and damaged plates for industrial usage. The target is to compensate at first for the effect of dispersion, reverberation, scattering, and boundary reflections for the healthy plate. The novelty of this work falls within the performance analysis of the demodulation algorithm based on cross-correlation combined with binary phase-shift keying (BPSK), using a finite-element simulation through healthy and damaged plates with different depths of symmetrical and asymmetrical notches (SN and AN) and steps based on the bit error percentage (BEP). Furthermore, another contribution related to the impact of multiple reflections and mode conversions caused by symmetrical and asymmetrical steps and notches is taken into account. After this, numerical results are validated using an ultrasonic guided wave experimental platform. Results based on BEP analysis prove that the algorithm has successfully compensated for the effect of dispersion and boundary reflections for the healthy plate and multiple reflections and mode conversions for the damaged ones. A highly effective data rate of up to 350 kb/s can be reached even in the presence of severe damage in the transmission channel.

关键词： Dispersion Binary phase shift keying signal processing algorithms Demodulation Aluminum Finite element analysis Solids Asymmetrical notches binary phase-shift keying (BPSK) modulation bit error percentage (BEP) channel characterization cross-correlation symmetrical notches ultrasonic lamb waves

来源：评论

学校读者我要写书评

暂无评论

Tear the Image Into Strips for Style Transfer

引用

IEEE TRANSACTIONS ON MULTIMEDIA 2022年 24卷 3978-3988页

作者： Huang, Yujie Liu, Yuhao Jing, Minge Zeng, Xiaoyang Fan, Yibo Fudan Univ Coll Microelect State Key Lab ASIC & Syst Shanghai 200000 Peoples R China Dalian Univ Technol Sch Comp Sci & Technol Dalian 116024 Peoples R China ZTE Corp State Key Lab Mobile Network & Mobile Multimedia Shenzhen 518057 Peoples R China

Recently, Deep Convolutional Neural Networks (DCNNs) have achieved remarkable progress in computer vision community, including in style transfer tasks. Normally, most methods feed the full image to the DCNN. Although high-quality results can be achieved in this manner, several underlying problems arise. For one, with the increase in image resolution, the memory footprint will increase dramatically, leading to high latency and massive power consumption. Furthermore, these methods are usually unable to integrate with the commercial image signal processor (ISP), which processes the image in a line-sequential manner. To solve the above problems, we propose a novel ISP-friendly deep learning-based style transfer algorithm: SequentialStyle. A brand new line-sequential processing mode is proposed, where the image is torn into strips, and each strip is sequentially processed, contributing to less memory demand. We further propose a Spatial-Temporal Synergistic (STS) mechanism that decouples the previously simplex 2-D image style transfer into spatial feature processing (in-strip) and temporal correlation transmission (in-between strips). Compared with the SOTA style transfer algorithms, experimental results show that our SequentialStyle is competitive. Besides, SequentialStyle has less demand for memory consumption, even for the images whose resolutions are 4 k or higher.

关键词： Strips Correlation Pipelines signal processing algorithms Memory management Task analysis Image resolution Style Transfer Image signal Processor Deep Learning

来源：评论

学校读者我要写书评

暂无评论

signal Reconstruction From Quantized Noisy Samples of the Discrete Fourier Transform

引用

IEEE TRANSACTIONS ON signal processing 2022年 70卷 2093-2104页

作者： Goyal, Mohak Kumar, Animesh Indian Inst Technol Dept Elect Engn Mumbai 400076 Maharashtra India Stanford Univ Dept Management Sci & Engn Stanford CA 94305 USA Twitter Toronto ON Canada

In this paper, we present two variations of an algorithm for signal reconstruction from one-bit or two-bit noisy observations of the discrete Fourier transform (DFT). The one-bit observations of the DFT correspond to the sign of its real part, whereas, the two-bit observations of the DFT correspond to the signs of both its real and imaginary parts. We focus on images for analysis and simulations, thus using the sign of the 2D-DFT. This choice of the class of signals is inspired by previous works on this problem. For our algorithm, we show that the expected mean squared error (MSE) in signal reconstruction is asymptotically proportional to the inverse of the sampling rate. The samples are affected by additive zero-mean noise of a known distribution. We solve this signal estimation problem by designing an algorithm that uses contraction mapping, based on the Banach fixed point theorem. Numerical tests with four benchmark images are provided to show the effectiveness of our algorithm. Various metrics for image reconstruction quality assessment such as PSNR, SSIM, ESSIM, and MS-SSIM are employed. On all four benchmark images, our algorithm outperforms the state-of-the-art in all of these metrics by a significant margin.

关键词： signal processing algorithms Discrete Fourier transforms signal reconstruction Noise measurement Sparse matrices Image reconstruction Compressed sensing Denoising single-bit sampling contraction mapping companding quantization image processing

来源：评论

学校读者我要写书评

暂无评论

Graphical Inference in Linear-Gaussian State-Space Models

引用

IEEE TRANSACTIONS ON signal processing 2022年 70卷 4757-4771页

作者： Elvira, Victor Chouzenoux, Emilie Univ Edinburgh Sch Math Edinburgh EH8 9YL Midlothian Scotland Alan Turing Inst London NW1 2DB England Univ Paris Saclay INRIA Cent Supelec Ctr Vis Numer F-1190 Gif Sur Yvette France

State-space models (SSM) are central to describe time-varying complex systems in countless signal processing applications such as remote sensing, networks, biomedicine, and finance to name a few. Inference and prediction in SSMs are possible when the model parameters are known, which is rarely the case. The estimation of these parameters is crucial, not only for performing statistical analysis, but also for uncovering the underlying structure of complex phenomena. In this paper, we focus on the linear-Gaussian model, arguably the most celebrated SSM, and particularly in the challenging task of estimating the transition matrix that encodes the Markovian dependencies in the evolution of the multi-variate state. We introduce a novel perspective by relating this matrix to the adjacency matrix of a directed graph, also interpreted as the causal relationship among state dimensions in the Granger-causality sense. Under this perspective, we propose a new method called GraphEM based on the well sounded expectation-maximization (EM) methodology for inferring the transition matrix jointly with the smoothing/filtering of the observed data. We propose an advanced convex optimization solver relying on a consensus-based implementation of a proximal splitting strategy for solving the M-step. This approach enables an efficient and versatile processing of various sophisticated priors on the graph structure, such as parsimony constraints, while benefiting from convergence guarantees. We demonstrate the good performance and the interpretable results of GraphEM by means of two sets of numerical examples.

关键词： Time series analysis Mathematical models Sparse matrices Numerical models Kalman filters signal processing algorithms State-space methods State-space modeling graphical inference sparsity proximal methods primal-dual algorithms Kalman filtering EM algorithm

来源：评论

学校读者我要写书评

暂无评论

Escaping Saddle Points for Successive Convex Approximation

引用

IEEE TRANSACTIONS ON signal processing 2022年 70卷 307-321页

作者： Bedi, Amrit Singh Rajawat, Ketan Aggarwal, Vaneet Koppel, Alec US Army Res Lab Adelphi MD 20783 USA IIT Kanpur Dept Elect Engn Kanpur 208016 Uttar Pradesh India Purdue Univ Sch Ind Engn W Lafayette IN 47907 USA Purdue Univ Sch Elect & Comp Engn W Lafayette IN 47907 USA Amazon Supply Chain Optimizat Technol Bellevue WA 98004 USA

Optimizing non-convex functions is of primary importance in modern pattern recognition because it underlies the training of deep networks and nonlinear dimensionality reduction. First-order algorithms under suitable randomized perturbations or step-size rules have been shown to be effective for such settings as their limit points can be guaranteed to be local extrema rather than saddle points. However, it is well-known that the practical convergence of first-order methods is slower than those which exploit additional structure. In particular, empirically, successive convex approximation (SCA) converges faster than first-order methods. However, to date, SCA in general non-convex settings converges to first-order stationary points, which could either be local extrema or saddle points whose performance is typically inferior. To mitigate this issue, we propose calibrated randomized perturbations of SCA, which exhibit the improved convergence rate as compared to the gradient descent counter part. In particular, our main technical contributions are to establish the non-asymptotic performance of SCA algorithm and its perturbed variant converges to an approximate second-order stationary point. Experiments on multi-dimensional scaling, a machine learning problem whose training objective is non-convex, substantiate the performance gains associated with employing random perturbations.

关键词： Convergence Optimization Perturbation methods Gradient methods Approximation algorithms Training signal processing algorithms Stochastic optimization non-convex optimization first order methods

来源：评论

学校读者我要写书评

暂无评论

Resource Allocation for IRS Assisted SGF NOMA Transmission: A MADRL Approach

引用

IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS 2022年第4期40卷 1302-1316页

作者： Chen, Jian Guo, Liang Jia, Jie Shang, Jianhui Wang, Xingwei Northeastern Univ Sch Comp Sci & Engn Shenyang 110819 Peoples R China Minist Educ Engn Res Ctr Secur Technol Complex Network Syst Shenyang 110819 Peoples R China

Non-orthogonal multiple access (NOMA) assisted semi-grant-free (SGF) transmission has been viewed as one of the promising technologies to meet massive connectivity requirements of the next-generation networks. A novel intelligent reconfigurable surface (IRS) assisted SGF NOMA transmission system is proposed, where the IRS is employed to satisfy the channel gain requirements for grant-based users (GBUs) and grant-free users (GFUs). The dynamic optimization on the sub-carrier assignment and power allocation for roaming GFUs, and the amplitude control and phase shift design for reflecting elements of the IRS, is formulated. Aiming at maximizing the long-term data rate of all GFUs, the optimization problem is first modeled as a multi-agent Markov decision problem. Then, three multi-agent deep reinforcement learning based frameworks are proposed to solve the problem under three different IRS cases, including the ideal IRS, non-ideal IRS with continuous phase shifts, and non-ideal IRS with discrete phase shifts. Specifically, for each GFU agent, a sub-carrier assignment deep Q-network (DQN) and a power allocation deep deterministic policy gradient (DDPG) are integrated to dynamically assign network resources for each GFU. For the only IRS agent, two DDPGs are integrated to dynamically assign phase shift and amplitude for each reflecting element of ideal IRS. The single DDPG for dynamically assigning continuous phase shifts, and parallel DQNs for dynamically assigning discrete phase shifts for non-ideal IRS with fixed amplitude are also proposed. Simulation results demonstrate that: 1) The network sum rates of all GFUs can achieve a significant improvement with the aid of IRS, comparing with the system without IRS. 2) The network sum rates of the NOMA assisted SGF transmissions are superior to that of OMA assisted GF transmissions.

关键词： NOMA Resource management Reinforcement learning Quality of service Silicon carbide signal processing algorithms Uplink DDPG DQN IRS NOMA resource allocation SGF

来源：评论

学校读者我要写书评

暂无评论

Novel Phase and CFO Estimation DSP for Photonics-Based Sub-Thz Communication

引用

JOURNAL OF LIGHTWAVE TECHNOLOGY 2022年第9期40卷 2710-2716页

作者： Rha, Hae Young Moon, Sang-Rok Lee, Joon Ki Cho, Seung-Hyun Elect & Telecommun Res Inst Opt Commun Res Sect Daejeon 34129 South Korea

Sub-terahertz (THz)-band communication system has drawn much attention as a promising technology to provide future-proof high data rate services. In a photonics-based sub-THz communication system, the generation of THz wave using free running lasers enables us to make a simple, cost effective and frequency tunable implementation. On the other hand, commercially available lasers have relatively broader linewidth and a large carrier frequency offset (CFO). To mitigate the performance degradations due to phase noise and CFO, a carrier recovery digital signal processing (DSP) algorithm is studied for a sub-THz transmission system. We propose a novel phase estimation algorithm to avoid cyclic slips while minimizing phase estimation error to improve BER performance. Our proposed phase recovery DSP algorithm is demonstrated in a 16-quadrature amplitude modulation (QAM) in a 0.3 THz band photonics-based transmission system. Experimental results show that the measured BER are improved from 8.8x10(-3) to 3.6x10(-3) in a 120 Gb/s 16-QAM transmission using the proposed algorithm. A wide range of CFO estimation is also supported for a sub-THz wireless transmission system using off-the-shelf lasers. Recovery of a CFO between -5 GHz and 5 GHz was also successfully demonstrated.

关键词： Phase estimation Phase noise Estimation AWGN Laser noise Photonics signal processing algorithms Carrier frequency offset cyclic slip phase estimation photonic technology sub-THz communication

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：