检索结果-内蒙古大学图书馆

Iterative Forward-Backward pursuit Algorithm for Compressed Sensing

JOURNAL OF ELECTRICAL AND COMPUTER ENGINEERING 2016年第Pt.1期2016卷 5940371.1-5940371.7页

作者： Wang, Feng Zhang, Jianping Sun, Guiling Geng, Tianyu Nankai Univ Coll Elect Informat & Opt Engn Tianjin 300350 Peoples R China Shanghai Jiao Tong Univ Sch Elect Informat & Elect Engn Shanghai 200030 Peoples R China

It has been shown that iterative reweighted strategies will often improve the performance of many sparse reconstruction algorithms. Iterative Framework for Sparse Reconstruction algorithms (IFSRA) is a recently proposed method which iteratively enhances the performance of any given arbitrary sparse reconstruction algorithm. However, IFSRA assumes that the sparsity level is known. Forward-Backward pursuit (FBP) algorithm is an iterative approach where each iteration consists of consecutive forward and backward stages. Based on the IFSRA, this paper proposes the Iterative Forward-Backward pursuit (IFBP) algorithm, which applies the iterative reweighted strategies to FBP without the need for the sparsity level. By using an approximate iteration strategy, IFBP gradually iterates to approach the unknown signal. Finally, this paper demonstrates that IFBP significantly improves the reconstruction capability of the FBP algorithm, via simulations including recovery of random sparse signals with different nonzero coefficient distributions in addition to the recovery of a sparse image.

关键词： Reconstruction algorithm pursuit algorithms iterative methods Iterative Compressed sensing

来源：评论

学校读者我要写书评

暂无评论

On Projection Matrices and Dictionaries in ECG Compressive Sensing - a Comparative Study 12

On Projection Matrices and Dictionaries in ECG Compressive S...

引用

Symposium on Neural Network Applications in Electrical Engineering (NEUREL)

作者： Fira, Monica Goras, Liviu Inst Comp Sci Bd Carol 1 22A Iasi 700505 Romania Gh Asachi Tech Univ Fac Elect & Telecommun Iasi 700505 Romania

ISBN: (纸本)9781479958887

In this communication we propose and discuss comparatively several techniques for ECG signal compression inspired from the fundamentals of compressed sensing (CS) theory, focusing on acquisition techniques, projection matrices and reconstruction dictionaries and on the effects of the preprocessing involved. Essentially, we investigate and discuss two approaches. The first approach for ECG signal compression relies on the direct CS acquisition of the signal with no preprocessing of the waveforms before taking the projections, neither for the construction of the dictionaries. This "genuine" CS we will call patient specific classical compressed sensing (PSCCS) since the dictionary is built from patient initial recordings. The second approach implements a specific preprocessing stage designed to enhance sparsity and improve recoverability, based on segmenting the signal into single heart beats (also known as cardiac patterns) - denoted further as cardiac patterns compressed sensing - (CPCS) since in this case the acquired signals and the dictionary atoms are preprocessed segmented cardiac beats without or with centering of the R wave.

关键词： Compressed sensing ECG compression electrocardiography pursuit algorithms

来源：评论

学校读者我要写书评

暂无评论

Approximate Guarantees for Dictionary Learning 32

Approximate Guarantees for Dictionary Learning

引用

32nd Conference on Learning Theory (COLT) part of the ACM Federated Computing Research Conference

作者： Bhaskara, Aditya Tai, Wai Ming Univ Utah Salt Lake City UT 84112 USA

In the dictionary learning (or sparse coding) problem, we are given a collection of signals (vectors in R-d), and the goal is to find a "basis" in which the signals have a sparse (approximate) representation. The problem has received a lot of attention in signal processing, learning, and theoretical computer science. The problem is formalized as factorizing a matrix X(d x n) (whose columns are the signals) as X = AY, where A has a prescribed number m of columns (typically m << n), and Y has columns that are k-sparse (typically k << d). Most of the known theoretical results involve assuming that the columns of the unknown A have certain incoherence properties, and that the coefficient matrix Y has random (or partly random) structure. The goal of our work is to understand what can be said in the absence of such assumptions. Can we still find A and Y such that X approximate to AY? We show that this is possible, if we allow violating the bounds on m and k by appropriate factors that depend on k and the desired approximation. Our results rely on an algorithm for what we call the threshold correlation problem, which turns out to be related to hypercontractive norms of matrices. We also show that our algorithmic ideas apply to a setting in which some of the columns of X are outliers, thus giving similar guarantees even in this challenging setting.

关键词： Dictionary learning sparse coding approximation algorithms pursuit algorithms

来源：评论

学校读者我要写书评

暂无评论

Sleep Spindle Detection Using Time-Frequency Sparsity

Sleep Spindle Detection Using Time-Frequency Sparsity

引用

IEEE Signal Processing in Medicine and Biology Symposium

作者： Parekh, Ankit Selesnick, Ivan W. Rapoport, David M. Ayappa, Indu NYU Sch Engn Dept Elect & Comp Engn Brooklyn NY 11201 USA NYU Sch Med Dept Med Div Pulm Crit Care & Sleep Med New York NY USA

ISBN: (纸本)9781479981847

This paper proposes an EEG processor for sleep spindle detection algorithms. It non-linearly separates the raw EEG signal into non-oscillatory transient and sustained rhythmic oscillation components using long and short windows for the short-time Fourier transform. The processor utilizes the fact that sleep spindles can be sparsely represented via the inverse of a short-time Fourier transform. Five sleep spindle detectors were tested on the EEG database with and without the proposed EEG processor. We achieved an improvement of 13.3% in the by-sample F-1 score, and 13.9% in the by-sample Matthews Correlation Coefficient score of these algorithms when the processed EEG was used for spindle detection. The processor was able to improve the scores by reducing the number of false positive spindles and increasing the number of true positive spindles detected.

关键词： Short time Fourier transform spectrogram convex optimization pursuit algorithms

来源：评论

学校读者我要写书评

暂无评论

ROBUST SPARSE IMAGE DENOISING

ROBUST SPARSE IMAGE DENOISING

引用

18th IEEE International Conference on Image Processing (ICIP)

作者： Obradovic, Radovan Janev, Marko Antic, Borislav Crnojevic, Vladimir Petrovic, Nemanja I. Univ Novi Sad Fac Engn Trg Dositeja Obradovica 6 Novi Sad 21000 Serbia Heidelberg Univ Interdisciplinary Ctr Sci Comp Heidelberg Germany

ISBN: (纸本)9781457713033

In this paper, we propose a novel method for denoising images corrupted by the mixture of the additive white Gaussian noise and the heavy tailed noise. The proposed method is based on robust statistical approach, i.e. application of M-estimators in the combination with l(1) sparse regularization technique. Thus, we perform the denoising by numerical solving of the convex optimization problem. Additionally, the developed method performs denoising adaptively in order to preserve the image details, by adjusting regularization coefficients according to the local variance. The proposed denoising scheme produces excellent results, both objectively (in terms of PSNR and MSSIM) and subjectively and outperforms state-of-the-art filters for denoising heavy tailed noise in images.

关键词： Image denoising estimation pursuit algorithms gradient methods

来源：评论

学校读者我要写书评

暂无评论

Computational Task Offloading in Mobile Edge Computing using Learning Automata

Computational Task Offloading in Mobile Edge Computing using...

引用

IEEE/CIC International Conference on Communications in China (ICCC)

作者： Abbas, Zahir Li, Jun Yadav, Nagendra Tariq, Irfan Nanjing Univ Sci & Technol Sch Elect & Opt Engn Nanjing 210094 Jiangsu Peoples R China

ISBN: (纸本)9781538670057

The widespread diffusion of mobile devices is triggering an exponential growth of mobile data causing an offloading problem. In this paper, a model has been proposed to investigate computing task offloading problem of small base stations (SBSs). We formulate an optimization problem to minimize utility function for energy and execution time consumptions, where mobile devices (MDs) can select local computation or offloading to SBSs. We use discrete generalist pursuit algorithm (DGPA) based on learning automata (LA) to optimize the computation allocation. First, we examine a faster-converging of discrete generalist pursuit algorithm (DGPA) for LA is based on the concept of conditional inaction discrete generalist pursuit algorithm (CIDGPA). Then, we design a reward function for the selection of optimal allocation of SBSs with minimum offloading latency. After convergence to optimal allocation, the offloading computing task of MDs has been performed to SBSs with minimum weighted sum of time and energy consumption. Finally, numerical results demonstrate the best improvement of DGPA algorithm.

关键词： Mobile handsets Task analysis Resource management Energy consumption pursuit algorithms Servers Learning automata

来源：评论

学校读者我要写书评

暂无评论

Identification of a multivariate fermentation process using constructive learning 7

Identification of a multivariate fermentation process using ...

引用

7th Brazilian Symposium on Neural Networks (SBRN 02)

作者： Meleiro, LAC Campello, RJGB Maciel, R Von Zuben, FJ Univ Estadual Campinas Sch Chem Engn Campinas SP Brazil

ISBN: (纸本)0769517099

In the present work, a constructive learning algorithm is employed to design an optimal one-hidden neural network structure that best approximates a given mapping. The method determines not only the optimal number of hidden neurons but also the best activation function for each node. Here, the projection pursuit technique is applied in association with the optimization of the solvability condition, giving rise to a more efficient and accurate computational learning algorithm. As each activation function of a hidden neuron is optimally defined for every approximation problem, better rates of convergence are achieved. Since the training process operates the hidden neurons individually, a pertinent activation function employing Hermite polynomials can he iteratively developed for each neuron as a function of the learning set. The proposed constructive learning algorithm was successfully applied to identify a large-scale multivariate process, providing a multivariable model that was able to describe the complex process dynamics, even in long-range horizon predictions.

关键词： Artificial neural networks Chemical engineering Convergence Design engineering Identity-based encryption Neural networks Neurons Polynomials Predictive models pursuit algorithms

来源：评论

学校读者我要写书评

暂无评论

Joint Channel Estimation and Pre-Equalization for Monobit Transmission System

Joint Channel Estimation and Pre-Equalization for Monobit Tr...

引用

2nd IEEE/CIC International Conference on Communications in China (ICCC)

作者： Li, Pengfei Wang, Zhiyong Yin, Huarui Univ Sci & Technol China Dept Elect Engn & Informat Sci Hefei 230027 Peoples R China

ISBN: (纸本)9781467328159

High date rate is one of the important targets in future communication systems, which requires large bandwidth. However, the multibit analog-to-digital converter (ADC) with such a high rate and large bandwidth will have a large power consumption and high implementation complexity. Therefore, monobit receivers have been proposed. In this paper, we concentrate on the channel estimation and equalization for monobit transmission system, which is not investigated up to now. To our surprise, the channel state information (CSI) can be estimated accurately in the condition of one bit quantization, with the proposed greedy pursuit algorithm and iterative method. The minimum mean square error (MMSE) pre-equalizer is then applied to suppress the effect of inter symbol interference (ISI) of the multipath channel. The bit error rate (BER) performance of the proposed algorithm is presented and compared with that of full CSI acknowledge. With the pre-equalizer, the ISI effect is canceled almost completely and "error floor" is eliminated as well. The BER performance can be boosted significantly.

关键词： channel estimation bit error rate performance COMMUNICATION SYSTEMS pre-equalization Transmission systems pre-equalizer Power consumption Equalization Mean square error analog digital converter pursuit algorithms

来源：评论

学校读者我要写书评

暂无评论

Towards a learning automata solution to the multi-constraint partitioning problem

Towards a learning automata solution to the multi-constraint...

引用

IEEE Conference on Cybernetics and Intelligent Systems

作者： Horn, Geir Ommen, B. John SIMULA Res Lab POB 134 N-1325 Lysaker Norway Carleton Univ Sch Comp Sci Ottawa ON K1S 5B6 Canada

ISBN: (纸本)9781424400225

We consider the problem of partitioning a set of elements (or objects) into mutually exclusive classes (or groups), where elements which are "similar" to each other are, hopefully, located in the same class. This problem has been shown to be NP-Hard, and the literature reports solutions in which the similarity constraint consists of a single index. For example, typical "similarity" conditions that have been used in the literature include those in which "similar" objects are accessed together, or when they communicate (as processes do) with each other. In this paper, we present the first reported solution(1) to the case when the objects could be linked together in a multi-constraint manner, and indeed, visit the scenario when the constraints could, themselves, be contradictory. The solution we propose is based on the theory of estimator-based Learning Automata (LA), operating in non-stationary environments. Rather than use traditional estimates, we advocate the use of stochastic weak-estimates [2] and the specific digraph properties of the relations between the elements. Although the solutions proposed perform admirably when the number of elements is small, the simulated results demonstrate that the quality of the final solution decreases with the number of elements. Thus, although this is the first reported solution to the problem which incorporates specific digraph properties of the objects, the scalability of the solution remains open.

关键词： object partitioning multi-constraint partitioning learning automata pursuit algorithms

来源：评论

学校读者我要写书评

暂无评论

Estimation of time-varying frequency-selective channels using a matching pursuit technique

Estimation of time-varying frequency-selective channels usin...

引用

IEEE Conference on Wireless Communications and Networking

作者： Yang Lui D.K. Borah Klipsch Sch. of Electr. & Comput. Eng. New Mexico State Univ. Las Cruces NM USA Klipsch School of Electrical and Computer Engineering New Mexico State University Las Cruces NM USA

Estimation of fast fading channels with large possible delay spread is considered. The time-varying fading channel taps are explicitly modeled by polynomial basis functions, and the basis coefficients are obtained via a matching pursuit (MP) algorithm. It is shown that, in the case of sparse channels, this approach outperforms the conventional least squares (LS) estimation with polynomial basis functions as well as the conventional adaptive matching pursuit, while still providing significant savings in complexity. Moreover, the MP is used to identify the appropriate polynomial model orders for channel taps that vary at different rates. A new least squares algorithm using matching pursuit (LSMP) is proposed. It is shown that the LSMP performs well for both sparse and non-sparse time-varying channels.

关键词： Frequency estimation Matching pursuit algorithms Fading Delay estimation Polynomials pursuit algorithms Least squares approximation Maximum likelihood detection Channel estimation Least squares methods

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：