负荷预测的线性稀疏数AR预测模型

对时间序列线性ARIMA模型的建模和预测方法进行了总结,根据其特征提出了采用稀疏数AR模型代替ARIMA模型作为通用线性预测模型,并给出了模型搜索算法,使得时间序列线性预测模型的建模过程无需人的干预而能通过程序自动完成。

稀疏阶数估计压缩频谱感知宽带认知通信研究

由于信号的稀疏度通常未知,因此需要按照样本数的上限进行采样.为了解决这个问题,提出一种稀疏阶数估计的方法压缩频谱感知宽带认知无线电通信技术,该技术采用一种统计学习的方法在稀疏信号恢复以前,利用极少的样本数据估计出信号的稀疏阶数.采用极少部分样本估计出宽带谱的稀疏阶数,然后根据估计的稀疏阶数调整所采集的样本数.采用这种方法能够自适应的调整数据获取量,从而减少数据获取代价而不会降低感知性能.

基于Spark框架的改进协同过滤算法

随着互联网数据量的不断膨胀,单机已经无法在可接受的时间范围内计算完基于大规模数据的推荐算法,也无法存放海量的数据。利用Spark平台内存计算的优点,设计了一种分布式的基于项目的协同过滤算法,利用Spark提供的RDD(resilient distributed dataset)算子完成算法的设计。针对由于数据稀疏而导致的相似度计算不准确的问题,提出了一种利用两项目间公共用户数目进行加权的相似度计算公式,提高了最终推荐结果的准确度。为了改善计算中涉及到的数据表等值连接操作耗时太长的问题,利用自定义的Hash_join函数替代Spark自带的连接操作算子,提高了计算效率。采用UCI的公用数据集MovieLens对算法进行测试,并分别与改进前的算法以及单机运行的算法进行对比,结果表明,改进的算法在准确度和效率方面都有更好的表现。

Geophysical data sparse reconstruction based on L0-norm minimization

Missing data are a problem in geophysical surveys, and interpolation and reconstruction of missing data is part of the data processing and interpretation. Based on the sparseness of the geophysical data or the transform domain, we can improve the accuracy and stability of the reconstruction by transforming it to a sparse optimization problem. In this paper, we propose a mathematical model for the sparse reconstruction of data based on the LO-norm minimization. Furthermore, we discuss two types of the approximation algorithm for the LO- norm minimization according to the size and characteristics of the geophysical data： namely, the iteratively reweighted least-squares algorithm and the fast iterative hard thresholding algorithm. Theoretical and numerical analysis showed that applying the iteratively reweighted least-squares algorithm to the reconstruction of potential field data exploits its fast convergence rate, short calculation time, and high precision, whereas the fast iterative hard thresholding algorithm is more suitable for processing seismic data, moreover, its computational efficiency is better than that of the traditional iterative hard thresholding algorithm.

Estimating primaries by sparse inversion of the 3D Curvelet transform and the L1-norm constraint

In this paper, we built upon the estimating primaries by sparse inversion （EPSI） method. We use the 3D curvelet transform and modify the EPSI method to the sparse inversion of the biconvex optimization and Ll-norm regularization, and use alternating optimization to directly estimate the primary reflection coefficients and source wavelet. The 3D curvelet transform is used as a sparseness constraint when inverting the primary reflection coefficients, which results in avoiding the prediction subtraction process in the surface-related multiples elimination （SRME） method. The proposed method not only reduces the damage to the effective waves but also improves the elimination of multiples. It is also a wave equation- based method for elimination of surface multiple reflections, which effectively removes surface multiples under complex submarine conditions.

Compressed sensing estimation of sparse underwater acoustic channels with a large time delay spread

The estimation of sparse underwater acoustic channels with a large time delay spread is investigated under the framework of compressed sensing. For these types of channels, the excessively long impulse response will significantly degrade the convergence rate and tracking capability of the traditional estimation algorithms such as least squares （LS）, while excluding the use of the delay-Doppler spread function due to huge computational complexity. By constructing a Toeplitz matrix with a training sequence as the measurement matrix, the estimation problem of long sparse acoustic channels is formulated into a compressed sensing problem to facilitate the efficient exploitation of sparsity. Furthermore, unlike the traditional l1 norm or exponent-based approximation l0 norm sparse recovery strategy, a novel variant of approximate l0 norm called AL0 is proposed, minimization of which leads to the derivation of a hybrid approach by iteratively projecting the steepest descent solution to the feasible set. Numerical simulations as well as sea trial experiments are compared and analyzed to demonstrate the superior performance of the proposed algorithm.

基于稀疏度阶数优化的杂波密度估计算法

针对杂波分布不均匀且密度未知的多目标跟踪问题,提出一种基于稀疏度阶数优化的杂波密度估计算法.首先,剔除在跟踪门内的潜在目标测量,获取杂波测量集;其次,从杂波测量集中构造"稀疏度阶数-超立方体容积"的样本,并利用支持向量回归机对样本拟合;再次,通过梯度法求得拟合曲线的极值点,实现稀疏度阶数在线优化;最后,将优化后的杂波稀疏度估计器嵌入高斯混合概率假设密度滤波器中,实现复杂杂波环境下目标状态与杂波密度联合估计.仿真结果验证了所提出算法的有效性.

Designing simultaneous multichannel receivers based on fast filter bank

A scheme to design a simultaneous multichannel receiver is proposed to process multichannel signals in parallel, which is achieved by exploiting the attractive characteristics of a fast filter bank（ FFB）, such as cascaded structure, high frequency selectivity and lowcomputational complexity. Based on the minimization of the objective function, quantified in terms of the total number of multiplications required, subject to prescribed allowable ripples in the passband and stopband, the impulse response coefficients of the prototype filter in each stage are obtained to meet the requirements of the overall specifications for each channel at the receiver side. Simulations and experimental results on the frequency modulation（ FM） broadcast mutlichannel signal receiving system with the FM range from88 to 108 MHz, built upon the proposed FFB structure, are performed to verify its performance. Those results indicate that the proposed scheme is efficient in FM audio indexing applications and has a lower computational complexity, which is approximately 66. 4% of the weighted overlap and add（ WOLA） filter banks based solution.

Three-dimensional interpretation of sparse survey line MT data: Synthetic examples

Currently, most of MT （magnetotelluric） data are still collected on sparse survey lines and interpreted using 2D inversion methods because of the field work cost, the work area environment, and so on. However, there are some 2D interpretation limitations of the MT data from 3D geoelectrical structures which always leads to wrong geological interpretations. In this paper, we used the 3D inversion method to interpret the MT sparse lines data. In model testing, the sparse lines data are the MT full information data generated from a test model and processed using the 3D conjugate gradients inversion code. The inversion results show that this inversion method is reasonable and effective. Meanwhile, we prove that for inversion results with different element parameters, the results by joint inversion of both the impedance tensor data and the tipper data are more accurate and closer to the test model.

Algorithm for reconstructing compressed sensing color imaging using the quaternion total variation

A new method for reconstructing the compressed sensing color image by solving an optimization problem based on total variation in the quaternion field is proposed, which can effectively improve the reconstructing ability of the color image. First, the color image is converted from RGB （red, green, blue） space to CMYK （cyan, magenta, yellow, black） space, which is assigned to a quaternion matrix. Meanwhile, the quaternion matrix is converted into the information of the phase and amplitude by the Euler form of the quatemion. Secondly, the phase and amplitude of the quatemion matrix are used as the smoothness constraints for the compressed sensing （CS） problem to make the reconstructing results more accurate. Finally, an iterative method based on gradient is used to solve the CS problem. Experimental results show that by considering the information of the phase and amplitude, the proposed method can achieve better performance than the existing method that treats the three components of the color image as independent parts.

Applying Analytical Derivative and Sparse Matrix Techniques to Large-Scale Process Optimization Problems

The performance of analytical derivative and sparse matrix techniques applied to a traditional dense sequential quadratic programming (SQP) is studied, and the strategy utilizing those techniques is also presented.Computational results on two typical chemical optimization problems demonstrate significant enhancement in efficiency, which shows this strategy is promising and suitable for large-scale process optimization problems.

Traffic danger detection by visual attention model of sparse sampling

A method to detect traffic dangers based on visual attention model of sparse sampling was proposed. The hemispherical sparse sampling model was used to decrease the amount of calculation which increases the detection speed. Bayesian probability model and Gaussian kernel function were applied to calculate the saliency of traffic videos. The method of multiscale saliency was used and the final saliency was the average of all scales, which increased the detection rates extraordinarily. The detection results of several typical traffic dangers show that the proposed method has higher detection rates and speed, which meets the requirement of real-time detection of traffic dangers.

Discriminative Structured Dictionary Learning for Image Classification

In this paper, a discriminative structured dictionary learning algorithm is presented. To enhance the dictionary's discriminative power, the reconstruction error, classification error and inhomogeneous representation error are integrated into the objective function. The proposed approach learns a single structured dictionary and a linear classifier jointly. The learned dictionary encourages the samples from the same class to have similar sparse codes, and the samples from different classes to have dissimilar sparse codes. The solution to the objective function is achieved by employing a feature-sign search algorithm and Lagrange dual method. Experimental results on three public databases demonstrate that the proposed approach outperforms several recently proposed dictionary learning techniques for classification.

Two-Level Bregman Method for MRI Reconstruction with Graph Regularized Sparse Coding

In this paper, a two-level Bregman method is presented with graph regularized sparse coding for highly undersampled magnetic resonance image reconstruction. The graph regularized sparse coding is incorporated with the two-level Bregman iterative procedure which enforces the sampled data constraints in the outer level and updates dictionary and sparse representation in the inner level. Graph regularized sparse coding and simple dictionary updating applied in the inner minimization make the proposed algorithm converge with a relatively small number of iterations. Experimental results demonstrate that the proposed algorithm can consistently reconstruct both simulated MR images and real MR data efficiently, and outperforms the current state-of-the-art approaches in terms of visual comparisons and quantitative measures.

Discriminant embedding by sparse representation and nonparametric discriminant analysis for face recognition

A novel supervised dimensionality reduction algorithm, named discriminant embedding by sparse representation and nonparametric discriminant analysis(DESN), was proposed for face recognition. Within the framework of DESN, the sparse local scatter and multi-class nonparametric between-class scatter were exploited for within-class compactness and between-class separability description, respectively. These descriptions, inspired by sparse representation theory and nonparametric technique, are more discriminative in dealing with complex-distributed data. Furthermore, DESN seeks for the optimal projection matrix by simultaneously maximizing the nonparametric between-class scatter and minimizing the sparse local scatter. The use of Fisher discriminant analysis further boosts the discriminating power of DESN. The proposed DESN was applied to data visualization and face recognition tasks, and was tested extensively on the Wine, ORL, Yale and Extended Yale B databases. Experimental results show that DESN is helpful to visualize the structure of high-dimensional data sets, and the average face recognition rate of DESN is about 9.4%, higher than that of other algorithms.

Sparse Approximations of the Schur Complement for Parallel Algebraic Hybrid Solvers in 3D

In this paper we study the computational performance of variants of an algebraic additive Schwarz preconditioner for the Schur complement for the solution of large sparse linear systems.In earlier works,the local Schur complements were computed exactly using a sparse direct solver.The robustness of the preconditioner comes at the price of this memory and time intensive computation that is the main bottleneck of the approach for tackling huge problems.In this work we investigate the use of sparse approximation of the dense local Schur complements.These approximations are computed using a partial incomplete LU factorization.Such a numerical calculation is the core of the multi-level incomplete factorization such as the one implemented in pARMS. The numerical and computing performance of the new numerical scheme is illustrated on a set of large 3D convection-diffusion problems;preliminary experiments on linear systems arising from structural mechanics are also reported.

Image denoising method with tree-structured group sparse modeling of wavelet coefficients

In order to enhance the image contrast and quality, inspired by the interesting observation that an increase in noise intensity tends to narrow the dynamic range of the local standard deviation (LSD) of an image, a tree-structured group sparse optimization model in the wavelet domain is proposed for image denoising. The compressed dynamic range of LSD caused by noise leads to a contrast reduction in the image, as well as the degradation of image quality. To equalize the LSD distribution, sparsity on the LSD matrix is enforced by employing a mixed norm as a regularizer in the image denoising model. This mixed norm introduces a coupling between wavelet coefficients and provides a tree-structured group scheme. The alternating direction method of multipliers (ADMM) and the fast iterative shrinkage-thresholding algorithm (FISTA) are applied to solve the group sparse model based on different cases. Several experiments are conducted to verify the effectiveness of the proposed model. The experimental results indicate that the proposed group sparse model can efficiently equalize the LSD distribution and therefore can improve the image contrast and quality.

Robust elastic impedance inversion using L1-norm misfit function and constraint regularization

The classical elastic impedance （EI） inversion method, however, is based on the L2-norm misfit function and considerably sensitive to outliers, assuming the noise of the seismic data to be the Guassian-distribution. So we have developed a more robust elastic impedance inversion based on the Ll-norm misfit function, and the noise is assumed to be non-Gaussian. Meanwhile, some regularization methods including the sparse constraint regularization and elastic impedance point constraint regularization are incorporated to improve the ill-posed characteristics of the seismic inversion problem. Firstly, we create the Ll-norm misfit objective function of pre-stack inversion problem based on the Bayesian scheme within the sparse constraint regularization and elastic impedance point constraint regularization. And then, we obtain more robust elastic impedances of different angles which are less sensitive to outliers in seismic data by using the IRLS strategy. Finally, we extract the P-wave and S-wave velocity and density by using the more stable parameter extraction method. Tests on synthetic data show that the P-wave and S-wave velocity and density parameters are still estimated reasonable with moderate noise. A test on the real data set shows that compared to the results of the classical elastic impedance inversion method, the estimated results using the proposed method can get better lateral continuity and more distinct show of the gas, verifying the feasibility and stability of the method.

Compression techniques of mechanical vibration signals based on optimal sparse representations

This paper presents the result of an experimental study on the compression of mechanical vibration signals. The signals are collected from both rotating and reciprocating machineries by the accelerometers and a data acquisition （DAQ） system. Four optimal sparse representation methods for compression have been considered including the method of frames （ MOF）, best orthogonal basis （ BOB）, matching pursuit （MP） and basis pursuit （BP）. Furthermore, several indicators including compression ratio （CR）, mean square error （MSE）, energy retained （ER） and Kurtosis are taken to evaluate the performance of the above methods. Experimental results show that MP outperforms other three methods.

Using Spatial Data Mining to Predict the Solvability Space of Preconditioned Sparse Linear Systems

The solution of large sparse linear systems is one of the most important problems in large scale scientific computing. Among the many methods developed, the preconditioned Krylov subspace methods [1] are considered the preferred methods. Selecting an effective preconditioner with appropriate parameters for a specific sparse linear system presents a challenging task for many application scientists and engineers who have little knowledge of preconditioned iterative methods. The purpose of this paper is to predict the parameter solvability space of the preconditioners with two or more parameters. The parameter solvability space is usually irregular, however, in many situations it shows spatial locality, i.e. the parameter locations that are closer in parameter space are more likely to have similar solvability. We propose three spatial data mining methods to predict the solvability of ILUT which make usage of spatial locality in different ways. The three methods are MSC （multi-points SVM classifier）, OSC （overall SVM classifier）, and OSAC （overall spatial autoregressive classifier）. The experimental results show that both MSC and OSAC can obtain 90% accuracy in prediction, but OSAC is much simpler to implement. We focus our work on ILUT preconditioner [2], but the proposed strategies should be applicable to other preconditioners with two or more parameters.