Network traffic prediction method based on wavelet transform and multiple models fusion

Accurate prediction of network traffic is an important premise in network management and congestion control. In order to improve the prediction accuracy of network traffic, a prediction method based on wavelet transform and multiple models fusion is presented. Mallat wavelet transform algorithm is used to decompose and reconstruct the network traffic time series. The approximate and detailed components of the original network traffic can be obtained. The characteristics of approximate components and detail components are analyzed by Hurst exponent. Then, according to the different characteristics of the components, autoregressive integrated moving average model (ARIMA) is chosen as the prediction model for the approximate component. Least squares support vector machine (LSSVM) is used to predict detail component. Meanwhile, an improved particle swarm optimization (PSO) algorithm is proposed to optimize the parameters of the LSSVM model. Gauss‐Markov estimation algorithm is adapted to fuse the predicted values of multiple prediction models. The variance of fusion prediction error is smaller than that of single prediction model, and the prediction accuracy is improved. Two actual datasets of network traffic are studied. Compared with other state‐of‐the‐art models, the case study results indicate that the proposed prediction method has a better prediction effect.     

基于频域多尺度Hurst指数的海杂波中目标检测方法

本文主要研究海杂波频谱的扩展自相似特性及多尺度Hurst指数在海杂波目标检测中的应用.作为分数布朗运动的一种推广,扩展自相似过程采用多尺度Hurst指数来描述分形信号.多尺度Hurst指数可以描述分形信号在各尺度下的细节信息,弥补了单一Hurst指数只能从整体上描述分形信号粗糙度的不足.首先,本文在实测雷达数据基础上研究了海杂波频谱的扩展自相似性以及影响参数;然后,利用在最优频域尺度下海杂波频谱的多尺度Hurst指数对目标相对较敏感的特点设计恒虚警检测方法,实现海杂波中的目标检测.实测数据分析表明,海杂波频谱的多尺度Hurst指数比时域单一Hurst指数、时域多尺度Hurst指数具有更好的海杂波与目标区分能力,且由于Fourier变换可以有效提升信杂比,该检测方法具有检测海杂波中微弱运动目标的潜力.     

Application of autoregressive models to fine arts painting analysis

This paper proposes algorithms for background texture removal in pictures. Two models of texture are considered: a multivariate autoregressive (AR) model (vector model) and a 2-D AR nonsymmetric half-plane model. Robinson's algorithm is used for vector model identification. The 2-D model is simply derived from the vector model, using the relationship established by Gambotto (1981).The two models are used to filter artificially textured test images, as well as naturally textured X-ray pictures of paintings from the Louvre Museum. The method proved to be very efficient to enhance the legibility of the paintings.     

有色噪声下基于Unscented粒子滤波的语音增强方法

针对含有色噪声的语音,提出了一种基于Unscented粒子滤波的单通道语音增强方法.采用时变自回归模型(TVAR)对干净语音建模,通过Unscented粒子滤波器估计AR模型的参数并滤除有色噪声.与大多数常用的粒子滤波选择的建议分布不同,Unscented粒子滤波器采用Unscented卡尔曼滤波器生成粒子滤波的建议分布.由于在粒子的更新过程中考虑了最近的观测值,Unscented粒子滤波器能够在粒子数少于传统粒子滤波算法所需粒子数目的基础上改善估计的性能.仿真实验结果表明,在有色噪声背景下该算法具有良好的语音增强效果.     

基于标签多伯努利滤波器的机动小目标检测前跟踪

标签多伯努利(LMB)滤波器在传统多伯努利滤波器基础上引入标签空间,能够实现真正意义上的多目标轨迹级滤波.文章对红外小目标的运动和量测进行建模,将标签多伯努利应用到红外小目标检测前跟踪领域.在此基础上,为了实现对运动模型时变目标的检测前跟踪,将交互式多模型(IMM)与LMB检测前跟踪算法相结合,提出IMM-LMB检测前跟踪算法.此外,给出了该算法的序贯蒙特卡罗实现.仿真结果表明,所提算法能够从输入的原始图像中直接实现轨迹级多目标检测和跟踪,且能够在线更新多模型概率,更好的适应多机动目标场景.     

Motion field modeling for video sequences

We propose a model for the interframe correspondences existing between pixels of an image sequence. These correspondences form the elements of a field called the motion field. In our model, spatial neighborhoods of motion elements are related based on a generalization of autoregressive (AR) modeling of the time-series. We also propose a joint spatio-temporal model by including spatial neighborhoods of pixel intensities in the motion model. A fundamental difference of our approach with most previous approaches to modeling motion is in basing our model on concepts from statistical signal processing. The developments in this paper give rise to the promise of extending well-understood tools of signal processing (e.g., filtering) to the analysis and processing of motion fields. Simulation results presented show the performance of our models in interframe prediction; specifically, on average the motion model performs 29% better in terms of the mean squared error energy over a commonly used pel-recursive approach. The spatio-temporal model improves the prediction efficiencies by 8% over the motion model. Our model can also be used to obtain estimates of the optical flow field as the simulations demonstrate.     

On the approximate decorrelation property of the discrete wavelet transform for fractionally differenced processes

In this correspondence, we develop an asymptotic theory for the correlations of wavelet coefficients of the discrete wavelet transform (DWT) for fractionally differenced processes. It provides a theoretical justification for the approximate decorrelation property of the DWT for fractionally differenced processes. In addition, it provides insights on how the length of the wavelet filter affects the within scale correlations and the between scale correlations differently; for within scale correlations, increasing the length of the wavelet filter increases the rate of decay as the two wavelet coefficients get further apart, while for between scale correlations, using a wavelet filter that is long enough can reduce the between scale correlations even for wavelet coefficients that are close together.     

Variable structure Human Intention Estimator with mobility and vision constraints as model selection criteria

In this paper, a novel method for early estimation of a human’s action intention is presented. Human intention, modeled as a goal location associated with a hand motion and eye gaze dynamics, is inferred by fusing information from collected hand motion and gaze motion data. The algorithm, called Human Intention Estimator with Variable Structure (HIEVS), uses two variable structure Interacting Multiple Model (VS-IMM) filters in parallel to process the hand motion and gaze data and generate posterior model probabilities associated with a finite set of action models. The posterior model probabilities from each filter are then fused at the end of each iteration and the current intention is estimated as the model, which has the highest fused posterior model probability. Two model set augmentation (MSA) algorithms are presented to select the active models for each VS-IMM during each iteration. For the hand motion filter, an MSA algorithm which computes the human’s reachable workspace is used. The MSA algorithm for the gaze filter utilizes the human’s visual span to determine the active models. This method allows for accurate early prediction of the human’s intention even when the total model set is large. A real world experiment is performed to validate the proposed method.     

海杂波AR谱多重分形特性及微弱目标检测方法

该文研究了海杂波功率谱的多重分形特性。为了克服频谱傅里叶分析的缺点,用现代谱估计的方法来计算海杂波的功率谱。AR模型是一个线性预测模型,它通过序列的自相关函数矩阵来估计功率谱,并且具有更精确的频谱分辨率。该文主要分析基于AR谱估计的海杂波功率谱的多重分形特性,以及在微弱目标检测中的应用。首先,以分数布朗运动(FBM)模型为例,证明其功率谱具有多重分形特性。其次,根据X波段雷达的实测海杂波数据,通过多重去趋势分析法(MF-DFA)验证了海杂波AR谱的多重分形特性。最后,分析了海杂波AR谱的广义Hurst指数以及影响参数,并提出一种基于局部AR谱广义Hurst指数的目标检测方法。实验结果表明,该种检测方法具有海杂波背景下微弱目标检测的能力。与现有的分形检测方法和传统的CFAR检测方法对比,该算法在低信杂比情况下具有较好的检测性能。     

TV系统中基于Kalman滤波器的自回归频谱空洞预测

研究了基于Kalman滤波器的自回归的频谱空洞预测模型。首先讨论了广播电视系统中授权用户和认知用户共存的系统模型;其次讨论了自回归预测原理和Kalman滤波预测的无偏性,并针对不同的通信模式提出不同的预测方法。理论分析和仿真结果均表明,基于Kalman滤波器的自回归预测值能很好地与理论值近似匹配,采用预测模型可以大大减少授权用户和认知用户的冲突,提高系统利用率。     

基于分形的音频频带盲扩展方法

本文提出了一种基于分形理论的音频频带盲扩展方法。首先,借助重标极差分析法估计频谱细节序列的Hurst指数,对不同音频信号谱细节序列的Hurst指数进行统计分析,验证了音频信号频域序列具有长相关特性。然后,运用坐标延迟法对谱细节序列进行相空间重构,基于夹角余弦选取预测中心相点的近邻点,根据预测中心相点和近邻相点的相关性,求取一个吸引子与预测中心相点相近的迭代函数系统,实现对高频谱细节的分形预测。最后,结合高斯混合模型的高频能量估计算法,实现宽带向超宽带的扩展算法。主、客观质量评测均表明本文算法优于传统的频带盲扩展方法。      

Extended fractal analysis for texture classification andsegmentation

The Hurst parameter for two-dimensional (2-D) fractional Brownian motion (fBm) provides a single number that completely characterizes isotropic textured surfaces whose roughness is scale-invariant. Extended self-similar (ESS) processes were previously introduced in order to provide a generalization of fBm. These new processes are described by a number of multiscale Hurst parameters. In contrast to the single Hurst parameter, the extended parameters are able to characterize a greater variety of natural textures where the roughness of these textures is not necessarily scale-invariant. In this work, we evaluate the effectiveness of multiscale Hurst parameters as features for texture classification and segmentation. For texture classification, the performance of the generalized Hurst features is compared to traditional Hurst and Gabor features. Our experiments show that classification accuracy for the generalized Hurst and Gabor features are comparable even though the generalized Hurst features lower the dimensionality by a factor of five. Next, the segmentation accuracy using generalized and standard Hurst features is evaluated on images of texture mosaics. For these experiments, the performance is evaluated with and without supplemental contrast and average grayscale features. Finally, we investigate the effectiveness of the Hurst features to segment real synthetic aperture radar (SAR) imagery.     

Asymptotic decorrelation of between-Scale Wavelet coefficients

In recent years there has been much interest in the analysis of time series using a discrete wavelet transform (DWT) based upon a Daubechies wavelet filter. Part of this interest has been sparked by the fact that the DWT approximately decorrelates certain stochastic processes, including stationary fractionally differenced (FD) processes with long memory characteristics and certain nonstationary processes such as fractional Brownian motion. It is shown that, as the width of the wavelet filter used to form the DWT increases, the covariance between wavelet coefficients associated with different scales decreases to zero for a wide class of stochastic processes. These processes are Gaussian with a spectral density function (SDF) that is the product of the SDF for a (not necessarily stationary) FD process multiplied by any bounded function that can serve as an SDF on its own. We demonstrate that this asymptotic theory provides a reasonable approximation to the between-scale covariance properties of wavelet coefficients based upon filter widths in common use. Our main result is one important piece of an overall strategy for establishing asymptotic results for certain wavelet-based statistics.     

卫星接收机非线性故障建模与仿真

非线性效应导致接收机设备故障以及性能降级一直是卫星通信系统面临的重要挑战。针对这一问题，本文提出了一种基于行为模型的接收机非线性故障分析方法。首先推导分析滤波器内反射干扰在幅频响应上的表现形式，以此构建滤波器故障描述模型。其次利用幅相转换关系对Saleh模型系数进行修正，建立表征放大器在过饱和点下不同程度的非线性故障模型。然后针对相位噪声相关性对振荡器频率稳定性的影响，利用一阶自回归模型表征在不同相关性相位噪声下，振荡器内存在的非线性故障。最后将滤波器、功率放大器和振荡器的故障模型级联，建立接收机射频前端整体的非线性故障模型，并通过星座图、功率谱以及电路仿真三方面对模型进行仿真分析和验证。实验结果表明，本文建立的故障行为模型能有效表征接收机射频前端存在的非线性故障。      

DFA在海杂波标度特性分析及目标检测中的应用

采用消除趋势波动分析(DFA)方法,研究了非平稳海杂波的标度特性及其参数表征问题,并提出了一种基于标度特性差异的目标检测算法。首先,在海杂波建模为分数布朗运动模型的基础上,对比分析了DFA和波动分析(FA)得出的标度特性,并初步解释了交叉标度现象出现的机理。然后,在假定相邻尺度范围内海杂波与理想分形模型相吻合的前提下,引入分段标度指数来表征海杂波标度特性随尺度的变化关系,分析结果表明,在特定的尺度范围内该参数可以有效区分目标单元与海杂波单元。为此,以该尺度范围内的分段标度指数均值为检验统计量设计了一种新的目标检测算法,并对尺度区间范围的选取、检测性能等问题进行了分析。实测数据中的检测性能证实了检测算法的有效性。      

Signal modeling with filtered discrete fractional noise processes

Filtered versions of fractionally differenced Gaussian noise (fdGn) processes are studied. Fractionally differenced Gaussian noise is a discrete-time equivalent of fractional Brownian motion. Filtered versions of such processes are ideally suited for modeling signals with different short-term and long-term correlation structure. Two iterative algorithms for estimating the parameters of filtered fdGn processes are described. The first technique is based on the expectation-maximization algorithm. It converges to a stationary point of the log-likelihood function corresponding to the parameters of the model. The second technique is a computationally efficient approximate approach. It is found to converge experimentally, but no proof of its convergence is given. The usefulness of filtered fdGn models and the performance of the proposed iterative algorithms are illustrated by fitting filtered fdGn models to speech waveforms and other data corresponding to natural phenomena     

多尺度H参数纹理特征的子带算法

扩展的自相似模型(ESS)是一种广义的分数布朗运动模型(fBm),它的多尺度H参数与粗糙度之间是对应的,因为不要求粗糙度的尺度不变性,所以能够区分大多数自然纹理.它的结构函数计算是基于图像在一定尺度上的灰度差,这可以用小波变换低频分量的一阶差分去定义.由于小波变换具有抑制噪声的能力,由此导出的特征具有更好的抗噪性能.实验证明对卫星遥感图像达到了较高的分类正确率.同时也说明,纹理的自相似特性在低频分量上的体现更突出.     

An adaptive ARMA four-line lattice filter for spectral estimationwith frequency weighting

A method for designing an adaptive four-line lattice filter which can perform frequency-weighting spectral estimation, which provides more accurate spectral estimation for some frequency bands than for others, is proposed. Using a suitable frequency-weighting function, denoted as an ARMA (autoregressive moving-average) model, an estimated spectrum is obtained by arbitrarily weighing some frequency bands more heavily than others. if the frequency-weighting function has the property of a low-pass filter, the spectrum of the reference model can be estimated accurately with a reduced ARMA order in the low-frequency band. Spectra of time-varying models can be estimated with an exponentially weighted sliding window, and the input signal of the reference model can be estimated by assumption. The order-update and the time-update recursive formulas and the frequency-weighting method for the filter are described. The algorithm is verified by experimental results     

Contradictory relationship between Hurst parameter and queueing performance (extended version)

Long Range Dependent (LRD) network traffic does not behave like the traffic generated by the Poisson model or other Markovian models. From the network performance point of view, the main difference is that LRD traffic increases queueing delays due to its burstiness over many time scales. LRD behavior has been observed in different types and sizes of networks, for different applications (e.g., WWW) and different traffic aggregations. Since LRD behaviour is not rare nor isolated, accurate characterization of LRD traffic is very important in order to predict performance and to allocate network resources. The Hurst parameter is commonly used to quantify the degree of LRD and the burstiness of the traffic. In this paper we investigate the validity and effectiveness of the Hurst parameter. To this end, we analyze the UCLA Computer Science Department network traffic traces and compute their Hurst parameters. Queueing simulation is used to study the impact of LRD and to determine if the Hurst parameter accurately describes such LRD. Our results show that the Hurst parameter is not by itself an accurate predictor of the queueing performance for a given LRD traffic trace.