基于Parzen窗的高阶统计量特征降维方法

高阶统计量通常能比低阶统计量提取更多原数据的信息,但是较高的阶数带来了较高的时间复杂度.基于Parzen窗估计构造了高阶统计量,通过论证得出:对于所提出的核协方差成分分析（KCCA）方法,通过调节二阶统计量广义D vs E的参数就能够达到整合高阶统计量的目的,而无需计算更高阶统计量.即核协方差成分分析方法能够对高阶统计量的特征降维的同时,又不增加计算复杂性.     

轴承数据的Matlab高阶谱分析

使用matlab高阶谱工具箱(HOSA)函数对轴承数据进行分析,介绍了若干在仿真中用到的HOSA函数及其数学基础.并验证,由于轴承振动信号的非高斯,非线性特性,低阶统计量不能分辨的振动特征,能够通过高阶统计量及其matlab工具进行分析.     

基于MCC的鲁棒高阶CKF在组合导航中的应用

针对高阶容积卡尔曼滤波器在非高斯噪声情况下滤波精度下降的问题,提出了一种新的基于Maximum Correntropy Criterion（MCC）的鲁棒高阶容积卡尔曼滤波算法。考虑到高阶容积规则可以较好地解决非线性问题,在高阶容积滤波的基础上,结合统计线性回归模型对量测更新过程进行重构,利用MCC估计算法实现状态的量测更新,同时解决了系统的非线性和非高斯问题。将所提算法应用到SINS/GPS组合导航系统中,仿真结果表明,核宽的选取对算法的滤波性能有较大的影响,在高斯混合噪声条件下,所提算法相比传统高阶容积卡尔曼滤波算法具有更强的鲁棒性和更高的滤波精度。     

高阶统计量方法及应用研究

高阶统计量方法是近几年国内外信号处理领域内的一个前沿课题，它包含了二阶统计量没有的大量丰富信息，广泛应用于所有需要考虑非高斯性、非最小相位、有色噪声、非线性或循环平稳性的各类问题中。凡是使用功率谱或相关函数进行分析与处理，而又未得到满意结果的任何问题都值得重新使用高阶统计量方法。高阶统计量的发展与应用是信号处理领域近年来一个十分重要的发展，是现代信号处理的核心内容之     

非高斯系统的建模方法研究*

近年来，高阶统计信息理论广泛应用于非高斯系统的建模问题中，由于高阶统计信息方法特殊的信息结构，难以充分利用具体问题中的统计分布信息，从而对该方法的应用效果产生了影响。针对这一问题，本文提出一种综合应用刘阶统计信息理论与非高斯系统的似然函数指标的建模的方法，为在非高斯系统建模系统建模问题中充分利用问题所提供的统计信息提供了新的途径，仿真研究和应用取得了令人满意的效果。     

基于高阶累积量的盲反卷积算法的比较

通过对高阶统计理论以及对建立在高阶统计理论基础上的累积量盲反卷积算法的分析与研究,研究了两种基于高阶累积量的盲反卷积算法的原理,并且通过这两种算法对信号的处理结果,比较了这两种算法的性能.     

ICA系数的高阶统计特征在纹理分类中的应用

根据独立成分分析系数满足非高斯分布的特点,研究高阶统计量在特征提取中的作用.提出利用方差、偏度、峭度的联合矩描述系数的分布特征,应用于纹理分类时取得较好效果.在矩估计时进一步提出利用L-矩代替常规矩进行估计,获得更好的纹理分类效果.     

基于高阶统计量和图像分割技术的图像盲复原

提出一种基于高阶统计量和图像分割技术的改进的非负支持域递归逆滤波算法(NAS-RIF)。该算法应用高阶统计量去噪,解决了原NAS-RIF算法敏感于噪声的特性,提高了退化图像的信噪比。其次,在每次迭代中利用图像分割技术进行图像支持域的自估计,改变了NAS-RIF算法中支持域必须是方形的不利限制。实验结果表明改进的NAS-RIF算法具有更好的噪声抑制和边缘细节恢复效果。     

信号分析与处理的几种新方法

论文介绍几种受到极大关注并取得迅速发展的信号分析与处理的新方法,即基于子波变换的方法、基于高阶统计量的方法和基于人工神经网络的方法。理论分析与实验结果表明它们适合于时变、非平稳、非高斯信号的分析与处理。     

高阶非完整系统的广义Birkhoff表示

通常方法构造的高阶非完整系统的运动微分方程不仅没有完整系统的辛几何结构和Lie代数结构,而且也不具备完整系统的自伴随性质.本文利用降阶方法,将高阶非完整系统变换为一阶动力学系统,并运用Cauchy-Kowalevski定理对其自伴随化,得到一种新的一阶动力学方程组-广义Birkhoff表示,这将为研究高阶非完整系统的若干动力学问题、几何结构、代数结构、几何数值积分以及工程应用提供了一个新的方法.     

Detecting abnormal situations using the Kullback–Leibler divergence

This article develops statistics based on the Kullback–Leibler (KL) divergence to monitor large-scale technical systems. These statistics detect anomalous system behavior by comparing estimated density functions for the current process behavior with reference density functions. For Gaussian distributed process variables, the paper proves that the difference in density functions, measured by the KL divergence, is a more sensitive measure than existing work involving multivariate statistics. To cater for a wide range of potential application areas, the paper develops monitoring concepts for linear static systems, that can produce Gaussian as well as non-Gaussian distributed process variables. Using recorded data from a glass melter, the article demonstrates the increased sensitivity of the KL-based statistics by comparing them to competitive ones.     

Socio-technical aspects of display units

The tree figure illustrates the concepts of this book. It addresses systems with displays of extremely social existence as compared with visual display units in the past. Given this recognition, the first chapter gives a socio-technical commentary on such systems using worldwide statistics. According to world statistics, some characteristic of the increasing use of display systems can be summarized in four items: appearance of liquid crystal displays, popularization of personal computer, diversification of display usage such as cell phones, and ergonomic consideration for elderly users. An example of the ethnic differences is illustrated.     

基于单链表和散列表比较的词频统计研究

针对操作系统课程中的词频统计实验,介绍两种分别采用单链表和散列表进行词频统计的方法,并着重说明后一种方法在效率方面的优越性。     

Statistical structures for Internet-scale data management

Efficient query processing in traditional database management systems relies on statistics on base data. For centralized systems, there is a rich body of research results on such statistics, from simple aggregates to more elaborate synopses such as sketches and histograms. For Internet-scale distributed systems, on the other hand, statistics management still poses major challenges. With the work in this paper we aim to endow peer-to-peer data management over structured overlays with the power associated with such statistical information, with emphasis on meeting the scalability challenge. To this end, we first contribute efficient, accurate, and decentralized algorithms that can compute key aggregates such as Count, CountDistinct, Sum, and Average. We show how to construct several types of histograms, such as simple Equi-Width, Average-Shifted Equi-Width, and Equi-Depth histograms. We present a full-fledged open-source implementation of these tools for distributed statistical synopses, and report on a comprehensive experimental performance evaluation, evaluating our contributions in terms of efficiency, accuracy, and scalability.     

基于卡方统计量的属性约简新方法

近年来粗糙集数据分析已经成为定性数据分析的一个常用方法,而信息系统的属性约简是粗糙集理论的一个基本问题.粗糙集方法虽然不需要数据之外的其它信息,但所得结果同时也缺乏统计证据.文中运用非参数统计的思想研究了信息系统的属性约简问题,首先将原始信息系统整理成任意两个属性子集之间的列联表,然后给出了基于卡方统计量的属性相关性的一个新度量,基于此度量给出了信息系统属性约简的新方法.数值例子说明了该方法的可行性和有效性.     

非线性系统的自适应推广的kalman滤波

本文提出了未知噪声统计的非线性系统中新的自适应推广的Kalman滤波算法.作者提 出了用虚拟时变噪声统计[1,2],补偿线性化模型误差的新思想. 在本文中,作者指出了文献[3]中,用Sage和Husa的常值噪声统计估值器来估计虚拟 噪声是不合理的.另外,即使原非线性系统的噪声统计是零均值,但线性化的模型的噪声统 计一般是非零均值的.两个数值模拟例子说明了本文方法的有效性.     

An adaptive strategy for statistics collecting in distributed database

Collecting statistics is a time- and resourceconsuming operation in database systems. It is even more challenging to efficiently collect statistics without affecting system performance, meanwhile keeping correctness in distributed database. Traditional strategies usually consider one dimension during collecting statistics, which is lack of adaptiveness. In this paper, we propose an adaptive strategy for statistics collecting(ASC), which well balances collecting efficiency, correctness of statistics and effect to system performance. We formally define the procedure of collecting statistics and abstract the relationships among collecting efficiency, correctness of statistics and effect to system performance, and introduce an elastic structure(ESI) storing necessary information generated during proceeding our strategy. ASC can pick appropriate time to trigger collecting action and filter unnecessary tasks, meanwhile reasonably allocating collecting tasks to appropriate executing locations with right executing models through the information stored at ESI. We implement and evaluate our strategy in a distributed database. Experiments show that our solutions generally improve the efficiency and correctness of collecting statistics, moreover, reduce the negative effect to system performance comparing with other strategies.     

Approximating feeder line reliability statistics with partial data collection in assembly systems

Due to the difficulty and high cost of data collection, many feeder lines in assembly systems lack full collection of data. However, reliability statistics of feeder lines are important in throughput analysis and continuous improvement of manufacturing systems. In this paper, a simple approximation approach is presented to estimate the reliability statistics of feeder lines from the associated assembly station's collected blocking and starving information. It is shown that the approach is helpful for accurate throughput estimation and sensitivity analysis. We also show how feeder line speed can be used to improve the approximation.     

Information sharing for distributed intrusion detection systems

In this paper, we present an information sharing model for distributed intrusion detection systems. The typical challenges faced by distributed intrusion detection systems is what information to share and how to share information. We address these problems by using the Cumulative Sum algorithm to collect statistics at each local system, and use a machine learning approach to coordinate the information sharing among the distributed detection systems. Our major contributions are two-fold. First, we propose a simple but robust scheme to monitor changes in the local statistics. Second, we present a learning algorithm to decide when to share information so that both the communication overhead among the distributed detection systems and the detection delay are minimized. We demonstrate the application of our information sharing model to a specific distributed intrusion detection scenario. We show that our approach is able to optimize the trade-off between the time required to detect an attack, and the volume of communication between the distributed intrusion detection systems.     

On the Kullback Leibler information for mixed systems

Entropy and Kullback–Leibler (KL) information for engineering systems have been studied in both statistical and reliability contexts. In this paper, we prove that the KL information between distributions of mixed system lifetimes and the corresponding component lifetimes and also the associated order statistics are distribution free and depends only on the signature of the system provided that lifetimes of components are independent and identically distributed (iid). The obtained results are used to find the closest and the farthest distribution of order statistics from the distribution of the system’s lifetime which is useful to approximate stochastic behaviour of mixed systems when the number of components is large. Finally, we provide bounds and also use the results to obtain a more preferable system among all systems. Some illustrative examples are also given.