自相似网络论坛的模型研究

为描述网络论坛文章数序列的自相似性,使用基于分数差分和ARMA过程的FARIMA模型,对论坛的文章数序列进行建模和分析。FARIMA模型具有描述精度高的特点,但时间复杂度较高。为此,提出?-ARMA模型。通过实验进行仿真分析,从模型的计算复杂度和拟合准确度方面进行对比,证明了?-ARMA模型的有效性。     

网络流量预测模型的研究与分析

网络流量模型以考察网络流量特性为出发点,以数学理论为基础,通过建立数学模型来反映真实的网络流量及其变化趋势。传统的泊松模型在现代数据网络中已经不再适用,不能真实地反映流量的趋势。但是自从网络流量的自相似性被发现后,网络流量的自相似模型不断涌现。应用了既能反映长相关性又能反映短相关性的FARIMA模型对真实网络流量数据进行了分析预测,经过研究和实践的验证,对模型进行了改进,提出了SFARIMA网络流量预测模型。     

FARIMA with stable innovations model of Great Salt Lake elevation time series

Great Salt Lake (GSL) is the largest salt lake in the western hemisphere, the fourth-largest terminal lake in the world. The elevation of GSL has critical effect on the people who live nearby and their properties. It is crucial to build an exact model of GSL elevation time series in order to predict the GSL elevation precisely. Although some models, such as ARIMA or FARIMA (fractional auto-regressive integrated moving average), GARCH (generalized auto-regressive conditional heteroskedasticity) and FIGARCH (fractional integral generalized auto-regressive conditional heteroskedasticity) have been proposed to characterize the variation of GSL elevation, which have been unsatisfactory. Therefore, it became a key point to build a more appropriate model of GSL elevation time series. In this paper a new model based on FARIMA with stable innovations is applied to analyze the data and predict the future elevation levels. From the analysis we can see that the new model can characterize GSL elevation time series more accurately. The new model will be beneficial to predict GSL elevation more precisely.     

基于FARIMA模型的日流量过程随机模拟研究

本文采用能反映时间序列长相关性的自回归分数整合滑动平均（FARIMA）模型,建立了日流量过程的随机模拟方法。首先提出概率权重矩-正态分位数变换（PWM-NQT）法进行日流量预处理,将预处理后得到的平稳、正态时间序列用于FARIMA模型的辨识和参数估计。基于FARIMA模型得到正态化日流量模拟值,并由正态分位数逆变换及逆标准化分别得到标准化日流量模拟值及日流量过程模拟值。将建立的模型应用于长江流域宜昌水文站日流量过程随机模拟,统计检验表明,建立的随机模拟方法能合理考虑日流量过程的季节性周期和长相关性,同时也能很好地保持日流量过程的其它统计特性。     

基于FARIMA模型的智能变电站通信流量异常分析

随着输变电设备自动化、变电站智能化建设的快速发展,电网信息安全隐患日益凸显。精确可靠的变电站通信网络流量模型建模和异常检测方法已成为预防网络安全问题和识别网络攻击的重要手段。文中在对变电站站控层网络流量行为特性进行分析的基础上,采用分形自回归积分滑动平均(FARIMA)模型对网络流量构建了阈值模型。针对变电站典型的网络攻击模式和流量异常特征,基于运行状态评估算法对某实际变电站站控层流量数据进行分析,并计算典型网络异常概率,从而实现了变电站在网络攻击情形下的安全态势评价。     

A Class of Antipersistent Processes

Abstract. We introduce a class of stationary processes characterized by the behaviour of their infinite moving average parameters. We establish the asymptotic behaviour of the covariance function and the behaviour around zero of the spectral density of these processes, showing their antipersistent character. Then, we discuss the existence of an infinite autoregressive representation for this family of processes, and we present some consequences for fractional autoregressive moving average models.     

互联网流量与两个相关模型

由于互联网具有多构性、突发连续性和自相似性等特征,使得用传统模型进行的排队分析、性能估计与实际网络有较大差距。本文介绍了两种互联网流量模型:AR IMA模型,FAR IMA模型,讨论了它们的适用范围、模型数学定义,描述了参数定阶推导等相关问题。     

Empirical likelihood confidence intervals for the mean of a long‐range dependent process

Abstract. This paper considers blockwise empirical likelihood for real‐valued linear time processes which may exhibit either short‐ or long‐range dependence. Empirical likelihood approaches intended for weakly dependent time series can fail in the presence of strong dependence. However, a modified blockwise method is proposed for confidence interval estimation of the process mean, which is valid for various dependence structures including long‐range dependence. The finite‐sample performance of the method is evaluated through a simulation study and compared with other confidence interval procedures involving subsampling or normal approximations.     

自相似网络业务的一个FARIMA模型

近来发现,高速网络业务具有自相似及长相关特性,分数噪声可描述该类业务。但它仅表现长相关特性,给出了利用ＦＡＲＩＭＡ模型拟合自似网络业务的一整套方法,该模型同时刻画了实际业务的长相关与短相关行性,通过对实测数据的实验, 模型的优效性。     

基于FARIMA过程的网络业务预报与应用

在高速网络控制与带宽分配研究中,网络业务量预报是一个重要的问题。本文首先介绍自回归分数整合滑动平均（FARIMA,Farctal Autoregressive Integrated Moving Average)过程的概念及其具体形式,并给出了FARIMA过程的预报方法,然后基于FARIMA过程的最小均方误差预报方法,提出了一个具有补偿功能的网络自相似业务的预报方法,最后给出这种预报方式在网络控制研究中的应用。