基于小波支持向量机的VBR视频流量预测模型

针对VBR视频流量的时变性、突发性和非线性等特点,提出一种基于小波支持向量机的VBR视频流量预测模型(WSVM)。首先对VBR视频流量时间序列进行相空间重构,然后将其输入到小波支持向量机进行学习,建立VBR视频流量预测模型,最后采用仿真实验对模型性能进行测试。结果表明,相对于对比模型,WSVM提高了VBR视频流量预测精度,更加准确地描述了VBR视频流量的复杂变化特点。     

VBR视频流量的小波包分解及其长时预测

长时预测是VBR视频流量预测领域中的难点问题.针对其时变、非线性以及长相关性等特点,提出一种多尺度分解的VBR视频业务的特征提取方法.选择具有任意多分辨分解特性的小波包,对其进行空间划分并求解适合视频信号特征提取的最优分解基.基于最优基对视频信号进行快速多尺度分解,得到了各级节点的小波系数矩阵,建立了基于最小二乘支持向量机与最小均方的小波系数预测方法.最后,根据预测小波系数,进一步提出了基于小波系数逆变换的视频流量长时预测方法.仿真结果验证了此算法的有效性.     

基于小波分解和支持向量机的网络流量组合预测

研究网络流量预测问题,网络流量具有突发性、周期性、非线性特点,传统网络流量预测模型无法建立准确预测模型,导致预测误差大,预测精度低.为了提高网络流量的预测精度,提出一种小波分解和支持向量机的网络流量预测模型.首先采用小波变换对网络流量进行分解,把网络流量不同特性成分分离出来,然后采用支持向量机对各分量进行预测,最后采用小波变换对各分量预测结果进行重构,得到网络流量的最终预测结果.仿真实验结果表明,相对其它预测模型,提高了网络流量的预测精度,为网络流量预测优化提供了可靠依据.     

小波消噪和蚁群优化支持向量机的网络流量预测模型

针对当前网络流量预测模型精度低的缺点,本文提出了一种新型的小波消噪和蚁群算法优化支持向量机的网络流量预测模型。首先采用小波阈值法对网络流量进行消噪处理;然后将网络流量输入到支持向量机中学习,并采用蚁群算法对支持向量机的参数进行优化,建立网络流量预测模型,最后采用实际网络流量数据进行仿真实验,结果表明,相对于其它网络流量预测模型,本文模型提高了网络流量的预测精度,具有更好的鲁棒性。     

智能集成VBR MPEG视频流量预测模型

作为数字媒体网络视频通信的主要方式,VBR MPEG视频流量的预测能力是直接关系缓冲区设计、动态带宽分配及拥塞控制等提高网络服务质量的关键因素.因此针对MPEG视频流的复杂特性,充分利用人工智能方法的优势,提出并建立了基于模糊神经网络的智能集成VBR MPEG 视频流量预测模型.采用模糊预测模型提高预测精度,利用神经网络解决预测的实时性问题.实验结果表明,与标准AR预测模型相比,该模型预测的准确度和可靠性显著提高,且算法简单易于推广到其他方法中使用.     

用支持向量机网络实现VBR视频通信量的预测

VBR(Variable Bit Rate)视频信号具有时变性、非线性和突发性等特点,实现该信号通信量的高精度预测是提高信息传输速度和提高网络带宽资源利用效率的重要手段.针对以上问题,本文提出了一种用于VBR视频通信量预测的差分输入支持向量机(SVM:Support Vector Machine)网络模型.该网络模型采用结构风险最小化准则,在最小化经验风险的同时,尽量缩小模型预测误差的上界,从而使网络模型具有更好的推广能力.实验结果表明:支持向量机网络模型的预测误差为0.0018,而梯度径向基函数(Gradient Radial Basis Function:GRBF)神经网络模型的预测误差为0.0029.可以看出,支持向量机网络模型的预测精度要比GRBF网络模型的预测精度高出大约40%.     

一种极端降水预测方法研究

为了提高极端降水量的预测精度,将小波分析、支持向量机以及遗传算法相结合,建立了一种极端降水预测模型。首先利用小波变换对极端降水数据进行分解,分离出序列中的低频信息和高频信息;然后对各子序列分别用遗传算法优化的支持向量机进行训练和预测;最后将各子序列的预测结果叠加,得到极端降水量的最终预测结果。实验表明,该组合模型能准确揭示极端降水的变化特性,具有更高的预测精度,从而为极端降水量的预测提供了一种有效方法。     

最小二乘支持向量机和预测误差校正的运动员成绩预测

为了获得更好的运动员成绩预测结果,提出最小二乘支持向量机和预测误差校正的运动员成绩预测模型。首先通过提升小波和最小二乘支持向量机对运动员成绩进行建模和预测,然后通过误差校正方式对运动员成绩的预测结果进行校正,最后通过运动员成绩预测实例对模型的有效性进行测试,并与其他运动员预测模型进行对比实验,验证其优越性,结果表明,所提模型降低了运动员成绩的预测误差,并且通过误差校正提高了运动员成绩预测结果的稳定性,预测精度要优于其他运动员成绩预测模型。     

野草算法和支持向量机相融合的短期负荷预测

短期负荷受到天气、季节的综合影响,具有一定的混沌性,为了准确描述短期负荷的变化趋势,以提高预测精度,提出一种野草算法和支持向量机相融合的短期负荷预测模型(WA-SVM)。首先收集大量的短期负荷历史数据,并对它们进行混沌分析和处理,建立支持向量机的训练和测试样本集;然后采用支持向量机建立短期负荷预测模型,并通过野草算法找到支持向量机最优参数;最后采用短期负荷预测仿真对比实验测试其性能。结果表明,WA-SVM获得了比其他模型更高的短期负荷预测精度,为短期负荷建模与预测提供了一种新的研究方法。     

基于小波分解和SVM的城市大气污染浓度预测

使用一种结合小波分解和支持向量机的方法建立城市大气污染物浓度预测模型。首先将大气污染物浓度数据序列小波分解和重构为不同频段的分解序列,然后对各分解序列使用不同的模型进行预测,最后将各分解序列的预测结果合成为浓度的最终预测结果。实验结果表明,该模型的预测精度优于RBF神经网络模型及一般支持向量机模型。     

DDoS detection based on wavelet kernel support vector machine

To enhance the detection accuracy and deduce false positive rate of distributed denial of service （DDoS） attack detection, a new machine learning method was proposed. With the analysis of support vector machine （SVM） and the wavelet kernel function theory, an admissive support vector kernel, which is a wavelet kernel constructed in this article, implements the combination of the wavelet technique with SVM. Then, wavelet support vector machine （WSVM） is applied to DDoS attack detections and as a classifying means to test the validity of the wavelet kernel function. Simulation experiments show that under the same conditions, the predictive ability of WSVM is improved and the computation burden is alleviated. The detection accuracy of WSVM is higher than the traditional SVM by about 4%, while its false positive is lower than the traditional SVM. Thus, for DDoS detections, WSVM shows better detection performance and is more adaptive to the changing network environment.     

基于小波支撑矢量机的陀螺仪漂移预测

提出了基于小波支撑矢量机的导弹惯性器件故障预报方法。将子波核函数应用于支撑矢量机，得到小波支撑矢量机（WSVM）方法，用于某陀螺仪漂移时间序列预测中，预测精度优于基于传统核函数的支撑矢量机。通过仿真实验证明了算法的有效性和可行性。     

Real time variable bit rate video traffic prediction

Real time variable bit rate (VBR) video traffic prediction plays an important role in dynamic bandwidth allocation schemes by providing an accurate estimation of the instantaneous bandwidth requirement of VBR video traffic and it has been widely used in dynamic bandwidth allocation. A number of prediction algorithms were proposed in the literature and they can be broadly classified into two categories: time domain approaches and wavelet domain approaches. In this paper, we first present a survey of the existing algorithms in the literature. On the basis of the survey, new algorithms are proposed in the time domain and in the wavelet domain, respectively. Simulations using real VBR video traces are conducted which show that the proposed algorithms achieve better performance than those in the literature. Copyright © 2006 John Wiley & Sons, Ltd.     

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.     

多因素影响下的忙时话务量预测方法的研究

为了提高节假日移动话务量的预测精度,提出了一种基于多因素影响的忙时话务量预测方法。首先对忙时话务量进行相关性分析,得到4个影响话务量的重要因子,然后把4个因子与忙时话务量训练数据一起作为输入变量,最后用改进的具有非线性惯性权重的粒子群算法优化的最小二乘支持向量机模型进行预测。实验结果表明该预测模型有更高的预测精度和较强的泛化能力。     

Resource allocation using adaptive linear prediction in WDM/TDM EPONs

This letter discusses dynamic-wavelength and bandwidth-allocation (DWBA) algorithm for differentiated services in hybrid WDM/TDM EPONs (Ethernet passive optical networks). We integrate the traffic estimation for variable-bit-rate (VBR) video traffic with the resource allocation to satisfy the increasing demand of video streaming services. A new DWBA algorithm using an adaptive linear prediction model is proposed to estimate the bandwidth required by VBR video traffic arriving during the next waiting period, so that the DWBA algorithm can assign wavelength and bandwidth more effectively. Simulation results show that the proposed DWBA-ALP algorithm can significantly improve the delay jitter performance for both CBR and VBR services.     

一种基于瑞利分布的VBR视频流的小波模型

本文提出了一种新型的视频业务流模型,以Haar小波的多分辨率分析为基础,在尺度空间和小波空间分别建模,然后通过小波反变换得出仿真业务流.在最"粗"的尺度空间里,我们根据视频流的概率分布特点,采用基于瑞利(Rayleigh)分布的AR模型对尺度系数建模;在各个小波空间里,采用一般的高斯不相关小波模型(WIG,Wavelet Independent Guassian)建模.由于在尺度空间和小波空间针对各自的特点作了不同的处理,本文模型不但能较好拟合复杂业务流在各个时间尺度的概率分布特性,也能拟合其长时相关的特性.另外,在多尺度排队分析(MSQ,MultiScale Queue)的框架下,我们还推导出了基于本文模型的排队分析的理论结果.最后,通过对实际视频业务流数据仿真实验与排队分析验证了本文模型的有效性.     

Efficient traffic prediction scheme for real-time VBR MPEG videotransmission over high-speed networks

The efficient transportation of real-time variable bit rate (VBR) video traffic in high-speed networks is currently an active area of research. The capability to predict VBR video traffic can significantly improve the effectiveness of numerous management tasks, including dynamic bandwidth allocation and congestion control. This paper proposes an adaptive traffic prediction method for VBR MPEG videos, a major multimedia application. Rapid traffic variations due to scene changes are analyzed, then a prediction scheme using the identification of scene changes related to I and P frames is presented. For predicting multiplexed MPEG traffic, a prediction interval is derived that represents a highly correlated traffic sequence. In addition, to reduce the prediction error, a less fluctuating signal instead of the original multiplexed traffic is used as the input for the predictor. Simulation results show that the proposed method is able to predict the original traffic more accurately than the conventional LMS method     

遗传算法优化支持向量机的交通流量预测

交通流量预测是智能交通系统中的关键技术,针对当前交通流量预测模型存在不足,提出一种遗传算法优化支持向量机的交通流量预测模型。首先收集交通流量历史数据,并基于混沌理想对其进行相空间重构,然后将训练样本输入到支持向量机中进行学习,并采用遗传算法优化支持向量机参数,建立交通流量预测模型,最后采用测试样本对模型的性能进行测试。结果表明,相对于经典交通流量预测模型,本文模型可以更加准确描述交通流量预测复杂的变化趋势,提高了交通流量的单步和多步预测精度。