基于卡尔曼滤波和小波的网络流量预测算法研究

流量预测是流量工程,拥塞控制和网络管理的核心问题。该文针对网络流量的特点,将卡尔曼滤波和小波分析混合的预测算法引入到网络流量预测领域中,对其进行了理论证明。仿真结果表明,该算法与传统的算法相比,具有较高的预测精度和较好的实时性与广谱性。     

基于SARIMA-LSTM组合模型的网络流量预测方法

针对网络流量预测问题，提出一种基于SARIMA和LSTM组合模型的网络流量预测方法。首先，利用S-HESD算法对异常流量数据进行检测，并通过滑动窗口均值进行数据平滑处理；然后，利用基于统计学习的SARIMA模型预测流量数据，并将其作为LSTM神经网络的输入，最终输出流量预测值。实验结果表明，SARIMA-LSTM组合模型能充分呈现骨干网、城域网、边缘接入网等不同层级网络的周期性和趋势性等特点，优于SARIMA、LSTM等单一模型。     

基于突发分离的自相似网络流量预测

网络流量预测在网络拥塞控制及资源分配中起着至关重要的作用。对于具有自相似性的网络业务流量,由于其存在较强突发,传统预测方法的预测精度普遍较低。本文针对存在高突发的网络流量数据,提出了一种基于数据分离的流量预测方法。在预测步骤前,本方法首先通过控制图将网络流量中难以预测的突发流量进行有效的分离,从而得到突发流量和非突发流量两部分数据。之后分别采用人工神经网络和自适应模板匹配方法实现对非突发流量和突发流量的预测。最后通过对两部分预测结果的合并得到最终的预测结果。基于实际流量数据的实验结果表明:相较于传统流量预测方法,本文所提出的方法具有更高的流量预测精度。      

组合神经网络的网络流量预测研究

提出一种组合神经网络的网络流量预测模型.首先采用SMOF网络对网络流量数据进行聚类,然后采用Elman网络对聚类后的流量数据进行训练并预测,同时采用遗传算法对Elman网络的网络结构进行优化,提高网络流量预测精度.仿真结果表明,组合神经网络加快了网络流量预测速度,提高了网络流量预测精度,克服了单一预测模型不足,为网络流量预测提供了新的思路,具有很好的应用前景.     

改进的基于时变FARIMA模型和小波变换的网络流量预测算法

网络的流量特性是反映网络实时状态的一个重要特征,对于网络流量的分析、预测一直是该领域的研究热点。传统的基于时间序列模型的方法在计算效率和多尺度分析能力方面存在一定的局限性。本文提出了一种改进的基于小波变换和时变FARIMA模型的流量预测方法,利用小波变换的多尺度分析特性将原有的流量数据进行分解,在使用时变FARIMA模型进行预测,可大大提高算法的执行效率和预测的准确性。最后,本文选取了Bellcore提供的真实的网络流量进行了仿真实验,验证了提出的预测方法的准确性和有效性。     

基于领域知识内嵌的深度学习网络流量预测研究

首先针对流量数据可能存在的时域分布突变问题设计了一种基于JS散度的异常检测算法，以获得能够反映流量变化规律的数据。而考虑到网络流量的自相似性、长时相关性和周期性特点，提出一种基于卷积生成网络和自适应注意力元网络组成的动态时域生成流量预测模型（GDTN），能够有效地降噪、生成和预测流量数据，并根据不同输入条件动态分配不同多时域流量特征的重要性。最后在智能城域网端口数据集上进行实验，验证了该算法相比于多种经典时序预测算法在预测准确度、存储资源占用和时间效用等方面具备明显优势。     

基于混沌理论与改进回声状态网络的网络流量多步预测

网络流量预测是网络管理及网络拥塞控制的重要问题,针对该问题提出一种基于混沌理论与改进回声状态网络的网络流量预测方法。首先利用0-1混沌测试法与最大Lyapunov指数法对不同时间尺度下的网络流量样本数据进行分析,确定网络流量在不同时间尺度下都具有混沌特性。将相空间重构技术引入网络流量预测,通过C-C方法确定延迟时间,G-P算法确定嵌入维数。对网络流量时间序列进行相空间重构之后,利用一种改进的回声状态网络进行网络流量的多步预测。提出一种改进的和声搜索优化算法对回声状态网络的相关参数进行优化以提高预测精度。利用网络流量的公共数据集以及实际数据进行了仿真,结果表明,提出的预测方法具有更高的预测精度以及更小的预测误差。     

基于机器学习的无线网络流量预测与增长潜力评估

提出一个基于机器学习的无线网络流量预测及流量增长潜力评估方案。该方案分析蜂窝网络中的实际业务流量数据在时间维度上的变化规律,并借助高斯过程的机器学习方法来预测业务变化趋势,从短期角度为运营商的网络优化部署提供指导。基于极限梯度提升(XGBoost)机器学习框架,建立网络中其他运营数据与业务流量的多维映射关系,应用改进的量子粒子群算法进一步寻找蜂窝小区所能承载的流量上限,从长期角度为网络优化部署提供指导,提升网络流量水平、释放流量增长潜力。     

IP网络流量流向分析与预测技术研究

概括了网络运营商进行IP网络流量流向分析与预测的主要方法,结合流量分析技术现状和IP网络规划的工程经验,提出一种基于因果分析的高效闭环预测模型,通过分析综合用户行为、流量分布、经济水平等影响因素,可有效评估中长期网络流量的增长趋势,为智能承载网络建设提供重要的设计工具.     

基于拓扑链路识别的光网络流量数据合成算法

基于深度学习的光网络流量诊断与预测等场景中,由于保密等原因,光链路的流量数据采集和存储工作受限。针对数据量少而无法支撑深度学习的问题,文章提出了一种基于拓扑链路识别的光网络流量数据合成算法,其核心思想是在生成对抗网络框架下,联合基于光网络拓扑的条件生成模型和基于光网络流量的数据合成模型,以自监督的方式合成指定光链路的流量数据。仿真结果表明,所提算法合成的光网络流量数据在自相关系数指标上与真实数据接近且使得基于全连接神经网络的流量预测模型准确率达到95%以上。     

Power Efficient Video Multipath Transmission over Wireless Multimedia Sensor Networks

This paper proposes a power efficient multipath video packet scheduling scheme for minimum video distortion transmission (optimised Video QoS) over wireless multimedia sensor networks. The transmission of video packets over multiple paths in a wireless sensor network improves the aggregate data rate of the network and minimizes the traffic load handled by each node. However, due to the lossy behavior of the wireless channel the aggregate transmission rate cannot always support the requested video source data rate. In such cases a packet scheduling algorithm is applied that can selectively drop combinations of video packets prior to transmission to adapt the source requirements to the channel capacity. The scheduling algorithm selects the less important video packets to drop using a recursive distortion prediction model. This model predicts accurately the resulting video distortion in case of isolated errors, burst of errors and errors separated by a lag. Two scheduling algorithms are proposed in this paper. The Baseline scheme is a simplified scheduler that can only decide upon which packet can be dropped prior to transmission based on the packet’s impact on the video distortion. This algorithm is compared against the Power aware packet scheduling that is an extension of the Baseline capable of estimating the power that will be consumed by each node in every available path depending on its traffic load, during the transmission. The proposed Power aware packet scheduling is able to identify the available paths connecting the video source to the receiver and schedule the packet transmission among the selected paths according to the perceived video QoS (Peak Signal to Noise Ratio—PSNR) and the energy efficiency of the participating wireless video sensor nodes, by dropping packets if necessary based on the distortion prediction model. The simulation results indicate that the proposed Power aware video packet scheduling can achieve energy efficiency in the wireless multimedia sensor network by minimizing the power dissipation across all nodes, while the perceived video quality is kept to very high levels even at extreme network conditions (many sensor nodes dropped due to power consumption and high background noise in the channel).     

Optimizing routing based on congestion control for wireless sensor networks

Along with the increasing demands for the applications running on the wireless sensor network (WSN), energy consumption and congestion become two main problems to be resolved urgently. However, in most scenes, these two problems aren’t considered simultaneously. To address this issue, in this paper a solution that sufficiently maintains energy efficiency and congestion control for energy-harvesting WSNs is presented. We first construct a queuing network model to detect the congestion degree of nodes. Then with the help of the principle of flow rate in hydraulics, an optimizing routing algorithm based on congestion control (CCOR) is proposed. The CCOR algorithm is designed by constructing two functions named link gradient and traffic radius based on node locations and service rate of packets. Finally, the route selection probabilities for each path are allocated according to the link flow rates. The simulation results show that the proposed solution significantly decreases the packet loss rate and maintains high energy efficiency under different traffic load.     

基于功率损耗均衡化控制机制的移动无线传感网数据传输算法

为了解决当前移动无线传感网数据传输中存在的同步寻址困难以及节点间功率交互难以均衡化的问题,提出了一种新的移动无线传感网数据传输算法。首先,采取广播机制实现同步控制分组传输,降低同步流量对寻址过程造成的压力;随后使用区域节点流量阀控制机制,且通过侦听方式记录并获取sink节点—区域节点链路间的数据流量,进一步采取流量—链路均衡方式促进流量均衡化;最后,通过基于轮数—sink 链路周期抖动筛选方式确认受限带宽,减少带宽受限导致的传输故障。仿真实验表明,与BLT-NB2R算法、NLSC算法和HT2C算法相比,所提出的算法可改善数据传输带宽,降低数据分组丢失频率,能够较好地满足实践需求。     

Network queue scheduling algorithm based on self-similar traffic level grading prediction

Self-similarity characteristic of network traffic will lead to the continuous burstness of data in the network.In order to effectively reduce the queue delay and packet loss rate caused by network traffic burst,improve the transmission capacity of different priority services,and guarantee the service quality requirements,a queue scheduling algorithm P-DWRR based on the self-similarity of network traffic was proposed.A dynamic weight allocation method and a service quantum update method based on the self-similar traffic level grading prediction results were designed,and the service order of the queue according was determined to the service priority and queue waiting time,so as to reduce the queuing delay and packet loss rate.The simulation results show that the P-DWRR algorithm can reduce the queueing delay,delay jitter and packet loss rate on the basis of satisfying the different service priority requirements of the network,and its performance is better than that of DWRR and VDWRR.     

基于多令牌桶流量整形算法的研究与设计

针对高速数据网络包交换系统中存在的拥塞问题,分析研究现有解决网络拥塞的流量整形技术,提出支持多组队列整形的多令牌桶算法,并通过VLSI设计方法实现了该算法．采用SMIC0．13μm标准CMOS工艺库综合,工作频率达到185MHz,行为级仿真及综合结果表明,该算法硬件实现简单,能有效的对包交换网络进行流量整形．     

Radio resource management for cellular CDMA systems supporting heterogeneous services

A novel radio resource management (RRM) scheme for the support of packet-switched transmission in cellular CDMA systems is proposed by jointly considering the physical, link, and network layer characteristics. The proposed resource management scheme is comprised of a combination of power distribution, rate allocation, service scheduling, and connection admission control. Power distribution allows individual connections to achieve their required signal-to-interference-plus-noise ratio, while rate allocation guarantees the required delay/jitter for real-time traffic and the minimum transmission rate requirement for non-real-time traffic. Efficient rate allocation is achieved by making use of the randomness and burstiness; of the packet generation process. At the link layer, a packet scheduling scheme is developed based on information derived from power distribution and rate allocation to achieve quality of service (QoS) guarantee. Packet scheduling efficiently utilizes the system resources in every time slot and improves the packet throughput for non-real-time traffic. At the network layer, a connection admission control (CAC) scheme based on the lower layer resource allocation information is proposed. The CAC scheme makes use of user mobility information to reduce handoff connection dropping probability (HCDP). Theoretical analysis of the grade of service performance, in terms of new connection blocking probability, HCDP, and resource utilization, is given. Numerical results show that the proposed RRM scheme can achieve both effective QoS guarantee and efficient resource utilization.     

面向带宽碎片最小化和QoS保障的数据中心网络流量调度算法

随着数据中心网络流量的迅速增长,如何提高数据中心网络性能和服务质量成为了研究热点。然而现有的流量调度算法在网络负载加大时,一方面会导致网络带宽碎片化从而使得网络吞吐量降低,另一方面忽视了流量应用需求导致网络服务质量较差。为此,该文提出一种面向带宽碎片最小化和QoS保障的动态流量调度算法,算法综合考虑了带宽敏感的大流、时延与丢包敏感的小流的不同需求,首先根据待调度流的源地址和目的地址建立最短路径集,其次从中筛选出满足待调度流的带宽需求的所有路径,然后根据路径剩余带宽信息和小流应用需求情况为每条路径建立权重函数,最后根据权重函数值利用轮盘赌算法选择转发路径。实验仿真结果显示,与其它算法相比,所提算法降低了小流的丢包率和时延,同时在网络负载较大时提升了网络吞吐量。     

A mode transformation algorithm based on traffic prediction in virtual multi‐OLT PON

In this paper, a mode transformation algorithm based on traffic prediction in virtual multiple optical line terminal (OLT) passive optical network (PON) is proposed. By proposing exponential smoothing algorithm based on weight update (WU‐ESA), user traffic is predicted well. WU‐ESA is a combination of two algorithms: exponential smoothing algorithm (ESA) and genetic algorithm (GA). The weight in ESA is optimized by GA based on real‐number encoding. By setting two periods, GA part and ESA part can be separated effectively. By presenting elastic packing algorithm (EPA), the load balance problem in virtual multi‐OLT PON is solved. EPA is implemented based on WU‐ESA. By the simulation and analysis, the effectiveness of the proposed algorithms is demonstrated. Compared with traditional mode transformation algorithm, EPA shows good performances in delay, throughput, and packet loss. Compared with traditional mode transformation algorithm, the EPA makes the packet loss decrease by at least 5% when the system load is greater than 0.9. Meanwhile, the delay of the two subsystems can be kept at a relatively balanced level by the implement of EPA. For throughput, the use of EPA improves the throughput by 20% when the system load is high.     

面向城市交通网络的一种新型动态路径寻优方法

 针对静态路径寻优方法中没有考虑到交通流随时间变化的问题,本文提出一种基于交通流量预测的路径寻优方法.首先,从实际交通路网本身的特点和人类对路网的认识出发,构建以"道路"为基本元素的新型路网模型;其次,采用基于单变量时间序列的预测方法对路网上各路段未来一段时间内的道路交通流量进行预测,并根据预测结果估计出行者在未来时段各路段上的交通代价;再次,通过估计的交通代价把车辆从起点到该路段时间内的道路交通变化状况融入改进A^*算法的路径搜索过程,从而建立一种高效的动态改进A^*路径搜索算法;最后,实例表明预测交通流量与实际交通流量具有很好的拟合度,路径搜索结果能够对出行者起到诱导作用,提高出行者行驶效率.     

Addressing self-similarity in optical switching networks by means of ant colony optimization

In this work, we propose the use of an Ant Colony Optimization (ACO) algorithm to mitigate packet loss in an optical packet switching network that carries self-similar traffic, which is known to have a great impact in the buffer performance in terms of loss probability as exemplified in this work. By adaptively routing the packets and balancing the network load, we demonstrate by some simulations the effectiveness of this approach when compared with a shortest-path routing scheme, achieving a performance that is comparable to the Poisson traffic scenario in some cases. The proposed algorithm can be used as a viable alternative to traffic shaping techniques.