面向5G/B5G的智能云化网络架构

5G/B5G技术的出现使移动网络的带宽、时延、连接等关键能力得到了极大的提升。面对移动接入侧能力的飞速发展,现有的集中式互联网云服务架构容易出现瓶颈,无法充分发挥 5G/B5G 的性能优势。网络5.0技术的提出,旨在通过一套完整的创新协议体系及控制管理架构,应对未来典型网络应用对数据网络的需求。基于网络5.0协议体系,提出一种新型的智能云化网络,通过云网深度融合,弥补互联网服务模型与5G/B5G接入能力之间的差距,推动网络持续演进发展。     

深度神经网络在风电功率预测中的应用研究

风电功率预测对于电网建设具有重要意义。文中提出一种基于深度神经网络的风电功率预测方法,该方法充分考虑了影响风电功率的若干因素,如风速、风向、空气密度和季节,通过深度神经网络训练、学习给出最佳的功率预测值。通过深度神经网络学习的特征能够更为有效地反映实际情况,因此提高了风电功率预测的稳健性。基于内蒙古某风电厂的实测数据进行了验证实验,结果表明了提出方法的有效性。     

Monitoring the Application-Layer DDoS Attacks for Popular Websites

Distributed denial of service (DDoS) attack is a continuous critical threat to the Internet. Derived from the low layers, new application-layer-based DDoS attacks utilizing legitimate HTTP requests to overwhelm victim resources are more undetectable. The case may be more serious when such attacks mimic or occur during the flash crowd event of a popular Website. Focusing on the detection for such new DDoS attacks, a scheme based on document popularity is introduced. An Access Matrix is defined to capture the spatial-temporal patterns of a normal flash crowd. Principal component analysis and independent component analysis are applied to abstract the multidimensional Access Matrix. A novel anomaly detector based on hidden semi-Markov model is proposed to describe the dynamics of Access Matrix and to detect the attacks. The entropy of document popularity fitting to the model is used to detect the potential application-layer DDoS attacks. Numerical results based on real Web traffic data are presented to demonstrate the effectiveness of the proposed method.      

基于电信大数据的5G网络海量用户复访行为预测模型

5G网络中的用户会产生大量的访问数据，导致用户复访行为难以精准预测，因此提出基于电信大数据的5G网络海量用户复访行为预测模型。从电信大数据中提取用户上网历史行为特征数据，构建数据集。引入多阶加权马尔可夫链模型，通过计算各阶自相关系数，得到模型权重值，计算模型的统计量。经过分析后得到各阶步长的马尔可夫氏链一步转移概率矩阵，从而实现对5G网络海量用户复访行为的精准预测。实验结果表明，该模型拥有最低的均值误差和标准差，以及最高的精度、查全率、查准率、F1指标，可证明该方法在预测用户复访行为方面有着非常明显的优势。     

An improved deep multiscale crowd counting network with perspective awareness

Crowd counting is a challenging task, which is partly due to the multiscale variation and perspective distortion of crowd images. To solve these problems, an improved deep multiscale crowd counting network with perspective awareness was proposed. This network contains two branches. One branch uses the improved ResNet50 network to extract multiscale features, and the other extracts perspective information using a perspective-aware network formed by fully convolutional networks. The proposed network structure improves the counting accuracy when the crowd scale changes, and reduce the influence of perspective distortion. To accommodate various crowd scenarios, data-driven approaches are used to fine-tune the trained convolutional neural networks (CNN) model of the target scenes. The extensive experiments on three public datasets demonstrate the validity and reliability of the proposed method.     

基于改进型深度学习的流量预测

为了解决无线网络中流量的预测精度不高的问题,提出了一种自适应分组的栈式自编码( AG-SAEs)深度学习预测方法。在数据的预处理过程中,首先使用最大最小方式对数据进行归一化处理,并提出一种新型的自适应分组方法,把归一化后的链路数据进行关联性分组;然后,基于深度学习方法建立了一个多输入多输出的预测模型,并将分组后的数据输入到预测模型中,对该模型进行训练来建立输入和输出流量之间的映射关系;最后,为了进一步提高预测精度,在模型的训练过程中,使用改进型的牛顿法来进行权值参数更新。仿真实验以及和其他算法对比的结果证实了所提方案具有更小的预测相对误差。     

基于连接两段激活模型的IEEE 802.16接纳控制算法

IEEE802.16系统中数据的传输是面向连接的,连接的建立服从两段激活模型。针对系统空闲资源耗尽但还为已接纳而未被激活的业务预留资源这一情况,该文提出了一种适当借用此预留资源去接纳新的立即被激活业务的接纳控制算法,并建立了此算法的3维马尔可夫链模型,进行了理论分析提出了找寻此算法中使用的两个门限的搜索方法。仿真结果表明,该接纳控制算法能显著地降低新业务阻塞率,提高系统的带宽利用率,已接纳业务的激活成功率只有少许下降。     

基于谱约束的生成对抗网络图像数据生成研究

数据匮乏是深度学习面临的一大难题。利用生成对抗网络(GAN)能够基于语义生成新的图像数据这一特性,提出一种基于谱约束的生成对抗网络图像数据生成方法,该方法针对卷积生成对抗网络模型易崩溃不收敛的问题,从每层神经网络的参数矩阵W的谱范数角度出发,引入谱范数归一化网络参数矩阵,将网络梯度限制在固定范围内,减缓判别网络收敛速度,从而提高GAN的训练稳定性。实验表明,通过该方法生成的数据相比原始GAN以及DCGAN、WGAN等生成的图像样本数据在图像识别网络中具有更高的准确率,能够对少量样本数据进行有效扩充。     

一种面向流量异常检测的概率流抽样方法

针对基于概率抽样的网络流量异常检测数据集构造过程中无法同时兼顾大、小流抽样需求及未区分flash crowd与流量攻击等问题,该文提出一种面向流量异常检测的概率流抽样方法。在对数据流按目的、源IP地址进行分类的基础上,将每类数据流抽样率定义为其目的、源IP地址抽样率的最大值,并在抽样过程中对数据流抽样数目向上取整,保证每类数据流至少被抽样一次,使抽样得到的数据集可有效反映原始流量在大、小流和源、目的IP地址方面的分布性。采用源IP地址熵刻画异常流源IP地址分散度,并基于源IP地址熵阈值设计攻击流抽样算法,降低由flash crowd引起的非攻击异常流抽样概率。仿真结果表明,该方法能同时满足大、小流抽样需求,具有较强的异常流抽样能力,可抽样到所有与异常流相关的可疑源、目的IP地址,并能在抽样过程中过滤非攻击异常流。     

5G无线网络智能规划与仿真

传统基于人工经验的规划方法已不能满足5G网络的需求,亟需引入基于大数据分析的5G无线网智能规划方法。提出5G智能站址规划、参数规划和后评估三步闭环规划方案,搭建完成5G无线网络智能规划系统,应用于实际5G无线网规划仿真,高效地支撑面向未来的5G无线网络组网部署,确保资源精准投入,为建设5G精品目标网络奠定基础。     

TSL:predicting popularity of Facebook content based on tie strength

The rapid development of online social networks leads to an explosion of information,however,there are great differences in the popularity of different messages,and accurate prediction is always a great difficulty is the current study.Popularity prediction of online content aims to predict the popularity in the future based on its early diffusion status.Existing models for popularity prediction were mostly based on discovering network features or fitting the equation into a varying time function that the accuracy of current popularity prediction model was not high enough.Therefore,with the help of the weak ties theory in sociology,the concept of tie strength was introduced and a multilinear regression equation was constructed combined with the early popularity.A TSL model to predict the popularity of Facebook’s well-known pages was proposed.The main contribution of this article was to solve the problem and few or no work based on sociology.A high linear correlation between the proportion of faithful fans was existed in Facebook homepage with frequent shares in the early and the future popularity.Compared with other baseline models,an experimental study of Facebook (including 1.54 million shares) illustrates the effectiveness of the proposed TSL model,and the performance is better than the existing similar methods.     

一种基于网格划分的有向传感网时空覆盖调度算法

针对有向传感器网络中的时空覆盖调度问题进行研究,从有向传感器节点感知模型出发,设计了基于网格划分的网络基本区域生成方法,在此基础上提出了节点最大覆盖调度迭代选择MaxGreedy算法.通过仿真实验验证了网格划分方法的有效性,设计了一系列的时空覆盖算法对比实验,深入评估MaxGreedy算法的性能.对比试验结果表明,MaxGreedy算法可以高效地生成网络的节点调度模式,并在一定程度上提高网络的时空覆盖率.     

Comparison of different congestion control strategies for low priority controllable traffic in packet switched backbone networks

It is expected that a significant part of the data flows of future multi‐service packet switched backbone networks will use low priority, non‐real‐time data transmission services of the networks. The common benefit for both user applications and network operators is that the data flows of the low priority services could use the free capacity of the networks, after the load of higher priority data flows. Congestion control methods are needed for these low priority data flows to reach an optimal utilisation level of the networks, high throughput and low packet loss ratios. This kind of low priority data transmission service which adjusts the data rates of the data flows according to the data rate changes of higher priority data flows, but does not guarantee any specific service for these data flows, is called a controlled load service. In this paper, we have compared the performance, efficiency and scalability of four different congestion control methods designed for the controlled load service. Two of these methods were based on very simple congestion control algorithms and the other two used relatively complex control algorithms based on control methods utilising computational intelligence. The principal aim of this study was to research how remarkable were the effects that the different complexities of the congestion control methods had on the achieved level of service. The simulation tests indicate that the complexity of the methods clearly affects the performance and efficiency of the methods. Copyright © 2001 John Wiley & Sons, Ltd.     

基于A3C的无线异构网络自适应视频流传输控制方法

比特率自适应（ABR）算法已经成为视频传输中研究的热点之一。然而,由于5G无线异构网络具有信道带宽波动大、不同网络间差异明显等特点,多终端协同的自适应视频流传输面临着巨大挑战。提出了一种基于深度强化学习的自适应视频流传输控制方法。首先,建立了视频流动态规划模型,对传输码率以及分流策略进行联合优化。由于该优化问题的求解依赖于精确的信道估计,这在信道状态动态变化的网络中很难实现。因此,将动态规划问题改进为强化学习任务,并采用A3C算法,动态决策视频码率和分流策略。最后,根据实测的网络数据进行仿真,与传统的优化方法相比,本文所提的方法较好地提高了用户QoE。     

远程遥控驾驶场景下基于网络信息开放的QoS预测

5G系统的飞速发展支持许多车联网用例对服务质量（quality of service,Qo S）的苛刻要求。但网络和应用的适配仍然存在很多问题。网络信息开放是一种潜在的解决方案,旨在实现网络向应用程序实时提供蜂窝无线网络信息,从而帮助服务提供方实现更好的策略控制并改善用户体验。提出一种基于网络信息开放的服务质量预测（predictive QoS,PQo S）方法,通过提前预测即将发生的网络变化来支持应用做出提前响应,提高用户的体验质量（qualityofexperience,Qo E）。介绍了网络信息开放及PQo S的背景,并介绍了PQo S的国内外研究、标准及落地现状;提出一种远程遥控驾驶（tele-operateddriving,To D）场景下基于网络信息开放的Qo S预测方法;对实际测试的数据进行分析,评估验证了PQo S的可行性。结果表明,基于网络信息开放的Qo S预测技术能够良好地支持包括5G ToD在内的车联网应用,为5G系统在智慧交通行业的落地提供了参考。     

一种5G网络用户与流量预测方法研究

在“基于市场推进策略流量预测方法”基础上，从网络建设与市场融合“推演”角度，提出一种“推进模型+分层预测”的联合5G用户与流量预测方法，能实现不同区域、不同策略下的4/5G联合分层预测。     

一种提高802.11无线Ad Hoc网络公平性的新机制-FFMA

实现多个数据流对无线信道的公平共享是802.11无线Ad Hoc网络中的一个重要议题,但802.11DCF机制在无线Ad Hoc网络中存在严重的公平性问题,甚至有可能出现单个节点或数据流独占信道而其他节点和数据流处于"饥饿"状态的情况.论文提出了一种新颖的保证数据流间公平性的MAC层接入机制FFMA(Flow rate-based Fair Medium Access),通过公平调度和公平竞争的方式,FFMA能够在数据流间公平地分配信道带宽资源.仿真结果表明,在无线Ad Hoc网络中,FFMA可以在保证信道吞吐量的前提下取得远优于802.11 DCF的数据流间的公平性.     

Saliency detection in human crowd images of different density levels using attention mechanism

The human visual system has the ability to rapidly identify and redirect attention to important visual information in high complexity scenes such as the human crowd. Saliency prediction in the human crowd scene is the process using computer vision techniques to imitate the human visual system, predicting which areas in a human crowd scene may attract human attention. However, it is a challenging task to identify which factors may attract human attention due to the high complexity of the human crowd scene. In this work, we propose Multiscale DenseNet — Dilated and Attention (MSDense-DAt), a convolutional neural network (CNN) using self-attention to integrate the result of knowledge-driven gaze in the human visual system to identify salient areas in the human crowd scene. Our method combines various state-of-the-art deep learning architectures to deal with the high complexity in human crowd image, such as multiscale DenseNet for multiscale deep features extraction, self-attention, and dilated convolution. Then the effectiveness of each component in our CNN architecture is evaluated by comparing different components combinations. Finally, the proposed method is further evaluated in different crowd density levels to appraise the effect of crowd density on model performance.     

一种用于卫星网络的业务预测方法

针对卫星网络业务具有自相似的特点,介绍了一种基于集合经验模式分解的业务组合预测方法(EEMD)。该方法利用EEMD的分解特性,将具有自相似的网络流量分解成多个只具备短相关性的本征模态函数(IMF),这样便可使用传统的流量预测算法进行预测。文中使用人工神经网络与自回归滑动平均模型(ARMA)这两种方法进行预测。最后将多个本征模态函数(IMF)预测的结果相加作为原始信号的预测结果,实验证明此方法有更高的精度。为了迎合卫星实时性的需求,给出了硬件的框架,该框架采用DSP与FPGA相结合的构架实现连续数据的EEMD实时处理。     

基于信息年龄的工业无线传感器网络混合数据调度方法

在工业无线传感器网络(IWSN)中，实时交付工业现场的周期性控制/传感数据流与非周期性事件数据流，是保障生产安全高效运行的关键。信息年龄(AoI)作为一种新兴的数据新鲜度衡量指标，能够从目标节点角度全面地度量网络数据交付的实时性。针对周期性和非周期性数据混合的工业无线传感器网络，该文在引入网络数据整体新鲜度指标的同时，考虑到周期性数据新鲜度在超过阈值后可能会对工业生产造成负面影响，建立了最小化系统平均AoI和周期性数据AoI逾期概率的联合优化模型，并将优化问题表述为马尔可夫决策过程(MDP)进行求解。由于传统基于相对值迭代的最优求解方法在大规模网络中因为维度灾难难以实施，因此采用深度强化学习(DRL)降低优化问题的状态空间维度，并改进决策探索机制以加快学习速度，提出了基于优化决策探索的深度强化学习(DRL-ODE)调度方法。仿真结果表明，所提方法能够提高网络数据交付的实时性，并有效减少周期性数据的AoI逾期概率。