基于决策树的P2P节点识别技术研究

通过对P2P应用的长期研究,根据P2P节点自身的特点选取了其中典型的特征属性,并提出了一种基于决策树模型的P2P节点识别方法。由于该方法是统计分析传输层数据分组的特征,因此对于采用加密或非加密的P2P应用的网络节点识别均有效。通过实验验证,与基于端口和基于负载特征的流量监测方法相比,所提出的方法体现出了较高的准确率和较低的漏报率及误报率。     

I2P匿名通信流量特征分析与识别

由于匿名通信网络具有非常强的匿名性,因此越来越多的人使用匿名通信网络发布违法信息,给监管带来了很大挑战。识别匿名通信网络的流量对监管匿名通信网络具有重要意义,但目前很少有针对I2P的流量识别研究。因此,通过对I2P数据流进行深入分析,得到I2P数据流的特征,并在此基础上设计实现了I2P流量识别算法。实验结果表明,该算法具有较高的精确度和召回率,可以有效识别I2P流量。     

一种P2P匿名控制身份鉴别机制

为解决一般的身份鉴别机制不适用于具有匿名通信要求的P2P网络系统模型的难题。通过对Diffle-Hellman密钥协商协议进行改进,并结合采用RSA数字签名协议,以及零知识证明GQ协议,提出了一种新的基于服务令牌对P2P匿名通信系统中的通信双方进行身份鉴别的机制。该机制在保证P2P匿名通信系统各种普遍特征的前提下,通过在P2P匿名通信系统中引入可信第三方节点,针对P2P匿名通信系统中各通信节点进行匿名控制和行为管理,同时既可抵御各种常见网络攻击的威胁,又有效实现了P2P匿名通信系统的身份鉴别,从而有效提高了P2P匿名通信系统的安全性,加强了此种系统的管理能力。     

PPLive协议分析及流量识别

近年来兴起的P2P流媒体应用造成了网络带宽的巨大消耗，精确识别P2P流量具有很强的实际意义。通过抓取PPLive的流量数据进行协议分析，主要关注系统获取节点列表方式、资源交换方式和数据传输方式等实现方法，进而总结协议特征，提出了一种基于协议分析的PPLive流量识别方法。     

基于市场机制的P2P流媒体网络节点选择算法

针对P2P流媒体网络中的节点选择问题,将移动代理引入P2P网络,建立起用于有效实现P2P网络节点的市场化节点选择模型.基于该模型,进一步提出了一种基于市场机制的自适应节点选择算法MANS:通过模拟市场经济中的价格机制适时调整节点的带宽价格,并动态划分节点角色,依据所建立的评价模型选择最优节点,进行数据服务.最后将该模型与MANS算法运用于直播场景,对仿真结果进行了分析并和静态选择算法进行了比较.     

一种新的基于弱中心的网络安全复合管理技术

首先,论文分析了常见网络管理技术的问题和不足。随后,结合信息技术的发展,提出了一种新的基于弱中心的网络复合管理技术。该技术采用P2P网络结构和通信模式,弱化了中心管理节点的功能,增强了各网络管理节点之间的直接联系和信息交互。和其它网络管理技术相比,文章提出的新技术具有更高的效率和灵活性。     

P2P流量的精细化识别方法研究

目前,已有多种方法可高效准确地完成对P2P流量的粗识别,但对P2P流量的精细化识别研究较少。该文首次将近邻传播(Affinity Propagation, AP)算法引入该领域,在Hi-WAP算法的基础上融合半监督聚类思想提出了一种基于分层加权半监督近邻传播(Hierarchical Weighted Semi-supervised AP, Hi-WSAP)算法的P2P流量精细化识别方法。该方法仅利用10个可快速计算获取的网络流特征对P2P流量按应用进行半监督聚类。两组数据集下的实验结果表明,该方法识别准确率高,时间复杂度低,为P2P流量的实时精细化识别提供了一种实现思路。     

P2P网络中基于Eigentrust算法的信任模型研究

在点对点(P2P)网络中,信任模型能够有效的降低恶意节点的影响。Eigen Trust是最具有权威性的信任度算法之一,本文提出了一个基于反馈的推荐信任度算法,通过区分服务信任值和推荐信任值,防止拥有较高服务信任值的节点对正常节点的诋毁和欺骗。引入请求信任值,对不积极参与文件共享的节点和恶意服务节点进行访问权限的限制,识别并拒绝使用该类恶意节点的推荐信息,有效提高交互成功率。     

Kad点对点网络中最佳结点标识选取研究

Kad网络是一种主流的文件共享对等网络（即P2P network）,不法信息在其上的传播导致对其进行网络监管的需求也与日俱增.Kad网络中广泛使用的节点标识Kad ID存在＂Kad ID别名＂问题,即单个节点具有多个Kad ID.为解决这一问题,设计实现Kad网络节点信息爬虫—Kcrawler,并提出了最佳节点标识概念,以获得实际Kad网络中节点的多个候选标识之间的对应关系信息.实验结果表明{userID}的稳定因子最大,是最佳节点标识.     

面向P2P特定信息的爬虫改进技术

传统的P2P流量监控与识别技术属于一种整体型识别技术,然而,却难以监测P2P特定信息的传播规律以及传播网络,然而,在广泛的应用了P2P下载软件后,当前的技术已经满足不了时代发展的要求.因此,文章以P2P特定信息爬虫改进技术为切入点进行了分析.     

互联网视频技术进展

目前在中国互联网视频每天收看次数在一亿次以上。互联网视频可以分为视频分享、在线点播和P2P流媒体3类。视频分享采用浏览器／服务器模式服务于个人制作内容;在线点播为收费业务,采用客户机／服务器模式;而P2P流媒体采用P2P模式用于热点内容。为解决P2P流量优化和内容监管问题,中国信息技术标准委员会成立P2P工作组制定相关标准。     

A distributed utility‐based scheduling for peer‐to‐peer video streaming over wireless networks

Interactive multimedia applications such as peer‐to‐peer (P2P) video services over the Internet have gained increasing popularity during the past few years. However, the adopted Internet‐based P2P overlay network architecture hides the underlying network topology, assuming that channel quality is always in perfect condition. Because of the time‐varying nature of wireless channels, this hardly meets the user‐perceived video quality requirement when used in wireless environments. Considering the tightly coupled relationship between P2P overlay networks and the underlying networks, we propose a distributed utility‐based scheduling algorithm on the basis of a quality‐driven cross‐layer design framework to jointly optimize the parameters of different network layers to achieve highly improved video quality for P2P video streaming services in wireless networks. In this paper, the quality‐driven P2P scheduling algorithm is formulated into a distributed utility‐based distortion‐delay optimization problem, where the expected video distortion is minimized under the constraint of a given packet playback deadline to select the optimal combination of system parameters residing in different network layers. Specifically, encoding behaviors, network congestion, Automatic Repeat Request/Query (ARQ), and modulation and coding are jointly considered. Then, we provide the algorithmic solution to the formulated problem. The distributed optimization running on each peer node adopted in the proposed scheduling algorithm greatly reduces the computational intensity. Extensive experimental results also demonstrate 4–14 dB quality enhancement in terms of peak signal‐to‐noise ratio by using the proposed scheduling algorithm. Copyright © 2015 John Wiley & Sons, Ltd.     

Resource assignment in short video accelerator system

Video accelerator is developed for better user experience in video sharing websites such as YouTube.PPLive video accelerator (PPVA),which has the largest number of users in China,is based on peer-to-pe...     

Scalable 3D video streaming over P2P networks with playback length changeable chunk segmentation

3D video distribution over P2P networks has been thought as a promising way for 3D video entering home. The convergence of scalable 3D video coding and P2P streaming can provide diverse 3D experiences for heterogeneous clients with high distribution efficiencies. However, the conventional chunk segmentation and scheduling algorithms originally aiming at the non-scalable 2D video streaming are not very efficient for scalable 3D video streaming over P2P networks due to the particular data characteristics of scalable 3D video. Based on this motivation, this paper first presents a playback length changeable 3D video chunk segmentation (PLC3DCS) algorithm to provide different error resilience strengths to video and depth as well as layers with different importance levels in the 3D video transmission. Then, a hybrid-priority based chunk scheduling (HPS) algorithm is proposed to be tied in with the proposed chunk segmentation algorithm to further promote the overall 3D video P2P streaming performance. The simulation results show that the proposed PLC3DCS algorithm with the corresponding HPS can increase the success delivery rates of chunks with more important levels, and further improve the user’s quality of 3D experience.     

Large-scale measurement experiments of P2P-TV systems insights on fairness and locality

P2P-TV is an emerging alternative to classical television broadcast systems. Leveraging possibilities offered by the Internet, several companies offer P2P-TV services to their customers. The overwhelming majority of these systems, however, is of closed nature, offering little insight on their traffic properties. For a better understanding of the P2P-TV landscape, we performed measurement experiments in France, Japan, Spain, and Romania, using different commercial applications. By using multiple measurement points in different locations of the world, our results can paint a global picture of the measured networks, inferring their main properties. More precisely, we focus on the level of collaboration between peers, their location and the effect of the traffic on the networks. Our results show that there is no fairness between peers and that is an important issue for the scalability of P2P-TV systems. Moreover, hundreds of Autonomous Systems are involved in the P2P-TV traffic and it points out the lack of locality-aware mechanisms for these systems. The geographic location of peers testifies the wide spread of these applications in Asia and highlights their worldwide usage.     

基于ARM的AVS-P2视频解码器

研究了AVS视频解码部分,介绍了AVS视频编码过程的关键技术.详细阐述了针对ARM平台的AVS-P2视频解码算法的优化,并在基于ARM-Linux的嵌入武平台上实现了AVS视频解码.     

VIDEO OBJECT SEGMENTATION BY 2-D MESH-BASED MOTION ANALYSIS

Video object extraction is a key technology in content-based video coding.A novel video object extracting algorithm by two Dimensional (2-D) mesh-based motion analysis is proposed in this paper.Firstly,a 2-D mesh fitting the original frame image is obtained via feature detection algorithm. Then,higher order statistics motion analysis is applied on the 2-D mesh representation to get an initial motion detection mask.After post-processing,the final segmenting mask is quickly obtained.And hence the video object is effectively extracted.Experimental results show that the proposed algorithm combines the merits of mesh-based segmenting algorithms and pixel-based segmenting algorithms,and hereby achieves satisfactory subjective and objective performance while dramatically increasing the segmenting speed.     

Inferring playback rate and rate resets of P2P video streaming transmissions by piecewise line envelope approximation

Conventionally, P2P video is regarded as CBR traffic. However, our measurements have shown that the rate reset cannot be neglected in a practical IPTV system because each rate reset often leads to performance degradation. Thus, addressing the problem of inferring playback rate and rate reset in a P2P video system is significant. In this article, an algorithm termed piecewise linear envelope approximation (PLEA) is proposed, in which a follow-up time is introduced to smooth rate fluctuations in a small time scale and to adapt rate jumps in a large time scale automatically. With the PLEA algorithm, discontinuity introduced by blind segmentations adopted by current methods is avoided. Furthermore, unlike existing algorithms in which both segmentation and combinations are performed in multiple runs, only a single computation path is involved in the PLEA algorithm. This makes PLEA algorithm amenable to implementation of low complexity by either software or hardware. Both theoretical analysis and experiment based on measured data show that the PLEA outperforms existing algorithms based on segmentation.     

基于ConferenceXP的P2P视频会议系统设计与实现

通过分析现有视频会议系统存在的问题,并结合P2P网络的技术特点,提出了一种基于P2P的视频会议系统体系结构。同时分析音视频同步和带宽自适应传输两个关键问题,并给出了解决方案,重点叙述基于Conference XP平台的P2P视频会议系统的设计实现过程。ConferenceXP融合了音频视频和网络技术中的新发展,在该平台上实现的视频会议系统具有很好的实用价值和市场潜力。     

Statistical characterization of P2P-TV services and users

P2P-TV systems will certainly contribute to innovate the broadcast TV concept by allowing ubiquitous access to a practically unlimited number of channels, representing an important step forward in the direction of the Anything/Anyone/Anywhere/Anytime communication paradigm of future Internet applications. Since its deployment, P2P-TV has already attracted millions of users and it is expected to grow exponentially in the near future, both in terms of the number and diversity of their clients. Analyzing the behavior of P2P-TV systems and users and characterizing the generated traffic are fundamental steps to understand and evaluate the potential impacts of these applications on the underlying transport network. In this paper, a set of intensive measurements was conducted in a residential Internet access environment in order to obtain a good insight over the mechanisms that govern data transfer on some of the most important P2P-TV applications (TVants, SOPCast and TVU Player) and understand the most important advances they have suffered in the latest years. Among other relevant findings, the analysis showed that peer location does not seem to be exploited by any of the studied applications and P2P-TV is already a worldwide phenomenon: using the three selected applications during 12 days of experiments, peers from 177 different countries all over the world were contacted, including countries where the interest for English-spoken channels was not so likely.