当前因特网多媒体视频流业务增长迅速,有效识别网络流量P2P(Peer to Peer)视频流服务是一项重要挑战。本文提出一种基于机器学习的P2P视频流识别技术。该技术中视频流特征分析系统包括数据报采集、处理、分流和属性特征计算模块。其次,...当前因特网多媒体视频流业务增长迅速,有效识别网络流量P2P(Peer to Peer)视频流服务是一项重要挑战。本文提出一种基于机器学习的P2P视频流识别技术。该技术中视频流特征分析系统包括数据报采集、处理、分流和属性特征计算模块。其次,提出了基于统计方法的P2P视频流数据报特征挖掘方法。仿真实验验证了所提出方法的高效性和准确性,能够较好的满足P2P视频流识别。展开更多
Pull-based P2P live streaming is a promising solution for the large scale streaming systems, like PPStream, PPlive, due to its high scalability, low cost and high resilience. However, they usually suffer from bad dela...Pull-based P2P live streaming is a promising solution for the large scale streaming systems, like PPStream, PPlive, due to its high scalability, low cost and high resilience. However, they usually suffer from bad delay performance. In this paper, we seek to improve the delay performance under ensuring video display quality stemming from chunk scheduling. And so we model Pull-based chunk scheduling problem as a multi-objective optimization problem to minimize the video delay and maximize video display quality in the environment of heterogeneous upload bandwidths, heterogeneous and dynamic propagation delays. Finally we put up with a greedy Pull-based scheduling approach(GPSA) to solve the optimization problem. The evaluation shows GPSA can outperform two classical chunk scheduling approaches and adapt to dynamic variance of propagation delays.展开更多
文摘当前因特网多媒体视频流业务增长迅速,有效识别网络流量P2P(Peer to Peer)视频流服务是一项重要挑战。本文提出一种基于机器学习的P2P视频流识别技术。该技术中视频流特征分析系统包括数据报采集、处理、分流和属性特征计算模块。其次,提出了基于统计方法的P2P视频流数据报特征挖掘方法。仿真实验验证了所提出方法的高效性和准确性,能够较好的满足P2P视频流识别。
基金supported by National Key Basic Research Program of China(973 Program)(2009CB320504)the Foundation for Innovative Research Groups of the National Natural Science Foundation of China (Grant No. 60821001)+1 种基金Beijing Municipal Commission of Education to build the project special,Research Fund for the Doctoral Program of Higher Education of China (20090005120012)National Natural Science Foundation (60672121)
文摘Pull-based P2P live streaming is a promising solution for the large scale streaming systems, like PPStream, PPlive, due to its high scalability, low cost and high resilience. However, they usually suffer from bad delay performance. In this paper, we seek to improve the delay performance under ensuring video display quality stemming from chunk scheduling. And so we model Pull-based chunk scheduling problem as a multi-objective optimization problem to minimize the video delay and maximize video display quality in the environment of heterogeneous upload bandwidths, heterogeneous and dynamic propagation delays. Finally we put up with a greedy Pull-based scheduling approach(GPSA) to solve the optimization problem. The evaluation shows GPSA can outperform two classical chunk scheduling approaches and adapt to dynamic variance of propagation delays.
