电子信息工程专业英语翻译
Cooperative Peer-to-Peer Streaming:An Evolutionary Game-Theoretic
Approach
YanChen,StudentMember,IEEE,BeibeiWang,StedengMember,IEEE,W.Sabrina
Lin,Member,IEEE,Yongle Wu,Student Member,IEEE,and K,.J.Ray Liu,Fellow,IEEE Abstrac t —While peer-to-peer(P2P)video streaming systems have achieved promising results,they introduce a large number of unnecessary traverse links,which consequengly leads to substantial network inefficiency.To address this problem and achieve better streaming performance,we propose to enable cooperation among “group peers,”which are geographically neighboring peers with large intra-group upload and download bandwidths.C onsidering the peers’selfish nature,we formulate the cooperative streaming problem as an evolutionary game and derive,for every peer,the evolutionarily stable strategy(ESS),which is the stable Nash equilibrium and no one will deviate from.Moreover,we propose a simple and distributed learning algorithm for the peers to converge to the ESSs. ed from the peer’sown padtWith the propo sed algorithm,each peer decides whether to be an agent who downloads data from the peers outside the group or a free rider who downloads data from the agents by simply tossing a coin,where the probability of being a head for the coin is learned from the peer’s own past payoff history.Simulation results shou to the traditional non-cooperative P2P schemes,the proposed cooperative scheme achieves much better performance in terms of social welfare,probability of real-time streaming,and video quality(source rate).
IndexTerms —Coopreative streaming,distributed learning,evolutionary,game theory,peer-to-peer(P2P),replicator dynamics.
ⅠIntroduction
With the rapid development of signal processing,communication, and networking technologies,video-over-IP applications become more and more popular and have attracted millions of users over the Internet.One simple solution to video streaming over Internet is the client-server service model,where the video is streamed directly from a server to clients.However with the client-server service model, the upload bandwidth of the server grows proportionally with the number of clients, whice makes the large-scale video streaming impractical.
To reduce the workload of the server,peer-to-peer(P2P)service model is proposed, where a peer not only acts as a client to download data from the network,but also acts as a server to upload data for the other peers in the network.The upload bandwidth of the peers reduces the workload placed on the server dramatically,which makes large-scale video streaming possible,Recently,several industrial large-scale P2P video streaming systems have been developed,including
Coolstreaming,PPLive,PPStream,UUSee,and Sopcast. Studies show that these systems can support hundreds of thousands of users simultaneously.
While P2P video streaming systems have schieved promising results, they have several drawbacks. First, there are a large number of unecessary traverse links within a provider’s network.As observed in ,each P2P bit on the Verizon network,As observed in,each P2P bit on the Verizon network traverses 1000 miles and takes 5.5 metro-hops on average,Second,there is a huge number of cross Internet service provider traffic,The studies showed that 50%-90%of the existing local pieces in active peers are downloaded externally.Third,the differences in playback time among peeers can be as high as 140s,and the lag can be greater if the source rate is higher,Fourth,most of the current P2P systems assume that all peers are willing to contribute their resources.however,this assumption may mot be true since the P2P systems are self-organizing networks and the peers are selfish by nature.Note that the selfish peers will act as free-riders if being free-riders can improve their utilities. 译文:
对等网络流:进化论的赛局的方法
杨晨,学生;王蓓蓓,学生;塞布丽娜林,IEEE 成员;永乐吴,学生;永乐刘,学生
摘要:而对等网络视频流系统取得了很有前途,但这样的技术引进大量不必要的导线连接, 从而导致实质性的网络无效率。为了解决这个问题, 达到更好的流的性能, 我们提出“使组流之间的能够合作同行, 即在物理地址上邻近组流之间共用上传和下载带宽. 考虑同流本身带宽小的缺陷, 我们将同行合作流问题作为演化博弈衍生出来, 对每一个同流, 进化稳定策略(ESS),这是稳定的纳什均衡, 没有任何事物能偏离了此算法,因为同流也包括在进化稳定策略中。此外, 我们提出了一种简单的算法和分布式学习,这符合进化稳定策略. 在既定的算法下, 由每个节点来通过简单的抛硬币来决定代理人同行下载数据以外的群体或搭便车的人是否可以从代理下载数据, 那里的概率作为一个头, 这个概率决定于同流节点之前的下载历史。仿真结果表明与传统的非合作的P2P 相比, 合作方案能够在社会福利、概率, 并实时流媒体视频质量(源速率) 达到更好的表现。
索引项目-合作数据流、分布式学习、进化、博弈理论、点对点(P2P),进行自我复制动态。
简介
随着通信信号处理技术和网络技术的应用的飞速发展, 通过网络地址传输的视频数据变得越来越受欢迎, 而且吸引了成千上万的互联网用户。一个很简单的解决方法是采用互联网视频流媒体服务模式, 在此模式下视频是直接从服务器流到客户机的。然而在此模式下,随着用户的增多,上传频宽的用户机器的数量增长,使得大量数据流传输变得无法实现。为了减少服务器工作负担, 点对点(P2P)服务模型, 在同流中不仅充当一个客户端从网络中下载数据, 并且也可作为服务器网络中上传数据。同流上传的带宽, 大大减少服务器的工作量, 这使得大量视频流可以实现。近期, 市场在中出现了一些大规模P2P 视频实时直播系统, 包括
Coolstreaming,PPStream UUSee,PPLive 和 Sopcast。研究表明, 这些系统可以支持成千上万客户同时使用。
P2P 流媒体系统视频方案已经达到了预期的效果,但也有一些不足。首先,
在一个提供商的网络中有大量不必要的网络连接。观察到, 在点对点比特率传输中会有1000里不必要的传输连接,平均要消耗掉5.5带宽。第二, 有大量的交叉的互联网服务提供商的, 这一研究表明, 在下载过程中本地内部网络连接中会损耗掉50%-90%。第三, 播放时间上的差异, 在同流中可以高达140s, 假如数据源效率较高, 滞后也很明显。第四, 现阶段多数的P2P 系统假设所有的同流都愿意贡献自己的资源. 然而, 这种假设很不真实,自P2P 网络中的节点是出现以来, 从本质上来说它就是排外的。值得注意的是自私的同流要担任无偿者如果被无偿享用则能大大提高