基于CCN的P2P流媒体内容分发技术研究

发布时间：2018-02-27 21:08

本文关键词： 内容中心网络流媒体内容分发节点协助　出处：《北京邮电大学》2014年硕士论文　论文类型：学位论文

【摘要】：随着宽带互联网技术和多媒体技术的发展,面向大规模用户的流媒体视频服务已经成为当今互联网上最流行的应用之一,如何有效地进行流媒体内容的大规模传输已成为网络运营商面临的主要挑战。由美国Palo Alto研究中心提出的内容中心网络(Content Centric Networking, CCN)是一种新型的以内容为中心的网络体系结构,引入了路由缓存和请求聚合的机制优化互联网内容分发,所有内容可以在网络中的任何CCN节点处缓存,以满足其他节点的内容需求,相对于目前的TCP/IP网络,与底层网络拓扑联系更紧密的CCN网络更适合进行大规模流媒体内容分发。但目前的CCN流媒体内容分发仍类似C／S架构,路由缓存空间有限,用户间缺乏交互,网络中仍会存在过多的冗余流量。现有的些优化方案集中于提升CCN节点的缓存空间利用率和命中率等,在定程度上优化了内容传输效率,但带来了过高的迭代代价,同时也没有考虑流媒体业务对实时性的要求以及用户访问特性。因此有必要对CCN流媒体内容分发技术进行进一步的研究。本文在CCN网络内容分发机制的基础上,从优化网络性能和应用性能的角度对CCN流媒体内容分发机制进行研究,引入了P2P节点共享的思想,提出一种在CCN网络中基于用户节点协助的流媒体内容分发(PAS over CCN, Peer-assisted Streaming over CCN)方案,由用户节点协助提供流媒体资源的缓存和共享,由位于网络边缘的CCN路由节点充当类似P2P系统中Tracker的角色,记录本地用户节点缓存的资源信息,并对边缘CCN路由节点模型进行改进以支持用户节点协助的缓存机制。最后,本文通过仿真验证了本方案在优化网络流量、降低源服务器负载和提升应用性能方面的优势。本文首先分析了CCN流媒体研究现状和不足,然后详细描述了本文提出的基于用户节点协助的流媒体内容分发方案,最后介绍了基于OmNeT++的仿真环境的设计和实现以及对方案性能的评估。
[Abstract]:With the development of broadband Internet technology and multimedia technology, streaming video service for large-scale users has become one of the most popular applications on the Internet. How to effectively transmit streaming media content on a large scale has become a major challenge for network operators. Content Centric networking, proposed by the Palo Alto Research Center in the United States, is a new content-centric network architecture. Routing caching and request aggregation mechanisms are introduced to optimize Internet content distribution. All content can be cached at any CCN node in the network to meet the content requirements of other nodes, as opposed to the current TCP/IP network. The CCN network, which is more closely connected with the underlying network topology, is more suitable for mass streaming media content distribution. However, the current CCN streaming media content distribution is still similar to the C / S architecture, the routing cache space is limited, and there is no interaction between users. There is still too much redundant traffic in the network. The existing optimization schemes focus on improving the cache space utilization and hit rate of CCN nodes, which optimizes the efficiency of content transmission to a certain extent, but brings too high iterative cost. At the same time, the real-time requirements and user access characteristics of streaming media services are not considered, so it is necessary to further study the CCN streaming media content distribution technology. Based on the CCN network content distribution mechanism, this paper studies the CCN streaming media content distribution mechanism from the point of view of optimizing network performance and application performance, and introduces the idea of P2P node sharing. In this paper, a scheme of streaming media content distribution (pas over, Peer-assisted Streaming over CCN) based on the assistance of user nodes in CCN network is proposed, in which user nodes assist in providing caching and sharing of streaming media resources. The CCN routing node located on the edge of the network acts as the role of Tracker in P2P system, records the resource information of the local user node cache, and improves the edge CCN routing node model to support the caching mechanism assisted by the user node. In this paper, the advantages of this scheme in optimizing network traffic, reducing source server load and improving application performance are verified by simulation. This paper first analyzes the current situation and shortcomings of CCN streaming media research, and then describes the proposed streaming media content distribution scheme based on user node assistance in detail. Finally, the design and implementation of simulation environment based on OmNeT and the evaluation of scheme performance are introduced.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.02

【参考文献】