未来网络信息分发技术研究

发布时间：2017-12-30 18:24

本文关键词：未来网络信息分发技术研究　出处：《北京邮电大学》2015年硕士论文　论文类型：学位论文

【摘要】：随着人们越来越依赖互联网来获取信息,互联网所承受的压力也越来越大。从用户的角度来看,网络数据的访问存在着时间和空间上的相关性,当某区域内有用户访问过一个内容后,该区域的其他用户很可能再次访问该内容,如果每次都从原始数据源获取内容,会造成同一内容地重复传输,浪费网络带宽,同时对源服务器也是很大的压力。网络缓存技术的出现在一定程度上缓解了上述问题。随着人们对网络的使用从基于主机位置(IP地址)向基于内容的逐渐转变,网络中缓存的重要性越来越明显。以内容为中心的新型未来网络架构应运而生。未来网络以信息的分发和获取为主要目的,信息内容分发技术是未来网络研究领域的一个核心问题。因此本论文重点研究了未来网络中的信息分发技术。近年来,未来网络领域的研究已经有很多,在数据分发技术方面的研究主要集中在网络内缓存相关的问题,未来网络特别是信息中心网络(Information-Centric Networking, ICN)中的缓存具有普遍存在、透明和细粒度的特点,在未来网络中构成了一个网络内置缓存网络,对于数据分发的性能有很大的影响,这方面研究的热点包括网络中各个结点缓存大小的设置、内容在哪些结点缓存的缓存决策策略、缓存满时更新哪些内容的缓存替换策略以及结点间的协作缓存策略等。但是现在的研究大多还局限在有线场景和静态拓扑的情况下,对于移动场景特别是车载网这种快速移动的动态拓扑场景下的缓存技术的研究还很少。本文针对未来网络特别是ICN在移动场景下的内容分发问题,设计了一种基于社会关系的协作缓存机制。该协作缓存机制能够提高移动网络中结点间的链路稳定性,保证数据分发的质量,同时扩大内容发现的范围和效率,提高结点稀疏环境下的内容分发效率,进而提高资源利用率,改善用户体验。本论文基于ndnSIM搭建仿真平台,针对车载网高速移动结点场景对仿真环境中的模块进行了更改,使之更加符合高速移动场景下的需要,然后将ICN中的普遍缓存策略和本论文提出的协作缓存机制进行了仿真对比,证明了协作缓存机制在未来网络数据分发过程中的有效性和优越性。最后,基于对全文总结的基础上,指出了本文工作的不足和可以改进的地方。
[Abstract]:As people rely more and more on the Internet to obtain information, the pressure on the Internet is also increasing. From the perspective of users, the access of network data has time and space relevance. When a user in a region has access to a content, other users in that area are likely to access the content again, and if the content is retrieved from the original data source each time, the same content will be transferred repeatedly. Wasting network bandwidth. At the same time, there is also a lot of pressure on the source server. The emergence of network caching technology alleviates the above problem to some extent. As people use the network from host location to IP address). A gradual shift to content-based. The importance of cache in the network is becoming more and more obvious. A new content-centered future network architecture emerges as the times require. The main purpose of the future network is to distribute and obtain information. The information content distribution technology is a core problem in the future network research, so this paper focuses on the information distribution technology in the future network. In recent years, there are a lot of research in the field of network in the future. The research on data distribution technology is mainly focused on the issues related to cache in the network. Caching in future networks, especially information center networks, is ubiquitous in Information-Centric networking (ICNs). The characteristics of transparency and fine granularity constitute a network with built-in cache network in the future, which has a great impact on the performance of data distribution. The research focus in this area includes the setting of the cache size of each node in the network and the cache decision strategy of which nodes are cached. Cache replacement policy and cooperative cache policy among nodes are updated when cache is full. However, most of the current researches are limited to wired scenarios and static topology. There is little research on cache technology in mobile scene, especially in vehicle network, which is a fast moving dynamic topology scenario. This paper aims at the content distribution of future networks, especially ICN in mobile scenarios. A cooperative caching mechanism based on social relations is designed, which can improve the link stability between nodes in mobile network, ensure the quality of data distribution, and expand the scope and efficiency of content discovery. This paper builds a simulation platform based on ndnSIM to improve the efficiency of content distribution, resource utilization and user experience. The module in the simulation environment is changed for the high-speed mobile node scene of the vehicle network to make it more suitable for the needs of the high-speed mobile scene. Then the general cache strategy in ICN and the cooperative cache mechanism proposed in this paper are simulated and compared, which proves the effectiveness and superiority of the cooperative cache mechanism in the future network data distribution process. Finally. Based on the summary of the paper, this paper points out the shortcomings of the work and the possible improvement.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TP393.02

【参考文献】