网络层流量识别与关键内容提取系统设计与实现

发布时间：2018-03-20 19:42

本文选题：流量识别　切入点：内容提取　出处：《电子科技大学》2014年硕士论文　论文类型：学位论文

【摘要】：多媒体业务和P2P业务等的蓬勃发展,使得在网络上传输的业务数据越来越多元化,不同的业务由于服务的类型和对象不同,所以对网络的要求也有明显差异,比如对传输时延的要求,对丢包率的要求以及各种各样的QoS需求,为了更好地控制网络流量,提高网络利用率,以及更好地为不同的业务按需服务,首先要解决的问题就是网络流量识别。网络流量识别技术是进行网络报文分类的基础,本文的主要研究工作就是实现对网络流量的识别和对网络关键内容的提取。本文首先通过阅读网络流量识别和关键内容提取相关领域的国内外文献,根据自身课题的具体要求,对网络流量识别和关键内容提取中的关键技术和算法进行了较为深入的研究,研究的重点主要有TCP/IP体系结构,报文分类算法和模式匹配算法,同时,在研究相关理论的基础上实现了一个用于网络流量识别和关键内容提取的Demo系统。由于实验条件的限制,虽然本文中实现的Demo系统可以完成对网络流量的识别和关键内容的提取,并且有直观的展示。但是,很难客观地评价其性能指标,所以,在本文中,还基于OPNET平台,对所实现的网络流量识别和关键内容提取算法进行了仿真和测试。总体来说,本文具有如下一些特点:1.较为全面的分析和研究了报文分类算法和模式匹配算法,这两类算法是网络流量识别和关键内容提取的技术基础。2.在研究相关算法的基础上,使用C++语言编程实现了一个网络流量识别和关键内容提取系统。3.在实现相关系统的基础上,又基于OPNET网络仿真平台对系统所采用的算法的性能做了测试和分析。4.本文所实现的网络流量识别与关键内容提取技术不仅可用于实际的网络环境中,作为软件系统使用,还可以单独将算法抽取出来,进行功能扩充和性能分析。
[Abstract]:With the rapid development of multimedia services and P2P services, the service data transmitted over the network is becoming more and more diversified. Because of the different types and objects of services, the requirements of the network are also obviously different. For example, the requirements of transmission delay, packet loss rate and various QoS requirements, in order to better control network traffic, improve network utilization, and better serve on demand for different services. The first problem to be solved is network traffic identification, which is the basis of packet classification. The main research work of this paper is to realize the identification of network traffic and the extraction of network key content. The key technologies and algorithms of network traffic identification and key content extraction are deeply studied. The research focuses on TCP/IP architecture, packet classification algorithm and pattern matching algorithm. Based on the research of related theories, a Demo system for network traffic identification and key content extraction is implemented. Although the Demo system realized in this paper can realize the identification of network traffic and the extraction of key contents, and has a visual display. However, it is difficult to evaluate its performance index objectively, so in this paper, it is also based on OPNET platform. The network traffic recognition and key content extraction algorithms are simulated and tested. In general, this paper has some characteristics as follows: 1.The packet classification algorithm and the pattern matching algorithm are analyzed and studied comprehensively. These two algorithms are the technical foundation of network traffic identification and key content extraction. Based on the research of related algorithms, a network traffic identification and key content extraction system is implemented by C language programming. The performance of the algorithm used in the system is tested and analyzed based on the OPNET network simulation platform. 4. The network traffic identification and key content extraction technology realized in this paper can be used not only in the actual network environment, but also as a software system. The algorithm can also be extracted separately for function expansion and performance analysis.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.06

【参考文献】