网络设备识别系统的设计与实现

发布时间：2018-04-25 17:12

本文选题：网络设备 + 特征提取　；参考：《北京邮电大学》2017年硕士论文

【摘要】：当今时代,计算机网络技术飞速发展。除了路由器、交换机等传统计算机网络设备,打印机、VoIP电话、网络摄像头等各类设备也开始出现在计算机网络上。随着物联网时代的到来,更多不同类型的设备也开始加入计算机网络,可以预见,未来的计算机网络将实现人与人、物与物、物与人的实时信息交换。然而发展总会伴随着挑战,网络设备数量庞大、鱼龙混杂,会暴露出更多的网络安全问题,并给网络管理带来更为严峻的挑战。如何准确、快速地识别出虚拟网络空间中的设备类型变得十分重要,为了解决这一问题,网络设备识别系统应运而生。网络设备识别系统在现有的网络空间搜索引擎的基础上获取网络设备数据(设备数据包括端口、操作系统、服务和Banner等数据项)作为分类模型的样本数据,然后对样本数据做进一步的分析处理得到网络设备类型特征数据。在特征数据基础上,使用不同的分类算法构建分类模型。通过对分类器进行比较,选择精确度较高的分类器。系统通过Nmap和Zmap扫描工具实时获取网络设备数据,然后使用系统的分类器来判断网络设备的设备类型。本文首先介绍网络空间探测中设备类型识别的需求,根据需求构建网络设备识别系统的总体架构。然后对总体架构进行模块分析、详细设计,完成网络设备识别系统的设计和实现。然后根据需求分析对系统进行测试,证明网络设备识别系统符合设计的原则和要求。本文最后总结了网络设备识别系统的工作和不足,并指出未来的工作方向。
[Abstract]:Nowadays, computer network technology develops rapidly. In addition to router, switch and other traditional computer network equipment, printer VoIP phone, network camera and other devices are also beginning to appear on the computer network. With the advent of the Internet of things era, more and more different types of equipment are also beginning to join the computer network. It can be predicted that the future computer network will realize the real-time information exchange between people, things and things, things and people. However, the development will always be accompanied by challenges, the large number of network equipment, mixed, will expose more network security problems, and bring more severe challenges to network management. It is very important to identify the types of devices in virtual cyberspace accurately and quickly. In order to solve this problem, the network equipment recognition system emerges as the times require. The network device identification system acquires the network equipment data (including port, operating system, service and Banner data items) on the basis of the existing network space search engine as the sample data of the classification model. Then the sample data are further analyzed and processed to get the network device type characteristic data. Based on the feature data, different classification algorithms are used to construct the classification model. By comparing the classifier, the classifier with higher accuracy is selected. The system acquires the network device data in real time through Nmap and Zmap scanning tools, and then uses the classifier of the system to judge the device type of the network device. This paper first introduces the requirements of device type identification in the detection of cyberspace, and constructs the overall architecture of the network device identification system according to the requirements. Then the overall architecture of the module analysis, detailed design, network equipment identification system design and implementation. Then the system is tested according to the requirement analysis to prove that the network equipment identification system meets the design principles and requirements. Finally, this paper summarizes the work and shortcomings of the network equipment identification system, and points out the future work direction.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP393.0

【相似文献】