ISSN 1000-1239 CN 11-1777/TP

计算机研究与发展 ›› 2019, Vol. 56 ›› Issue (3): 467-479.doi: 10.7544/issn1000-1239.2019.20170473

• 网络技术 • 上一篇    下一篇

基于网络节点聚类的目标IP城市级定位方法

李明月1,2,罗向阳1,2,柴理想1,2,袁福祥1,2,甘勇3   

  1. 1(中国人民解放军战略支援部队信息工程大学 郑州 450001); 2(数学工程与先进计算国家重点实验室(中国人民解放军战略支援部队信息工程大学) 郑州 450001); 3(郑州轻工业大学计算机与通信工程学院 郑州 450001) (lmypretty@163.com)
  • 出版日期: 2019-03-01
  • 基金资助: 
    国家重点研发计划项目(2016YFB0801303,2016QY01W0105);国家自然科学基金项目(U1636219,61572052,61672354,61772549);河南省科技创新人才计划项目(2018JR0018);河南省科技攻关项目(162102210032)

City-Level IP Geolocation Method Based on Network Node Clustering

Li Mingyue1,2, Luo Xiangyang1,2, Chai Lixiang1,2, Yuan Fuxiang1,2, GanYong3   

  1. 1(Zhengzhou Information Science and Technology Institute, Zhengzhou 450001); 2(State Key Laboratory of Mathematical Engineering and Advanced Computing (Zhengzhou Information Science and Technology Institute), Zhengzhou 450001); 3(School of Computer and Communication Engineering, Zhengzhou University of Light Industry, Zhengzhou 450001)
  • Online: 2019-03-01

摘要: 现有经典的基于网络拓扑启发式聚类的目标IP城市级定位方法(HC-Based定位方法)通过网络结构的集群划分对网络IP节点进行聚类,定位结果误差较大,为此提出了一种基于网络节点聚类的IP定位方法(简记为NNC方法).该方法首先利用同一个网络社区往往位于同一个城域网的规律,考虑模块度能够可靠衡量网络社区结构强度的特点,基于模块度最优化进行网络拓扑聚类,得到模块度最高的网络社区划分结果;然后,基于IP地理位置数据库投票规则确定网络社区所处位置;最后,根据目标IP所处的网络社区,确定其所处的城市.基于中国河南、山东、陕西、广东、浙江5个省的15 000个互联网IP节点的实验结果表明:NNC方法与HC-Based定位方法相比,能够明显提升对目标IP的城市级定位的准确率和召回率,并降低地标错误对定位结果的影响.

关键词: IP定位, 网络拓扑聚类, 模块度, 社区发现, 城市级定位

Abstract: Existing city-level target IP geolocation method based on network topology heuristic clustering (HC-Based method) clusters IP nodes by simple voting rules, which is liable to cause a lot of errors in geolocation results. This paper presents a city-level IP geolocation method based on network node clustering, referred to as the NNC method. This method firstly uses the principle that the same network community locates in the same metropolitan area network. Considering the characteristics of the module that can accurately measure the strength of the network community structure, the network topology is clustered based on the modular optimization, and the network community with the highest module degree is obtained. Then the IP geography database voting rules is used to determine the location of the network community. Finally, depending on the network community where the target IP is located in, the city where the target IP is located in can be determined. Experimental results of 15 000 Internet IP nodes in five provinces (Henan, Shandong, Shaanxi, Guangdong and Zhejiang) of China show that compared with HC-Based method, the proposed method can significantly improve the accuracy and recall rate of the target IP, and reduce the effect of the inaccurate landmarks on the location results.

Key words: IP geolocation, network topology clustering, modularity, community detection, city-level geolocation

中图分类号: