欢迎访问《兵工学报》官方网站,今天是 分享到:

兵工学报 ›› 2019, Vol. 40 ›› Issue (7): 1450-1459.doi: 10.3969/j.issn.1000-1093.2019.07.015

• 论文 • 上一篇    下一篇

基于功能链的融合网络功能抗毁性评估

李尔玉, 龚建兴, 黄健   

  1. (国防科技大学 智能科学学院, 湖南 长沙 410073)
  • 收稿日期:2018-06-04 修回日期:2018-06-04 上线日期:2019-09-03
  • 通讯作者: 龚建兴(1976—),男,副研究员,硕士生导师 E-mail:fj_gjx@qq.com
  • 作者简介:李尔玉(1994—),男,硕士研究生。E-mail:lieryu18@yeah.net

Analysis about Functional Invulnerability of Convergent Network Based on Function Chain

LI Eryu, GONG Jianxing, HUANG Jian   

  1. (College of Intelligence Science and Engineering, National University of Defense Technology, Changsha 410073, Hunan, China)
  • Received:2018-06-04 Revised:2018-06-04 Online:2019-09-03

摘要: 为解决现有作战网络模型抽象度过高、网络功能复杂性研究不够充分的问题,提出了一种 针对融合网络的功能抗毁性分析方法。建立作战体系融合复杂网络模型,在单层网路中处理异质、融合的节点;提出新的功能链概念,并基于功能链定义了节点的功能参与度和业务强度;在此基础上,定义功能完整度为网络功能抗毁性测度,对融合网络模型进行抗毁性评估。在抗毁性实验中,采用3类攻击方式对网络进行打击:网络信息完全未知情况下的随机攻击;网络信息不完全已知情况下的蓄意攻击——按节点度攻击;网络信息完全已知情况下的蓄意攻击——按功能重要度攻击和按贪婪规则攻击。实验结果显示:网络对抗随机攻击能力较强,对抗信息不完全已知的蓄意攻击能力稍弱,对抗信息已知的蓄意攻击能力较弱,验证了所提算法合理性。

关键词: 复杂网络, 功能链, 抗毁性, 融合节点

Abstract: As the existing combat network model is abstracted and the study on network functional complexity is not sufficient enough, an analysis method for functional invulnerability of convergent networks is presented. A complex convergent network model of combat system is established, which deals with hetero- geneous and convergent nodes in a single-layer network, and a new concept of function chain is proposed. The functional participation and functional importance of nodes are defined by function chain. On this basis, the functional integrity is defined as the measure of network functional invulnerability. Three attack modes are adopted in experiment: the first is random attack when network information is unknown, the second is intentional attack as network information is known incompletely, including attacking according to node degree, and the third is intentional attack within network information is known, including attacking according to functional importance and greedy algorithm. Experimental results show that the network has a higher robustness against random attack; it has less robustness against intentional attack in the case of incompletely known information ; and it has more less robustness against intentional attack in the case of known information.Key

Key words: complexnetwork, functionchain, invulnerability, convergentnode

中图分类号: