Telecommunications Science ›› 2021, Vol. 37 ›› Issue (11): 75-85.doi: 10.11959/j.issn.1000-0801.2021212

• Research and Development • Previous Articles     Next Articles

SDN security prediction method based on bayesian attack graph

Yanshang YIN, Tongpeng SUO, Ligang DONG, Xian JIANG   

  1. School of Information and Electronic Engineering (Sussex Artificial Intelligence Institute), Zhejiang Gongshang University, Hangzhou 310018, China
  • Revised:2021-09-11 Online:2021-11-20 Published:2021-11-01
  • Supported by:
    Zhejiang Province Key Research and Development Program(2020C01079);Zhejiang Province Key Research and Development Program(2021C01036);The Natural Science Foundation of China(61871468);Zhejiang Provincial Natural Science Foundation of China(LY18F010006);Zhejiang Provincial Key Laboratory of New Network Standards and Application Technology(2013E10012);University Students' Scientific and Technological Achievements Promotion Project(1120KZN0220031G)

Abstract:

Existing researchers use threat modeling and security analysis system to evaluate and predict SDN (software defined network) security threats, but this method does not consider the vulnerability utilization of SDN controller and the location of devices in the network, so the security evaluation is not accurate.In order to solve the above problems, according to the probability of device vulnerability utilization and device criticality, combined with PageRank algorithm, a algorithm to calculate the importance of each device in SDN was designed; according to SDN attack graph and Bayesian theory, a method to measure the success probability of device being attacked was designed.On this basis, a SDN security prediction method based on Bayesian attack graph was proposed to predict the attacker's attack path.Experimental results show that this method can accurately predict the attacker's attack path and provide more accurate basis for security defense.

Key words: SDN security prediction, vulnerability utilization probability, attack graph, PR algorithm

CLC Number: 

No Suggested Reading articles found!