Automatic identification of covert channels inside Linux kernel based on source codes

  • Authors:
  • Gaoshou Zhai;Yufeng Zhang;Chengyu Liu;Na Yang;MinLi Tian;Hengsheng Yang

  • Affiliations:
  • Beijing Jiaotong University, Hai Dian District, Beijing, China;Beijing Jiaotong University, Hai Dian District, Beijing, China;Beijing Jiaotong University, Hai Dian District, Beijing, China;Beijing Jiaotong University, Hai Dian District, Beijing, China;Beijing Jiaotong University, Hai Dian District, Beijing, China;Beijing Jiaotong University, Hai Dian District, Beijing, China

  • Venue:
  • Proceedings of the 2nd International Conference on Interaction Sciences: Information Technology, Culture and Human
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

Operating systems and their security are the foundation to underlay and assure the security of information systems. While effective analysis and processing mechanism for covert channels inside operating systems is one of the bottleneck issues to build up operating systems of high security levels. In this paper, some methods for identification of covert channels such as syntactic information flow method, semantic information flow method, shared resources matrix method and covert flow tree method are studied at first. Then a prototype is designed and implemented to identify covert channels inside Linux kernel based on source codes by the way of integrating above methods together. The prototype is tested on the source codes of Linux kernel 2.4.18, potential covert channels are identified and analyzed from the view of application scene so that some actual covert channels are distilled and it is verified about the validity of the prototype and the related method. Finally, research work in this paper is summarized and further research directions in future are pointed out.