Chinese Journal of Computers   Full Text
Title	Stochastic Modeling and Evaluation for Network Security
Authors	LIN Chuang1) WANG Yang1) LI Quan-Lin2)
Address	1)(Department of Computer Science and Technology, Tsinghua University, Beijing100084) 2)(Department of Industrial Engineering, Tsinghua University, Beijing100084)
Year	2005
Issue	No.12(1943—1956)
Abstract &   Background	Network systems are becoming more complex and larger, and specifically network attacks and security destroy is also increasingly prevalent and multiple. Network security is being confronted with many significant challenges. Network security is one of the most important scientific issues in the world and has been an important factor which effects on social improvement and national develop strategy. In this paper, we provide a complete overview on stochastic models and evaluation techniques for network security. The authors indicate the present status and evolvement in the area of network security. The authors provide several effective and efficient methods for modeling and analysis of network security, and also analyze network survivability. Finally, the authors give some concluding remarks on new and challenging directions for future and potential research of network security. keywords network security； stochastic modeling； attack modeling； survivability analysis； evaluation technique background This research is partly supported by the National Basic Research Program of China (973 Program) under grant No.2003CB314804, the National Natural Science Foundation of China under grant No.90412012, 60218003, 60503052 and 60573144, and Special Research Foundation of Ph.D. Study in High School under grant No.20020003027. Network security is a rather important direction with many significant scientific and practical issues in the world, which is due to that network attacks are increasingly prevalent. Since it is hardly to build a completely secure system, how to quantify network security become more and more important. Inspired by the study of dependability evaluation, stochastic models and performance evaluation should lead to sound and promising methods in security evaluation. The authors’ main objectives are to provide methodology and techniques for quantifying network security, including stochastically analytic methods. In this paper, the authors introduce status and evolvement in the study of stochastic models and evaluation techniques of network security. Also, they give effective and efficient methods to model structure of attackers and cyber attack behavior, which are shown to be a key for security evaluation. Finally, they provide a number of directions for potential future research in the stochastic study of network security.