BLP模型是最为经典的强制访问控制模型,它基于一个信息流策略,通过允许低安全级别到高安全级别的单向信息流动,来实现多级的强制访问控制.但对实际中普遍存在的高安全级别到低安全级别的下向信息流动,BLP模型并没有给出理论解决方法.本文对BLP模型提出一种量化的分析方法以分析下向信息流动的安全性.这一方法基于Shannon信息论中条件熵的概念,提出用安全门限来度量信息安全的方法.特别地,这一分析方法给出了下向信息流安全的条件,并证明了该条件下系统仍保持其保密性.
BLP model is the most classical model in mandatory access control, it is based on a multi-level information flow policy to implement access control. However, the information flow policy of BLP model does not support downward information flow from high security level to low security level. This paper proposed an analysis method based on the conditional entropy which gives quantitative analysis to the security of BLP model. With this analysis method, we can define the confidentiality of an information system with secure thresholds defined by confidential information and its conditional entropy. We proved that under certain condition, a downward information flow won't let the system be not secure.