Evaluating Security Policies in Pervasive Mobile Environments Using Context Information

Due to both the number of entities and the nature of the interactions and collaborations amongst them, conventional security models are inadequate for regulating access to data and services in a pervasive mobile computing environment. Since many of these interactions occur between entities that have not interacted with each other previously, new security paradigms rely on context information in order to arrive at a security and decision. However, these new systems fail to take into account the variability, correlation and uncertainty of the context variables composing a security policy when making a security decision. In this paper, we propose a Monte Carlo based framework to evaluate security policies that are based on the changes in multiple context variables. In this framework, context variables are modeled and risk in security decisions is measured.