Overview
In today's world, security policies are used in almost every systems to make decisions regarding access-control, authentication, etc. However, these security policies are almost invariably defined using bi-valent (two-valued) traditional logic. Therefore, decisions have to be either true or false, with nothing in between.
However, real life is often functioning in gray, i.e. there are many possibilities due to lack of information, uncertainty, and imprecisions in data. Often, we have situations where the trust is uncertain, the data provided to the policy engine are imprecise, and therefore traditional logic is unable to handle such situations.
In this project, we introduce FlexPolicies - a flexible framework of building security policies that can handle uncertainty and imprecisions of real life. FlexPolicies is intended to allow vagueness, uncertainty and imprecision in data, and to ensure trust management accordingly.
Currently, the project is a work-in-progress.
Collaboration
Publications
- Ragib Hasan and Marianne Winslett, "Flexpolicies: Flexible Security Policies with Vagueness and Uncertainty", manuscript under preparation.