Improving privacy in identity management systems for health care scenarios

Abstract

Privacy is a very complex and subjective concept with different meaning to different people. The meaning depends on the context. Moreover, privacy is close to the user information and thus, present in any ubiquitous computing scenario. In the context of identity management (IdM), privacy is gaining more importance since IdM systems deal with services that requires sharing attributes belonging to users’ identity with different entities across domains. Consequently, privacy is a fundamental aspect to be addressed by IdM to protect the exchange of user attributes between services and identity providers across different networks and security domains in pervasive computing. However, problems such as the effective revocation consent, have not been fully addressed. Furthermore, privacy depends heavily on users and applications requiring some degree of flexibility. This paper analyzes the main current identity models, as well as the privacy support presented by the identity management frameworks. After the main limitations are identified, we propose a delegation protocol for the SAML standard in order to enhance the revocation consent within healthcare scenarios.