Author:

Abstract:

In today's society, privacy is subject to a lively debate. The growing connectivity and new technologies make linking and profiling easier and more accurate. Hence, the protection of privacy becomes a necessity unless we believe that our privacy is lost. Therefore, it is important to work on solutions that can improve our privacy.In this dissertation, we start with a critical assessment of electronic identity technology currently deployed, in particular the Belgian electronic identity card. The results clearly show that the protection of privacy is inadequate, especially when the card is used across both the public and private domains.Anonymous credential systems promise an alternative, supporting privacy and strong authentication. Unfortunately, these credential systems are still mainly a research topic, and have not yet found their way towards the general public. A major drawback of anonymous credential systems is that they are considerably more complex than the technologies currently used. In this dissertation, we provide a solution based on mobile devices as a platform, possibly extended with a secure element, for hosting the anonymous credential system. Another issue with anonymous credential systems is the lack of an efficient and practical revocation strategy. Multiple schemes have been presented, but their complexity is much more involved than revocation schemes used in traditional PKI-based systems. We present a pragmatic assessment of revocation schemes for anonymous credential schemes, of which some have been implemented as a basis for an in depth evaluation.Anonymous credential systems are complex systems supporting privacy-friendly transactions. However, to make sense, anonymous credentials should be accompanied by an infrastructure that supports privacy-preserving applications and new protocols will need to be designed. We analyze how simulation-based security models can be applied for building such larger complex systems. We provide a number of building blocks in order to help and guide protocol designers. As a validation of the framework and of our building-blocks, we model the concept of Oblivious Trusted Third Parties and present an actual implementation.