Members of the defense committee:
- Prof. Dr Ulrich Sorger, University of Luxembourg, Chairman
- Prof. Dr Thomas Engel, University of Luxembourg, Vice-Chairman
- A-Prof. Dr Radu State, University of Luxembourg, Supervisor
- Dr Thomas Scherer, Telindus, Luxembourg, Member
- Dr Stefan Hommes, ZF Friedrichshafen, Germany, Member
- Prof. Dr Omar Cherkaoui, University of Quebec in Montreal, Canada, Expert in Advisory Capacity
Abstract:
Ever since the inception of Bitcoin, blockchains have been receiving more and more attention. Blockchains are interesting in that they allow the easy collaboration of distrusting entities at large scale. Hundreds to millions of users can easily interact within a given blockchain system, maintained by the consensus algorithm.
However, blockchains are still roaming the lands and trying to find their place. As such, they have lent themselves to support many different applications, from payments to healthcare to botnet reinforcement or detection. Blockchains are championed as the way to revolutionize many different sectors by challenging solutions based on previous technologies.
Specifically to this work, blockchains provide advantages related to security. Among them are auditable and immutable logs, user identification, verifiable data modifications according to the protocol and the consensus built among participants. The latter makes it hard for any adversary to modify data that the blockchain keeps. Nonetheless, all of them are properties that security critical applications such as access control can benefit from.
Therefore, in this thesis, blockchain-based access control is investigated along three axes.
First, access to a single logical resource, the blockchain itself, is examined. This is the simplest case and is analogous to granting and revoking access to a single room, protected by a community forming a consensus on who is allowed to enter and thus access the resource.
Second, access to multiple, distributed computing resources is considered. This builds on the previous discussion, by adding more objects that require access control. It is in the object owner’s interest to maintain access, as is the case in botnets.
Third, blockchain-based access control with respect to user data is discussed. A system is proposed that allows only the users themselves to modify the permissions on their data. This is a prime example of blockchain-based access control, because the permissions are decentralized, as is the blockchain itself.
This talk ends with a discussion of blockchain-based access control while considering the notions learned throughout this thesis.