|
A key property of modern day network environments such as the Internet is the possibility of multiple processes running simultaneously, concurrently and unaware of each other. However, the same property also allows an attacker for a coordinated attack in which an adversary controls many parties, interleaving the executions of the various protocol instances and creating rogue interactions between protocols. With changing network environments and new-emerging paradigms such as cloud computing, we need to assess the threat model in order to capture a broader class of attacks. This research project identifies strong notions of security to capture previously unforeseen threats and the intrinsic cryptographic complexity of constructing protocols that are secure against such threats.
This project focuses on a strong form of corruption model referred to as adaptive corruption, which considers an adversary that can hijack a host anytime during the course of computation. This models hacking attacks where an external attacker breaks into parties' machines in the midst of a protocol execution, and captures additional threats. This research identifies the minimal cryptographic assumptions to design protocols that are secure against adaptive corruption in the concurrent setting with and without setup. The research focuses on black-box construction wherever possible and relies on non-black-box constructions as a stepping stone towards establishing feasibility. The project bridges the gap between theory and practice, by creating a wiki that will enable a web-based interaction to retrieve relevant information regarding current techniques for cryptographic constructions and security notions associated with these constructions.
|
TWC: Small: Understanding the Complexity of Concurrent Security