CSAR: A Practical and Provable Technique to Make Randomized Systems Accountable

We describe CSAR, a novel technique for generating cryptographically strong, accountable randomness. Using CSAR, we can generate a pseudo-random sequence and a proof that the elements of this sequence up to a given point have been correctly generated, while future values in the sequence remain unpredictable. CSAR enables accountability for distributed systems that use randomized protocols. External auditors can check if a node has deviated from its expected behavior without learning anything about the node’s future random choices. In particular, an accountable node does not need to leak secrets that would make its future actions predictable. We demonstrate that CSAR is practical and efficient, and we apply it to implement accountability for a server that uses random sampling for billing purposes.

• Type: Conference
• Year: 2009
• Where: NDSS
• Authors: Michael Backes, Peter Druschel, Andreas Haeberlen, Dominique Unruh
• Keywords: Accountable Randomness, Computer Networks, CSAR Enables Accountability, NDSS 2009, Pseudo-random Sequence