Paper 2019/1193 Original: eprint.iacr.org/2019/1193 ↗

Security models for everlasting privacy

Abstract

We propose security models for everlasting privacy, a property that protects the content of the votes cast in electronic elections against future and powerful adversaries. Initially everlasting privacy was treated synonymously with information theoretic privacy and did not take advantage of the information available to the adversary and his behavior during or after the election. More recent works provided variations of the concept, limiting the view of the future adversary to publicly available data. We consider an adversary that potentially has insider access to private election data as well. We formally express our adversarial model in game based definitions build on top of a generic voting scheme. This allows us to define a stronger version of everlasting privacy and contrast the two main proposals to achieve it, namely perfectly hiding commitment schemes and anonymous channels.