International Association
for Cryptologic Research

Distributed randomness is very useful for many applications, such as smart contract, the proof-of-stake-based blockchain, elliptic curve generation and lottery. Randomness beacon protocols are proposed, which are aimed at continuously distributed randomness generation. However, a reliable source of distributed randomness is gained with difficulty because of Byzantine behavior which may lead to bias for distributed randomness. These Byzantine behaviors include, but not limited to, the “last actor” problem, DoS attack, and collusion attack. Various cryptography schemes have been used to generate distributed randomness. Current constructions face challenging obstacles due to high communication overheads and collusion problems. Given these barriers, we propose a new protocol that is the first precept to utilize attribute-based encryption for distributed randomness (ABERand). Compared to existing state- of-the-art public distributed randomness protocols, ABERand possesses distinguished scalability, security and efficiency. More specifically, we resolve the “last actor” problem and make ABERand an intensive output randomness beacon with com- munication complexity O(n2), computation complexity O(1), verification complexity O(n), and communication complexity O(n) of nodes adding/removing.