A Construction of Robustly Reusable Fuzzy Extractors over Blockchains

Fuzzy extractors (FEs) turn a noisy source of high entropy, such as biometric information, into a uniformly distributed key. FEs can extract cryptographically desirable keys without storing the keys. FE was first proposed by Dodis et al. [1], and it has been studied actively since. Moreover, a robust FE to achieve higher security and a reusable FE to achieve more functionality were proposed. Wen et al. constructed a robustly reusable FE (rrFE) that satisfies robustness and reusability simultaneously [2]. However, only few rrFEs have been proposed and all are based on the Common Reference String (CRS) model.Recently, Goyal and Goyal [3] defined blockchain formally as a model for use as components of cryptographic protocols. In this paper, we introduce an rrFE based on the blockchain proposed by Goyal et al. instead of the CRS model. In contrast to the CRS model, a blockchain trusts not a specific third party but a distributed system. Therefore, a protocol based on CRS may lose security by a betrayal of TTP. Meanwhile, our protocol tolerates malicious acts of parties as long as the majority are honest.