Efficient Negative Databases From Cryptographic Hash Functions

A negative database is a privacy-preserving storage system that allows to efficiently test if an entry is present, but makes it hard to enumerate all encoded entries. We improve significantly over previous work presented at ISC 2006 by Esponda et al. [9], by showing constructions for negative databases reducible to the security of well understood primitives, such as cryptographic hash functions or the hardness of the Discrete-Logarithm problem. Our constructions require only O(m) storage in the number m of entries in the database, and linear query time (compared to O(l center dot m) storage and O(l center dot m) query time, where l is a security parameter.) Our claims are supported by both proofs of security and experimental performance measurements.