Fully Private Revocable Predicate Encryption.
Juan Manuel González Nieto, Mark Manulis, Dongdong Sun
Abstract
Juan Manuel González Nieto, Mark Manulis, Dongdong Sun
Abstract
We introduce the concept of Revocable Predicate Encryption (RPE), which extends current predicate encryption setting with revocation support: private keys can be used to decrypt an RPE ciphertext only if they match the decryption policy (defined via attributes encoded into the ciphertext and predicates associated with private keys) and were not revoked by the time the ciphertext was created. We formalize the notion of attribute hiding in the presence of revocation and propose an RPE scheme, called AH-RPE, which achieves attribute-hiding under the Decision Linear assumption in the standard model. We then present a stronger privacy notion, termed full hiding, which further cares about privacy of revoked users. We propose another RPE scheme, called FH-RPE, that adopts the Subset Cover Framework and offers full hiding under the Decision Linear assumption in the standard model. The scheme offers very flexible privacy-preserving access control to encrypted data and can be used in sender-local revocation scenarios.
OpenAlex reports 9 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
We introduce the concept of Revocable Predicate Encryption (RPE), which extends current predicate encryption setting with revocation support: private keys can be used to decrypt an RPE ciphertext only if they match the decryption policy (defined via attributes encoded into the ciphertext and predicates associated with private keys) and were not revoked by the time the ciphertext was created. We formalize the notion of attribute hiding in the presence of revocation and propose an RPE scheme, called AH-RPE, which achieves attribute-hiding under the Decision Linear assumption in the standard model. We then present a stronger privacy notion, termed full hiding, which further cares about privacy of revoked users. We propose another RPE scheme, called FH-RPE, that adopts the Subset Cover Framework and offers full hiding under the Decision Linear assumption in the standard model. The scheme offers very flexible privacy-preserving access control to encrypted data and can be used in sender-local revocation scenarios.
Key concepts: Revocation, Computer science, Ciphertext, Encryption, Attribute-based encryption, Semantic security, Predicate (mathematical logic), Theoretical computer science