Password-Authenticated Public-Key Encryption
Password-Authenticated Public-Key Encryption
复制标题
DOI:
10.1007/978-3-030-21568-2_22
复制
发表时间:
2019-06
期刊:
影响因子:
--
通讯作者:
Tatiana Bradley;J. Camenisch;Stanislaw Jarecki;Anja Lehmann;G. Neven;Jiayu Xu
中科院分区:
文献类型:
--
作者:
Tatiana Bradley;J. Camenisch;Stanislaw Jarecki;Anja Lehmann;G. Neven;Jiayu Xu
We introduce password-authenticated public-key encryption (PAPKE), a new cryptographic primitive. PAPKE enables secure end-to-end encryption between two entities without relying on a trusted third party or other out-of-band mechanisms for authentication. Instead, resistance to man-in-the-middle attacks is ensured in a human-friendly way by authenticating the public key with a shared password, while preventing offline dictionary attacks given the authenticated public key and/or the ciphertexts produced using this key.Our contributions are three-fold. First, we provide property-based and universally composable (UC) definitions for PAPKE, with the resulting primitive combining CCA security of public-key encryption (PKE) with password authentication. Second, we show that PAPKE implies Password-Authenticated Key Exchange (PAKE), but the reverse implication does not hold, indicating that PAPKE is a strictly stronger primitive than PAKE. Indeed, PAPKE implies a two-flow PAKE which remains secure if either party re-uses its state in multiple sessions, e.g. due to communication errors, thus strengthening existing notions of PAKE security. Third, we show two highly practical UC PAPKE schemes: a generic construction built from CCA-secure and anonymous PKE and an ideal cipher, and a direct construction based on the Decisional Diffie-Hellman assumption in the random oracle model.Finally, applying our PAPKE-to-PAKE compiler to the above PAPKE schemes we exhibit the first 2-round UC PAKE’s with efficiency comparable to (unauthenticated) Diffie-Hellman Key Exchange.