DEVELOPMENT ON SECURITY MODELS FOR PROXY RE-ENCRYPTION SCHEMES

A proxy re-encryption is a type of public key encryption where a proxy is given the ability to transform ciphertext for one user into ciphertext for another user, without having access to either the plaintext or the decryption keys. Proxy re-encryption schemes, since their introduction in 1998 by Blaze et al. [3], are often proposed with specific security notions. We propose here to recall the most commonly used security notions and to propose standard security models for security analysis of proxy reencryption scheme. In this article, we consider the goals of confidentiality and nonmalleability, each in the context of a parametric family of attack models. For each of the resulting pairs of definitions, we prove either an implication or a separation from the other security definitions.