Verifiable proxy re-encryption from indistinguishability obfuscation

Abstract

Proxy re-encryption scheme allows a semi-trusted proxy to re-encrypt ciphertexts of a client into ciphertexts of a receiver. However, the proxy might not as honest or reliable as supposed in practice. Ohata et al. [16] recently introduced a new functionality for proxy re-encryption with verifiability of the re-encryption soundness. However, careful inspection reveals that the construction in their work can not resist against normal collusion attack. Specifically, if the proxy and the receiver collude, the master key of the client will be leaked. We consider this as a serious weakness. Moreover, the storage of keys for a receiver in that work will increase linearly with the number of clients. In this paper, we present a novel generic scheme for verifiable proxy reencryption from indistinguishability obfuscation. It can ensure the security of master secret key even when the proxy and the receiver collude. In addition, our scheme possesses the advantage that any receiver’s key storage will remain constant, no matter how many clients he deals with. Furthermore, the re-encryption mechanism in our construction is very succinct in that the size of re-encrypted ciphertext relies only on the size of the encrypted message and the security parameter, compared to that in [16], which relies on the size of the original ciphertext and the receiver’s public key as well.