The notion of Public-key Encryption with Keyword Search (PEKS) was first proposed by Boneh et al. in 2004. However, almost all PEKS schemes cannot resist offline Keyword Guessing Attacks (KGA). To address this issue, Huang and Li introduced the notion of Public-key Authenticated Encryption with Keyword Search (PAEKS) in 2017. Recently, Qin et al. put forward a new security model named Multi-Ciphertext Indistinguishability (MCI), in which an adversary aims to distinguish two tuples of ciphertexts. They found that Huang and Li’s scheme cannot achieve MCI-security, so they proposed a new scheme which is able to achieve MCI-security. Furthermore, Qin et al. refered to another security model named Multi-Trapdoor Indistinguishability (MTI). They stated that the future work direction is to design a scheme which can achieve both MCI-security and MTI-security. In this paper, we present a new PAEKS scheme and prove that it is capable of achieving MCI-security and MTI-security simultaneously with the help of random oracles. Finally, we compare our scheme with Huang and Li’s PAEKS scheme and Qin et al.’s PAEKS scheme using PBC library and provide experimental results. It turns out that our scheme achieves a higher security level with a little more cost.

Boneh等人首先提出了带有关键字搜索的公钥加密（PEKS）的概念。在2004年。但是，几乎所有PEKS方案都无法抵抗脱机关键字Guessing Attacks（KGA）。为了解决这个问题，Huang和Li在2017年引入了带有关键字搜索的公共密钥身份验证加密（PAEKS）的概念。提出了一种新的安全模型，称为多密文不可区分性（MCI），其中的对手旨在区分两个密文元组。他们发现Huang和Li的方案无法实现MCI安全，因此他们提出了一种能够实现MCI安全的新方案。此外，秦等。指的是另一个名为“多重陷阱门不可区分性（MTI）”的安全模型。他们表示，未来的工作方向是设计一种既可以实现MCI安全又可以实现MTI安全的方案。在本文中，我们提出了一种新的PAEKS方案，并证明它能够在随机预言的帮助下同时实现MCI安全性和MTI安全性。最后，我们使用PBC库将我们的方案与Huang和Li的PAEKS方案以及Qin等人的PAEKS方案进行了比较，并提供了实验结果。事实证明，我们的方案以更高的成本实现了更高的安全级别。