Cryptanalysis and improvement of "group public key encryption scheme supporting equality test without bilinear pairings"

Public key encryption with equality test (PKEET) is a novel primitive which supports equality comparisons on two encrypted messages. Currently, most of the existing PKEET schemes are based on bilinear pairing and require heavy computational overheads. To address this issue, Shen et al. recently proposed an efficient group public key encryption supporting equality test without bilinear pairings scheme. Compared with other schemes, their scheme reduces the usage of expensive bilinear pairing operations and enjoys higher computation efficiency. They claimed that their scheme achieved one-wayness security in the random oracle model and resisted offline message recovery attack. In this letter, we analyze Shen et al.'s scheme through two concrete attacks and demonstrate that their scheme can not support the above two security requirements. An improved scheme is provided to overcome the security vulnerabilities in their scheme. Performance analysis shows that our improved scheme has certain advantages in both computation overhead and storage overhead.