Ciphertext-policy attribute-based encryption (CP-ABE) schemes are an appropriate cryptographic technique to enable privacy along with access control in the cloud, but the existing CP-ABE schemes do not directly apply for big data because they have the issue of long ciphertext and long secret key size (LC-LS). To address LC-LS, the constant size ciphertext and secret key (CSC-S) schemes proposed. However, the existing CSC-S schemes suffer from the key escrow security issue and efficiency issue. To address both simultaneously, the authors propose an efficient escrow-free CP-ABE with constant size ciphertext and secret key (EEF-CPABE) for big data storage in the Cloud. The EEF-CPABE scheme reduces the encryption and decryption computation overhead by designing CSC-S. Further, the data owner generates the decryption global key to decrypt the data along with user secret key which solves the key escrow issue. Security and performance analysis demonstrate that the EEF-CPABE scheme resists against authority, and chosen plain-text attacks and more efficient than CSC-S schemes.
Turning Big data into tiny data: Constant-size coresets for k-means, PCA and projective clustering