A Homomorphic Searching Scheme For Sensitive Data In NoSQL Database
Baohua Huang, Sheng Liang, Dongdong Xu, Zhuohao Wan · 2018
NoSQL database system is widely used in Big Data environment for its high performance and scalability. Due to the width of usage of NoSQL database, it is inevitable to store sensitive data in it. Utilizing homomorphic encryption, sensitive data can be encrypted and calculated in the form of ciphertext to preserve privacy. Although calculating on homomorphically encrypted data is possible, searching on these data is not simple for the calculating results are encrypted and the search engine doesn't know what data should be sent back to the requester. A homomorphic searching scheme is proposed to handle this problem. In this scheme, the searching function is divided into two parts named homomorphic comparing and homomorphic decryption. The homomorphic comparing is executed by the encrypted data holder, and the homomorphic decryption is executed by the key holder. With this scheme, encrypted data holder can only get encrypted data and calculate the encrypted differences of these data, and the key holder can only get the encrypted differences and decrypt them. A sample implementation of the scheme with MongoDB is given. Analysis and experimental results show that the proposed scheme is secure and practical.