A Rankable Boolean Searchable Encryption Scheme Supporting Dynamic Updates in a Cloud Environment

At present, three problems exist in searchable encryption in cloud storage services: firstly, most traditional searchable encryption schemes only support single-keyword search while fail to perform Boolean searches; even if a few schemes support Boolean searching, the storage efficiency is also unsatisfactory. Secondly, most existing schemes do not support dynamic keyword updates, so the update efficiency is low. Thirdly, most existing schemes cannot meet all demands of users, to perform rankable searching over search files according to the importance of keywords. To solve these problems, a rankable Boolean searchable encryption scheme supporting dynamic updates in a cloud environment (RBDC) is proposed. By using Paillier and GM encryption algorithms, secure indices supporting dynamic updating are established. Based on applicable knowledge gleaned from cryptography and set theory, the indices of keyword intersections and the intersection search trapdoors are constructed to achieve multi-keyword Boolean search. With assistance of the SCP, score indices of each file are constructed according to the TF-IDF index, which allow ranking of files. Security analysis proofs that our scheme can ensure security in the known ciphertext model and the known background model. Experimental results prove that the scheme improves the search efficiency and the index storage efficiency.