Research On Inverted Index For Ciphertext In Cloud Environment

With the rapid development and reform of information technology, digital information grows exponentially and full-text retrieval technology is applied more and more widely. However, the demand of information security increase substantially since cloud computing and Internet evolve quickly and data is migrated to the cloud environment gradually. Faced with the untrusted cloud service, users need to ensure the integrity and privacy of data. Data encryption can effectively guarantee the security of data, but how to retrieve ciphertext in the cloud environment is becoming hot research topic at home and abroad. This thesis is mainly to study how to construct efficient and safe dynamic inverted index for ciphertext in the cloud environment.After the analysis of existing plaintext retrieval mechanism, HDFS file system and ciphertext retrieval, a secure inverted index structure for ciphertext based on cloud storage is constructed by transforming traditional inverted index step by step. In order to resist the attack of semantic analysis and statistical attack, the entries are encrypted and the traditional index is transformed into segment index. Besides two-level index structure and the partitioning encryption mechanism are adopted to strike the balance between efficiency and security. Meanwhile, according to the features of HDFS, words are divided into two categories based on word frequency. Based on the structure of new ciphertext inverted index, the corresponding mechanism is designed, including construction, maintenance and retrieval.Based on the structure of ciphertext index designed above, an encryption solution of ciphertext index is also provided, especially the encryption algorithm. Take the security and key management into consideration, this thesis presents an efficient scheme of searchable public key encryption with conjunctive keyword based on parings. The improved algorithm only use two bilinear operators, which reduces the computational expense greatly, and use two pair of public and private keys owned by server and client respectively,which guarantee the security. Then this thesis presents a cipher-text searchable encryption application system in the cloud storage and the framework of the whole system is provided, including client, cloud server and cloud storage.At last, this thesis analyzes the performance and security about the improved searchable public key encryption algorithm. A performance testing is also conducted to verify the feasibility of this scheme. The inverted index structure of ciphertext based on cloud storage designed by this thesis strikes a balance between security and performance well.