Research And Implementation Of Ciphertext Full-text Retriveal Technology In Cloud Storage Environment

With the rapid development of cloud computing,the data on the network is exploding,cloud storage as the important application of cloud computing,it has been widely used due to its low service fee,convenience and strong extensibility,etc.Although the application of cloud storage solves the problem of storing and interacting massive data,because the cloud service provider is semi trusted,the data owner loses the absolute control of the data,and the security problem is increasingly prominent.Therefore,it is a mainstream scheme to encrypt data before uploading data to protect the privacy of data,which results in the traditional clear text retrieval operation is difficult to carry on,how to achieve efficient full-text retrieval over encrypted data under the premise of protecting the privacy of the user and data,and support access control at the same time has become an urgent problem to be solved.Therefore,this paper mainly focuses on the research of privacy-preserved full-text retrieval technology over encrypted data in cloud storage environment,the main contents include the following:1.This paper designs an improved ciphertext full-text retrieval scheme based on privacy protection,aiming at the traditional full-text index build strategy is not applicable in the environment of cipher text,using IKAnalyzer,maximum word length restriction and punctuation mark to control the semantic and index file size of index words,effectively reduce the physical space occupied by index files,lighten the communication load between users and the cloud server.Finally,according to the index word length to calculate the relevancy score between keywords and documents and realize the retrieval results are sorted by the score.2.This paper designs an improved index words processing algorithm based on user identity information and MD5 hash algorithm,and realizes access control strategy for "multi-owner multi-user" in cloud storage environment,while guaranteeing the security of index words.In addition,to resist the only ciphertext attack,known plaintext attack and select plaintext attack,the position of the ciphertext index word is disturbed by pseudo random algorithm.3.Based on the full-text indexing strategy over encrypted data,the prototype of full text retrieval system is designed and implemented,which consists of client and server side,the main function of the system include document encryption and decryption,indexing construction(word segmentation,index word encryption,index generation),full-text retrieval of ciphertext,user management,document upload and download,and the document content is handled by the symmetric encryption AES algorithm,and the index words are processed using the improved hash algorithm,so that the content of the two is not correlated to ensure the security of data.Finally,the prototype of the full-text retrieval system over encrypted data is tested and the experimental results are compared,including the index construction time,the index file size,retrieval time,the result sorting,and the experiment result is analyzed,the result verifies the system's efficiency,security and scalability.