Research On Security Deduplication Technology Of Cloud Storage Encrypted Data

With the advent of the era of big data,a large amount of data makes users' storage resources increasingly tense.Cloud storage technology attracts people to store data on remote servers by virtue of its convenience,fast,low cost,and pay-as-you-go characteristics,in order to free up local memory space.Because the same data will be stored multiple times by different users,resulting in a decrease in the storage efficiency of cloud servers,data deduplication technology has begun to attract widespread attention from academia and enterprises.For plaintext data,the cloud server can directly delete the duplicate data and keep only one copy,but because the cloud server is semi-trusted,users usually protect the privacy of the data in an encrypted form,but due to the difference between the keys differences,so that the cloud server cannot distinguish whether different ciphertexts are from the same plaintext data.In addition,when the encrypted data is deduplicated,it will cause a series of security problems.Therefore,how to design a safe and reliable encrypted data deduplication scheme is of great research significance.In response to the problems in the current program,this article proposes two effective security deduplication solutions,the main content and contributions are as follows:1.A ciphertext deduplication scheme based on Bloom Filter is proposed.The design of deduplication detection labels in the scheme is based on the difficulty of discrete logarithm,which ensures that the detection tags will not leak any information about the content of the plaintext.At the same time,the Bloom Filter is introduced to use its own the advantages in time and space overhead improve the efficiency of the deduplication detection stage.During the key transfer process,no other users are required to participate online,and a trusted third-party server is used to securely transfer the encryption key of the initial uploader to subsequent users,while allowing subsequent users to perform data on the data stored by the initial uploader verify.The safety and experimental analysis show that the scheme proposed in this article has great performance advantages.2.This paper proposes a ciphertext deduplication scheme based on popularity division,in which the popularity is divided by the number of times the data is stored in the cloud server.The purpose is to reduce the storage overhead of the user and improve the efficiency of deduplication.In the system initialization stage,a threshold T is set,and non-popular data whose storage times are less than T is implemented with a reliable and secure encryption method,and popular data greater than T uses an efficient convergent encryption algorithm.Using the pairing feature of bilinear mapping to design deduplication detection tags,it can detect whether the same data is repeatedly uploaded under the premise of ensuring data privacy.Introducing the Proof of Ownership protocol,non-data owners cannot pass the identity verification of the cloud server,and no third-party server is required to participate,making the solution more feasible.Using the modular arithmetic mechanism,the key of the data is distributed to all authorized users in an offline form.The security analysis and experimental performance show that the scheme in this paper costs less computational overhead when generating deduplication detection labels and key transmission.