Research On Key Technologies Of Relevance Deletion For Cloud Storage

With the development of cloud storage,more and more people store data in cloud storage so that they can save the cost of the purchase of additional storage devices.So how to ensure the privacy and integrity of data in the cloud storage has become a security issue that people need to consider now.Particularly,when the user requests to delete the data in the cloud storage,how to ensure that the data stored in the cloud storage device is removed that the deleted data can not be accessed any more.In order to solve the above problems,this paper firstly introduces the cryptography and the related techniques of data deletion.After studying and analyzing the existing assurance data deletion scheme,we find that the existing schemes do not pay attention to the problem of file association in cloud storage.Files stored in the cloud storage are not isolated,files stored in the cloud storage may be related to other files in cloud storage on the content or other aspects.Each file stored on the cloud will have many corresponding copy.There is also a relation between copies of the same file.If the user delete files and they doesn`t delete the file associated with this file,cloud storage providers may be dishonest by stealing associated files with the original file to restore the original file,which will lead to user information leakage.In order to solve the above problem,I propose a relevance deletion scheme for cloud storage.The main idea of this scheme is when user wants to delete the file stored on the cloud,this scheme give user feedback about which file on the cloud is related with the file that user want to delete through analyzing file content and how user use file on the cloud.Then user can decide the file list they want to delete on their own.So that this scheme can help user delete file..In this thesis,I also need to solve the problem of assurance deletion.After the user removes the data from the cloud,it needs to be sure that the cloud storage is honest about whether they have deleted the cloud file,rather than just sending a successful fake feedback to the user.To solve this problem,we first study the hourglass model.Then,we propose a scheme called assurance deletion based on hourglass scheme.The scheme is based on how the hourglass model ensure the correct encoding of cloud files.