面向云存储的关联性删除关键技术研究

发布时间：2019-01-01 20:15

【摘要】：云存储技术的发展使将文件存储在云端的人们的数量变得越来越大。人们将数据存储在云端的目的是为了节约存储设备的花费。那么如何保证数据在云存储中的隐私性和完整性就成了人们现在所需考虑的安全问题。尤其是,当用户请求删除云存储中的数据时,如何保证该数据确实从云存储的存储设备中被删除从而无法被他人获取数据内容的问题。针对上述安全问题,本文首先介绍了密码学和现有的云存储中关于数据删除的相关技术。研究和分析了现有的数据确定性删除方案之后,现有的方案对云存储中文件的关联性问题关注的比较少。云存储中保存的文件之间并不是孤立的,用户存储在云存储中的文件可能在内容或者其他方面存在关联。不光是云存储上的文件之间,用户的文件存储在云上,每个文件都会存在其对应的副本。同一个文件的副本之间也存在着关联性。如果用户在删除文件的同时,不将云存储上与该文件关联的其他文件删除,不诚实的云存储提供商可能会通过窃取关联文件的方法还原用户的原文件,导致用户信息的泄露。为了解决上一段提出的问题,本文提出了面向云存储的关联性删除方案。该方案主要思想是在用户需要对云端文件执行删除操作时,方案通过对文件内容以及用户操作云文件行为的分析,将与待删除文件内容关联、访问关联的文件列表反馈给用户,让用户决定最终删除文件的列表,从而辅助用户达到确定性删除的目的。本文还需要解决数据的确定性删除问题。用户在对云端的数据进行删除操作之后,需要确定云存储诚实的将云端的文件删除了,而不是只给用户发送了一个删除成功的伪造的反馈。为了解决这个问题,我们首先研究了沙漏模型。然后借鉴沙漏模型确定云端文件被正确编码的思想提出了基于沙漏模型的确定性删除方案。
[Abstract]:With the development of cloud storage technology, more and more people store files in the cloud. People store data in the cloud to save money on storage devices. Therefore, how to ensure the privacy and integrity of data in cloud storage becomes a security issue that people need to consider now. In particular, when the user requests to delete the data in cloud storage, how to ensure that the data is deleted from the storage device of cloud storage, so that the data content can not be obtained by others. In view of the above security problems, this paper first introduces the cryptography and the existing cloud storage technology about data deletion. After studying and analyzing the existing data deterministic deletion schemes, the existing schemes pay less attention to the problem of file correlation in cloud storage. Files stored in cloud storage are not isolated, and files stored by users in cloud storage may be related in content or other aspects. Not only the files stored in the cloud, but also the files of the users are stored in the cloud, and each file has its corresponding copy. There is also a correlation between copies of the same file. If the user does not delete the other files associated with the file while deleting the file, the dishonest cloud storage provider may restore the user's original file by stealing the associated file, resulting in the disclosure of user information. In order to solve the problem mentioned in the previous paragraph, this paper proposes an association deletion scheme for cloud storage. The main idea of the scheme is that when the user needs to delete the cloud file, the scheme will associate the file content with the file content to be deleted and access the associated file list to the user by analyzing the file content and the behavior of the user operating the cloud file. Let the user decide to delete the list of files, thus assisting the user to achieve the purpose of deterministic deletion. This paper also needs to solve the problem of deterministic data deletion. After deleting the data from the cloud, the user needs to make sure that the cloud store honestly removes the files from the cloud, instead of sending the user a false feedback to delete successfully. In order to solve this problem, we first study the hourglass model. Then, a deterministic deletion scheme based on the hourglass model is proposed to determine the correct coding of cloud files by using the hourglass model.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP333

【参考文献】