Dynamic database filling method based on outlier detection technology and bitmap index

Abstract

The invention provides a dynamic database filling method based on outlier detection technology and bitmap index. The method includes the following steps: firstly, a filling database is generated, and pseudo-files that meet the conditions are selected to fill the real database by using outlier detection technology, and the corresponding relationship between files and keywords is represented by a bitmap index. Then dynamically modify the filling database according to the update operation. If the update operation is to add, if the number of files corresponding to the keyword is the largest after the update, fill other keywords according to this number; otherwise, reduce the filling of the keyword until The keyword's frequency is reduced to the same as before the update operation. If the update operation is a delete, then increase the padding of the key until the frequency of the key is increased to the same as before the update operation. This padding method is suitable for dynamic searchable encryption schemes, which can prevent data leakage, such as resisting attacks based on keyword frequency inference, and improve the security of the scheme.

Description

technical field [0001] The invention relates to the technical field of searchable encryption, in particular to a dynamic database filling method based on outlier detection technology and bitmap index. Background technique [0002] With the rapid development of the network and the advent of the era of big data, local storage and computing resources cannot meet people's needs, and cloud storage and computing technologies have developed rapidly. In order to ensure data security, the data is usually encrypted and uploaded to the cloud server. However, performing searches on ciphertext data is very difficult. Therefore, Searchable Encryption (SE, Searchable Encryption) technology is proposed. Early SE solutions were static and did not support dynamic database updates, such as adding or deleting uploaded files, which limited the application of SE solutions. In order to solve this problem, Dynamic Searchable Symmetric Encryption (DSSE, Dynamic Searchable Symmetric Encryption) te...

AI Technical Summary

Problems solved by technology

[0003] However, there is a problem of data leakage in the DSSE scheme. For example, attacks based on keyword frequency inference (such as counting attacks) can easily identify query keywords from access patterns, that is, query results. Database filling technology is a simple and effective solution to the above problems. measures, and is suitable for real large datasets

In recent years, the research focus of filling algorithms is mainly on how to design efficient filling schemes. However, none of the existing filling algorithms are suitable for dynamic databases.

Images