Research On Fine-grained Data Integrity Check Method

Integrity is one of the important properties of data. Hash check is an important method to ensure data integrity. Because of the characteristic of hash function, any small change of the data will affect the usability and creditability of the whole data. In order to protect the integrity of data, we shouldn't restrict ourselves just in testing its integrity. To solve this problem, one ideal method is to separate the wrong data and keep using the other data. Fine-grained data integrity check becomes an important way to support the usability of data in information security field.In order to alleviate the problem of huge hash check data, on the basis of error correction coding, we can accomplish fine-grained data integrity check with a few Hash data by using crossed integrity check method. Compare with the method of using one hash for each data object, the crossed integrity check method incarnates as compression. Now, the investigators have constructed several kind of fine-grained data integrity indication code. One kind of these indication codes is called complex rotary integrity indication code. This code can indicate sever errors of data well and truly. But when facing with more errors, this code has low compression ration, which couldn't meet with the actual requirement.Based on the existing theory and tool, using the Finite Projective Geometry theory, the author proposes a novel multi-error integrity indication method, which is called Finite Projective Integrity Indication Code. In this code, each hash in correspondence with a point of the finite projective space and each data object in correspondence with a line of the space. This code generates uniformly crossed hash data by using the correlative relation of the point and line. This code can find all the errors well and truly if only the user set the parameters properly, as a result, the finite projective integrity indication code could mitigate the disastrous effect that all the data become invalid caused by accidental errors or intentional forging modification. The code has high compression ratio, which can save storage space and network transmitting bandwidth efficiently.The author develops a tool for fine-grained data integrity check based on the existing Finite Projective Integrity Indication Code. This tool has three modules, including hash data generating module, hash data storing module and hash check module. This tool can accomplish fine-grained data integrity check.