Multi-dimensional space information hiding method and system

Abstract

The invention relates to an information hiding method and system of a multi-dimensional space, and belongs to the technical field of information security, at least one subclass set is obtained according to an OOXML composite document set, documents in the subclass set are ranked according to a set rule to form a document matrix, the document matrix comprises a plurality of transverse rows and a plurality of vertical columns, and the transverse rows and the vertical columns are arranged in the document matrix. At least one of the number of the transverse rows and the number of the vertical columns is a positive integer power of 2; binary coding is carried out on the transverse rows and the vertical columns to obtain row codes of the transverse rows and column codes of the vertical columns, binary code words of all documents in the document matrix are combinations of the corresponding row codes and column codes, and the mapping relation between the documents and the binary code words is obtained; and converting the to-be-hidden information into a binary character string, and hiding the binary character string in the sub-category set through the mapping relationship. Compared with an existing single space mapping strategy, content features can be dynamically and randomly selected to construct the mapping relation, and the problems of hidden capacity and mapping flexibility of an existing algorithm are solved.

Description

Technical Field

[0001] The invention relates to a method and system for hiding information in a multi-dimensional space, belonging to the technical field of information security. Background Art

[0002] Existing carrier-free information hiding technology generally constructs a mapping between carrier difference characteristics and secret information under a single type of carrier set based on the hidden space formed by the differences between carriers to achieve information hiding. There is usually a problem of limited hiding capacity. At the same time, in order to achieve a 100% hiding success rate, it often requires the support of large-scale data sets.

[0003] The patent document with the publication number CN111027080A discloses an information hiding method and system based on the position arrangement order of the data area of ​​the source file of the OOXML compound document, which mainly includes: step 1: establishing and maintaining the OOXML compound document library C, the OOXML compound document library C is used to store the mapping between the secret information and the position arrangement order of the data area of ​​the source file of the OOXML compound document, and the OOXML compound document that conforms to the position arrangement order of the data area of ​​the source file; step 2: the sender searches the OOXML compound document library C according to the secret information, obtains the position arrangement order of the data area of ​​the source file of the OOXML compound document corresponding to the mapping, and further obtains the corresponding OOXML compound document d, and sends d to the receiver; step 3: the receiver obtains the OOXML compound document d, obtains the position arrangement order p of the data area of ​​the source file of the OOXML compound document corresponding to d; and queries the OOXML compound document library C according to p and d to obtain the mapped secret information. This invention carries out information hiding for the redundancy caused by the position arrangement of the data area of ​​the source file of the OOXML compound document, but there is still a problem that the hiding capacity is limited and the hiding capacity cannot meet the needs of information hiding. Summary of the invention

[0004] The purpose of the present invention is to provide a method and system for hiding information in a multi-dimensional space, so as to solve the problem that the existing hiding capacity cannot meet the information hiding requirements.

[0005] To achieve the above object, the solution of the present invention includes:

[0006] The present invention provides a method for hiding information in a multidimensional space. At least one subcategory set is obtained according to an OOXML composite document set. Documents in the subcategory set are sorted according to set rules to form a document matrix, wherein the document matrix includes a plurality of horizontal rows and a plurality of vertical columns, and at least one of the numbers of the horizontal rows and the vertical columns is a positive integer power of 2. The horizontal rows and the vertical columns are binary-coded to obtain a row code for each horizontal row and a column code for each vertical column, and the binary codeword of each document in the document matrix is ​​a combination of its corresponding row code and column code, thereby obtaining a mapping relationship between the document and the binary codeword. The information to be hidden is converted into a binary string, and the binary string is hidden in the subcategory set through the mapping relationship.

[0007] The present invention establishes a mapping relationship between documents and binary codewords in a subcategory set, which is embodied in a codeword matrix of equal-length binary codewords. For example, the codewords in any row and any column of the codeword matrix are combined to form a 4-bit binary codeword, and one 4-bit binary codeword corresponds to one document. When hiding secret information, for example, a 20-bit binary string that embodies the secret information is divided into 4-bit segments to obtain 5 segments of 4-bit binary codewords. The row and column relationship of the 5 segments of 4-bit binary codewords and their corresponding documents are found in the codeword matrix. The sender provides the secret information to the receiver through a public channel in the form of a document set, thereby realizing information hiding.

[0008] The documents in the subcategory set of the present invention are sorted according to the set rules to form a document matrix, and the rows and columns where each document in the document matrix is ​​located are row-coded and column-coded. The combination of the row code and column code corresponding to the document is the binary codeword of the document. The mapping relationship between the binary codeword and the document is embodied in the codeword matrix, and the information to be hidden (secret information) is converted into a binary string, and the binary string is hidden in the subcategory set through the mapping relationship. The present invention obtains at least one subcategory set according to the OOXML compound document set, and also sorts the documents in the subcategory set according to the set rules to form a document matrix, and then obtains the mapping relationship between the document and the binary codeword according to the relationship between the row code, the column code and the document. Compared with the prior art that establishes a mapping relationship by a single fixed source file data area position arrangement order for the OOXML compound document, the present invention flexibly utilizes the advantages of the OOXML compound document, provides a flexible mapping relationship between the document and the binary codeword, and improves the hiding capacity of information hiding.

[0009] Furthermore, according to the differences in source file types and / or differences in the order of source file locations, the OOXML compound document set is divided into several subcategory sets, each subcategory set corresponds one-to-one to the recipient identity, and for different recipients, the mapping relationship corresponding to the corresponding subcategory set is selected for information hiding.

[0010] The present invention divides the OOXML compound document set into several subcategory sets according to source file differences (differences in source file types and / or differences in source file position order), each subcategory set corresponds one-to-one to a receiver identity ID (recipient identity or receiver identity information), and for different receivers, a mapping relationship corresponding to the subcategory set is selected flexibly and reliably for information hiding.

[0011] The present invention obtains a mapping relationship (mapping algorithm) based on source file differences and multiple spaces, which not only retains the dynamic selection characteristics of the organizational structure based on the identity ID, but also fully considers the multiple space redundancy characteristics corresponding to the multiple content features of the OOXML compound document. Compared with the existing single space mapping strategy, this method can dynamically and randomly select content features to construct a mapping relationship. While improving the algorithm's hidden capacity, it also gives full play to the advantages of the OOXML compound document's complexity, solving the problems of the existing algorithm's hidden capacity and mapping flexibility.

[0012] Furthermore, the differences in source file types include differences in source file types of the same type of OOXML compound documents and differences in source file types of different types of OOXML compound documents; the differences in source file position order include differences in source file position order of the same type of OOXML compound documents and differences in source file position order of different types of OOXML compound documents.

[0013] Furthermore, the rule sorting is set as follows: the documents in the subcategory set are sorted in rows or columns in a set order according to the musical note feature of the pronunciation of the first character of the document.

[0014] Furthermore, the musical note feature is the first letter of the pinyin corresponding to the pronunciation of the character, and the setting order is letter a to letter z.

[0015] Furthermore, when there are documents with the same first note feature in the subcategory set, the documents with the same first note feature are classified into the corresponding note set, and this set is sorted according to the second note feature, and so on, until all documents in the subcategory set are sorted.

[0016] The present invention provides a method for corresponding sorting of documents with the same first note feature, so that documents and code words (binary code words) correspond one to one, and provides a reliable mapping relationship between documents and binary code words for information hiding.

[0017] A multi-dimensional space information hiding system includes a processor, and the processor is used to execute instructions for implementing the multi-dimensional space information hiding method as described above.

[0018] The present invention establishes a mapping relationship between documents and binary codewords in a subcategory set, which is embodied in a codeword matrix of equal-length binary codewords. For example, the codewords in any row and column of the codeword matrix are all 4-bit binary codewords, and one 4-bit binary codeword corresponds to one document. When hiding secret information, the 20-bit binary string embodying the secret information can be divided into 4-bit segments to obtain 5 segments of 4-bit binary codewords. The row and column relationship of the 5 segments of 4-bit binary codewords and the corresponding documents are found in the codeword matrix. The sender provides the secret information to the receiver through a public channel in the form of a document set, thereby realizing information hiding. The present invention obtains a mapping relationship (mapping algorithm) based on source file differences and multiple spaces, which not only retains the dynamic selection characteristics of the organizational structure based on the identity ID, but also fully considers the multiple space redundancy characteristics corresponding to the multiple content features of the OOXML compound document. Compared with the existing single space mapping strategy, the present method can dynamically and randomly select content features to construct a mapping relationship. While improving the algorithm hiding capacity, it also gives full play to the advantages of the OOXML compound document's complexity, solving the existing algorithm hiding capacity and mapping flexibility problems. The present invention conducts security research from different dimensions in view of the multi-attribute characteristics of OOXML compound documents. The results show that the hidden spaces of different dimensions are independent of each other, and the hidden capacity of multiple spaces can be regarded as the arithmetic superposition of the hidden capacity of each dimensional space. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a flowchart of information hiding in multi-dimensional space based on OOXML compound document set of the present invention;

[0020] Figure 2 It is a structural schematic diagram of a dual space type tree structure of the present invention;

[0021] Figure 3 It is a flowchart of information hiding in multi-dimensional space based on the synchronous identity information of the receiver according to the present invention. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0023] An embodiment of a method for hiding information in a multidimensional space:

[0024] In order to meet the hiding capacity requirements of information hiding, the present invention provides a multi-dimensional space information hiding method, which establishes a mapping relationship between documents of different subcategory sets and binary codewords based on OOXML compound documents, and hides the binary string of hidden information in the subcategory set through the mapping relationship.

[0025] OOXML (Open Office XML) compound document is a technical specification developed by Microsoft in Office 2007 products. It has become a widely used document format standard internationally. The essence of this document format is a ZIP file. OOXML compound documents include three document types: pptx documents, xlsx documents, and docx documents. The ZIP file structure includes content source data, directory source data, and directory end identifiers. Each component of the content source data corresponds to an item in the directory source data. The content source data is composed of several sub-data areas, which store the source file (source XML file) header and the corresponding compressed data. Therefore, a OOXML compound document can be regarded as several source XML file compressed data organized in a certain position order. The organization of the OOXML compound document is determined by the type of its source XML files and the position order of these source XML files.

[0026] According to the difference in the types and position order of XML source files, 00XML compound documents can be generally divided into the following three categories: differences in the types of source files of different types of 00XML compound documents; differences in the types of source files of the same type of 00XML compound documents; differences in the position order of source files of the same type of 00XML compound documents.

[0027] As an important form of compound documents, OOXML compound documents can contain multiple content types such as text, pictures, audio and video at the same time, so that they can have redundant information in multiple spaces. For example, when the content of an OOXML compound document contains both text and pictures, it may have both text content redundant features and picture redundant features. In addition, OOXML compound documents may also have redundant features in their organizational structure. In other words, for a certain OOXML compound document data set, there may be multiple redundant spaces.

[0028] Hidden space based on the perspective of OOXML compound document organizational structure: This hidden space is generated by the differences in the types and position orders of the document's source files; Hidden space based on the perspective of document text content: This hidden space is generated by the characteristics or differences of the document text content, such as: keywords, tone characteristics, character characteristics, and note characteristics; Hidden space based on the perspective of pictures: There may be hidden space in the pixel brightness value, color, texture, edge, contour, and high-level semantic information of the document picture content; Hidden space based on the perspective of audio and video: In audio and video, redundant information such as time redundancy, space redundancy, and visual redundancy can be used to embed information.

[0029] With the development of information technology, OOXML compound documents, web pages and other composite digital carriers containing various types of content have become common, which can provide multiple hidden spaces at the same time. Mining and making full use of these hidden spaces is conducive to alleviating the problem that the existing hidden capacity cannot meet the needs of information hiding: a carrier-free information hiding method for OOXML compound documents based on multiple spaces, starting from the two perspectives of document organizational structure characteristics and text content characteristics, establishes a dual mapping between secret information and the position arrangement order of the OOXML compound document source file and the document note characteristics, and realizes information hiding based on OOXML compound documents.

[0030] Theoretical analysis and experimental results show that: from the perspective of multi-attribute characteristics of OOXML composite documents, security research is conducted from different dimensions. It is found that hidden spaces of different dimensions are independent of each other, and the hidden capacity of multiple spaces can be regarded as the arithmetic superposition of the hidden capacity of each dimensional space.

[0031] The carrier-free algorithm is constructed from the multi-dimensional redundant space existing in the OOXML compound document itself. This method not only fully combines the advantages of the compound document's own composite digital carrier, but also fully utilizes the hidden performance of each space while avoiding the performance limitations of a single space. From the perspective of mapping structure, the multi-space mapping method is an extension of the method based only on a single mapping tag, and the existing method is a special case of a certain form of the method of the present invention. The method of the present invention comprehensively considers the multi-space characteristics contained in the OOXML compound document itself, designs a carrier-free information hiding algorithm from the perspective of multi-dimensional space, and constructs a multi-space mapping strategy to significantly improve the hiding capacity of the algorithm.

[0032] In order to better demonstrate the performance improvement of the proposed method on existing information security, based on the above-mentioned carrier-free information hiding model, considering the two redundant features of document organization structure and text note features, a mapping sequence structure based on dual space is constructed. The mapping structure mainly includes two parts: 1) Mapping based on OOXML compound document organization structure: mapping construction is based on the differences in the types and position order of source files within the compound document; 2) Mapping based on text note features: mapping construction is performed using Chinese character note features. For better identification, the following is expressed in a type tree structure, such as Figure 2 As shown:

[0033] This part uses the composite characteristics of composite documents to construct a mapping between secret information and documents based on multiple hidden spaces. In order to ensure the correct transmission and reception of secret information, the mapping strategy is based on the sharing of the receiver ID. Figure 1 As shown. Figure 3 As shown, the process is as follows:

[0034] Document set organizational structure category analysis: The document set W is analyzed according to the source file category and order based on the multi-tree model, and the category features are used as the category mark of the document of this type. Each subcategory set is obtained and numbered in sequence.

[0035] Note feature coding matrix: The two communicating parties use the receiver's identity ID to obtain the subcategory set W. Each subcategory set constructs a feature matrix based on the text note features and encodes the number of rows and columns to establish a mapping between codewords and documents:

[0036] Step 1: Identify the note features of each document in the subcategory set W and sort them by note sequence to obtain a set H, where the sorting process is performed according to "a"..."z". If the first phonetic letter of the document (the first letter of the pinyin of the character pronunciation) is the same, they are grouped into one set and the set is sorted according to the second letter, and so on, until all documents are sorted;

[0037] Step 2: Let the number of documents contained in the subcategory set be, set the number of rows m and the number of columns n, which must satisfy, fill the subcategory set according to the structure matrix T, if there are excess documents, match them directly with the codewords of the upper hidden space (the upper hidden space is the first space, that is, the source file organization structure); let the number of documents contained in the subcategory set be len(W i ′ ), set the number of rows and columns of the structure matrix T0 to m, and the number of rows and columns to n, where m≤n; m,n∈2 l , l∈N * ; m·n is less than or equal to len(W i ′ ),

[0038] Step 3: Encode the number of rows and columns of the document category matrix T with equal-length binary codewords to obtain the note feature coding matrix and complete the mapping between codewords and documents. The documents, codeword sequences, and storage paths are saved in a multi-space mapping table to complete the mapping construction process.

[0039] For example, the number of rows m (number of rows) is 2 2 , the number of columns n (number of columns) is 2 3 , the codeword matrix is:

[0040] or

[0041] When there are documents with the same first note feature in the subcategory set, the documents with the same first note feature are classified into the corresponding note set, and the set is sorted according to the second note feature, and so on, until all documents in the subcategory set are sorted;

[0042] For example: Document 1 with the pronunciation of letters aasc, document 2 with abcc, document 3 with babb, ..., document 32 with zcab; the 32 documents are sorted by row and the codeword matrix Match, that is, the code word corresponding to document 1 of aasc is 00000, the code word corresponding to document 2 of abcc is 00001, the code word corresponding to document 3 of babb is 00010, ..., and the code word corresponding to document 32 of zcab is 11111.

[0043] For example: the number of rows m is 2 2 , the number of columns n is 2 2 , the codeword matrix is:

[0044] or

[0045] For example: Document 1 with the pronunciation letters aasc, Document 2 with baac, Document 3 with babb, ..., Document 16 with facv; the 16 documents are sorted by columns and the codeword matrix Match, that is, the code word corresponding to document 1 of aasc is 0000, the code word corresponding to document 2 of baac is 0100, the code word corresponding to document 3 of babb is 1000, ..., and the code word corresponding to document 16 of facv is 1111.

[0046] If the codewords in any row and column of the codeword matrix are all 4-bit binary codewords, and one 4-bit binary codeword corresponds to one document, when hiding secret information, the 20-bit binary string 1001 001110100001 0101 that reflects the secret information can be divided into 4-bit segments to obtain 5 segments of 4-bit binary codewords. The row and column relationship of the 5 segments of 4-bit binary codewords and their corresponding documents are found in the codeword matrix. The sender provides the secret information of the 5 documents (document 7, document 13, document 11, document 5 and document 6) to the receiver through a public channel in the form of a document set, thereby realizing information hiding.

[0047] From the above process construction process, we can see that after the construction of the multi-space mapping algorithm, it not only retains the dynamic selection characteristics of the organizational structure based on the identity ID, but also fully considers the multi-space redundant characteristics corresponding to the multi-content characteristics of the OOXML compound document. Compared with the existing single-space mapping strategy, the method in this paper can dynamically and randomly select content features for mapping construction. While improving the algorithm's hidden capacity, it also gives full play to the advantages of the OOXML compound document's complexity, solving the problems of the existing algorithm's hidden capacity and mapping flexibility.

[0048] In order to better demonstrate the hidden capacity performance of the algorithm in this paper, this paper randomly selects document sets of different sizes for testing, and selects the information hiding method in a single-dimensional space for comparison. The results are shown in the following table:

[0049]

[0050] Note: The unit of performance indicator is: bit / document

[0051] The carrier-free information hiding method of OOXML compound document based on multiple spaces starts from the two perspectives of document organizational structure characteristics and text content characteristics, establishes a dual mapping between secret information and the position arrangement order of the source file of OOXML compound document and text notes, and realizes information hiding based on OOXML compound document.

[0052] Theoretical analysis and experimental results show that: from the perspective of multi-attribute characteristics of OOXML composite documents, security research is conducted from different dimensions. It is found that hidden spaces of different dimensions are independent of each other, and the hidden capacity of multiple spaces can be regarded as the arithmetic superposition of the hidden capacity of each dimensional space.

[0053] An embodiment of an information hiding system in a multi-dimensional space:

[0054] A multidimensional space information hiding system includes a processor, the processor is used to execute instructions to implement the multidimensional space information hiding method as described above, wherein the multidimensional space information hiding method has been described in detail in a multidimensional space information hiding method, which will not be repeated here.

Claims

1. A method for hiding information in a multidimensional space, characterized in that: At least one subcategory set is obtained according to the OOXML composite document set, and the documents in the subcategory set are sorted according to the set rules to form a document matrix, wherein the document matrix includes a plurality of horizontal rows and a plurality of vertical columns, and at least one of the number of the horizontal rows and the number of the vertical columns is a positive integer power of 2; the horizontal rows and the vertical columns are binary-coded respectively to obtain a row code of each horizontal row and a column code of each vertical column, and the binary codeword of each document in the document matrix is ​​a combination of its corresponding row code and column code, thereby obtaining a mapping relationship between the document and the binary codeword; The information to be hidden is converted into a binary string, and the binary string is hidden in the subcategory set through the mapping relationship.

2. The multi-dimensional space information hiding method according to claim 1, characterized in that: According to the differences in the types of source files and / or the differences in the location order of source files, the OOXML compound document set is divided into several subcategory sets. Each subcategory set corresponds to the identity of the receiver one by one. For different receivers, the mapping relationship corresponding to the corresponding subcategory set is selected for information hiding.

3. The multi-dimensional space information hiding method according to claim 2, characterized in that: The differences in source file types include differences in source file types of the same type of OOXML compound documents and differences in source file types of different types of OOXML compound documents; the differences in source file position order include differences in source file position order of the same type of OOXML compound documents and differences in source file position order of different types of OOXML compound documents.

4. The multi-dimensional space information hiding method according to claim 1, characterized in that: The set rule sorting is: the documents in the subcategory set are sorted in rows or columns in a set order according to the musical note feature of the pronunciation of the first character of the document.

5. The multi-dimensional space information hiding method according to claim 4, characterized in that: The note feature is the first letter of the pinyin corresponding to the character pronunciation, and the setting order is letter a to letter z.

6. The multi-dimensional space information hiding method according to claim 4, characterized in that: When there are documents with the same first note feature in the subcategory set, the documents with the same first note feature are classified into the corresponding note set, and this set is sorted according to the second note feature, and so on, until all documents in the subcategory set are sorted.

7. A multi-dimensional space information hiding system, characterized in that: The method comprises a processor, wherein the processor is used to execute instructions for implementing the information hiding method in a multi-dimensional space as claimed in any one of claims 1 to 6.

Images