CN102497270B - Encryption method of standardized document - Google Patents

Abstract

A large number of documents in reality are based on various languages, and there are many defects. For example, it is difficult to understand and process natural language, it is difficult to be recognized, counted and summarized by computers, and it is also extremely irregular, and there are a lot of redundant documents. Remain. It is difficult to find the pseudo-key for the ciphertext obtained after encryption by the traditional method, and the unique solution distance is very short. Facing the dual requirements of computer processing and encryption security, a standardized document is proposed. A normalized document is composed of tags and their elements. For all tags, there are fixed elements. Assuming that the number of elements is limited, this kind of document can completely replace the general text content document in many cases. In some cases, it can be combined with other documents to effectively express information. The present invention designs an encryption method for this kind of standardized document, the encryption method is simple, and it is easy to find a false key, which can also guarantee the traditional security requirements.

Description

An Encryption Method for a Class of Normalized Documents

technical field

The invention belongs to the field of information security, and relates to an encryption method for a class of standardized documents, in particular to an encryption method which is easy to find false keys and has a misleading function for such standardized documents.

Background technique

A large number of documents in reality are based on various languages, and there are many defects. For example, it is difficult to understand and process natural language, it is difficult to be recognized, counted and summarized by computers, and it is also extremely irregular, and there are a lot of redundant documents. Remain. Because the language is irregular and there is redundancy, it is difficult to find the pseudo-key for the ciphertext encrypted by the traditional method, and the unique solution distance is very short. In this way, when encountering soft and hard foam attacks, the key holder often has to hand over the real key. At the same time, it is also vulnerable to ciphertext-only analysis under unlimited computing power. The inventor has designed some encryption methods with misleading functions, but they also have their own limitations. Due to the irregularity and standardization of the language, the calculation amount of encryption is large, and sometimes misjudgment of keywords also occurs. causing some pseudo-keys to become invalid. Therefore, facing the dual requirements of computer processing and misleading encryption, a normalized document is proposed. A normalized document is completely composed of tags and their elements. For all tags, there are fixed elements. Assuming that the number of elements is limited, this kind of document can completely replace the general text content document in many cases. In some cases, it can be combined with other documents to effectively express information. To give a simple example, time, location, weather, people, and activities can all be used as a tag to gather all possible times, which is the element of the time tag, and the same is true for other tags. Some xml, tables, and excel documents in reality can become such standardized documents after being constrained. Such documents are easier to compress, redundancy can be well eliminated, special encryption can increase the number of pseudo-keys, and this normalization is more suitable for computers to process information. In order to express complex semantics, more specifications can be made on the content between tags.

Contents of the invention

The present invention aims to overcome the limitation that the existing encryption method cannot be misleading, and the existing encryption method with misleading function is complicated and other defects, and after standardizing the document, it provides a simple and practical normalized document with misleading function encryption algorithm. For various safety considerations, the present invention includes two layers of encryption, the outer layer and the inner layer. The inner layer encryption can have a misleading function, while the outer layer encryption is to further encrypt some information such as labels that may need to be kept secret. Layer encryption is optional.

Based on our definition of this type of standardized document, each label has limited fixed elements, and a database of label elements can be established. This database stores all the elements in each label, and the encoding of these elements. In the database Number all the elements. In the most critical inner layer encryption, map the elements in each label in the plaintext to be encrypted to the corresponding number in the database, and encrypt the number to obtain the ciphertext c, which is used for encryption and decryption. The operation of is closed, that is to say, for the encrypted ciphertext c of each label number, and then decrypting c with different keys can traverse all the numbers, and then write the ciphertext number c into the corresponding element position of the label.

The complete encryption process is as follows:

(1) Read the content of each label in the document one by one, that is, the elements in the label, query the database, get the number corresponding to this element in the corresponding label, and then intercept the corresponding key from the inner key one by one. The key is used to encrypt the content of this tag to obtain the ciphertext c of the element, and the encryption and decryption algorithm ensures the closure. Write the ciphertext of the elements in the tag to the content corresponding to the corresponding tag in the document one by one.

(2) Save the encrypted document in the inner layer to obtain the transitional ciphertext.

(3) Optionally, perform an outer encryption on this document to obtain the final ciphertext. This encryption is carried out using traditional encryption methods, and a symmetric encryption method can be selected.

In the decryption process, firstly, the final ciphertext is decrypted with the traditional symmetric encryption algorithm to obtain the transitional ciphertext file, and secondly, the inner decryption of the transitional file is carried out. When this level of decryption is performed, the ciphertext in each label is read, and then the key corresponding to this label is obtained. After decryption, the number corresponding to the plaintext of the label content is obtained, and then according to the database, you can check corresponding content.

This encryption method can easily find the false key, that is, the wrong key can decrypt meaningful plaintext. The method of obtaining the false key: A) There are two keys in the encryption and decryption. To obtain the false key, you can use The key of the outer symmetric cipher remains unchanged, and then arbitrarily selects an inner key sequence to intercept, assigns it to the corresponding label in turn, and decrypts it. At this time, when reading the text, whether there is any dissatisfaction (such as unreasonable, exposed flaws, misleading, etc.) effect is not good), if you are not satisfied, you can try other keys in the same way, and you can modify it according to your own needs according to the position of the keyword, thus forming the effective part of the final inner key, Using this key with the correct outer symmetric encryption key is a pseudo-key; B) There is a more free way to choose a pseudo-key, which can be determined according to the content of the misleading label you want, so as to determine The pseudo-key sequence of the inner layer encryption, the method is to keep the outer key unchanged, according to the keywords of the text, based on the database, select the misleading element to be replaced, and then determine the corresponding pseudo-subkey of this element , connect the pseudo-subkeys corresponding to each label in turn, which is the pseudo-key of the inner layer, and use the constant outer key to obtain the total pseudo-key.

When subjected to cryptanalysis by cryptanalysts, cryptanalysts can obtain different meaningful plaintexts based on different inner keys, and the plaintexts obtained by adding these different keys may also be similar to the original plaintexts, and vice versa. meaning, so it fits well with the communication background, and cryptanalysts cannot determine the real key.

Detailed ways

The following are examples of the present invention, however, the present invention is not limited to the examples.

In the actual implementation, the two parties agree that different parts of the total key are used for inner and outer encryption respectively, and the outer key is a fixed length, which is determined according to the algorithm used in the outer encryption. For example, the outer encryption algorithm uses 256 bits. AES encryption. The first 256 bits of the key are outer keys, and the rest are inner keys. The inner key is truncated to equal length. The truncated length is determined according to the maximum number N of possible elements in all tags in the database, and the value should be greater than or equal to log ₂ N. The inner key is intercepted and distributed to each label in turn as a subkey for encryption.

The file encryption process is as follows:

1) The communication parties share the key, and the key is relatively long. Quantum key distribution can be used to realize the distribution of long random keys. Both parties stipulate to use the same encryption algorithm and share the same database.

2) During the encryption process, according to the type or format of the file, use the corresponding software to open the file to read its content. For example, to encrypt an xml or excel file, first use the corresponding software to open the file and read the content in each label. Prepare for inner encryption.

3) Inner layer encryption, the number we choose here is to number all elements from 0 to n-1, and n is the number of all elements of this label. The encryption algorithm is a modular addition operation. According to the fixed structure of the key, the subkey k is intercepted from the sequence. The subkey interception methods are: A) Use the method of intercepting equal-length subkeys to intercept from the random sequence, The intercepted length is a positive integer equal to or greater than log ₂ N. When log ₂ N is a positive integer, the intercepted length is log ₂ N; when log ₂ N is a non-integer, add 1 to the integer part of log ₂ N is the maximum number of elements in all tags, and the encryption and decryption are intercepted according to such rules, which can ensure the synchronization of the subkeys intercepted by encryption and decryption. This method requires a large number of keys, but the operation speed is fast; B ) is intercepted according to the shortest length required for traversing all elements of each label, the intercepted length is a positive integer equal to or greater than log ₂ n, when log ₂ n is a positive integer, the intercepted length is log ₂ n; log ₂ n is not When it is an integer, add 1 to the integer part of log ₂ n, where n is the number of all elements of this label. When encrypting and decrypting, intercept according to this rule, which can ensure the synchronization of the subkeys intercepted by encryption and decryption. The operation speed of this method is slow, but it can save the key demand to the greatest extent.

The method of calculating the ciphertext number is c=m+k mod n, the following is the relevant information of a very simple text:

Table 1 Example

Label element number m no key value Cipher text number Week Monday 1 7 76 0 matter work 0 5 56 1

4) The inner encryption gets the new text, and saves the text in the transitional ciphertext file.

5) Encrypt the outer layer to get the final ciphertext. In order to strengthen the security, the outer encryption is performed on the transitional ciphertext file. This encryption is encrypted using the traditional 256-bit AES algorithm.

Decryption is a reverse process, and the file decryption process is as follows: firstly, corresponding decryption is performed on the outer layer of symmetric password encryption to obtain the transitional ciphertext file, and secondly, the obtained transitional file is opened with relevant software to read its content, according to Intercept the corresponding subkey obtained, decrypt it, obtain the m value, query the database, and determine the corresponding database element, for example, decrypt the number of the week to get 1, and the element numbered 1 in the query database is Monday , the content of the corresponding label in the plain text is Monday.

In this way, we can also obtain the pseudo-key according to one of the two methods mentioned above, obtain meaningful plaintext, and obtain satisfactory "pseudo-plaintext" according to our own intentions.

Claims (4)

1. the encryption method of a class standardization document, comprise that A. opens file, read each label substance, B. the content of all labels is carried out to interior infill layer with sub-key independently successively, C. preserve file, obtain transition ciphertext document, D. carries out the outer infill layer of traditional symmetry algorithm to transition document, E. calculate pseudo-key, it is characterized in that:

Be expressly normalized, document is made up of label and element thereof completely, for all labels, has fixing element, and the quantity of supposing element is limited;

Internal layer encryption key adopts random sequence to carry out, after successively internal layer key being intercepted, distribute to each label, interior infill layer is mainly mapped as the element in each label in plaintext to be encrypted numbering corresponding in database, numbering is encrypted and obtains ciphertext c, then ciphertext is numbered to c and write the element position that label is corresponding, enciphering and deciphering algorithm has closure, by the content of all possible sub-key enabling decryption of encrypted, finally can travel through all elements that same label comprises, encrypting and decrypting depends on a database, each label is limited fixed element in database, this database is being deposited all elements in each label, and these elements are any codings, and in database, all elements are numbered,

The division methods of the sub-key in internal layer key is, adopt the method that intercepts equal length sub-key that the sub-key k of interior infill layer is intercepted out from internal layer key sequence, or adopt the sub-key k that intercepts current label according to the needed shortest length of each label traversal all elements;

Outer infill layer is optional;

The method of calculating the pseudo-key of generation is: A) by constant the key of outer symmetric cryptography, then choosing arbitrarily an internal layer key sequence intercepts, distribute to successively corresponding label, be decrypted, form the live part of final internal layer key, coordinated correct outer symmetric cryptography key with this key; Or B) according to the label substance of desirable misleading, in determining, the pseudo-key sequence of infill layer, is connected in turn pseudo-sub-key corresponding each label, adopts constant outer key, the pseudo-key that gets final product always.

2. the method for claim 1, is characterized in that: the method that the method for the intercepting equal length sub-key below described employing intercepts the sub-key k of interior infill layer out from internal layer key sequence is: the length of intercepting is for being equal to, or greater than log ₂the positive integer of N, works as log ₂when N is positive integer, intercepted length is log ₂n; log ₂when N is non-integral, intercept log ₂the integer part of N adds the maximum that 1, N is element number in all labels.

3. the method for claim 1, is characterized in that: the method that described employing intercepts the sub-key k of current label according to the needed shortest length of each label traversal all elements is: the length of intercepting is for being equal to, or greater than log ₂the positive integer of n, works as log ₂when n is positive integer, intercepted length is log ₂n; log ₂when n is non-integral, intercept log ₂the integer part of n adds the number that 1, n is this label all elements.

4. the method for claim 1, it is characterized in that: at the method A that calculates the pseudo-key of generation) in, read text while being decrypted, judge whether keyword is satisfied with, if there is unsatisfied keyword, attempt other key with same method, and modify according to the needs of oneself according to the position at keyword place;

At the method B that calculates the pseudo-key of generation) in, according to the keyword of text, taking database as foundation, select to want the misleading element of replacement, determine the corresponding pseudo-sub-key of this element.