JP2001101055A - Data base managing device, data base system, enciphering device and ercording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To ensure the security of a data base, and to attain the quick retrieval of an item to be used for retrieval. SOLUTION: At the enciphering of a data base, the data of column items to be used for retrieval are enciphered by using a column key common to the column items stored in a key storing device 15, and the data of the other column items are enciphered by using a line key specific to each line stored in the key storing device 15, and the resultant data are preserved in a data base storage area 16a of a data storage device 16. Thus, it is possible to increase security by making keys for each line different. In the retrieval, data inputted as data for retrieval are enciphered by using the column key common to the prescribed column items, and the enciphered data for retrieval are compared with the enciphered data base in a data base storage area 16a so that high speed retrieval can be realized.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a database management device for encrypting and managing a database, a database system in which a search device having an encrypted database is arranged on a network, an encryption device for encrypting the database, and It relates to a recording medium.

[0002]

2. Description of the Related Art In a database management apparatus, in order to ensure the security of a database, it is required that a database held as a management target be encrypted and stored.

[0003] Here, in order to enhance security, it can be realized by performing more complicated cryptographic processing. However, it is inevitable that the time required for the complicated cryptographic processing becomes longer.

A database has a huge amount of data, and a data search is a process of selecting a certain item from the data that satisfies a condition and outputting a record (line data) including the item data that satisfies the condition. become. Therefore, in a data retrieval system that handles a large amount of data, an increase in the operation time causes a reduction in processing efficiency.

[0005]

As described above, there is a demand for security in a database storing highly confidential data, and encryption processing for enhancing security hinders the usability of the database. There is.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and has a database management device, a database system, an encryption device, and a recording medium which ensure the security of a database and enable high-speed retrieval of data. The purpose is to provide.

[0007]

According to the present invention, when encrypting a database, data of a column item used for retrieval is encrypted using a column key common to the column item, and other column items are encrypted. Is encrypted using a unique row key for each row.

Specifically, the database management apparatus of the present invention encrypts data of a predetermined column item of the database using a column key common to the column item, and encrypts data of other column items for each row. It comprises an encrypting means for encrypting using a unique row key, and a storage means for storing a database encrypted by the encrypting means.

According to this configuration, when encrypting the database, security can be improved by using a different key for each row, and at the time of retrieval, data input for retrieval is stored in the predetermined column. A high-speed search can be realized by encrypting an item using a common column key and comparing the encrypted search data with each item data of the encrypted database.

[0010] Further, if data of a column item other than a column item used for retrieval is encrypted by using a unique row key for each row and a column key common to the column item in a complex manner. , Can further enhance security.

In addition, if a multidimensional space rotation method is used as an encryption method in which vectors are sequentially generated in a multidimensional space based on a predetermined function and these vector components are used as a key stream of encryption, complicated arithmetic processing can be performed. Since the database can be encrypted without the need for the information processing, the present invention can be applied to an information processing apparatus having a low computing capacity.

It is also possible to construct a database system in which the storage locations of the databases are separated from each other and a search is requested from another information terminal via a network. In this case, as described above, data of a predetermined column item (column item used for search) in the database is encrypted using a column key common to the column item, and data of the other column items is encrypted for each row. Encrypt your unique row key. Then, when requesting a database search from another information terminal, the search data is encrypted using a common column key for the column item, and the encrypted search data is sent via a network. I do. Upon receiving the search data, the encrypted database is searched, and the data obtained as a result of the search is returned to the information terminal via the network in an encrypted state.
Therefore, data can always be exchanged in an encrypted state, so that the security of the database can be ensured.

[0013]

Embodiments of the present invention will be described below with reference to the drawings.

(First Embodiment) FIG. 1 is a diagram showing a configuration of a database management apparatus according to a first embodiment of the present invention. This device has a matrix-type database consisting of rows and columns, and with the function of encrypting and managing the database, encrypting the input search data,
It has a function of searching a database based on the encrypted search data, and reads a program recorded on a recording medium such as a magnetic disk, for example,
It is realized by a computer whose operation is controlled by this program.

As shown in FIG. 1, a CPU 1
1, a display device 12, an input device 13, a program storage device 14, a key storage device 15, and a data storage device 16 are provided.

The CPU 11 controls the entire apparatus, reads a program stored in the program storage device 14, and executes various processes according to the program. In the present embodiment, the CPU 11 executes a database encryption process as shown in FIG. 2 and a database search process as shown in FIGS.

The display device 12 is a device for displaying data, for example, an LCD (Liquid Crystal Displ.).
ay) or CRT (Cathode-ray tube). The input device 13 is a device for inputting data, and for example, a keyboard, a mouse, and the like are used.

The program storage device 14 is, for example, a ROM
Alternatively, it is constituted by a RAM or the like, and stores a program necessary for this apparatus. The necessary programs for this device are:
There are a database management program, an encryption program, and the like.

The program storage device 14 can be constituted by a magnetic or optical recording medium other than the semiconductor memory. The recording medium includes a portable medium such as a CD-ROM and a fixed medium such as a hard disk. Further, the program stored in the recording medium may be configured such that a part or all of the program is received from a transmission control unit via a transmission medium such as a network line from a server or a client. It may be a recording medium of a server constructed in the above. Further, the program may be transmitted to a server or a client via a transmission medium such as a network line and installed in these devices.

The key storage device 15 is composed of a RAM, for example, and stores keys (row key and column key) used for encrypting the database.

The data storage device 16 is a device for storing various data necessary for the present device.
It is composed of an external storage device such as an AM or a magnetic disk device. The data storage device 16 stores a database storage area 16a for storing a database and information (search target items, non-encrypted items, etc.) set by an operator when encrypting the database. Setting information storage area 16c for storing information (object column items, search character strings, etc.) set by an operator when searching a database, and an encryption setting information storage area 16c for searching a database. Comparison character string storage area 16d for storing comparison character strings
And so on.

Here, before describing the operation of the present apparatus, a database encryption method realized by the present apparatus will be described.

When encrypting a database using a different key for each row (each record), decryption of the key becomes difficult, and security can be improved. However, when searching the database, the encrypted data must be decrypted with a key for each row, or data (keyword) input for search must be encrypted with a key for each row. , It takes time to get search results. On the other hand, if the database is encrypted using a different key for each column, the database search can be performed at high speed because only the search data needs to be encrypted using the key corresponding to the column item to be searched. . However, if there is the same data in the same column, the encryption result will be the same, so that the possibility of decrypting the key therefrom increases.

Therefore, according to the present invention, when encrypting a database, data of a column item frequently used for retrieval is encrypted with a column common key, and data of other column items is encrypted with a different key for each row. And encrypting. In other words, by making the key different for each line,
In addition to enhancing security, at the time of retrieval, high-speed retrieval is enabled by encrypting data input to an item used for retrieval with a column key and comparing the data with encrypted data in a database.

FIGS. 5A and 5B are diagrams for explaining the configuration of the database in the first embodiment. FIG. 5A shows a state before encryption, FIG. 5B shows a state after encryption, and FIG. (C)
Indicates the state after decoding. FIG. 6 is a diagram illustrating a configuration of a column key and a row key according to the first embodiment.

As shown in FIG. 5A, the present apparatus is provided with a matrix database composed of rows and columns. Here, a case is shown in which personal data is made into a database. In this database, one record has "nu
mber "," name "," state "," wei
ght ”,“ height ”,“ age ”,“ phon ”
e ".

Such a database is encrypted using a column key and a row key. That is, the column items frequently used in the search are “name”, “state”, “ag”.
When "e" is set, as shown in FIG. 6, "apple", "ora"
nge "," lemon ", and so on, using a common column key for the other column items, and the other column items" weigh ".
The data in each row of “t”, “height”, and “phone” is encrypted using a unique key for each row.

It is assumed that the "number" column is not encrypted. As the row key, "tiger",
"Dog", "cat", "mouse", "elep
hunt, "cow", "pig", "rabbi"
Keys such as "t" and "lion" are used. These keys can be generated mathematically for each column or row using a predetermined function.

FIG. 5B shows the result of encrypting the database of FIG. 5A using the column key and the row key. In the database storage area 16a of the data storage device 16, FIG.
The database is stored in a state as shown in FIG.

When searching the database, the search data is encrypted using the column key corresponding to the column item used for the search, and then the search process is performed. For example, "stat
When searching for data such as “Florida” in “e”, first, “F” input as search data is searched.
"lorida" is changed to the column key "apple" of "state".
To obtain “h * / fDD”. This "h * /
Data such as “fDD” is retrieved from each row of the “state” column. As a result, “number2” and “nu”
It can be seen that there is data corresponding to “mber8”.

When the encrypted database is restored, the same column key and row key as in the encryption are used. When the database in FIG. 5B is decrypted using the same column key and row key as in the encryption, the original data can be obtained as shown in FIG. 5C.

The operation of the present apparatus will be described below.

Here, a description will be given of (a) a process for encrypting the database and (b) a process for searching the database. The program for realizing each function shown in this flowchart is in the form of a program code readable by a CPU.
4 recording medium. This program can also be transmitted in the form of a program code via a transmission medium such as a network line.

(A) Encrypting the database FIG. 2 is a flowchart showing the operation of the database encryption processing executed by the present apparatus. Now, it is assumed that the database is stored in the database storage area 16a of the data storage device 16 in an unencrypted state.
This state is shown in FIG.

First, a database to be encrypted is specified on a database encryption setting screen (not shown) (step A11).

Next, among the column items provided in the database, a column item to be used for retrieval and a column item not requiring encryption are set (step A12). In the example of FIG. 5A, the column item used for the search is “nam
e "," state ", and" age ", and the column item that does not require encryption is" number ". The setting information here is stored in the encryption setting information storage area 16b of the data storage device 16.

Next, a row key and a column key used for encrypting the database are determined (step A13). The information of the row key and the column key determined here is stored in the key storage device 15.

After such a setting operation, each column item in the database is sequentially designated (step A1).
4) The encryption method for the designated column item is determined based on the setting information (step A15). In this case, nothing is performed for the column item of “number” in the database because it is set as an unencrypted item. That is, the item of “number” remains the original data.

If the designated column item is set as a column item to be used for searching, the key storage device 15
Is read out (step A15 → A16), and the data of each row of the column item is encrypted with the column key (step A17). That is, "name", "state",
As shown in FIG. 6, the data of each line of the item of “age” is “apple”, “orange”, “lemo”.
n "is encrypted using a key unique to each column.

If the designated column item is not set as a column item to be used in the search, that is, if it is another column item, a row corresponding to each row stored in the key storage device 15 The key is read (step A15 → A1)
8) The data of each row of the column item is encrypted with a unique row key (steps A19 and A20). That is, "state", "wei" in the database
As for the data of each item of “ght” and “height”, as shown in FIG.
r, the data on the second line is “dog”, the data on the third line is “cat”, the data on the fourth line is “mouse”, the data on the fifth line is “elephant”, and the data on the sixth line is “cat”. cow, the data in the seventh row is “pig”, the data in the eighth row is “rabbit”, and the data in the ninth row is “lio”.
For example, n is encrypted using a row key corresponding to each row.

Such an encryption process is repeatedly performed for each column item of the database. When the encryption of the data in each row of all the column items is completed, the encrypted database is overwritten and stored in the database storage area 16a of the data storage device 16 (step A22). This state is shown in FIG.

(B) Searching the database FIG. 3 is a flowchart showing the operation of the database search process executed by the present apparatus. Now, it is assumed that the database is encrypted and stored in the data storage device 16 in the encryption processing described in (a).

First, as shown in the flowchart of FIG. 3A, search information is input on a database search setting screen (not shown) (step B11). Inputting the search information means inputting a column item to be searched and a character string (keyword) for search. These pieces of input information are stored in the search setting information storage area 16c of the data storage device 16. When search information is input through the input device 13, pre-search processing is executed (step B12).

In the pre-search process, as shown in the flowchart of FIG. 3B, it is determined whether or not the column item input as a search target is a predetermined column item (step C).
11) If it is determined that the column item is a predetermined column item (Yes in step C11), a character string for search is encrypted with a column key common to the column item (step C12).

The predetermined column items are search target items (items used for search) set at the time of encrypting the database, and specifically, “name”, “stat”
e) and “age”. Information on the search target item is stored in the encryption setting information storage area 16b of the data storage device 16. Therefore, in step C11, this encryption setting information storage area 1
6b, it is determined whether or not the item is a predetermined column item. The column key common to the column items is the key storage device 1
5 is stored. Therefore, step C12
Then, a column key corresponding to the column item is read from the key storage device 15 and a character string for search is encrypted. For example, if the designated item is “state”, “o”
A character string for search will be encrypted using a column key such as “range”.

If the column item input as a search target is not a predetermined column item (N in step C11).
o), the search character string is not encrypted as described above.

After such a pre-search process, a database search process (see FIG. 4) is performed (step B13).
The data obtained as a result of the search is displayed on the display device 12 (step B14). FIG. 4 shows a database search process.

FIG. 4 is a flowchart specifically showing the operation of the search processing in step B13.

First, as shown in the flowchart of FIG. 4A, the search character string is used as a comparison character string with the database as a comparison character string storage area 16d of the data storage device 16.
Is set to (Step D11). In this case, if the column item input as a search target is a predetermined column item (“name”, “state”, “age”) as described above, the search The column is encrypted with the column key corresponding to the column item and set in the comparison character string storage area 16d. In the case of other column items, they are set in the comparison character string storage area 16d as they are without being encrypted.

Next, the encryption method based on the column number of the encrypted database stored in the database storage area 16a of the data storage device 16 is determined (step D1).
2). Accordingly, when the search target corresponds to a predetermined column item encrypted with the column key, the data in each row of the target column is sequentially scanned (step D12 → D13) and included in the designated row. A comparison process is performed between the character string of the data of the target item and the search character string (encrypted character string) set in the comparison character string storage area 16d (step D14).

In this comparison processing, as shown in the flowchart of FIG. 4B, the encrypted character string of the data of the target item retrieved from the database is compared with the encrypted character string for search, and both are compared. It is determined whether they match (step E11). If they match (step E1)
1; Yes), record data including the matching item is extracted as a database search result (step E1).
2).

This process is repeated until the end of the encrypted database, and the corresponding data is sequentially extracted (step D).
15), and output the extracted data as a search result (step D20).

More specifically, in the example of the encrypted database shown in FIG. 5B, if it is specified to search for data such as "Florida" in the item "state", first, "Florida" input as search data is replaced with the "state" column key "ap
ple ”to obtain“ h * / fDD ”. Data such as “h * / fDD” is searched from the column of “state”. As a result, “number2” and “n”
It can be seen that there is data corresponding to “number8”.

On the other hand, if the search target corresponds to another column item encrypted with the row key, the data in each row of the target column is sequentially scanned (step D12 → D16),
After the data of the target item included in the designated line is decrypted with the line key unique to each line (step D17), the search character string (non-encrypted character) set in the comparison character string storage area 16d Is performed (step D18).

In this comparison processing, as shown in the flowchart of FIG. 4B, by comparing the decrypted character string of the data of the target column retrieved from the database with the non-encrypted character string for search, both are compared. It is determined whether they match (step E11). If they match (step E
11 Yes), record data including the matched item is extracted as a database search result (step E).
12).

This process is repeated until the end of the encrypted database, and the corresponding data is sequentially extracted (step D).
19), and outputs the extracted data as a search result (step D20).

More specifically, in the example of the encrypted database shown in FIG. 5B, if it is specified to search for data such as "163" in the item "weight", first, The data in the first row of “weight” is decrypted with a row key such as “tiger”. Similarly, the data in the second row is “dog”, the data in the third row is “cat”, the data in the fourth row is “mouse”,
The data in the line is “elephant”, the data in the sixth line is “cow”, the data in the seventh line is “pig”, the data in the eighth line is “rabbit”, and the data in the ninth line is “lio”.
After decrypting each using a row key corresponding to each row, such as “n”, the corresponding data is searched from the “state” column based on “163” input as search data. This indicates that data corresponding to “number3” and “number9” exists.

As described above, when encrypting the database, predetermined column items used for search are encrypted with the column common key, and at the time of search, the search data is encrypted with the column common key and stored in the database. High-speed search can be realized by comparing with the encrypted data. Further, security is enhanced by encrypting column items other than the predetermined column items by giving different keys to each row. In this case, at the time of retrieval, decryption using a key for each row is required. Therefore, although it takes more time than retrieval for the predetermined column item, it is not a problem because the item is not frequently used for retrieval. .

(Second Embodiment) In the first embodiment, the data of the column items other than the predetermined column items is encrypted using a unique row key for each row. In the second embodiment, in order to further increase security,
It is characterized in that each row is encrypted using a unique row key and a column key common to the column item in a complex manner.

FIG. 7 is a diagram showing the configuration of a database according to the second embodiment. FIG. 7A shows a state before encryption, FIG. 7B shows a state after encryption, and FIG. Indicates the state after decoding. FIG. 8 is a diagram illustrating a configuration of a synthetic key according to the second embodiment.

As shown in FIG. 7A, the present apparatus is provided with a matrix database composed of rows and columns. Here, a case is shown in which personal data is made into a database. In this database, one record has "nu
mber "," name "," state "," wei
ght ”,“ height ”,“ age ”,“ phon ”
e ".

Such a database is encrypted using a synthetic key. That is, the column items frequently used in the search are “name”, “state”, “age”
In this case, as shown in FIG. 8, the data of each row of these column items is "apple", "orang".
e "," lemon ", and the like, using a common column key for each column item, and encrypting the other column items" weight "," lemon ".
The data of each row of “height” and “phone” is encrypted by combining a column key and a row key, such as “banana + row key”, “lyche + row key”, and “apricot + row key”.

It is assumed that the "number" column is not encrypted. As the row key, "tiger",
"Dog", "cat", "mouse", "elep
hunt, "cow", "pig", "rabbi"
Keys such as "t" and "lion" are used. These keys can be generated mathematically for each column or row using a predetermined function.

FIG. 7B shows the result of encrypting the database shown in FIG. 7A using the synthetic key. In the database storage area 16a of the data storage device 16, FIG.
The database is saved in the state as shown in FIG.

When searching the database, the first
As in the embodiment, the search processing may be performed after encrypting the search data using the column key common to the column items used for the search. For example, "Flor" in "state"
When searching for data such as "ida", first, "Florida" input as search data is replaced with "s
"h *" encrypted with the column key "apple"
/ FDD ". Data such as “h * / fDD” is searched from each row in the column of “state”. This indicates that data corresponding to “number2” and “number8” exists.

When the encrypted database is restored, the same composite key as that used for the encryption is used. FIG.
By decrypting the database of (b) using the same synthetic key as in the encryption, the original data can be obtained as shown in FIG. 7 (c).

In the process for encrypting the database and the process for searching the encrypted database, a column key and a row key are applied to the data of each row of the column items other than the predetermined column items. The processing is the same as the processing of the first embodiment (FIGS. 2 to 4) except that the encryption is performed in combination, and the description is omitted here.

As described above, for a column item frequently used for retrieval, a high-speed retrieval can be realized as in the first embodiment by encrypting the column item using a column key common to the column item. For other column items, security can be further enhanced by encrypting the column item and the row key in combination.

(Third Embodiment) In the first or second embodiment, the present invention is configured as a single device. However, the storage location of the database is separated, and another information terminal is connected via a network. It is also possible to construct a database system that requests a search.

Hereinafter, such a database system will be described.

FIG. 9 is a block diagram showing the configuration of a database system according to the third embodiment of the present invention. This system has a first terminal device 20 and a second terminal device 30. The first terminal device 20 and the second terminal device 30 are connected via a network 40.

The first terminal device 20 is used as a server computer for performing a database service, and is a search device 21 for performing a database search process.
And a data storage device 22 for storing a database. The second terminal device 30 is used as a client computer that requests the first terminal device 20 to search the database and receives the result from the first terminal device 20.
And the decoding device 32.

In such a database system, as described with reference to FIG. 2, in the first terminal device 20, data of each row of a predetermined column item of the database is written using a column key common to the column item. The data in each row of the other column items is encrypted using a unique row key for each row and stored in the data storage device 22.

Here, when the second terminal device 30 requests the first terminal device 20 to search the database, the processing up to the search pre-processing shown in FIG. 3 is executed on the second terminal device 30 side. That is, the search request device 31 of the second terminal device 30 determines whether or not the column item input as the search target is a predetermined column item. A column (keyword) is encrypted using a column key common to the column item. If other column items are to be searched, such encryption is unnecessary.

After the pre-search processing, the second terminal device 30 sends a search character string to the first terminal device 20 via the network 40. The first terminal device 20 executes the search processing as described in FIG. 4 by receiving the search character string.

That is, the search device 21 of the first terminal device 20 determines whether or not the search target is a predetermined column item.
If it is a predetermined column item, the search character string (encrypted character string) acquired from the second terminal device 30 and the data storage device 22
Then, a process of comparing the data of each row of the corresponding column item of the encrypted database in the table and extracting the corresponding data is performed. If the search target is a column item other than the predetermined column item, the data of the column item in the encrypted database in the data storage device 22 is decrypted using the key for each row, and then the second
The search character string (unencrypted character string) obtained from the terminal device 30 of the above is compared with the data of each decrypted line,
Processing such as extracting relevant data is performed.

When the search result is obtained in this way, the first terminal device 20 transmits the data obtained as the search result to the second terminal via the network 40 in the encrypted state.
To the terminal device 30. The second terminal device 30 has a common encryption key with the first terminal device 20. Therefore, when the search result is received from the first terminal device 20,
The data can be decrypted by the internal decryption device 32 using the encryption key. In this case, since the data is always exchanged between the first terminal device 20 and the second terminal device 30 in an encrypted state, the security of the database can be ensured.

As described above, even in a database system in which a database is provided on the first terminal device 20 side and a database search is performed by access from the second terminal device 30, it is frequently used for search. Enhance security and perform high-speed search by encrypting column item data using a common column key for the column item and encrypting other column item data using a unique row key for each row. Can be realized.

As for the data of the column items other than the predetermined column item, as described in the second embodiment, a unique row key for each row and a column key common to the column item are combined. Alternatively, the encryption may be performed by using the encryption in this way, and in this case, the security can be further enhanced.

(Encryption System) Next, a database encryption system applied to the present invention will be described.

In the present invention, a multidimensional space rotation method (multidimensional space vector method) is used as a database encryption algorithm. In the multidimensional space rotation method, vectors are sequentially generated in a multidimensional space based on a predetermined function, and these vector components are used as a key stream for encryption. This multidimensional spatial rotation method is suitable for application to a portable terminal because it can be calculated even with an information processing device having a low computing capacity. In other words, in an environment where the database of the present invention is accessed from the outside, it is desirable to employ such an encryption method in order to process the data while maintaining the confidentiality.

Hereinafter, referring to FIGS. 10 and 11,
2 shows encryption and decryption processing using a multidimensional spatial rotation method.

FIG. 10A is a flowchart showing an encryption processing operation when the multidimensional space rotation method is used.

The original data to be encrypted is set to M (step F11). This data M is binary data. First, scramble 1 is applied to this data M in bit units (step F12). The data thus obtained is defined as M '(step F13).

Here, the data M 'is subjected to XOR (exclusive OR) with a random number generated mathematically in sequence to perform encryption (step F14). At this time, the use of the multidimensional vector r as a function for generating random numbers is a feature of the multidimensional spatial rotation method. In this case, a function for generating the multidimensional vector r or a constant used for the function is determined by an encryption key (a secret / public key).

That is, a secret key is used for encryption.
(P₁, P_Two) And public key (P_Three, P _Four) With a constant
Generates a multidimensional vector r according to the function used
Perform logical operation such as' XOR r 'to encrypt data M'
I do. The encrypted data thus obtained is designated as C (step
F15).

More specifically, as shown in FIG. 11, for example, it is assumed that r is a three-dimensional vector (x, y, z) and the calculation accuracy of the vector components x, y, z is 16 bits. The three-dimensional vector r (x, y, z) is sequentially generated as r ₀ , r ₁ , r ₂ , r _3, ... In accordance with the following equation (1).

Now, when the data M is given as m ₀ m ₁ m ₂ m ₃ m ₄ m ₅ m ₆ ... As a sequence of 8-bit data, M is based on the calculation accuracy (16 bits). It is decomposed into two (8 bits) each. And 3
If the dimension vector is r ₀ , the data M and r
₀ (x ₀ , y ₀ , z ₀ ) and XOR (exclusive OR) with (x XOR m ₀ m ₁ ) (y XOR m ₂ m ₃ ) (z X
OR m ₄ m ₅ )..., And as a result of this calculation, encrypted data C such as C ₀ C ₁ C ₂ C ₃ C ₄ C ₅ .

The data C thus obtained is further scrambled in bit units (step F16). The data thus obtained is set as C 'and output as final encrypted data (step F17).

In the above-described processing, the level of difficulty in decryption can be increased by re-assuming C 'as M' and repeating the same encryption as described above. Further, if the form of the function for generating the multidimensional vector r is changed, the difficulty of decoding will be further increased.

Next, the decoding operation will be described.

FIG. 6B is a flowchart showing the operation of the decoding process.

For decryption, basically, the reverse process of the above-described encryption process may be performed.

That is, assuming that the encrypted data is C '(step G11), first, the data C is subjected to scrambling 2 which is the reverse of the scrambling 2 performed at the time of encryption in units of bits (step G12). This allows
Data C before scrambling 2 is obtained (step G13).

Next, data C is decoded by performing calculations such as C XOR r (step G14). This allows
Data M 'before encryption is obtained (step G1).
5).

The data M 'is also bit
A scramble 1 reverse to the scramble 1 performed at the time of the encryption is applied in units (step G12). As a result, data before scrambling 1, that is, original data M is obtained (step G17).

In the encryption process, C ′ is changed to M ′ again.
If processing for repeating encryption or changing the form of the function for generating the multidimensional vector r is added, the decryption processing is performed in accordance with the processing.

By the way, in the encryption using the multidimensional vector r, a set P of parameters (constants) for determining a function for handling the multidimensional vector r is divided into two parts, and P = {P _s , P _p } It is expressed as Here, P _s is a secret parameter, and corresponds to an encryption key (secret keys P ₁ and P ₂ ). P _p is a public parameter, and an encryption key (public keys P ₃ and P ₄ )
Is equivalent to

Next, the multidimensional spatial rotation method will be described in more detail.

A vector extending an n (n ≧ 1) -dimensional space is denoted by r, and a new vector r is sequentially obtained from its initial value r _0.
Let R be a function that generates _i (i = 0, 1, 2, 3...). At this time, r _i is represented by the following equation (1).

[0102] _{r i = A · R (P} , r i-1) r i-1 + c ... (1) where, A is appropriate constant factor. P is a set of constants used for the function, and secret keys P ₁ and P ₂ and public keys P ₃ and P ₄ are used. c is a constant vector that translates the vector.

In the above equation (1), the coefficient A is a condition for each vector to exist in a closed space region of a multidimensional space when an appropriate constraint (for example, | R | ≦ 1) is provided for the function R. give. When the constant vector c vector r _i converges, (also permitted course c = 0) guaranteed that not a "trivial" point (e.g., meaningless terms such as r = 0).

In an n-dimensional space, a vector r has n elements
have. r = (x₁, X_Two... x_n). On the calculator, the number
Value data Generally, the bit length (m) defined by the compiler
Expressed in precision (for example, 8 bytes or 64 bits)
I have. Therefore, at some moment in the sequential vector generation method,
If the vector r cannot be reproduced with the accuracy of nxm data,
The following vector r cannot be reproduced exactly (or
Defines such a function R). This is a vector
Initial value r of₀The same applies to ₀Is nx
only when reproduced with the accuracy of m data
Kutor r₁, R _Two, R_ThreeIs guaranteed.

The vector r obtained by using the above equation (1)
Are arranged one or more according to the data definition length, a character string corresponding to the total bit length (usually 8 bits per character) and an exclusive logical operation (XOR) for each bit
I do. This is referred to as first encryption. This is as described in FIG.

This procedure is a measure against decryption. In this case, a new vector may be generated by changing equation (1) and the function R again, and encryption may be performed by the same method as the first encryption. This is referred to as second encryption.

As a specific example, consider the case where n = 2.

First, if R is defined as an operation in which only θ rotates around a normal line where r _i−1 is set on this plane, R
Becomes a 2 × 2 matrix, and is expressed as follows.

[0108]

(Equation 1)

In this case, θ is a kind of parameter. That is, when the parameter is given as a function of r _i−1 and θ (r) = f (P, r) (3), the conversion represented by the above equation (2) becomes
Formally represented by the above equation (1).

In the above equation (3), P is a function f
Is intended to be defined as a set of used is constant, the secret key P _1, P _2, the public key P _3, P ₄ are used.

As described above, by using a vector r that spans a multidimensional space generated sequentially for encryption, encryption independent of the processing accuracy and processing ability of a computer can be performed in comparison with encryption such as RSA. realizable.

Now, such a multidimensional space rotation method is described in the book.
When applied to database encryption of the invention, the row key
And the column key as r '= AR (P, r) r + C (4) Used as the P of the function R. P in the above equation (4) is a function
A set of constants used for R. In the case of the present invention,
"Apple", "orange", "lemon", etc.
Is used as a secret key. That is, in the above description,
P as secret key ₁, P_TwoSo, for example,
When "apple" is used as a secret key, P₁To "a
p ", P_TwoIs “ple”. This is an example
The key setting method is not limited to this, and various combinations
Can be considered.

As described above, when the multidimensional space rotation method is applied to the encryption of the database of the present invention, the row key and the column key are used as constant components of a predetermined function of the multidimensional space rotation method. As a result, a cryptographic result that is difficult to decipher can be obtained without requiring high-precision arithmetic capability.

In the multidimensional space rotation method, as the number of dimensions increases, the number of elements in the rotation matrix R (function) also increases.
There is a problem that the calculation load at the time of encryption / decryption increases.
As a method for solving such a problem, there is a method in which a multidimensional space rotation matrix in an encryption method using the multidimensional space rotation method is replaced with a pseudo-space rotation matrix having a small degree for calculation.

More specifically, for example, the elements of the rotation matrix R in a six-dimensional space are as shown in Expression (5).
The amount of calculation increases enormously.

[0116]

(Equation 2)

Therefore, the rotation matrix R is calculated by replacing the rotation matrix R with the pseudo rotation matrix Q. The pseudo rotation matrix Q is a pseudo matrix in which elements of a spatial rotation matrix having a small degree are arranged diagonally, and the remaining elements are set to 0. For example, in the case of a six-dimensional space, a pseudo rotation matrix Q as shown in Expression (6) is used.

[0118]

(Equation 3)

In the equation (6), A and B are three-dimensional rotation matrices.

When comparing the pseudo rotation matrix Q with the elements of the rotation matrix R, the elements of Q have many 0s and the amount of calculation is small. Moreover, the function as the encryption is sufficiently performed. Generally, a multidimensional spatial rotation matrix Q may be set as in equation (7).

[0121]

(Equation 4)

As described above, the rotation matrix in the multidimensional space rotation method is calculated by arranging the elements of a plurality of rotation matrices having a small number of dimensions diagonally and replacing the remainder with a pseudo rotation matrix having 0 elements. The amount of calculation can be significantly reduced and the encryption or decryption process can be executed quickly.

[0123] In addition, as another method, may be used P obtained as ^{P = S · Q · S T} ... (8) as a new pseudo-space rotation matrix.

In the equation (8), Q is the pseudo rotation matrix. S is a permutation matrix, and is a square matrix including one element in each row and each column, as in Expression (9).

[0125]

(Equation 5)

For example, when the pseudo rotation matrix Q is expressed by the above equation (6) (in the case of six dimensions), the pseudo spatial rotation matrix P can be expressed by the following equation (10).

[0127]

(Equation 6)

This is the case where the permutation matrix S is as shown in equation (11).

[0129]

(Equation 7)

As described above, the rotation matrix in the multidimensional space rotation method can be newly formed by arranging elements of a plurality of rotation matrices having a small number of dimensions diagonally and combining the permutation matrix with a pseudo rotation matrix having the remaining 0 elements. If the calculation is performed by replacing the pseudo rotation matrix, the calculation process becomes complicated, so that the decryption of the encryption can be made more difficult.

[0131]

As described above in detail, according to the present invention, when encrypting a database, data of a column item used for retrieval is encrypted using a column key common to the column item. The data of other column items is encrypted using a unique row key for each row, so the security can be improved by using a different key for each row, and the data is input for search at the time of search. High-speed search can be realized by encrypting data using a common column key for the predetermined column item and comparing the encrypted search data with an encrypted database.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration of a database management device according to a first embodiment of the present invention.

FIG. 2 is a flowchart showing an operation of a database encryption process executed by the database management device.

FIG. 3 is a flowchart showing an operation of a database search process executed by the database management device.

FIG. 4 is a flowchart specifically showing an operation of a search process in step B13 of FIG. 3;

FIGS. 5A and 5B are diagrams for explaining the configuration of a database according to the first embodiment. FIG. 5A shows a state before encryption, FIG. 5B shows a state after encryption, and FIG. () Is a diagram showing a state after decoding.

FIG. 6 is a diagram illustrating a configuration of a column key and a row key according to the first embodiment.

FIG. 7 is a diagram showing a configuration of a database according to a second embodiment of the present invention. FIG. 7A shows a state before encryption,
FIG. 2B shows a state after encryption, and FIG. 2C shows a state after decryption.

FIG. 8 is a diagram showing a configuration of a synthetic key according to the second embodiment.

FIG. 9 is a block diagram showing a configuration of a database system according to a third embodiment of the present invention.

FIG. 10 is a flowchart showing an operation of an encryption process and a decryption process when a multidimensional space rotation method is used.

FIG. 11 is a diagram for explaining a calculation method of encryption when a multidimensional space rotation method is used.

[Explanation of symbols]

 11 ... CPU 12 ... Display Device 13 ... Input Device 14 ... Program Storage Device 15 ... Key Storage Device 16 ... Data Storage Device 16a ... Database Storage Area 16b ... Encryption Setting Information Storage Area 16c ... Search Setting Information Storage Area 16d ... Comparison Character Column storage area 20 first terminal device 21 search device 22 data storage device 30 second terminal device 31 search request device 32 decoding device

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 5B017 AA03 BA07 CA16 5B075 KK54 KK63 NK02 NR03 NR20 PP22 QP10 QT06 5B082 GA11 5J104 AA01 AA18 FA00 JA04 NA02 NA27 PA07 PA14

Claims (9)

[Claims] 1. An encryption method for encrypting data of a predetermined column item of a database using a common column key for the column item, and encrypting data of other column items using a unique row key for each row. A database management device comprising: an encrypting unit; and a storage unit that stores a database encrypted by the encrypting unit. 2. When the predetermined column item is set as a search target, data input for search is encrypted using a column key common to the predetermined column item, and the encrypted search data is used. 2. The database management apparatus according to claim 1, further comprising a database search unit that performs a search process by comparing the data with each item data of the encrypted database stored in the storage unit. 3. The encryption means encrypts data of a predetermined column item in a database using a column key common to the column item, and encrypts data of other column items with a unique row key for each row. 2. The database management apparatus according to claim 1, wherein the column item is encrypted using a common column key in combination. 4. The encryption means according to claim 1, wherein said encrypting means sequentially generates vectors in a multidimensional space based on a predetermined function, and uses an encryption method in which these vector components are used as a key stream of encryption. 2. The database management apparatus according to claim 1, wherein the database is encrypted using a key as a constant of the function. 5. A database system comprising: a first information terminal having a database; and a second terminal requesting the first information terminal to search the database, and connecting these via a network. On the side of the first information terminal, data of a predetermined column item of the database is encrypted using a column key common to the column item, and a unique row key is provided for each row for data of other column items. When the second information terminal requests a database search with the predetermined column item as a search target, data input for search is encrypted using a column key common to the column item. And sends the encrypted search data to the first information terminal via the network, and on the first information terminal side, encrypts the encrypted search data based on the search data. A database system which performs a database search process and returns data obtained as a result of the search in an encrypted state to the second information terminal via the network. 6. A database management apparatus for managing a database in which data of a predetermined column item is encrypted using a column key common to the column item, wherein a data search for the predetermined column item is performed. And encrypting the data input for search using the column key, and comparing the encrypted search data with each item data of the encrypted database. A database management device comprising: a search unit that performs a search process. 7. An encryption device for encrypting at least one column item data of a database using a column key common to the column items, wherein the original data acquisition means acquires original data to be encrypted. Vector generating means for sequentially generating vectors defined in a closed region of an n-dimensional (n ≧ 1) space using at least a function determined by the column key; and the original data obtaining means. A logical operation means for performing a logical operation on the bit data of the original data obtained by the above and the components of the vector generated by the vector generation means to generate encrypted data. 8. A computer encrypts data of a predetermined column item of a database using a common column key for the column item, and encrypts data of other column items using a unique row key for each row. A computer-readable recording medium storing a program for executing an encryption function to be encrypted and a search function to search data in an encrypted database obtained as a result of executing the encryption function. 9. A computer for managing a database in which data of a predetermined column item is encrypted using a column key common to the column item, when a data search for the predetermined column item is performed. An encryption function for encrypting data input for search using the column key, and a search process by comparing the encrypted search data with each item data of the encrypted database. And a computer-readable recording medium recording a program for executing a search function.