KR20190022344A - Method and apparatus for securing data - Google Patents

Abstract

The present invention may provide an efficient data polymorphic dispersion technique for providing the confidentiality and the stability of the entire data and minimizing amount of computation required to protect data since the entire original data is prevented from being restored when an attacker obtains a part of distributed data and a legitimate user may accurately restore original data when distributed some data is damaged by providing a data protection device. The data protection device comprises: a memory storing information for data processing; a processor dividing original data into a plurality of pieces of partial data and generating a plurality of pieces of partitioned data by arbitrarily determining positions in each original data of the partial data; and a communication unit transmitting the partitioned data to each of the servers.

Description

[0001] METHOD AND APPARATUS FOR SECURING DATA [0002]

The present invention relates to a data protection method and an apparatus for protecting data.

In an IoT-based big data environment where all human lives are gathered and utilized as information, attack techniques against data collected on servers are getting more intelligent and diverse. Known countermeasures against existing attack techniques have a limitation in completely blocking data threats. In addition, as can be seen from the frequent occurrence of personal information leakage, it is very difficult to prevent hacking at its source. Therefore, even if an attacker acquires the authority of the server on which data is managed, .

As a method to fundamentally defend against attacks on data or leakage of data, various types of encryption technologies have been studied. For example, searchable ciphers, order preservation ciphers, cipher data duplication processing, and proprietary management techniques, and providing additional functions for cipher data based on data encryption are the main research topics. However, the encryption-based protection methods as described above have a disadvantage in that they are limited in their application range because they have a great restriction to utilize the stored data, and they are ineffective in recent data attack such as Raman software.

On the other hand, research related to the verification of stability of data stored in the server is also being studied. Typically, there are PDP (Provable Data Possession) and PoR (Proof of Retrievability) technologies. While the PDP technology provides only the presence of the data that the server actually stores, PoR technology provides a restore function to the original data when the data is partially damaged. Data restoration technology of PoR technology is based on applying error correcting code (ECC).

The terminal (data owner) processes its own data and generates recoverable encoded data even if the data is partially damaged. Various error correcting code schemes have been used as such coding schemes. This coded data provides original data restoration function for some data corruption, and is used as a data protection function for a network transmission error or a physical error in a part of the disk.

However, since an attacker can compromise all the data stored in the server in the situation where an attacker has acquired the privilege of a specific server through an attack such as a hacking, a general error correcting code application method can not cope with such an attack. In addition, since the error correction code does not guarantee the confidentiality of data, there is a disadvantage that the original data can be easily restored when the entire encoded data is acquired by an attacker. In particular, in case of Ransomware, which is a recent problem, it is difficult to prevent the Ransomware attack by applying the general error correction code because it corrupts all the data of the attacked server.

In other words, although the data encryption technology provides original protection function for the data stored in the server, it has a problem of deteriorating the data utilization, and recently it can not provide a solution against the data corruption attack such as Ransomware.

Embodiments of the present invention provide a method and apparatus for solving the problem of security for data distributed and stored in a server and the problem of usability for distributed data.

A data protection apparatus according to an embodiment of the present invention includes a memory for storing information for data processing; A processor for dividing original data into a plurality of partial data and randomly determining positions of the plurality of partial data in the original data to generate a plurality of divided data; And a communication unit for transmitting the plurality of divided data to a plurality of servers, respectively.

The processor may replace the positions of different partial data among the plurality of partial data with each other.

The processor may obtain a secret key stored in the memory, generate a pseudo-random number using the secret key, and randomly determine the location of the plurality of partial data using the pseudo-random number.

Wherein the processor obtains encoded data by applying an error correcting code (ECC) to the original data, generates a plurality of noise fragments using the secret key, and randomly assigns the noise fragments to the encoded data The encoded data may be divided into a plurality of encoded partial data, and the plurality of divided data may be generated by randomly determining the positions of the plurality of encoded partial data.

The processor may generate the noise string using the secret key, and determine, based on the noise string, the location and value at which the noise fragments are to be inserted.

Wherein the communication unit receives the plurality of divided data from the plurality of servers and the processor can restore the position of the plurality of partial data included in the plurality of divided data to a position before the change in the original data have.

The processor may remove the plurality of noise fragments from the plurality of partitioned data using the secret key.

According to the embodiment of the present invention, similar to the document shredder, digital information is decomposed into small pieces and the data is distributed and stored in a plurality of places, so that even if the attacker acquires a part of the distributed data, And a legitimate user can accurately restore the original data even if some distributed data is damaged, thereby providing the confidentiality and resilience of the entire data.

In addition, by not using the data encryption technique, it is possible to provide an efficient data polymorphic dispersion technique that minimizes the amount of computation required to protect data.

Figure 1 illustrates a data protection system in accordance with an embodiment of the present invention.
2 is a flowchart illustrating a data dividing method according to an embodiment of the present invention.
3 is a flowchart illustrating a method of adding a noise piece according to an embodiment of the present invention.
4 is a flowchart illustrating a data replacement method according to an embodiment of the present invention.
5 is a flowchart illustrating a data restoration method according to an embodiment of the present invention.
FIG. 6 illustrates a data protection process according to an embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present invention. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. In order to clearly illustrate the present invention, parts not related to the description are omitted, and similar parts are denoted by like reference characters throughout the specification.

Figure 1 illustrates a data protection system in accordance with an embodiment of the present invention.

1, a data protection system 100 according to an embodiment of the present invention includes a terminal 110 that owns data and a plurality of servers 121, 122, 123, and 124 that store data. can do. The terminal 110 may include a communication unit 111, a memory 113, a communication unit 111, and a processor 112 for controlling the memory 113.

The communication unit 111 sets a communication method (for example, a wireless communication) with each of the servers 121, 122, 123 and 124 and transmits data to the servers 121, 122, 123, 124).

The memory 113 can acquire and store data from the outside. The memory 113 may store information (e.g., secret keys, etc.) for processing data according to the data protection method of the present invention.

The processor 112 may process and divide the data stored in the memory 113 and send it to the respective servers 121, 122, 123, To this end, the processor 112 applies an error correcting code (ECC) to the data stored in the memory 113, divides the encoded data to which the error correction code is applied, Data can be transmitted to each of the servers 121, 122, 123, Various error correcting code applying methods are used as the error correcting code applying method (encoding method), and the present invention is not limited to the specific error correcting code applying algorithm.

Each of the servers 121, 122, 123, and 124 may store the divided divided data in a storage device (not shown) included in each server.

Each of the servers 121, 122, 123, and 124 transmits the divided data to the terminal 110. The divided data is then transmitted to the terminal 110, 110). The terminal 110 can recover the original data by collecting the received divided data.

2 is a flowchart illustrating a data dividing method according to an embodiment of the present invention.

As shown in FIG. 2, the data protection method according to the embodiment of the present invention may include an original data acquisition step (S201), a separation step (S203), and a data transmission step (S205).

In step S201, the terminal (e.g., the terminal 110 in FIG. 1) can acquire original data in which a plurality of partial data are arranged in a predetermined order.

First, the processor 112 may obtain from the memory 113 a secret key (SK) of X-bits for use in processing data.

The processor 112 may then select the values of the variables S (first variable), L (second variable), K (third variable), N (fourth variable). S (first variable) means the number of servers for storing original data. L (second variable) denotes the size of the minimum unit bit string for processing the original data. That is, the processor 112 treats the bit string of L-bits as one character constituting the partial data. The processor 112 divides the original data into partial data of L-bits units and processes the partial data. K (third variable) is a variable for an error correction code, and is the number of partial data included in one block data. That is, one block data includes a bit string of K * L bits, which is a bit string as many as the number K of partial data included in one block data and the size L of the minimum unit bit string per partial data, Can be a multiple of K * L. N (fourth variable) is a parameter for an error correction code, which is a parameter for determining an error correcting bound and is determined from a desired error correction range and K (third variable). N (fourth variable) is generated based on the error correction code.

The processor 112 applies the error correcting code to the original data to generate encoded data including a plurality of blocks including N partial data. That is, each block composed of KL-bits is extended to NL-bits, so that the length of the entire encoded data is extended to N / K ratio as compared with the original data.

The processor 112 may add noise fragments generated according to an embodiment of the present invention to the encoded data.

In step S203, the processor 112 may randomly divide the partial data to generate a plurality of divided data.

For example, the processor 112 may divide the original data into a plurality of partial data, randomly determine positions of the plurality of partial data, and generate a plurality of divided data using the plurality of partial data.

For example, if the encoded data F 'generated by adding noise fragments is F' = F '[1] || F '[2] || F '[3] || ... F '[S] || F '[S + 1] || F '[S + 2] .... Each F '[i] means partial data.

For example, the processor 112 may determine the partitioning scheme for the server to store each partial data.

For example, the processor 112 may select a sequential partitioning scheme. The sequential division scheme is a scheme in which the processor 112 sequentially allocates partial data composed of partial data of L-bits length to each server based on a predetermined initial sequence in the original data of the partial data.

In a sequential partitioning scheme, the processor 112 divides the first partitioned data to be transferred to the first server into F'1 = F '[1] || F '[S + 1] || ... || F '[tS + 1] || ... < / RTI > Similarly, the processor 112 converts the a-th partial data to be transmitted to the a-th server into F'a = F '[a] || F '[S + a] || ... || F '[tS + a] || ... can be configured.

Also, unlike the sequential division method, the processor 112 can determine a random division method.

For example, according to the random partitioning scheme, the processor 112 may obtain the function P using a pseudorandom generating function. The function P is a function for generating a fixed bit stream which can not be cryptographically distinguished from the random bit string from the secret key SK obtained from the memory 113. The function P can be generated using the unidirectionality of a cryptographic based technique such as a hash function, symmetric key cryptography, and the like.

P: {0,1} ^X -> {1, 2, ..., S}, and X can be defined as the length of the secret key (SK).

The terminal is F '= F' [1] || F '[2] || F '[3] || ... F '[S] || F '[S + 1] || F '[S + 2] || ..., i-th partial data F '[i] to the P (SK + i) th server. That is, the index set A = {i | Assuming that A = {a1, a2, a3, ...} for P (SK + i) = a}, the ath partial data is F 'a = F' [a1] || F '[a2] || F '[a3] || ... < / RTI >

If the function P is a cryptographically secure pseudorandom function and the original data F has a sufficient length (e.g., a size greater than a predetermined bit), then the amount of partial data stored in each server is similarly divided.

In step S307, the processor 112 may transmit the plurality of divided data to the servers (e.g., the servers 121, 122, 123, and 124 of FIG. 1). The processor 112 may delete the partitioned data stored in the memory 113 after transmitting the partitioned data, and the server 120 may store the partitioned data transmitted.

3 is a flowchart illustrating a method of adding a noise piece according to an embodiment of the present invention.

As shown in Fig. 3, in step S301, the processor 112 can acquire the encoded data and obtain the secret key stored in the memory 113. Fig. First, the processor 112 converts the encoded data EF = EF [1] || EF [2] || EF [3] || ..., | EF [i] | = L, and then obtain the secret key to add noise fragments to the encoded data.

In step S303, the processor 112 may apply the secret key to a pseudo-random number generation function to generate a noise string. First, the processor 112 determines a pseudorandom generating function P. Where P is a function that generates a noise string from a given secret key (SK) and is defined as P: {0,1} ^X -> {0,1} ^* for convenience. Processor 112 generates a noise string RT = P (SK) from the secret key SK.

In step S305, the processor 112 may divide the noise string to generate a plurality of noise fragments. Processor 112 selects variable Y, and the noise string RT = j1 || v1 || j2 || v2 || j3 || v3 ... to generate noise fragments. Each j is a Y-bits noise string, and each v is an L-bits noise string. The processor 112 generates a set of noise fragments RS = {(j1, v1), (j2, v2), (j3, v3) ...} from the noise string RT.

In step S307, the processor 112 may insert noise fragments into the encoded data. First, the processor 112 defines an INS function. The processor 112 determines EF = EF [1] || EF [2] || EF [3] || (J, v) is added to EF [i] = EF [i-1] for all i> j, Add a value with EF [j] = v. That is, the processor 112 inserts a noise piece having a value of v at the position of the jth partial data, and increments the index of all the partial data thereafter by one. The processor 112 sequentially adds the encoded data to the encoded data starting from the first element of the noise fragments RS and performs INS (E1, (j1, v1)) for the first element j1, v1 added first. The processor 112 may add the second and subsequent elements of the noise fragments to the encoded data by accumulating the first index of each element. In other words, the processor 112 performs INS (EF, (j1 + j2, v2) for the second element j2, v2 and INS . + ji, vi).

The processor 112 may divide the encoded data F 'to which noise fragments have been added according to the division scheme described above with reference to FIG. 2, and then transmit the divided data F' to the server.

On the other hand, since inserting the individual noise fragments into the existing encoded data can increase the amount of computation of the terminal, the processor 112 calculates the value of EF [j] = v for one partial data (j, v) And the data may not be changed for other partial data.

4 is a flowchart illustrating a data replacement method according to an embodiment of the present invention.

As shown in FIG. 4, in step S401, the processor 112 selects a secret key, and in step S403, the secret key may be used to generate a noise string. The processor 112 determines a pseudorandom generating function P. P is defined as P: {0, 1} ^X -> {0, 1, ..., M}. M is F '= F' [1] || F '[2] || F '[3] || ... || Is the number of all partial data included in F '[M].

In step S405, the processor 112 may generate the pseudo-random number Pi using the noise string. The processor 112 may apply the noise string to a pseudorandom number generation function to generate a pseudo-random number Pi = P (SK + i).

In step S407, the processor 112 may change the data value of each partial data using the pseudo-random number. Specifically, the processor 112 may replace the position of each partial data. Processor 112 may replace the positions of the values of F '[i] partial data and F' [Pi] partial data. Processor 112 may replace F '[i] with F' [Pi] values and F '[Pi] with F' [i] values.

5 is a flowchart illustrating a data restoration method according to an embodiment of the present invention.

As shown in FIG. 5, the processor 112 applies error correction coding to the original data in step S501, applies noise to the encoded data that is original data to which error correction coding is applied using the secret key in step S503 , A plurality of partial data arranged in a predetermined order in step S505 is randomly divided using a secret key, and in step S507, the divided data including a plurality of partial data can be transmitted to a plurality of servers.

After the original data is divided and transmitted to the respective servers through the steps S501 to S507, the processor 112 may receive a plurality of partial data included in the plurality of divided data from the plurality of servers in step S509 .

In step S511, the processor 112 can restore the order among the plurality of partial data in the original order in the original data using the secret key. If the data partitioning is sequential partitioning, the processor 112 determines that the a-th partial data F'a = F'a [1] || F'a [2] || F'a [3] ... F '= F'1 [1] || F'2 [1] || F'3 [1] || ... || F'S [1] || F'1 [2] || ... and so on to reconstruct F '. If a random partitioning scheme is applied, the processor 112 generates a pseudorandom generating function P: {0,1} ^X -> {1, 2, ..., S} (SK). For example, the processor 112 calculates P (SK + 1) = a to obtain the first partial data of F 'and sets the value of F'a [1] as the first partial data of F'a to F' After setting to [1], delete F'a [1] from F'a. Similarly, the processor 112 calculates P (SK + i) = ai to obtain the first partial data of F'ai to obtain the i-th partial data F '[i]. This process can be performed sequentially from F '[1] to F' [M] to reconstruct the entire F '.

In step S513, the processor 112 removes a noise piece from F 'to construct encoded data. For example, if a noise piece is added between partial data, the processor 112 generates the same noise string RT = P (SK) using P and the secret key SK used in step S503. The processor 112 further decomposes the noise string RT into noise fragments RS = {(j1, v1), (j2, v2), (j3, v3), ...} ), The removal of the noise fragments from the partial data is performed from the last element of the noise pieces RS, and the piece of F '[j1 + j2 + ... ji] is removed with respect to the element ji, vi, To obtain the encoded data.

In step S515, the processor 112 can acquire the original data by equally applying the error correction code applied to the original data in step S501 to the encoded data. For example, if step S501 is performed by applying an erasure code instead of a general error correction code applying method, the entire data can be restored from a predetermined number of partial data among the data stored in a plurality of servers.

FIG. 6 illustrates a data protection process according to an embodiment of the present invention.

6, the terminal 110 may acquire the data 610 and generate the encoded data 620 by encoding (for example, applying an error correction code) to the data 610. [

The terminal 110 may obtain the secret key 630 and generate the plurality of noise fragments 650 through the pseudo random number generation function 640 using the secret key 630. [

Next, the terminal 110 may add the noise fragments 650 to the encoded data 620.

Lastly, the terminal 110 may divide the encoded data 660 to which noise fragments have been added into a plurality of divided data for transmission to each of the servers 670. However, in the division, the terminal 110 has a sequential division method in which a plurality of partial data of the encoded data 660 are firstly arranged in order and transmitted to the servers, but in the present invention, It can be randomly divided and transmitted to each of the servers 670 irrespective of the arranged order.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, It belongs to the scope of right.

Claims (7)

A memory for storing information for data processing;
A processor for dividing original data into a plurality of partial data and randomly determining positions of the plurality of partial data in the original data to generate a plurality of divided data; And
And a communication unit for transmitting the plurality of divided data to a plurality of servers, respectively
Data protection device. The method according to claim 1,
Wherein the processor replaces the positions of different partial data among the plurality of partial data with each other
Data protection device. 3. The method of claim 2,
Wherein the processor obtains a secret key stored in the memory, generates a pseudo-random number using the secret key, and randomly determines a position of the plurality of partial data using the pseudo-random number
Data protection device. The method of claim 3,
Wherein the processor obtains encoded data by applying an error correcting code (ECC) to the original data, generates a plurality of noise fragments using the secret key, and randomly assigns the noise fragments to the encoded data And dividing the encoded data into a plurality of encoded partial data, and generating the plurality of divided data by randomly determining a position of the plurality of encoded partial data
Data protection device. 5. The method of claim 4,
The processor generates a noise string using the secret key, and based on the noise string, determines a position and a value at which the noise fragments are to be inserted
Data protection device. The method according to claim 1,
Wherein the communication unit receives the plurality of divided data from the plurality of servers,
Wherein the processor restores the position of the plurality of partial data included in the plurality of divided data to a position before the change in the original data
Data protection device. The method according to claim 6,
Wherein the processor removes the plurality of noise fragments from the plurality of divided data using the secret key
Data protection device.

Images