A method for processing personal information using a trusted execution environment of a smart contract platform, a computer-readable recording medium including instructions for executing the personal information processing method, and a system

Abstract

This invention provides a personal information processing method and system that utilizes a trusted execution environment of a smart contract platform capable of restoring personal information to an encryption key. [Solution] A personal information processing method using a trusted execution environment of a smart contract platform according to an embodiment of the present invention includes the steps of: a data processing platform server responding to a data processing request received from a data processing request terminal in accordance with a smart contract distributed on a blockchain and generating a trusted execution environment including a data processing code; the trusted execution environment decrypting first data obtained from a data generation terminal; the trusted execution environment processing the decrypted first data based on the data processing code to generate a data processing result; and the trusted execution environment providing the data processing result to the data processing request terminal in accordance with the smart contract.

Description

Technical Field

[0001] The present invention relates to a personal information processing method and system using a trusted execution environment of a smart contract infrastructure.

Background Art

[0002] Blockchain is a decentralized distributed system. Decentralization means that there is no single central agency to perform functions, and a large number of participants perform together. Basically, a blockchain network is composed of a P2P network, and uses a consensus algorithm to generate blocks through a common process. The use of hash is core in blockchain, and the combination of hash and the fact that a large number of participants execute the same process through the same data prevents data tampering or forgery.

[0003] A smart contract is a series of programming codes that enable the automatic execution of contract fulfillment based on a blockchain that handles parts related to assets and trust. The content of the code created by a developer is stored in a block of the blockchain, and a user can access the smart contract address to execute the code.

[0004] The reason it is called a smart contract is that, like general transaction contents, the content of the smart contract code is included in block information and thus cannot be manipulated, and the program operates according to the coded content, so it is suitable for working on fixed content like a contract.

[0005] Smart contracts created on Ethereum utilize the Solidity language and operate on the EVM. The virtual machine specifically designed to execute Ethereum smart contracts is called the Ethereum Virtual Machine (EVM). Ethereum smart contracts are created using languages ​​such as Solidity and Viper and are widely applied and used in the blockchain field. Various monitoring web applications exist to provide information on smart contracts on the Ethereum network. The most representative are Etherescan, which is provided by Ethereum itself, Alvio, which provides information in a visual form, and Remix, a Solidity IDE.

[0006] On the other hand, personal information de-identification is a technology that removes or replaces some or all of personal information through methods such as data deletion, pseudonymization, categorization, and data masking, so that a specific individual cannot be identified even when combined with other information. It is used as a measure to minimize the risk of personal information leakage and privacy violations during the data utilization process, and the issue of de-identification has recently attracted attention in the context of changes in the information and communication technology (ICT) environment, exemplified by Big Data, and the accompanying discussions surrounding the protection of personal information.

[0007] When personal information is anonymized, the generated data becomes unidentifiable, making it impossible to further identify individuals. However, in certain situations, such as tracking criminal targets or identifying intruders, identified personal information is still necessary. Therefore, personal anonymization requires technology that can hide sensitive data information that could potentially violate privacy, while still allowing personal information to be restored under limited conditions. [Overview of the project] [Problems that the invention aims to solve]

[0008] The technical problem that this invention aims to solve is to provide a personal information processing method and system that uses a trusted execution environment of a smart contract platform that protects an individual's privacy by de-identifying identified personal information using an encryption key, and that can restore personal information to an encryption key under restricted conditions.

[0009] Furthermore, the technical problem that the present invention aims to solve is to provide a personal information processing method and system that uses a trusted execution environment of a smart contract platform that fundamentally blocks access to personal information by pre-de-identifying personal information identified before a series of data processing processes such as data storage, streaming, and transfer.

[0010] Furthermore, the technical problem that the present invention aims to solve is to provide a personal information processing method and system that uses a trusted execution environment of a smart contract platform that can block data processing entities from viewing personal information, prevent leakage of personal information to external parties, and securely protect the privacy of entities that have provided data. [Means for solving the problem]

[0011] A personal information processing method using a trusted execution environment of a smart contract platform according to an embodiment of the present invention includes the steps of: a data processing platform server responding to a data processing request received from a data processing request terminal in accordance with a smart contract distributed on a blockchain and generating a trusted execution environment including a data processing code; the trusted execution environment decrypting first data obtained from a data generation terminal; the trusted execution environment processing the decrypted first data based on the data processing code to generate a data processing result; and the trusted execution environment providing the data processing result to the data processing request terminal in accordance with the smart contract.

[0012] According to the embodiment, the first data may include at least one of video data and audio data from which personal information has been de-identified, metadata, and first encryption key information.

[0013] According to the embodiment, the metadata may include first terminal information of the data generation terminal that generated the first data, the generation time of at least one of the video data and audio data before the deidentification process of the personal information, and the location of the data generation terminal.

[0014] According to the embodiment, the first encryption key information may include a second terminal information of the encryption key supply terminal that generated the first encryption key, and a first public key used to de-identify the personal information.

[0015] According to the embodiment, the step of decrypting the first data can be performed by identifying the data area corresponding to the personal information with the first personal key obtained from the encryption key supply terminal.

[0016] According to the embodiment, the first encryption key is an asymmetric key consisting of a pair of a first personal key and a first public key generated by an encryption key supply terminal, the first public key is generated based on the first personal key, and the first encryption key can be updated by the encryption key supply terminal according to a predetermined schedule.

[0017] According to the embodiment, the step of generating the data processing result may include: analyzing the decoded first data to determine the data processing target corresponding to the data processing request; selecting target video data containing the data processing target from the decoded first data; extracting metadata corresponding to the selected target video data; and processing the extracted metadata according to the data processing code to generate the data processing result.

[0018] According to the embodiment, the step of determining the data to be processed includes the step of recognizing an object in the video data within the decoded first data using an object recognition model included in the data processing code, and the step of determining the degree of agreement between the object characteristic information included in the data processing request and the recognized object, and identifying an object having a value equal to or greater than a standard agreement rate as the data to be processed.

[0019] According to the embodiment, the step of generating the data processing result may include: analyzing the decoded first data to determine the data processing target corresponding to the data processing request; selecting target acoustic data containing the data processing target from the decoded first data; extracting metadata corresponding to the selected target acoustic data; and processing the extracted metadata according to the data processing code to generate the data processing result.

[0020] According to the embodiment, the step of determining the data to be processed may include the steps of classifying the acoustic data in the decoded first data using an acoustic signal analysis model included in the data processing code, and determining the agreement rate between the object characteristic information included in the data processing request and the classified acoustic data, and identifying the acoustic data having a value equal to or greater than a standard agreement rate as the data to be processed.

[0021] According to the embodiment, the step after the step of providing the data processing request terminal may further include a step in which the trusted execution environment is destroyed in accordance with the smart contract.

[0022] According to the embodiment, the computer-readable recording medium storing one or more programs is configured to be executed by one or more processors of an electronic device, and the one or more programs include instructions for executing any one of the personal information processing methods using the trusted execution environment of the smart contract platform.

[0023] A personal information processing system using a trusted execution environment of a smart contract infrastructure according to another embodiment of the present invention includes a plurality of data generation terminals that capture images in different areas from each other, or collect sounds to generate temporary data, recognize personal information in the temporary data, and generate first data obtained by anonymizing the personal information based on a first encryption key, and a data processing platform server that generates a trusted execution environment that responds to a data processing request received from a data processing request terminal according to a smart contract distributed on a blockchain and provides a data processing result generated based on the first data to the data processing request terminal.

[0024] According to an embodiment, the first data may include at least one of video data and audio data obtained by anonymizing personal information, metadata, and first encryption key information.

[0025] According to an embodiment, the trusted execution environment can decrypt the first data obtained from each data generation terminal through the blockchain, and process the decrypted first data according to a data processing code to generate a data processing result.

[0026] According to an embodiment, the trusted execution environment can determine a first public key used for anonymization processing based on the first encryption key information, and decrypt the first data by identifying the personal information with a first personal key corresponding to the first public key.

[0027] According to an embodiment, the trusted execution environment analyzes the decrypted first data to determine a data processing target corresponding to the data processing request, selects target video data including the data processing target from the decrypted first data, and processes metadata corresponding to the selected target video data according to a data processing code to generate the data processing result.

[0028] According to the embodiment, the trusted execution environment uses the object recognition model included in the data processing code to recognize an object from the video data in the decrypted first data, determines the matching rate between the object characteristic information included in the data processing request and the recognized object, and can identify an object having a value equal to or higher than the reference matching rate as the data processing target.

[0029] According to the embodiment, the trusted execution environment analyzes the decrypted first data to determine a data processing target corresponding to the data processing request, selects target audio data including the data processing target from the decrypted first data, and can generate the data processing by processing metadata corresponding to the selected target audio data according to the data processing code.

[0030] According to the embodiment, the trusted execution environment uses the acoustic signal analysis model included in the data processing code to classify the acoustic data in the decrypted first data, determines the matching rate between the object characteristic information included in the data processing request and the classified acoustic data, and can identify acoustic data having a value equal to or higher than the reference matching rate as the data processing target.

[0031] According to the embodiment, a personal information processing system using the trusted execution environment of the smart contract platform may further include an encryption key supply terminal that generates the first encryption key according to a predetermined schedule and provides it to each data generation terminal.

[0032] According to the embodiment, each of the data generation terminals can receive a second encryption key having a different value from each other.

Advantages of the Invention

[0033] According to the personal information processing method and system using the trusted execution environment of the smart contract platform according to the embodiment of the present invention, even if non-identification of personal information is performed, it can be restored and used so that personal information can be identified in a restricted situation, and after deriving the data processing result, it can be safely erased to protect personal privacy.

[0034] Furthermore, according to the personal information processing method and system using the trusted execution environment of the smart contract platform according to the embodiment of the present invention, personal information identified prior to a series of data processing processes such as data storage, streaming, and transfer can be de-identified in advance, thereby fundamentally blocking access to personal information at the hardware level.

[0035] Furthermore, according to the personal information processing method and system using the trusted execution environment of the smart contract platform according to the embodiment of the present invention, when a smart contract is terminated, the trusted execution environment is destroyed and the stored data is deleted at the same time, thereby preventing the collected data from being leaked to the outside, fundamentally blocking the operator of the data processing platform server from viewing the data, and protecting the privacy of the data subject.

[0036] Furthermore, according to the personal information processing method and system using the trusted execution environment of the smart contract platform according to the embodiment of the present invention, personal information generated by data generation terminals such as cameras, CCTVs, and home cameras can be fundamentally encrypted, not disclosed to the data subject, and used restrictively only under specific analytical conditions, thereby protecting the privacy of individuals.

[0037] Furthermore, according to the personal information processing method and system using the trusted execution environment of the smart contract platform according to the embodiment of the present invention, a large number of smart contracts can be executed simultaneously using a trusted execution environment that is executed independently for each smart contract.

[0038] Furthermore, according to the personal information processing method and system using the trusted execution environment of the smart contract platform according to the embodiment of the present invention, by recording a destruction order flag on the blockchain, it is possible to ensure that the trusted execution environment to which the data was provided is stably removed and that the data has been securely deleted without being leaked to the outside.

[0039] Furthermore, according to the personal information processing method and system using the trusted execution environment of the smart contract platform according to the embodiment of the present invention, the data processing request terminal can obtain the desired data processing result without directly collecting or processing the data necessary for data processing, and since the data necessary for data processing is not leaked to the data processing request terminal, the privacy of the entity that provided the data can be securely protected. [Brief explanation of the drawing]

[0040] [Figure 1] This is a schematic block diagram of a personal information processing system using a trusted execution environment of a smart contract platform according to an embodiment of the present invention. [Figure 2] This is a block diagram illustrating the first data generation process of multiple data generation terminals according to an embodiment of the present invention. [Figure 3] This is a block diagram illustrating a method for generating a data processing process for a trusted execution environment according to an embodiment of the present invention. [Figure 4] This is a block diagram illustrating a method for generating data processing results in a trusted execution environment according to an embodiment of the present invention. [Figure 5a] This figure illustrates a method for processing personal information using a trusted execution environment of a smart contract platform according to an embodiment of the present invention. [Figure 5b] This figure illustrates a method for processing personal information using a trusted execution environment of a smart contract platform according to an embodiment of the present invention. [Figure 5c] This figure illustrates a method for processing personal information using a trusted execution environment of a smart contract platform according to an embodiment of the present invention. [Figure 6] This figure illustrates a method for processing personal information non-identifiable on a data generation terminal according to another embodiment of the present invention. [Figure 7] This figure illustrates a method for identifying personal information in a trusted execution environment according to an embodiment of the present invention. [Figure 8] This figure illustrates a smart contract according to an embodiment of the present invention. [Figure 9]This is a flowchart illustrating a method for processing personal information using a trusted execution environment of a smart contract platform according to an embodiment of the present invention. [Figure 10] This is a flowchart illustrating a method for processing personal information using a trusted execution environment of a smart contract platform according to another embodiment of the present invention. [Modes for carrying out the invention]

[0041] The advantages, features, and methods for achieving them of the present invention will become apparent by referring to the embodiments described below in detail with accompanying figures. However, the present invention is not limited to the embodiments disclosed below and can be embodied in a variety of different forms, and these embodiments are provided merely to complete the disclosure of the present invention and to fully inform a person ordinary in the art to which the present invention belongs of the scope of the present invention, and the present invention is defined only by the claims.

[0042] The terms used herein are for illustrative purposes only and are not intended to limit the invention. In this specification, singular forms include plural forms unless otherwise specified in the context. The terms “comprises” and / or “comprising” used herein do not exclude the presence or addition of one or more other components in addition to those mentioned. Throughout the specification, the same figure number refers to the same component, and “and / or” includes each component mentioned and all combinations of one or more. Even though terms such as “first,” “second,” etc., are used to describe various components, these components are not limited by these terms. These terms are used to distinguish one component from others. Therefore, the first component mentioned below may be the second component within the technical concept of the invention.

[0043] Unless otherwise specified, all terms used herein (including technical and scientific terms) should be used in the sense that would be ordinarily understood by those skilled in the art in which the present invention pertains. Furthermore, terms defined in commonly used dictionaries should not be interpreted ideally or excessively unless explicitly defined otherwise.

[0044] Figure 1 is a schematic block diagram of a personal identification information de-identification processing system using a trusted execution environment of a smart contract platform according to an embodiment of the present invention. Referring to Figure 1, the personal identification information de-identification processing system 10, which uses the trusted execution environment of the smart contract platform, includes a data processing platform server 100, a data generation terminal 200, an encryption key supply terminal 300, and a data processing request terminal 400.

[0045] The data processing platform server 100 is a device capable of hosting an online network and specifying a network address, and can provide a data processing platform through which data is shared online with the data generation terminal 200, the encryption key supply terminal 300, and the data processing request terminal 400. The data processing platform server 100 can perform a series of processes on the blockchain, such as generating, distributing, transferring, and terminating smart contracts, through the data processing platform.

[0046] The data processing platform server 100 can generate smart contracts for data processing and distribute them to the blockchain. Smart contracts for data processing are electronic contracts that are automatically concluded when pre-set conditions are met on the blockchain. The data processing platform server 100 can generate smart contracts that record a series of contract details necessary for data processing, such as data collection, analysis, and result derivation, and distribute them to the blockchain.

[0047] Smart contracts are recorded in blockchain blocks and can be executed on the blockchain. For example, a blockchain can be embodied as the Ethereum blockchain, and smart contracts created with Solidity, Viper, etc., can be distributed to the blockchain through the Ethereum Virtual Machine (EVM). Furthermore, all transactions within the blockchain are hashed using SHA-256 and the hash value is stored, and the hash value may also be stored in the blockchain block to protect individual privacy and prevent overloading.

[0048] The smart contract distributed by the data processing platform server 100 is automatically concluded when a data processing request that conforms to the prescribed terms and conditions is received from the data processing request terminal 400, and the contract can be terminated when the data processing result is returned in response to the data processing request.

[0049] Here, the data processing request refers to a data processing request that can be processed by the data processing platform server 100, and the data processing code necessary for data processing may be provided from the data processing request terminal 400 or generated by the data processing platform server 100.

[0050] For example, when a smart contract is concluded between the data processing platform server 100 and the data processing request terminal 400, the data processing platform server 100 can obtain the data processing code provided by the data processing request terminal 400 and use it for data processing.

[0051] For example, when a smart contract is concluded between the data processing platform VALUE 100 and the data processing request terminal 400, the data processing platform server 100 can determine the processing model necessary for the data processing request, and then lead the determined processing model for use in data processing.

[0052] The data processing platform operated by the data processing platform server 100 generates a Trusted Execution Environment (TEE) for data processing when a smart contract is concluded. Specifically, when the data processing platform server 100 receives a data processing request from the data processing request terminal 400, it generates a trusted execution environment including a data processing code and an encryption key in response to the data processing request.

[0053] Specifically, the data processing platform server 100 can generate a primitive trust execution environment, which serves as the source for generating other trust execution environments. A primitive trust execution environment refers to a virtual execution environment in which no arbitrary data processing code or encryption key has been generated.

[0054] The original trust execution environment is a secure execution environment provided by an independent security area, and can be generated as an image file to embody a virtual execution environment. However, it is not limited to this; the original trust execution environment can also be implemented as either a hardware-based solution or a hardware / software-based solution.

[0055] According to the embodiment, when a primitive trust execution environment embodied as a primitive image file is generated, the data processing platform server 100 can generate a hash value for the primitive image file, generate a transaction including the hash value, and distribute it to the blockchain. This allows the data processing platform server 100 to publicly announce that a primitive trust execution environment for generating a trust execution environment has been generated.

[0056] The data processing platform server 100 can generate a trusted execution environment that includes a data processing code and a second encryption key that respond to a data processing request, based on the original trusted execution environment. The trusted execution environment, like the original execution environment, is a secure execution environment provided by an independent security area, and can be generated as a first image file for realizing a virtual execution environment.

[0057] According to the embodiment, the data processing platform server 100 can determine the authenticity of the source image file that forms the basis for generating the source trust execution environment before generating or executing a trust execution environment. The data processing platform server 100 can generate a hash value of the source image file and compare it with the hash value of the source image file distributed on the blockchain to see if they match, and if they match, the authenticity of the source image file can be ensured. Then, based on the source trust execution environment whose authenticity is guaranteed, the data processing platform server 100 can generate a trust execution environment that includes a data processing code and a second encryption key corresponding to the data processing request.

[0058] The second encryption key of a trusted execution environment includes the account address of the trusted execution environment, and a public-private key pairing mechanism can be used to generate the second encryption key to indicate the uniqueness of the blockchain participant. The second encryption key includes a second public key, which is the account address, and a second private key that controls the said second public key. For example, if the trusted execution environment operates on the Ethereum blockchain, the public key can be an Eternally Owned Address (EOA).

[0059] The trusted execution environment generates a 256-bit random number when generating an account address, sets this as the second personal key, and can extract a unique second public key by applying an elliptic curve cryptography algorithm to the second personal key.

[0060] For example, a trusted execution environment can generate a second personal key by generating random 256-bit data and encoding the generated 256-bit data into a 64-digit hexadecimal sequence. Based on this second personal key, a second public key can be generated using an elliptic curve cryptography algorithm. The trusted execution environment can then convert the public key into a Keccak256 hash value to generate 256-bit binary data, remove the first 96 bits of the binary data, and then encode the remaining 160 bits of binary data into a hexadecimal sequence to generate the account address, which is the second public key.

[0061] However, as an encryption algorithm that generates a public key based on a personal key, it is possible to use not only elliptic curve cryptography algorithms, but also other encryption algorithms such as RSA (Rivest, Shamir, and Adleman) and ElGamal encryption algorithms.

[0062] The second public key of the trusted execution environment, generated through this process, is made publicly available via the blockchain, and data encrypted with the second public key is configured to be decryptable only with the second private key of the trusted execution environment.

[0063] According to the embodiment, when a trusted execution environment embodied as a first image file is generated, the data processing platform server 100 can generate a hash value for the first image file, generate a transaction including the hash value, and record it on the blockchain. This allows the data processing platform server 100 to publicly demonstrate that the trusted execution environment corresponding to the smart contract was executed securely without tampering or forgery.

[0064] The trusted execution environment can request data collection from the data generation terminal 200 in order to perform data processing corresponding to the data processing code. To this end, the trusted execution environment can generate a first transaction requesting first data from the data generation terminal 200 and record it on the blockchain. At this time, the trusted execution environment can distribute another smart contract for data collection on the blockchain, or it can generate and record only a transaction to notify data collection on the blockchain.

[0065] According to the embodiment, when a data collection target is specified from the data processing request terminal 400 at the time of concluding a smart contract, the trusted execution environment can request data transfer from the data collection target. That is, the trusted execution environment can specify the account of a particular data generating terminal and record a transaction on the blockchain requesting data transfer from the specified data generating terminal.

[0066] In other embodiments, the trusted execution environment can record transactions, including the type and content of data necessary for data processing, on the blockchain and request data transfer from any data generation terminal 200. That is, the trusted execution environment can publish data collection conditions without specifying a data collection entity and collect data from all data generation terminals 200 that possess data that satisfies the data collection conditions.

[0067] On the other hand, the first data is a data area corresponding to personal information that has been de-identified by the first encryption key, and identification processing of the said data area is necessary for data processing. Therefore, the trusted execution environment can generate a second transaction requesting the first encryption key from the encryption key supply terminal 300 in order to execute data processing corresponding to the data processing code, and record it on the blockchain.

[0068] The trusted execution environment can perform data processing based on data collected from the data generation terminal 200. Since the trusted execution environment has inbound and outbound rules configured, it receives only data in a format corresponding to the data processing code, and the data processing results are returned externally in a fixed format.

[0069] The trusted execution environment can perform data processing if the first data collected from the data generation terminal 200 meets the criteria required for data processing. However, since some data areas of the first data collected from the data generation terminal 200 that correspond to personal information are de-identified using the first encryption key of the encryption key supply terminal, the trusted execution environment can perform data processing according to the data processing code after identifying the data using the first encryption key.

[0070] Here, the first encryption key is an asymmetric key consisting of a pair of the first personal key and the first public key of the encryption key supply terminal 300, and means an encryption key generated by the encryption key supply terminal 300 according to a predetermined schedule.

[0071] On the other hand, if the first personal key among the first encryption keys is leaked to the outside, there is a risk that the personal information of the first data will be made public and privacy will be violated. Therefore, the trusted execution environment can obtain the first personal key, which has been encrypted with the second public key, from the encryption key supply terminal 300. The trusted execution environment can then decrypt the encrypted second personal key using the first personal key and decrypt the first data based on the first personal key.

[0072] The trusted execution environment can perform data processing if the decrypted first data meets the criteria required for data processing. In particular, the trusted execution environment can generate data processing results by aggregating the data processing of the first data received from multiple data generation terminals 200.

[0073] Specifically, the trusted execution environment can process the decoded first data using a variety of artificial intelligence models included in the data processing code. Here, the artificial intelligence model can utilize at least one of the following as the artificial intelligence algorithm: a multi-layered perceptron, which is a type of artificial neural network (ANN), and a recurrent neural network (RNN).

[0074] An artificial neural network model can include an input layer, a hijacking layer, and an output layer, each consisting of multiple nodes, and each layer can be expanded or contracted to reflect changes over time. An artificial neural network can consist of a convolutional neural network, but is not limited to this; other artificial intelligence algorithms such as multi-layered perceptrons and cyclic neural networks (RNNs) can also be used.

[0075] On the other hand, while a trusted execution environment is provided with an artificial intelligence model that has completed training when generating data processing code, it is possible to directly train the artificial intelligence model in order to generate data processing results.

[0076] The trusted execution environment can recognize objects within the video data in the first data set, which has been decoded using an object recognition model. In particular, the object recognition model may be, but is not limited to, any of the following: a face detection algorithm using feature points (landmarks), a deep learning-based YOLO (You Look Only Once) model, MobileNet, R-CNN (Recursive Convolutional Neural Network), Faster-RCNN, or SSD (Single Shot Multibox Detector). Known algorithms and models for object recognition may also be used.

[0077] The trusted execution environment can identify the object to be processed by recognizing an object through an object recognition model and then determining the degree of agreement between the object characteristic information included in the data processing request and the recognized object.

[0078] Here, the object characteristic information included in the data processing request is information that allows for the recognition of a specific object, and may include, for example, feature point information and image information of the object. The trusted execution environment can determine the degree of agreement between the object characteristic information and the recognized object through the data processing target identification model. If the trusted execution environment determines that the degree of agreement between the object characteristic information and the recognized object is equal to or greater than the standard agreement rate, it can identify the recognized object as a data processing target.

[0079] If the trusted execution environment determines that the matching rate between the object characteristic information and the recognized object is below the standard matching rate, it may determine that the object is not related to the data processing request and may not use the video data containing the object for data processing.

[0080] The trusted execution environment can classify video data containing the data to be processed and select the classified video data as target video data. In other words, since video data that does not contain the data to be processed does not contain the information necessary for processing, the trusted execution environment can select only the target video data and process it.

[0081] The trusted execution environment can generate data processing results based on target video data and metadata selected through the data processing result generation model.

[0082] According to the embodiment, the trusted execution environment can extract metadata from target video data and process it according to the data processing code. For example, if a data processing request terminal requests the location and movement path of a specific X from a trusted execution environment, the trusted execution environment can extract metadata from video data containing the specific X and use the video recording time and video recording location of the data generation contained in the metadata to generate the time-series location of the specific X as a data processing result.

[0083] When a data processing request terminal requests video data containing a specific X from a trusted execution environment, the trusted execution environment can generate an image frame containing that specific X and metadata as the data processing result.

[0084] Furthermore, the trusted execution environment can classify the acoustic data within the first data set that has been decoded using an acoustic signal analysis model. The trusted execution environment can use, but is not limited to, MFCC (Mel-Frequency Cepstral Cofficient) as the acoustic signal analysis model, and can use any known algorithm model for acoustic signal analysis.

[0085] The trusted execution environment can classify acoustic data through an acoustic signal analysis model, and then determine the degree of agreement between the object characteristic information included in the data processing request and the classified acoustic data to identify the data to be processed.

[0086] Here, the object characteristic information included in the data processing request is information that can recognize a specific acoustic signal, and the trusted execution environment can determine the degree of agreement between the object characteristic information and the classified acoustic data through the data processing target identification model. If the trusted execution environment determines that the degree of agreement between the object characteristic information and the classified acoustic data by the data processing target identification model is equal to or greater than the standard agreement rate, it can identify the classified acoustic data as a data processing target.

[0087] For example, a trusted execution environment can select acoustic data containing the data to be processed as target acoustic data, and generate data processing results based on the selected target acoustic data and metadata.

[0088] In this way, the trusted execution environment can comprehensively analyze the initial data generated by different data generation terminals and generate data processing results that correspond to data processing requests.

[0089] Once data processing is complete, the trusted execution environment can return the data processing results to the data processing request terminal 400. The trusted execution environment can obtain the third public key provided by the data processing request terminal 400 when the smart contract is concluded, and can encrypt the data processing results using the third public key of the data processing request terminal 400 and return them to the blockchain.

[0090] The smart contract between the data processing platform server 100 and the data processing request terminal 400 is automatically concluded when a data processing request that matches the pre-configured conditions is entered, and the contract can be terminated when the data processing result corresponding to the data processing request is returned. Therefore, when the trusted execution environment encrypts the data processing result and generates a transaction containing the encrypted data processing result and records it in a blockchain block, it means that the smart contract has been fulfilled.

[0091] On the other hand, the trusted execution environment has a temporary status, maintaining its validity only for the duration that the smart contract is in existence. In other words, the trusted execution environment is generated only when a smart contract distributed on the blockchain is established and a data processing request is received from the data processing request terminal 400. The trusted execution environment is discarded when data processing in accordance with the contents of the smart contract is completed, or when the smart contract is terminated for other reasons.

[0092] The destruction of a trusted execution environment means that the environment constituting the trusted execution environment is initialized, and a series of data related to the trusted execution environment are erased. Therefore, not only the first data received by the trusted execution environment from the data generation terminal 200 for data processing, but also the data processing code received from the data processing request terminal 400 are erased together when the trusted execution environment is destroyed.

[0093] If the data processing platform server 100 were to store data even after data processing is complete, there is a risk that the data could be leaked due to a malicious external attack, and that it could be viewed by the operator of the data processing platform server 100.

[0094] However, the data processing platform server 100 according to the embodiment of the present invention stores data received from the data generation terminal 200 only during the period in which the smart contract is maintained, and when the smart contract is terminated, it destroys the trusted execution environment and simultaneously deletes the stored data. As a result, the data processing platform server 100 can prevent data from being leaked to the outside and fundamentally block the operator of the data processing platform server 100 from viewing the data, thereby protecting the privacy of the entity that provided the data.

[0095] According to the embodiment, the trusted execution environment can record a flag on the blockchain indicating that a discard command has been executed. When a discard command is executed, a flag that cannot be arbitrarily modulated is generated. When a discard command is executed, the trusted execution environment can generate a transaction including the aforementioned flag, record it on the blockchain, and then proceed to the discard process.

[0096] The trusted execution environment generated for data processing is a virtual execution environment, and at least one is generated on the data processing platform server 100 to execute smart contracts. It can be generated independently depending on whether a contract is formed or not, and can be destroyed and initialized. In this way, the data processing platform server 100 can execute a large number of smart contracts simultaneously using trusted execution environments that run independently for each smart contract.

[0097] The data generation terminal 200 is a communication device that can connect to a data processing platform, and can store data necessary for data processing and run user data management applications necessary for managing the stored data.

[0098] In particular, the data generation terminal 200 is a data generation terminal that includes personal information such as that from a home camera, CCTV, or vision camera, and can generate temporary data that includes video footage of a specific area or acoustic signals collected in a specific area. In this case, the data generation terminal 200 can detect personal information in the temporary data and perform de-identification processing on the personal information before saving, streaming, or transmitting the data.

[0099] In other words, the data generation terminal 200 can perform the process of saving, streaming, or transmitting first data generated based on the temporary data after de-identifying personal information within the temporary data generated in a specific area. This fundamentally blocks access to personal information at a hardware level, and the original data saved, streamed, or transmitted by the data generation terminal 200 is first data from which personal information has been de-identified.

[0100] Here, de-identification processing refers to processing techniques such as de-identification, masking, and face synthesis transformation of personal information such as faces, body parts, license plates, and resident registration numbers within video frames. The de-identification processing standards can be presented through the HIPAA (Health Insurance Portability and Accountability Act) Privacy Rules linked to the HIPAA, but are not limited to these and can be modified in various ways by the designer of the data generation terminal 200.

[0101] Furthermore, de-identification processing is a technique for modulating the acoustic components corresponding to a specific acoustic signal, and known acoustic modulation programs such as PowerDirector, VoiceMod, AV Voice Changer, and Audacity can be used. Here, de-identification processing of acoustic components means converting a specific acoustic component into a pre-configured composite tone, monotone, white noise, etc., and the de-identification processing method for acoustic components may change depending on the administrator's settings.

[0102] The first data may include at least one of video data and audio data from which personal information contained in the temporary data has been de-identified, and may also include metadata and first encryption key information related thereto.

[0103] The metadata may include the first terminal information of the data generation terminal that generated the first data, the generation time of the temporary data, and the location of a specific area. In other words, the metadata may include the first terminal information of the data generation terminal that generated the first data, the generation time of at least one of the video data and audio data before the anonymization process for personal information, and the location of the data generation terminal.

[0104] According to the embodiment, the metadata may include determination results such as whether or not an object is present in the video data and whether or not de-identification processing is performed.

[0105] For example, if the data generation terminal 200 recognizes objects within the video frame when generating video data and performs de-identification processing on personal information, the metadata may include information indicating that de-identification processing has been performed on the video data.

[0106] For example, if the data generation terminal 200 analyzes or classifies the types of objects within a video frame when generating video data, the metadata can include the types of objects contained in the video data and the classification results.

[0107] The first encryption key information may include the second terminal information of the encryption key supply terminal 300 that generated the first encryption key, and the first public key used for the de-identification process of personal information.

[0108] For example, if the first personal key is generated by encoding 256 bits of data into a 64-digit hexadecimal sequence, and the first public key is generated using an elliptic curve cryptography algorithm based on the first personal key, then the first encryption key information can contain only the information of the first public key.

[0109] The data generation terminal 200 identifies personal information each time it generates temporary data and performs deidentification using the first encryption key. The data generation terminal 200 can perform deidentification on all temporary data using the same first encryption key, but to improve security, it can perform deidentification using a new first encryption key at predetermined schedules.

[0110] Since the entity that decrypts the first data is the trusted execution environment, not the terminal that generated the first encryption key, the trusted execution environment must be able to recognize the first public key used in the deidentification process and obtain the corresponding first personal key from the encryption key supply terminal 300. Therefore, when the data generation terminal 200 generates temporary data, it can include information about the first public key used in the deidentification process of personal information as first encryption key information in the first data.

[0111] Furthermore, since the data generation terminal 200 is provided with only the first public key of the first encryption key from the encryption key supply terminal 300, it cannot arbitrarily decrypt the first data to identify personal information. This is to fundamentally prevent the leakage of personal confidential information to the outside.

[0112] When the data processing platform server 100 requests first data for data processing, the data generation terminal 200 can provide the first data to the data processing platform server 100 through the user data management application.

[0113] The data generation terminal 200 runs an application that scans the blockchain, allowing it to verify transactions recorded on the blockchain and confirm data requests from the trusted execution environment. For example, the data generation terminal 200 can use applications such as Etherescan and Remix to verify data requests from the trusted execution environment recorded on the blockchain.

[0114] The data generation terminal 200 can provide the first data to a trusted execution environment run on the data processing platform server 100. The data generation terminal 200 can also transfer encrypted data to the trusted execution environment through the data processing platform, but to ensure reliability, it can also generate a transaction containing the hash value of the first data and record it on the blockchain.

[0115] The data generation terminal 200 can acquire blockchain coins as compensation for providing data to the data processing platform server 100, and the amount of coins that can be acquired can be set by the trusted execution environment.

[0116] The owner of the data generation terminal 200 provides the data processing platform server 100 with personally identifiable information such as faces and body parts included in the first data, and therefore may have concerns about the leakage of personal information.

[0117] To solve this problem, the data generation terminal 200 can scan the blockchain through an application and check for the flag of a discard command recorded on the blockchain. If the data generation terminal 200 scans for the flag of a discard command for the trusted execution environment, it can be guaranteed that the trusted execution environment and data have been securely erased.

[0118] The encryption key supply terminal 300 is a terminal that generates a first encryption key according to a predetermined schedule and provides it to the data generation terminal 200. The encryption key supply terminal 300 may be the administrator terminal of the data generation terminal 200, or it may be a terminal that generates only encryption keys separately. The encryption key supply terminal 300 can generate different first encryption keys according to a predetermined schedule and can sequentially supply the generated first public keys to the data generation terminal 200.

[0119] As a result, the data generation terminal 200 can perform deidentification processing on personal information using one first public key per unit time. When the data generation terminal 200 saves the first data, it also saves the first public key information used for deidentification processing. Thereafter, the trusted execution environment can perform decryption using the first personal key corresponding to the first public key information contained in the first data.

[0120] The data processing request terminal 400 is a terminal that generates a data processing request that conforms to the smart contract, provides the transaction including the data processing request to the blockchain, and concludes a smart contract with the data processing platform server 100. For example, the data processing request terminal 400 may be at least one of the following: a PC (personal computer), a smartphone (smart phone), a tablet PC, a mobile internet device (MID), an internet tablet, an IoT (Internet of Things) device, an IoE (Internet of Everything) device, a desktop computer, a laptop computer, a workstation computer, a Wibro (Wireless Broadband Internet) terminal, and a PDA (Personal Digital Assistant).

[0121] Since the data processing request terminal 400 will obtain identified personal information or the results of data processing generated using personal information, it may be a terminal of an entity that has obtained permission to view personal information from the data generation terminal 200, or a terminal of an entity that has been delegated permission to view personal information by a government agency.

[0122] The data processing request terminal 400 can provide the data processing code necessary for data processing along with the data processing request to the transaction when concluding a smart contract, and can also provide the account of a specific data generation terminal 200 to the transaction in order to specify the data collection entity.

[0123] The data processing request terminal 400 can also provide its public key to the transaction along with the data processing request when concluding a smart contract, in order to provide encrypted data processing results.

[0124] The data processing request terminal 400 can obtain the data processing results from the blockchain once the smart contract is fulfilled. If the data processing results are encrypted, the data processing request terminal 400 can decrypt them using its personal key to obtain the data processing results that match the data processing request.

[0125] Thus, the data processing request terminal 400 can obtain the desired results without directly collecting or processing the data necessary for data processing. Furthermore, since personal information or information requiring security other than the data necessary for data processing is not leaked to the entity providing the data, the privacy of the entity providing the data can be securely protected.

[0126] Figure 2 is a block diagram illustrating the first data generation process of multiple data generation terminals according to an embodiment of the present invention. In Figure 2, for convenience, only the first to third data generation terminals are shown to illustrate the process by which the encryption key supply terminal provides the first encryption key to multiple data generation terminals. However, the number of data generation terminals to which the first encryption key is provided by the encryption key supply terminal is not limited to these.

[0127] Referring to Figure 2, the encryption key supply terminal 300 can generate the first to firstc encryption keys KEY1a to KEY1c according to a predetermined schedule and provide them to the first to third data generation terminals 200A to 200C.

[0128] Each of the first to third data generation terminals 200A to 200C may receive the first a to first c encryption keys KEY1a to KEY1c from the encryption key supply terminal 300 at the same time, but is not limited to this, and may also receive them at a time individually set for each terminal.

[0129] Furthermore, the first encryption keys KEY1a to KEY1c may be first public keys having the same value, but they may also be designed to have different values ​​from each other by the encryption key supply terminal 300.

[0130] As a result, the first to third data generation terminals 200A to 200C can perform de-identification processing on personal information using the first to firstc encryption keys KEY1a to KEY1c respectively, generate the first to firstc data DAT1a to DAT1c, and provide them to the data processing platform server 100.

[0131] Here, the first data includes first public key information used for personal information anonymization processing as first encryption key information, but the firsta data DAT1a may contain information about the firsta encryption key KEY1a, the firstb data DAT1b may contain information about the firstb encryption key KEY1b, and the firstc data DAT1c may contain information about the firstc encryption key KEY1c.

[0132] The first to third data generation terminals 200A to 200C can transmit encrypted data to a trusted execution environment through the data processing platform, but at least some of them can also generate a transaction containing the hash value of the first data and record it on the blockchain to ensure reliability.

[0133] Figure 3 is a diagram illustrating a method for generating a data processing process for a trusted execution environment according to an embodiment of the present invention. Referring to Figure 3, the trusted execution environment can decrypt the first data based on the first personal key and identify the area corresponding to the personal information in the temporary data. Since the first public key information is also stored in the first data, the trusted execution environment can perform decryption using the first personal key corresponding to the first public key information contained in the first data.

[0134] If data DAT1a to DAT1c are received, the trusted execution environment can use the first encryption key information contained in each of the data DAT1a to DAT1c to determine the information of the first public key used for the de-identification process of personal information.

[0135] The trusted execution environment can obtain the first personal key corresponding to the first public key from the encryption key supply terminal, and can decrypt the firsta data DAT1a using the firsta personal key P_KEY1a, the firstb data DAT1b using the firstb personal key P_KEY1b, and the firstc data DAT1c using the firstc personal key P_KEY1c.

[0136] The trusted execution environment can then process the decrypted 1a-1c data DAT1a'-DAT1c' according to the data processing code to generate data processing results.

[0137] Figure 4 is a diagram illustrating a method for generating data processing results in a trusted execution environment according to an embodiment of the present invention. Referring to Figure 4, the trusted execution environment can process the first data DAT1', which has been decoded using various artificial intelligence models included in the data processing code.

[0138] The data processing code is an artificial intelligence processing model generated by the data processing platform server 400 after determining the processing model necessary for the data processing request, and may include an object recognition model, a data processing target identification model, and a data processing result generation model.

[0139] The object recognition model can recognize objects within the video data contained in the decoded first data DAT1'. In particular, the object recognition model may be any of the following, but is not limited to them: a face detection algorithm using feature points (landmarks), a deep learning-based YOLO (You Look Only Once) model, MobileNet, R-CNN (Recursive Convolutional Neural Network), Faster-RCNN, or SSD (Single Shot Multibox Detector). Known algorithms for object recognition can also be used.

[0140] The object recognition model provides information about the recognized object OB for each video data set as input to the data processing target identification model, which can then determine whether the recognized object OB in the video data is an object necessary for data processing.

[0141] According to the embodiment, when a trusted execution environment decodes acoustic data, an acoustic signal analysis model can be used instead of an object recognition model. The acoustic signal model can classify the acoustic data contained in the decoded first data DAT1'. The acoustic signal analysis model can use, but is not limited to, MFCC (Mel-Frequency Cepstral Cofficient), and other known algorithmic models for acoustic signal analysis can also be used.

[0142] The acoustic signal analysis model provides information about the classified acoustic data as input to the data processing target identification model, which can then determine whether the acoustic components within the acoustic data are necessary for data processing.

[0143] The data processing target identification model can identify data processing targets based on information about the recognized object OB. To this end, the data processing target identification model can identify data processing targets TA by determining the degree of agreement between the object characteristic information included in the data processing request and the recognized object.

[0144] Here, the object characteristic information included in the data processing request is information that allows for the recognition of a specific object, and may include, for example, feature point information and image information of the object. The trusted execution environment can determine the degree of agreement between the object characteristic information and the recognized object through the data processing target identification model. If the trusted execution environment determines that the degree of agreement between the object characteristic information and the recognized object is equal to or greater than the standard agreement rate, it can identify the recognized object as a data processing target (TA).

[0145] According to the embodiment, when the trusted execution environment decodes the acoustic data, the data processing target model can identify the data to be processed based on the classified acoustic data. To this end, the data processing target identification model can identify the data to be processed TA by determining the degree of agreement between the object characteristic information included in the data processing request and the classified acoustic data.

[0146] The data processing result generation model can generate data processing result RES based on target video data including the target TA and metadata. Alternatively, it can extract only metadata from the target video data and process it according to the data processing code, or it can generate video frames and metadata as the data processing result RES.

[0147] According to the embodiment, the data processing result generation model can generate a data processing result RES based on target acoustic data including the TA to be processed and metadata. In this way, the trusted execution environment can comprehensively analyze the initial data generated by different data generation terminals and generate a data processing result RES that corresponds to the data processing request.

[0148] Figures 5a and 5b illustrate a method for anonymizing personal information using a trusted execution environment of a smart contract platform according to an embodiment of the present invention. Referring to Figure 5a, the encryption key supply terminal 300 can generate a first encryption key and provide it to the data generation terminal 200.

[0149] The first encryption key KEY1 generated by the encryption key supply terminal 300 is an asymmetric key consisting of a pair of a first personal key and a first public key, and the data generation terminal 200 is provided only with the first public key. This is to prevent the first data DAT1, which has been deidentified by the data generation terminal 200, from being arbitrarily decrypted, and to prevent the first personal key from being leaked to the outside due to malicious attacks such as hacking.

[0150] The encryption key supply terminal 300 can generate different first encryption keys KEY1 according to a predetermined schedule and sequentially supply the first public key to the data generation terminal 200. If there are multiple data generation terminals 200 to which the first encryption key KEY1 is provided by the encryption key supply terminal 300, the encryption key supply terminal 300 can either provide the same first encryption key KEY1 to the data generation terminals 200, or provide different first encryption keys KEY1 to each data generation terminal 200, according to the user's design.

[0151] The encryption key supply terminal 300 can provide the first encryption key KEY1 to the data generation terminal 200 via blockchain BC.

[0152] However, the first encryption key KEY1 provided to the data generation terminal 200 contains only the first public key, and even if the first public key is leaked to any third party, no damage to personal information will occur. Therefore, the encryption key supply terminal 300 can either provide the first public key directly to the data generation terminal 200 via network communication, or it can provide it through the data processing platform.

[0153] The data generation terminal 200 can capture a specific area, de-identify personal information contained within the video frame IM, and then perform data storage, streaming, or transfer operations.

[0154] The data generation terminal 200 can identify objects containing personal information within the video frame IM and perform de-identification processing on these objects using a first public key. De-identification processing is a processing technique that includes de-identification, masking, and face synthesis transformation of personal information such as faces, body parts, license plates, and resident registration numbers within the video frame. The de-identification processing standards can be presented through the HIPAA Privacy Rule, which is linked to the HIPAA (Health Insurance Portability and Accountability Act), but are not limited to this and can be modified in various ways by the designer of the data generation terminal 200.

[0155] The data generation terminal 200 can perform deidentification processing on personal information using one first public key per unit time, but is not limited to this; it can also perform deidentification processing on all personal information based on a single first public key. In the first data DAT1 generated by the data generation terminal 200, the data area corresponding to personal information has been deidentified based on the first public key, and no separate deidentification processing is performed on data that is not related to personal information.

[0156] According to the embodiment, after recognizing an object in the video frame, the data generation terminal 200 can select only the personal information relating to a pre-configured target from the recognized object and perform de-identification processing. Therefore, in the first data DAT1, only the data area corresponding to the personal information of the pre-configured target is de-identified, and the rest is not de-identified.

[0157] According to the embodiment, if the data size of the video frame is greater than or equal to a standard size, the data generation terminal 200 can temporarily de-identify the data area corresponding to personal information through a third encryption key using a symmetric key algorithm such as the Advanced Encryption Standard (AES), and encrypt the third encryption key based on the first encryption key KEY1. In this case, the third encryption key can be stored in the first data DAT1 together with the first public key information.

[0158] Referring to Figure 5b, the data processing platform server 100 can generate smart contracts SC for data processing and distribute them to the blockchain BC. The smart contracts SC are recorded in the blocks of the blockchain BC and can be executed on the blockchain BC.

[0159] The smart contract SC automatically concludes when a transaction containing a document processing request QUE1 that conforms to the prescribed terms is provided to the blockchain BC from the data processing request terminal 400, and terminates when a document processing result that conforms to document processing request QUE1 is returned.

[0160] The data processing request terminal 400 can provide a data processing request QUE1 that matches the smart contract SC to the blockchain BC in order to obtain data processing results. Once the smart contract SC is concluded between the data processing platform server 100 and the data processing request terminal 400, the data processing platform server 100 generates a trusted execution environment VM in response to the data processing request QUE1, which includes a data processing code and a second encryption key. Here, the data processing code can mean the data processing model that the data processing request terminal 400 has provided to the blockchain.

[0161] The trusted execution environment VM can generate a transaction including a data request QUE2 to the data generation terminal 200 and record it on the blockchain BC in order to perform data processing corresponding to the data processing code.

[0162] Referring to Figure 5c, the data generation terminal 200 can obtain the data request QUE2 of the trusted execution environment VM from the blockchain BC, record the stored first data DAT1 on the blockchain, or generate a transaction containing the hash value of the first data DAT1 to provide the encrypted data with a hash value to the trusted execution environment VM, and record it on the blockchain BC.

[0163] The data processing request terminal 400 can obtain the recorded second encryption key KEY2 from the blockchain BC along with the data request QUE2 of the trusted execution environment VM. The second encryption key KEY2 obtained by the data processing request terminal 400 represents the second public key of the trusted execution environment VM.

[0164] On the other hand, since the first personal key of the first encryption key KEY1 is required to decrypt the first data DAT1, the data processing request terminal 400 must encrypt the first personal key and securely provide it to the trusted execution environment VM.

[0165] Therefore, the data processing request terminal 400 can encrypt the first encryption key KEY1 based on the second encryption key KEY2 and provide the encrypted first encryption key KEY1 to the trusted execution environment. That is, it can encrypt the first personal key based on the second public key of the trusted execution environment VM, generate a transaction for the second personal key encrypted with a hash value, and record it on the blockchain BC.

[0166] The trusted execution environment VM can perform data processing if the data collected from the data generation terminal 200 meets the criteria required for data processing.

[0167] The first data DAT1 obtained from blockchain BC is deidentified using the first public key of the first encryption key KEY1, so the trusted execution environment VM must prioritize obtaining the first personal key. The trusted execution environment VM can then use the second personal key to decrypt the first personal key, which has been encrypted with the second public key, and obtain the first personal key.

[0168] The trusted execution environment VM can decrypt the first data based on the first personal key and identify the data area corresponding to the personal information. On the other hand, since the first public key information is also stored in the first data, the trusted execution environment VM can perform decryption using the first personal key corresponding to the first public key information contained in the first data.

[0169] The trusted execution environment VM can then perform data processing according to the data processing code. Once data processing is complete, the trusted execution environment VM can provide the data processing result RES to the data processing request terminal 400.

[0170] The trusted execution environment VM can record the data processing result RES on the blockchain without separately encrypting it. However, to prevent the data processing result RES from being leaked externally, it can also encrypt the data processing result using the third public key of the data processing request terminal 400 and return it to the blockchain BC.

[0171] The smart contract SC terminates when a document processing result RES matching the document processing request QUE1 is returned. The trusted execution environment VM can verify that the document processing result RES is recorded in the blockchain BC and confirm that the smart contract SC has been successfully executed.

[0172] When the data processing result RES is returned and the smart contract SC is terminated, the trusted execution environment VM running on the data processing platform server 100 enters discard mode. As the trusted execution environment VM is discarded, the data received from the data generation terminal 200 for data processing is also deleted from the data processing platform server 100.

[0173] The data processing request terminal 400 can obtain encrypted data processing results RES recorded on the blockchain BC, and by decrypting the encrypted data processing results RES with a third-party personal key, it can obtain data processing results that match the data processing request.

[0174] Figure 6 is a diagram illustrating a method for de-identifying personal information in a data generation terminal according to another embodiment of the present invention. Referring to Figure 6, the data generation terminal 200 is a terminal that generates data containing personal information such as home cameras, CCTV cameras, and other cameras, and can capture a specific area and generate first data including video frames.

[0175] The data generation terminal 200 can perform data storage, streaming, or transfer operations after capturing a specific area and pre-de-identifying personal information contained within the video frame, in order to fundamentally block access to personal information through hardware.

[0176] De-identification processing can be performed on user-defined areas within a video frame, such as information about people, text, or objects. According to the embodiment, the deidentification process can be performed only on targets that have already been registered, according to the user's settings, and not on targets that have not been registered.

[0177] For example, if the data generation terminal 200 is a home camera installed in a home, the user can set the de-identification process to apply only to residents of the home. In this case, outsiders who do not reside in the home may be present in a state where they are identifiable in the video frame.

[0178] Furthermore, the deidentification process is performed only on a portion of the object, and the deidentification method can be varied according to user settings, including masking and face synthesis transformation.

[0179] Specifically, the data generation terminal 200 may be a home CCTV that films the inside of a home in real time, and the first video frame IM1 may be generated by the data generation terminal 200. The data generation terminal 200 can recognize objects contained in the first video frame IM1 through an artificial intelligence object recognition algorithm and can distinguish between the first user HM1 and the object OB.

[0180] If the first user HM1 is set as a target for deidentification processing, the data generation terminal 200 can deidentify the data area corresponding to the first user HM1 based on the first encryption key KEY1. At this time, the data generation terminal 200 can selectively execute a deidentification processing method based on the administrator's settings, such as applying a mosaic effect to a part of the first user HM1's body, performing face synthesis or transformation, or inserting a character into the body.

[0181] In the de-identified first video frame IM1', all or part of the body of the first user HM1' is de-identified BR, while other objects OB remain identified. The data generation terminal 200 can only save, stream, or transfer the first data containing the de-identified first video frame IM1' after it has been generated.

[0182] As a result, even if a third party obtains the primary data, the personal information has been anonymized and de-identified, making it impossible to further identify the individual.

[0183] Figure 7 is a diagram illustrating a method for identifying personal information in a trusted execution environment according to an embodiment of the present invention. Referring to Figure 7, the trusted execution environment VM can identify and process a portion of the data area corresponding to personal information from the first data collected from the data generation terminal 200 for data processing by the data processing code.

[0184] Since the first data has been deidentified using the first public key of the first encryption key, the trusted execution environment VM can perform identification processing using the first personal key of the first encryption key obtained from the encryption key supply terminal 300.

[0185] On the other hand, since the first data also stores the first public key information used for de-identification processing, the trusted execution environment VM can perform decryption using the first personal key corresponding to the first public key information contained in the first data.

[0186] For example, the first data may include the second video frame IM2, and the second video frame IM2 may contain an anonymized second user HM2. The trusted execution environment VM can perform identification operations on the anonymized second user HM2 in order to perform data processing. At this time, the trusted execution environment VM may selectively identify all or part of the anonymized second user HM2, which may vary depending on the data processing code.

[0187] The trusted execution environment VM can perform data processing using the decrypted first data, i.e., the first data including the identified second video frame IM2'. When the data processing results are returned in accordance with the smart contract, the trusted execution environment is discarded, and a set of personal information associated with the trusted execution environment is simultaneously deleted, thereby protecting individual privacy.

[0188] In conclusion, according to the personal information identification processing method for a trusted execution environment VM as embodied in the present invention, even after de-identifying personal information, it can be restored and used in a way that allows for identification of personal information under restricted circumstances, and after the data processing results are obtained, they can be securely deleted to protect the privacy of individuals.

[0189] Figure 8 is a diagram illustrating a smart contract according to an embodiment of the present invention. Referring to Figure 8, the data processing platform server 100 can generate smart contracts for data processing and distribute them to the blockchain. Once a smart contract is concluded with the data processing request terminal 400, it can generate a trusted execution environment for data processing.

[0190] The data processing platform server 100 can generate a large number of smart contracts and distribute them to the blockchain. However, if a single smart contract requires multiple different data processing operations, multiple different trust execution environments may be generated.

[0191] The trusted execution environment generated for data processing is a virtual execution environment, and at least one can be generated on the data processing platform server 100 for the execution of smart contracts. In other words, even if 10 smart contracts are distributed on the blockchain, if only 2 smart contracts have been concluded, the data processing platform server 100 can execute two trusted execution environments to execute the contracts.

[0192] Furthermore, since each smart contract can be configured to have different settings for whether or not performance has been successfully completed, the performance period, and the performance method, each smart contract undergoes an independent process of generation and termination depending on whether or not the contract is formed and performed.

[0193] For example, the data processing platform server 100 can conclude a first smart contract SC1 with a first data processing request terminal 400A and a second smart contract SC2 with a second data processing request terminal 400B. The data processing platform server 100 can generate a first trust execution environment VM1 corresponding to the conclusion of the first smart contract SC1 and a second trust execution environment VM2 corresponding to the conclusion of the second smart contract SC2. Even if there are other distributed smart contracts on the blockchain BC besides the first and second smart contracts SC1 and SC2, the data processing platform server 100 will only execute the trust execution environment for the smart contracts that have been concluded.

[0194] The first trusted execution environment VM1 and the second trusted execution environment VM2 are virtual machines created based on different contract terms and run independently of each other; therefore, one trusted execution environment does not affect the other. Consequently, even if the first trusted execution environment VM1 is destroyed in the course of contract fulfillment, the second trusted execution environment VM2 can continue to execute the data processing process in accordance with the terms of the second smart contract SC2 without being affected.

[0195] Thus, the data processing platform server 100 can simultaneously execute a large number of smart contracts using a trusted execution environment that runs independently for each smart contract, and by using an independent data processing space that is impossible to intrude upon, it is possible to prevent the risk of data leakage or mixing.

[0196] Figure 9 is a flowchart illustrating a method for de-identifying personal identification information using a trusted execution environment of a smart contract platform according to an embodiment of the present invention. Referring to Figure 9, the trusted execution environment running on the data processing platform server 100 can be generated in response to a data processing request received from a data processing request terminal 300 in accordance with a smart contract distributed on the blockchain BC, and can include data processing code (S100).

[0197] Furthermore, the trusted execution environment can obtain the first data from the data generation terminal 200 and decrypt the first data using the first personal key among the first encryption keys (S110).

[0198] Furthermore, the trusted execution environment can process the first data decoded based on the data processing code and generate data processing results (S120).

[0199] Furthermore, the trusted execution environment can provide the data processing results to the data processing request terminal 400 in accordance with the smart contract (S130).

[0200] The trusted execution environment can then be destroyed by executing a destruction order once the performance of the smart contract is completed (S140). When the trusted execution environment is destroyed, the environment constituting the trusted execution environment is initialized, and a series of data, including the first data and data processing code received from the data generation terminal 200, are erased.

[0201] Figure 10 is a flowchart illustrating a method for processing personal information using a trusted execution environment of a smart contract platform according to another embodiment of the present invention.

[0202] Referring to Figure 10, the trusted execution environment can perform object recognition on the video data in the first data that has been decoded using the object recognition model (S200).

[0203] The trusted execution environment determines the degree of agreement between the object characteristic information included in the data processing request and the object recognized by the object recognition model, and can identify objects that are determined to have an agreement rate equal to or greater than the standard agreement rate as targets for data processing (S210).

[0204] The trusted execution environment can select the video data containing the data to be processed from the video data included in the first data as the target video data (S220).

[0205] The trusted execution environment can extract metadata corresponding to the selected target video data (S230), and process the extracted metadata according to the data processing code to generate data processing results (S240).

[0206] The user data management method or algorithm steps using data processing of the trusted execution environment of the smart contract platform described in relation to embodiments of the present invention may be embodied directly by hardware, as a software module executed by hardware, or by a combination thereof. The software module can be stored in RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), flash memory, hard disk, removable disk, CD-ROM, or any type of computer-readable recording medium known in the art to which the present invention belongs.

[0207] Although embodiments of the present invention have been described above, it will be understood that those with ordinary skill in the art to which the present invention pertains can implement the invention in various ways without departing from the scope of the claims. [Industrial applicability]

[0208] According to the present invention, even after personal information has been anonymized, it can be restored and used in a way that allows for identification of personal information under limited circumstances, and after the data processing results are obtained, the data can be securely deleted to protect the privacy of individuals.

[0209] Furthermore, according to the present invention, by de-identifying identified personal information before a series of data processing processes such as data storage, streaming, and transfer, it is possible to fundamentally block access to personal information at the hardware level.

[0210] Furthermore, according to the present invention, when a smart contract is terminated, the trusted execution environment is destroyed and the stored data is also deleted, thereby preventing the collected data from being leaked to the outside and fundamentally blocking the data subject's privacy by blocking the operator of the data processing platform server from viewing the data.

[0211] Furthermore, according to the present invention, personal information generated by data generation terminals such as cameras, CCTVs, and home cameras can be fundamentally encrypted and used restrictively only under specific analytical conditions without being disclosed to the data subject, thereby protecting the privacy of individuals.

[0212] Furthermore, according to the present invention, a large number of smart contracts can be executed simultaneously using a trusted execution environment that runs independently for each smart contract.

[0213] Furthermore, according to the present invention, by recording a destruction order flag on the blockchain, it is possible to ensure that the trusted execution environment in which the data was provided is stably removed and that the data has been securely erased without being leaked to the outside.

[0214] Furthermore, according to the present invention, the data processing request terminal can obtain the desired data processing result without directly collecting or processing the data necessary for data processing, and since the data necessary for data processing is not leaked to the data processing request terminal, the privacy of the entity that provided the data can be securely protected. [Explanation of Symbols]

[0215] 10. Personal Identification Information De-identification Processing System 100 Data Processing Platform Servers 200 data generation terminals 200A First Data 200B Second Data 200C Third Data 300 encryption key supply terminals 400 Data Processing Request Terminals 400A First Data Processing Request Terminal 400B Second Data Processing Request Terminal

Claims

1. The data processing platform server responds to data processing requests received from data processing request terminals in accordance with smart contracts distributed on the blockchain, and generates a trusted execution environment including data processing code. The steps include: the first step in which the trusted execution environment decrypts the first data obtained from the data generation terminal; The steps include: the trusted execution environment processing the decoded first data based on the data processing code to generate a data processing result; A method for processing personal information using a trusted execution environment of a smart contract platform, characterized by comprising the step of providing the data processing results to the data processing request terminal in accordance with the smart contract.

2. The personal information processing method using the trusted execution environment of the smart contract platform according to claim 1, characterized in that the first data includes at least one of video data and audio data from which personal information has been de-identified, metadata, and first encryption key information.

3. The method for processing personal information using the trusted execution environment of a smart contract platform according to claim 2, characterized in that the metadata includes first terminal information of the data generation terminal that generated the first data, the generation time of at least one of the video data and audio data before the deidentification processing of the personal information, and the location of the data generation terminal.

4. The method for processing personal information using the trusted execution environment of a smart contract platform according to claim 2, characterized in that the first encryption key information includes second terminal information of the encryption key supply terminal that generated the first encryption key, and a first public key used for de-identifying the personal information.

5. The step of decoding the first data is: A method for processing personal information using the trusted execution environment of a smart contract platform according to claim 4, characterized in that the data area corresponding to the personal information is identified using a first personal key obtained from the encryption key supply terminal.

6. The personal information processing method using the trusted execution environment of the smart contract platform according to claim 2, characterized in that the first encryption key is an asymmetric key consisting of a pair of a first personal key and a first public key generated by an encryption key supply terminal, the first public key is generated based on the first personal key, and the first encryption key is updated by the encryption key supply terminal according to a predetermined schedule.

7. The step of generating the aforementioned data processing results is: The steps include: analyzing the decoded first data to determine the data to be processed in response to the data processing request; The steps include selecting target video data from the decoded first data that includes the data to be processed, The stage of extracting metadata corresponding to the selected target video data, A method for processing personal information using a trusted execution environment of a smart contract platform according to claim 1, characterized by comprising the step of processing the extracted metadata in accordance with the data processing code to generate the data processing result.

8. The step of determining the documents to be processed is: The steps include: recognizing an object in the video data within the decoded first data using the object recognition model included in the data processing code; A method for processing personal information using the trusted execution environment of a smart contract platform according to claim 7, characterized by comprising the steps of: determining the degree of agreement between the object characteristic information included in the data processing request and the recognized object, and identifying objects having a value equal to or greater than a standard agreement rate as targets for data processing.

9. The step of generating the aforementioned data processing results is: The steps include: analyzing the decoded first data to determine the data to be processed in response to the data processing request; The steps include selecting target acoustic data from the decoded first data that includes the data to be processed, The steps include extracting metadata corresponding to the selected target acoustic data, A method for processing personal information using a trusted execution environment of a smart contract platform according to claim 1, characterized by comprising the step of processing the extracted metadata in accordance with the data processing code to generate the data processing result.

10. The step of determining the documents to be processed is: The steps include classifying the acoustic data in the decoded first data using the acoustic signal analysis model included in the data processing code, A method for processing personal information using the trusted execution environment of a smart contract platform according to claim 9, characterized in that it includes the step of determining the degree of agreement between the object characteristic information included in the data processing request and the classified acoustic data, and identifying acoustic data having a value equal to or greater than a standard agreement rate as the data to be processed.

11. After the stage of providing the data to the aforementioned data processing request terminal, A method for processing personal information using a trusted execution environment of a smart contract platform according to claim 1, further comprising the step of the trusted execution environment being destroyed in accordance with the smart contract.

12. A computer-readable recording medium for storing one or more programs, wherein the one or more programs are configured to be executed by one or more processors of an electronic device, and the one or more programs include instructions for executing the method according to any one of claims 1 to 11.

13. Multiple data generation terminals that generate temporary data by taking images or collecting sounds in different areas, recognize personal information within the temporary data, and generate first data by de-identifying the personal information based on a first encryption key, A personal information processing system using a trusted execution environment of a smart contract platform, characterized by including a data processing platform server that generates a trusted execution environment that responds to data processing requests received from a data processing request terminal in accordance with smart contracts distributed on the blockchain, and provides the data processing results generated based on the first data to the data processing request terminal.

14. The personal information processing system using the trusted execution environment of the smart contract platform according to claim 13, characterized in that the first data includes at least one of video data and audio data from which the personal information has been de-identified, metadata, and first encryption key information.

15. The aforementioned trusted execution environment is A personal information processing system using the trusted execution environment of a smart contract platform according to claim 14, characterized in that it decrypts the first data obtained from each data generation terminal via the blockchain, processes the decrypted first data according to a data processing code, and generates the data processing result.

16. The aforementioned trusted execution environment is A personal information processing system using the trusted execution environment of a smart contract platform according to claim 14, characterized in that it determines the first public key used for de-identification processing based on the first encryption key information, and decrypts the first data by identifying the personal information with the first personal key corresponding to the first public key.

17. The aforementioned trusted execution environment is A personal information processing system using the trusted execution environment of a smart contract platform according to claim 14, characterized in that it analyzes the decrypted first data to determine the data processing target corresponding to the data processing request, selects target video data containing the data processing target from the decrypted first data, and processes metadata corresponding to the selected target video data according to a data processing code to generate the data processing result.

18. The aforementioned trusted execution environment is A personal information processing system using the trusted execution environment of a smart contract platform according to 17, characterized in that it recognizes an object in the video data within the decoded first data using an object recognition model included in the data processing code, determines the degree of agreement between the object characteristic information included in the data processing request and the recognized object, and identifies objects having a value equal to or greater than a standard agreement rate as targets for data processing.

19. The aforementioned trusted execution environment is A personal information processing system using the trusted execution environment of a smart contract platform according to claim 14, characterized in that it analyzes the decrypted first data to determine the data processing target corresponding to the data processing request, selects target acoustic data including the data processing target from the decrypted first data, and processes the metadata corresponding to the selected target acoustic data according to the data processing code to generate the data processing result.

20. The aforementioned trusted execution environment is A personal information processing system using the trusted execution environment of a smart contract platform according to 19, characterized in that it classifies the acoustic data in the decoded first data using an acoustic signal analysis model included in the data processing code, determines the agreement rate between the object characteristic information included in the data processing request and the classified acoustic data, and identifies acoustic data having a value equal to or greater than a standard agreement rate as the target of data processing.

21. A personal information processing system using a trusted execution environment of a smart contract platform according to claim 13, further comprising an encryption key supply terminal that generates the first encryption key according to a predetermined schedule and provides it to each of the data generation terminals.

22. A personal information processing system using the trusted execution environment of the smart contract platform according to 21, characterized in that each data generation terminal receives a second encryption key having a different value from each other.

Images