Private data cross-domain coprocessing method based on trusted data space

Abstract

The invention relates to the technical field of information security, and discloses a private data cross-domain cooperative processing method based on a trusted data space, a data provider firstly encrypts own original data by using symmetric encryption, and a key is stored in an own key management system to ensure that plaintext data does not go out of a domain; the computing power provider can only receive the ciphertext data of the user and cannot obtain any privacy from the ciphertext data. When the privacy computing system in the trusted data space carries out privacy computing, the trusted service program intercepts operations such as file reading of the privacy computing system, generates a corresponding authentication report and sends the authentication report to a data provider to request a corresponding secret key, the data provider can carry out remote authentication on each secret key request, and after the authentication is passed, the data provider can send the authentication report to the private computing system. A one-time key is generated and sent to the trusted service program, the trusted service program carries out decryption operation on the corresponding ciphertext data in the encryption memory of the trusted execution environment, and the key is destroyed immediately after decryption is completed.

Description

Technical Field

[0001] This invention relates to the field of information security technology, and more specifically to a method for cross-domain collaborative processing of privacy data based on a trusted data space. Background Technology

[0002] Against the backdrop of rapid growth in the digital economy, data, as a new factor of production, is the foundation of digitalization, networking, and intelligence. It has rapidly integrated into all aspects of production, distribution, circulation, consumption, and social service management, profoundly changing production methods, lifestyles, and social governance. However, in reality, data often contains various private information of enterprises and users, making the collection and circulation of data elements extremely difficult.

[0003] Most existing solutions for processing user privacy data typically require storing the user's privacy data in plaintext at a designated storage location within the privacy computing system. This means that the user's privacy data is stored in plaintext until it leaves the domain for computation. Since existing privacy computing systems are deployed locally, users may not need to worry about the security risks associated with plaintext storage. However, this approach requires all users providing privacy data to deploy their own privacy computing systems locally. Local deployment of privacy computing systems requires significant computing power, resulting in substantial deployment costs for users and thus presenting some limitations.

[0004] Therefore, in response to the problems of insufficient computing power and data out-of-domain security in the production and use of privacy computing systems, this invention provides a method for cross-domain collaborative processing of privacy data based on a trusted data space. Summary of the Invention

[0005] In view of the above problems, the present invention is proposed to provide a method for cross-domain collaborative processing of privacy data based on a trusted data space that overcomes or at least partially solves the above problems.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, embodiments of the present invention provide a method for cross-domain collaborative processing of privacy data based on a trusted data space, comprising: The data provider generates a public-private key pair and sends the public key to the computing power provider via a secure channel; The computing power provider allocates trusted storage space and privacy computing roles to the data provider in the trusted execution environment, and establishes a mapping between public keys and trusted storage space, and saves the public keys to trusted storage space; The data provider encrypts the original data using a locally randomly generated key to obtain ciphertext, and sends the ciphertext, key access address and signature to the computing power provider. The computing power provider verifies the signature using a public key and stores the ciphertext and key access address in a trusted storage space, and establishes a mapping between the ciphertext and the key access address. Depending on whether they have computing power, data processors can choose to initiate data processing requests locally or in a trusted data space. When data from other data providers needs to be used, after authorization, the trusted service program intercepts the reading operation, finds the corresponding key call address through the mapping between the stored ciphertext and the key call address, generates an authentication report with a TEE signature, stores it on the blockchain, and requests a one-time decryption key from the data provider. After remotely verifying the TEE signature authentication report, the data provider distributes a one-time decryption key. The trusted service program decrypts the ciphertext within a trusted execution environment, and the calculation result is transparently encrypted by the trusted service program and stored in the trusted storage space of the data processor.

[0007] Preferably, the key access address includes the service address requesting the use of the key and the index of the key.

[0008] Preferably, the TEE signature authentication report includes the data processor's identity, TEE metric value, trusted service program hash value, timestamp, and request ID, and constructs a blockchain transaction to submit to the consortium blockchain before sending the key request.

[0009] Preferably, after remotely verifying the TEE signature authentication report, the data provider distributes a one-time decryption key, and the trusted service program decrypts the ciphertext within a trusted execution environment, including: After receiving the one-time decryption key acquisition request from the trusted service program, the data provider remotely authenticates the TEE signature authentication report sent by the trusted service program: The digital signature of the authentication report signed by the TEE is verified using the public key of the Trusted Execution Environment (TEE) hardware, and the content of the certification report is also verified. After successful verification, the data provider finds the corresponding key based on the key index in the key call address, constructs a one-time key for use by the trusted service program, and records it on the blockchain.

[0010] Preferably, the one-time key includes basic key data, a timestamp, an expiration date, and an integrity check code, and is used only once in a specified TEE instance and destroyed immediately after use.

[0011] Preferably, the trusted service program decrypts the ciphertext within a trusted execution environment, including: After receiving the returned one-time key, the trusted service program queries the blockchain records to verify the validity of the one-time key; After successful verification, the ciphertext is decrypted using a one-time key within the trusted storage space of the trusted execution environment.

[0012] Preferably, the calculation results are transparently encrypted by a trusted service program and then stored in the trusted storage space of the data processor, including: The privacy computing system reads the decrypted data through the transparent decryption system of the trusted service program, performs the corresponding calculation operations, and writes the calculation results into the trusted storage space of the data processor. At the same time as writing, the trusted service program is triggered to intercept the data and transparently encrypts the written data.

[0013] Preferably, the trusted service program transparently encrypts the written data, including: The trusted service program finds the public key corresponding to the user of the trusted storage space based on the mapping between the public key and the trusted storage space; The trusted service program generates a random result key for the computation result; The trusted service program performs encryption operations on the calculation results and the result key, and stores the encrypted results in the trusted storage space of the data processor. The trusted service program uses the corresponding public key to encrypt the result key and sends the encrypted result key to the data processor through a secure channel.

[0014] Preferably, the computing power provider deploys trusted service programs in a trusted execution environment, which includes Intel SGX, ARM TrustZone, or domestically produced trusted computing modules.

[0015] Preferably, the entire process involves recording key requests, distribution, and usage logs via blockchain to achieve secure control and traceability auditing throughout the data lifecycle.

[0016] As can be seen from the above technical solution, compared with the prior art, the present invention discloses a method for cross-domain collaborative processing of privacy data based on a trusted data space, which has the following beneficial effects: 1. Even if a data provider lacks sufficient computing power, it can leverage the trusted data space of a computing power provider to conduct secure privacy-preserving computations with other entities, while ensuring the secure isolation of user data. Simultaneously, it supports multiple participants with insufficient computing power to perform privacy-preserving computations.

[0017] 2. All outbound data from the data provider is encrypted, and the keys are stored in the local key management service. All key requests must be remotely authenticated by the local key management service before being approved. All keys sent are one-time keys with a limited validity period and are destroyed after use. Usage records are stored using blockchain technology, which effectively ensures the data provider's cross-domain collaborative management of its own privacy data and realizes encrypted computation and encrypted transfer of data.

[0018] 3. The decryption of encrypted data is performed by the trusted service program in encrypted memory within a trusted execution environment. The data provider can remotely authenticate the encryption requests of the trusted service to ensure that the trusted service program has not been tampered with by adversaries and is running in the expected environment, thereby ensuring the privacy and security of its own data.

[0019] 4. The trusted service program can transparently encrypt the calculation results of the privacy computing system, realize data encryption on disk, and ensure the security of the storage of calculation results. The data processor can only obtain the specified calculation results.

[0020] 5. The privacy computing system is unaware of transparent encryption and decryption operations. All encryption and decryption operations in the trusted data space are transparently implemented by the trusted service program, which intercepts the underlying system-level operations. Therefore, no modification is required to the existing privacy computing system, and any privacy computing system can be ported to it. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of a cross-domain collaborative processing method for privacy data based on a trusted data space, provided in an embodiment of the present invention.

[0023] Figure 2 The flowchart is for Embodiment 1 of the present invention.

[0024] Figure 3 The flowchart is for Embodiment 2 of the present invention. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] This invention discloses a method for cross-domain collaborative processing of privacy data based on a trusted data space. This solution combines blockchain technology, secure multi-party computation, and cryptographic computation technology to ensure the full lifecycle security of privacy data from multiple users during cross-domain collaborative processing, including data transmission security, storage security, and runtime security. This invention primarily includes three entities: data providers, computing power providers, and data processors. Data providers mainly provide valuable data elements but may not possess sufficient computing power for privacy computation; for example, data providers may be small and medium-sized enterprises or government / enterprise units with high-value data. Computing power providers are mainly responsible for constructing a trusted data space and can provide the trusted computing power, trusted service programs, privacy computation systems, and storage services required for data processing. For example, computing power providers can be local big data bureaus or other credible computing power providers, whose main function is to provide a trusted platform for the secure flow of data elements; data processors are companies or organizations that need to use data for processing operations. They may have certain computing power and have deployed their own privacy computing systems locally. For example, a data processor may be a large technology company that wants to use a large amount of data for large model training or other processing operations that require a large amount of data. Data processors may also be other enterprises or organizations that do not have computing power and can use the computing power of computing power providers to process data. At the same time, data processors may also provide data themselves during the data use and processing process. Therefore, the identities of data users and data providers can coexist.

[0027] like Figure 1 As shown, a method for cross-domain collaborative processing of privacy data based on a trusted data space includes: The data provider generates a public-private key pair and sends the public key to the computing power provider via a secure channel; The computing power provider allocates trusted storage space and privacy computing roles to the data provider in the trusted execution environment, and establishes a mapping between public keys and trusted storage space, and saves the public keys to trusted storage space; The data provider encrypts the original data using a locally randomly generated key to obtain ciphertext, and sends the ciphertext, key access address and signature to the computing power provider. The computing power provider verifies the signature using a public key and stores the ciphertext and key access address in a trusted storage space, and establishes a mapping between the ciphertext and the key access address. Depending on whether they have computing power, data processors can choose to initiate data processing requests locally or in a trusted data space. When data from other data providers needs to be used, after authorization, the trusted service program intercepts the reading operation, finds the corresponding key call address through the mapping between the stored ciphertext and the key call address, generates an authentication report with a TEE signature, stores it on the blockchain, and requests a one-time decryption key from the data provider. After remotely verifying the TEE signature authentication report, the data provider distributes a one-time decryption key. The trusted service program decrypts the ciphertext within a trusted execution environment, and the calculation result is transparently encrypted by the trusted service program and stored in the trusted storage space of the data processor.

[0028] The entire process uses blockchain to record key requests, distribution, and usage logs, enabling secure control and traceable auditing of the entire data lifecycle.

[0029] The specific implementation process of this invention is described in more detail below: Step 1: The data provider generates a public / private key pair locally. sk , pk The public and private key pairs use an asymmetric encryption algorithm.

[0030] Step 2: The data provider applies to the computing power provider for the use of trusted data space and transmits its public key through a secure channel (the data provider and the computing power provider establish a secure communication channel using advanced encryption protocols such as TLS / SSL). pk Send to the computing power provider.

[0031] Step 3: The trusted service program deployed by the computing power provider in the trusted execution environment, upon receiving the data provider's usage request, allocates corresponding trusted storage space for it, denoted as... O i And assign it a privacy computing role deployed in a trusted data space. U i And establish the corresponding mapping relationship. O i , pk > Store its public key in a trusted storage space pk 'i' is the identity identifier assigned to this data provider by the trusted service program. Because there may be multiple data providers in the trusted data space, the data provider can... U i Manage your own data and tasks.

[0032] Step 4: The data provider uses its own key management system to randomly generate a key. k and the index of the key idk Key k A symmetric encryption algorithm is used. A random key is used. k Original plaintext data M Encryption yields ciphertextC The data provider will encrypt the text C and keys k call address d splicing C|d and using the private key sk splicing information C|d Obtain by signing s = sig ( C|d The data provider will encrypt the data. C Key access address d and signature s Send to the computing power provider, where the key access address is... d Including requests to use keys k The service address and key k index idk It can be done through the service address and key. k The index sends a key retrieval request to the key management system of the data provider.

[0033] Step 5: The trusted service program in the computing power provider uses the pre-stored public key of the data provider. pk For signature s Verification will be performed; once verification is successful, the encrypted message will be sent. C Save to the trusted storage space corresponding to the data provider. O i And establish a mapping relationship. C , d And save it.

[0034] Step 6: There are two possible scenarios for the data processor: one is that the data processor itself lacks computing power and needs to rely on computing power provided by a computing power provider for data processing; the other is that the data processor itself has a certain amount of computing power, and all calculations are completed locally. The operations required for each of these two scenarios are listed below: (1) If the data processor does not have computing power and also provides data, steps 1-5 above need to be repeated to securely transmit and store its privacy data in the trusted data space. The privacy computing system in the trusted data space initializes an independent privacy computing platform for both the data processor and the data provider. Each participant can use the privacy computing role assigned to them. U i Log in to perform the corresponding privacy calculation operations.

[0035] (2) If the data processor has its own computing power, no changes are required. It can simply operate normally in its own local privacy computing system.

[0036] Step 7: If the data processor wishes to use data from other data providers, it needs to send a cooperation request to the data provider and establish a cooperative relationship once the provider agrees.

[0037] Step 8: The data processing system sends its data processing request to the privacy computing system in the trusted data space. The request includes the specific data processing operations and the required partner data.

[0038] Step 9: After the data provider logs into their own privacy computing platform and agrees to the computation, both parties begin to perform the computation operation. When reading the corresponding input file, the transparent encryption and decryption function of the trusted service program will be triggered.

[0039] Step 10: The trusted service program intercepts all file operations of the privacy computing system. When the privacy computing system attempts to read a file, the trusted service program uses the stored mapping relationship... C,d Find the corresponding key access address d The trusted service program generates an authentication report for the trusted execution environment (TEE) using a trusted metric. The TEE-signed authentication report includes: the identity of the data processor, the unique identifier of the TEE (metric value), the hash value of the trusted service program, and the request time. d Send a request to the data provider to obtain the key. k .

[0040] This section combines blockchain technology to enhance audit security. Tamper-proof technologies like blockchain are used to document key usage: Before sending a key request, the trusted service program constructs a blockchain transaction containing a unique request ID, TEE metric, trusted service program hash value, data processor identity information, request timestamp, request data identifier, and an authentication report of the TEE signature. This transaction is then submitted to the blockchain network for recording. Simultaneously, after the data provider verifies and distributes the key, the verification result and key distribution certificate are also recorded on the blockchain, including the associated request ID, newly generated key ID, distribution timestamp, key validity period, usage restrictions, and the data provider's digital signature. This constructs a complete key usage audit chain, ensuring the immutability and traceability of the entire key request, distribution, and usage process.

[0041] Step 11: After receiving the one-time decryption key acquisition request from the trusted service program, the data provider remotely authenticates the TEE signature authentication report sent by the trusted service program. The public key of the Trusted Execution Environment (TEE) hardware (TEE vendors publish their verification services, from which the public key can be obtained) is used to verify the digital signature of the report, ensuring the report's authenticity and integrity. The content of the verification report is also verified to confirm that: the trusted service program has not changed, the current environment metrics are consistent with expectations, and the request is within the valid timeframe.

[0042] After successful verification, the data provider parses the key request and calls the address according to the key. d Key index in idk Find the corresponding key k And construct a one-time key for use by the trusted service program. k tmp A one-time key is defined as {basic key data, timestamp, expiration date, integrity checksum} and recorded on the blockchain. The data provider then transmits the one-time key via a secure channel. k tmp Send to a trusted service program.

[0043] Step 12: After receiving the returned one-time key information, the trusted service program queries the blockchain record to verify the validity of the one-time key. If the verification is successful, it uses the one-time key to decrypt the ciphertext in the encrypted memory of the trusted execution environment. M = Dec ( k tmp , C Once decryption is complete, the key is immediately securely erased, and the key usage record is uploaded to the blockchain. The key usage count is then automatically updated via a smart contract.

[0044] For privacy-preserving computing systems deployed in a trusted execution space, the transparent decryption of trusted service programs is imperceptible to them. The privacy-preserving computing system can directly read the decrypted data through the transparent decryption system of trusted service programs and perform corresponding computational operations.

[0045] Step 13: After the privacy computing system completes the corresponding data processing operations, it will output the calculation results. r Write to the trusted storage space of the data processor O i During the writing process, the trusted service program will simultaneously trigger an interception operation, and the trusted service program will transparently encrypt the written data.

[0046] Specifically, the trusted service program operates based on the mapping relationship. O i , pk Find the public key corresponding to the user of this trusted storage space. pk The trusted service program generates a random result key for it. k r The trusted service program performs encryption operations on the calculation result and the result key to obtain... R = Enc ( k r , r The encrypted result is then stored in the data processor's trusted storage space. O i In this process, the trusted service program uses the corresponding public key to encrypt the result key and sends the encrypted result key to the data processor through a secure channel.

[0047] In the above embodiments, the data processor wishes to perform privacy-preserving computations with the data provider for data processing and analysis. However, the data provider lacks the computing power to deploy a privacy-preserving computation platform and also desires cross-domain control over its own privacy data. When using a privacy-preserving computation system based on a trusted data space, the data provider can remotely authenticate each call request to the real-time audit trusted service program and use blockchain technology for audit evidence retention, thereby ensuring cross-domain control over its own privacy data. In the above embodiments, the user ensures the security of the data provider's privacy data throughout its entire lifecycle by using technologies such as symmetric encryption, asymmetric encryption algorithms, blockchain, and a trusted execution environment, while also solving the problem of insufficient computing power preventing the data provider from deploying a privacy-preserving computation system locally. A secure privacy-preserving computation system can be deployed in a trusted data space for data providers with insufficient computing power. Traditional peer-to-peer privacy computing systems require both data providers and data processors to possess sufficient computing power, and all plaintext data is stored within their own privacy computing systems. However, for participants with insufficient computing power, deploying a privacy computing system is costly. A common solution is to leverage the computing power of a third-party provider. However, in this case, the data provider needs to transmit its own plaintext privacy data to the privacy computing system deployed by the computing power provider, posing a significant security challenge to user privacy data. In this invention, even data providers with insufficient computing power can deploy a privacy computing system with the help of a third-party provider. Instead of transmitting its own plaintext data to the privacy computing system, the data provider first encrypts its original data using symmetric encryption, and the key is stored in its own key management system. This ensures that the plaintext data remains within the domain, and the computing power provider can only receive the user's encrypted data, unable to extract any privacy information from it. When a privacy computing system performs privacy computation in a trusted data space, the trusted service program intercepts file reading and other operations of the privacy computing system and generates a corresponding authentication report to send to the data provider to request the corresponding key. The data provider can remotely authenticate each key request and use smart contracts to store audit logs on the blockchain. After successful verification, a one-time key is generated and sent to the trusted service program. The trusted service program decrypts the corresponding encrypted data in the encrypted memory of the trusted execution environment and destroys the key immediately after decryption.The entire transparent decryption process is imperceptible to the privacy computing system. Therefore, the data read by the privacy computing system is already plaintext, allowing it to execute all subsequent calculations normally. After the execution is complete, the trusted service program also performs transparent encryption on the result, achieving disk encryption of all data. In this entire scheme, a secure channel built through protocols such as TLS ensures the secure transmission of data from the data provider; symmetric encryption and the trusted service program's transparent encryption algorithm ensure the secure storage of privacy data; and trusted execution environment technology ensures that both the data decryption process and the privacy computing system's computation process are performed in encrypted memory, guaranteeing the security of user data throughout its entire lifecycle. This invention combines blockchain and other technologies to store relevant requests for key services and trusted service programs, ensuring the immutability of audit logs and enabling traceability of the entire data flow process. This is of great significance for the promotion and use of trusted data spaces and encrypted computing technology.

[0048] Furthermore, the symmetric encryption algorithm used supports multiple symmetric encryption algorithms such as the Chinese national standard SM4 and AES. The asymmetric encryption algorithm also supports multiple asymmetric encryption algorithms, including SM2, as well as quantum-resistant cryptographic algorithms.

[0049] Example 1: In this example, there is a data provider A, a data processor B (who also provides data), and a trusted data space constructed by a computing power provider. In this embodiment, data processor B does not possess computing power. It is assumed that all data providers and data processors have pre-negotiated and synchronized the necessary information offline, as well as the metrics of the trusted execution environment and trusted service program for subsequent verification, and have established a secure communication channel with the trusted data space using security protocols such as TLS. Figure 2 As shown.

[0050] Data provider A and data processor B each generate a public-private key pair locally. , and their respective public keys and Send it to the trusted service program via a secure channel.

[0051] The trusted service program allocates corresponding trusted storage space to data provider A and data processor A. and and the corresponding privacy computing system and And establish the corresponding mapping relationship. , >, < , > Save it.

[0052] Data provider A and data processor B each generate symmetric keys locally. and and the corresponding key access address and .

[0053] Data provider A and data processor B each use symmetric keys locally. and For its own plaintext data and Encrypt to obtain and .

[0054] Data provider A and data processor B each use their private keys locally. and Sign the encrypted data and the key access address to obtain and and each of them will , , as well as , , Self-sent to trusted service programs.

[0055] Trusted service programs use pre-stored public keys from data providers. and Each party verifies the signatures sent. Once verification is successful, the ciphertext is sent. and Save to the trusted storage space corresponding to the data provider. and And establish a mapping relationship. , >and< , And save it.

[0056] Data processor B uses a privacy-preserving computing system allocated to itself within a trusted data space. Send your own data processing request, which includes the specific data processing operations and the required partner data.

[0057] After data provider A logs into its privacy computing platform and agrees to the computation, both parties begin performing the privacy computing operation.

[0058] The trusted service program intercepts all file operations of the privacy computing system. When any participating party's privacy computing system attempts to read a file, the trusted service program uses the stored mapping relationship... , >and< , Find the corresponding key access address and Next, the trusted service program will generate an authentication report for the trusted execution environment using a trusted metric. The authentication report includes: the identity information of the data processor, the unique identifier (metric value) of the trusted execution environment, the hash value of the trusted service program, the request time, and other information. The trusted service program signs the trusted execution environment using its built-in unique private key and calls the address using the key. and Send a request to data provider A and data processor B to obtain the key.

[0059] After receiving the key acquisition request from the trusted service program, data provider A and data processor B first remotely authenticate the authentication report sent by the trusted service program. They verify the digital signature of the report using the public key of the trusted execution environment hardware to ensure the report's authenticity and integrity. They also verify the content of the report, confirming that the trusted service program has not changed, the current environment metrics are consistent with expectations, and the request is within a valid timeframe. After successful verification, the data provider parses the key request, finds the corresponding key based on the key index in the key access address, and constructs a one-time key for use by the trusted service program. and The data is then recorded on the blockchain for verification. Data provider A and data processor B communicate via a secure channel and pass on the one-time key. and Send to a trusted service program.

[0060] After receiving the returned key information, the trusted service program first queries the blockchain record to verify the validity of the one-time key. If the verification is successful, it uses the one-time key to decrypt the ciphertext in the encrypted memory of the trusted execution environment. = as well as = Once decryption is complete, the key is immediately securely erased, and the key usage record is uploaded to the blockchain. The key usage count is then automatically updated via a smart contract.

[0061] For privacy-preserving computing systems deployed in a trusted execution space, the transparent decryption of trusted service programs is imperceptible to them. The privacy-preserving computing system can directly read the decrypted data through the transparent decryption system of the trusted service programs and perform corresponding computational operations.

[0062] After the privacy-preserving computation system completes the corresponding data processing operations, it writes the computation result r into the trusted storage space of the data processor B. During the writing process, the trusted service program simultaneously triggers an interception operation, and the trusted service program transparently encrypts the written data. First, the trusted service program, based on the mapping relationship... , Find the public key corresponding to the user of this trusted storage space. Next, the trusted service program generates a random result key for it. Next, the trusted service program performs encryption operations on the calculation result to obtain... = The encrypted result is then stored in a trusted storage space. In, and using public keys Encryption result key The encrypted result key Enc( The data is sent to the data processor. If the data provider A also needs the result, the trusted service performs the same operation.

[0063] Example 2 In this example, there is a data provider A, a data processor B (which possesses its own computing power), and a trusted data space constructed by the computing power provider. It is assumed that all data providers and data processors have pre-negotiated and synchronized the necessary information offline, as well as the metrics of the trusted execution environment and trusted service programs for subsequent verification, and have established a secure communication channel with the trusted data space via secure protocols such as TLS. Figure 3 As shown.

[0064] Data provider A generates a public-private key pair locally. and public key Send it to the trusted service program via a secure channel.

[0065] The trusted service program allocates corresponding trusted storage space to data provider A. and the corresponding privacy computing system And establish the corresponding mapping relationship. , Save it.

[0066] Data provider A generates a symmetric key locally. and the corresponding key access address .

[0067] Data provider A uses a symmetric key locally. For its own plaintext data Encrypt to obtain .

[0068] Data provider A uses a private key locally. Sign the encrypted data and the key access address to obtain and will , , Send to a trusted service program.

[0069] Trusted service programs use pre-stored public keys from data providers. Verify the sent signature; if the verification is successful, the ciphertext will be sent. Save to the trusted storage space corresponding to the data provider. And establish a mapping relationship. , And save it.

[0070] Data processor B uses its own privacy computing system Send your own data processing request, which includes the specific data processing operations and the required partner data.

[0071] After data provider A logs into its privacy computing platform and agrees to the computation, both parties begin performing the privacy computing operation.

[0072] Unlike Example 1, in this example, data processor B has its own computing power, so it does not need to apply for a privacy computing system and a trusted storage space in the trusted data space. It only needs to use its own privacy computing system to perform privacy computing with data provider A. During this process, all of data processor B's data is stored in its own storage space, and the data that interacts with the trusted data space is encrypted data after privacy computing processing, ensuring complete control over its own data.

[0073] The subsequent steps are the same as in Example 1. Only data provider A, which is deployed in the trusted data space, will trigger the same steps, while data processor B only needs to perform the calculations locally.

[0074] After completing the corresponding data processing operations, the privacy-preserving computation system writes the computation result *r* into the local storage space of data processor B, which can then retrieve the result from its own storage space. Simultaneously, privacy-preserving computation system A, deployed by data provider A in a trusted data space, also writes the result *r* into its own trusted storage space. During the writing process, the trusted service program simultaneously triggers an interception operation, and the trusted service program transparently encrypts the written data. First, the trusted service program, based on the mapping relationship... , Find the public key corresponding to the user of this trusted storage space. Next, the trusted service program generates a random result key for it. Next, the trusted service program performs encryption operations on the calculation result and the result key to obtain... = The encrypted result is then stored in a trusted storage space. In this context, the trusted service program uses the corresponding public key. Result key Encryption is performed and the encrypted result key is transmitted through a secure channel. Send to data provider A.

[0075] Example 3 In this example, there are data provider A, data provider B, data processor C (who also provides data), and a trusted data space constructed by a computing power provider. It is assumed that all data providers and data processors have pre-negotiated and synchronized the necessary information offline, as well as the metrics of the trusted execution environment and trusted service programs for subsequent verification, and have established a secure communication channel with the trusted data space using secure protocols such as TLS.

[0076] This example is similar to Example 1, except that there are multiple participants in this example. Each data provider only needs to perform the same steps as in Example 1 to complete the entire process. If a participant has its own computing power, refer to Example 2. There is no need to deploy a corresponding privacy computing system instance in the trusted data space. The detailed process will not be repeated here.

[0077] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0078] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for cross-domain collaborative processing of private data based on a trusted data space, characterized in that, Comprise: The data provider generates a public-private key pair and sends the public key to the computing power provider through a secure channel; The computing power provider allocates a trusted storage space and a privacy computing role for the data provider in a trusted execution environment, and establishes a mapping between the public key and the trusted storage space, and saves the public key to the trusted storage space; The data provider encrypts the original data using a locally randomly generated key to obtain ciphertext, and sends the ciphertext, key call address and signature to the computing power provider, which stores the ciphertext and key call address in the trusted storage space after verifying the signature of the public key pair, and establishes a mapping between the ciphertext and the key call address; The data processing party selects to initiate a data processing request locally or in the trusted data space according to whether it has computing power; When other data provided by the data provider is needed, after authorization, the trusted service program intercepts the read operation, finds the corresponding key call address through the stored mapping between the ciphertext and the key call address, generates an authentication report containing a TEE signature and stores it through a blockchain, and requests a one-time decryption key from the data provider; The data provider distributes the one-time decryption key after remotely verifying the TEE signed authentication report, and the trusted service program decrypts the ciphertext in the trusted execution environment, and the calculation result is transparently encrypted by the trusted service program and stored in the trusted storage space of the data processing party.

2. The method of claim 1, wherein, The key call address includes the service address requesting to use the key and the index of the key.

3. The method of claim 1, wherein, The TEE signed authentication report contains the identity of the data processing party, the TEE measurement value, the trusted service program hash value, the timestamp and the request ID, and is constructed into a blockchain transaction before sending the key request to the consortium chain.

4. The method of claim 2, wherein, The data provider distributes the one-time decryption key after remotely verifying the TEE signed authentication report, and the trusted service program decrypts the ciphertext in the trusted execution environment, including: After receiving the one-time decryption key request sent by the trusted service program, the data provider remotely authenticates the TEE signed authentication report sent by the trusted service program: Verify the digital signature of the TEE signed authentication report through the public key of the trusted execution environment hardware, and verify the content of the proof report; After verification, the data provider finds the corresponding key according to the key index in the key call address, and constructs a one-time key for the trusted service program to use, while recording to the blockchain.

5. The method of claim 4, wherein, The one-time key contains basic key data, timestamp, validity period and integrity check code, and is used only once in the specified TEE instance, and is destroyed immediately after use.

6. The method of claim 1, wherein, The trusted service program decrypts the ciphertext in the trusted execution environment, including: After receiving the returned one-time key, the trusted service program queries the blockchain record to verify the validity of the one-time key; After verification, the one-time key is used to decrypt the ciphertext in the trusted storage space of the trusted execution environment.

7. The method of claim 1, wherein, The calculation result is transparently encrypted by the trusted service program and stored in the trusted storage space of the data processing party, including: The privacy computing system reads the decrypted data through the transparent decryption system of the trusted service program, performs corresponding computing operations, and writes the computing result into the trusted storage space of the data processor. When writing, the interception operation of the trusted service program is triggered, and the trusted service program performs transparent encryption on the written data.

8. The method of claim 7, wherein, The transparent encryption of the trusted service program on the written data comprises: The trusted service program finds the public key corresponding to the user of the trusted storage space according to the mapping of the public key and the trusted storage space; The trusted service program generates a random result key for the computing result; The trusted service program performs an encryption operation on the computing result and the result key, and stores the encrypted result into the trusted storage space of the data processor. The trusted service program encrypts the result key using the corresponding public key and sends the encrypted result key to the data processor through a secure channel.

9. The method of claim 1, wherein, The computing power provider deploys the trusted service program in a trusted execution environment, which includes Intel SGX, ARM TrustZone or a domestic trusted computing module.

10. The method of claim 1, wherein, The whole process records the key request, distribution and use log through the blockchain, realizes the safe management and traceable audit of the whole life cycle of data.

Images