A large-scale model corpus data interaction system, interaction method, terminal, and media based on a trusted execution environment.

Abstract

This invention discloses a large-scale model corpus data interaction system, interaction method, terminal, and medium based on a trusted execution environment. The system includes: a data provider, a data requester, and a trusted data space service platform. The data provider encrypts the corpus data and preprocesses, catalogs, and packages the corpus data in its connector to obtain data capsules. The data requester submits computation tasks in its connector and, after authorization, obtains the corpus data from the data capsules for computation. The trusted data space service platform stores the encrypted corpus data, controls access for data users, schedules computation tasks, manages keys, and generates audit logs. This invention, through hardware isolation and end-to-end encryption within the trusted execution environment, ensures that the corpus data exists only in plaintext form within the trusted execution environment throughout its entire lifecycle, from access to destruction, preventing unauthorized access and leakage.

Description

Technical Field

[0001] This invention relates to the field of trusted execution environment technology, and in particular to a large model corpus data interaction system, interaction method, terminal and medium based on a trusted execution environment. Background Technology

[0002] With the rapid development of generative AI and large-scale model technology, the performance of large models (such as comprehension ability and generation quality) is highly dependent on the support of high-quality, large-scale corpus data. Currently, large-scale model trainers (such as technology companies and research institutions) need to obtain corpora from multiple channels (such as data providers and third-party data platforms), but the following core technologies and challenges exist in the corpus delivery process.

[0003] Currently, data delivery typically takes the following methods:

[0004] 1. Direct plaintext delivery: extremely high risk, easily leading to data leakage and misuse.

[0005] 2. Traditional encrypted delivery: Data needs to be decrypted before computation, and even after decryption, the data is still at risk of being stolen while in memory (e.g., through advanced malware or attacks within cloud platforms).

[0006] 3. Secure multi-party computation or homomorphic encryption: Although these cryptographic schemes can guarantee computational encryption, they have huge computational overhead and high communication complexity, making them difficult to apply to processing scenarios with massive large model corpora, thus limiting their practicality.

[0007] Although some data sharing platforms based on software encryption and access control exist in the existing technology, while these platforms provide a certain level of access control, their trust is built on the software stack and operating system, resulting in a huge attack surface that cannot defend against basic attacks from platform administrators or underlying systems.

[0008] Existing trusted technologies lack sufficient security protection, specifically as follows:

[0009] 1. Standalone TEE (Trusted Execution Environment) technology: Although it can provide a hardware-isolated trusted execution environment, existing solutions mostly focus on the computational trust of a single node and do not cover the entire link of the corpus from the data provider to the data demander (such as access verification, transmission control, and usage auditing). Furthermore, it is not optimized for the "massive and heterogeneous" characteristics of large model corpora. Directly importing TB-level corpora into TEE will lead to memory overflow and excessive processing latency.

[0010] 2. Traditional Trusted Data Space: Existing trusted data space solutions focus on the trustworthiness of data storage and sharing, but their core mechanisms rely on software-level identity authentication and access control, lacking hardware-level trust assurance for the execution environment. When data requesters decrypt corpora in the trusted data space and use them for large model training, they still face the risk of leakage due to an untrustworthy computing environment.

[0011] Therefore, existing technologies still have shortcomings. Summary of the Invention

[0012] The technical problem to be solved by this invention is to provide a large-scale model corpus data interaction system, interaction method, terminal, and medium based on a trusted execution environment, addressing the aforementioned deficiencies of the prior art. The technical solution adopted by this invention is as follows:

[0013] In a first aspect, the present invention provides a large-scale model corpus data interaction system based on a trusted execution environment, wherein the system comprises:

[0014] The data provider is used to encrypt the corpus data using encryption algorithms and upload it to the trusted execution environment. The data provider connector built in the trusted execution environment preprocesses, organizes, and packages the corpus data to obtain data capsules.

[0015] The data requester is used to submit computational tasks in the data requester connector built in the trusted execution environment, and after obtaining authorization from the data provider, to obtain the corpus data of the data capsule for computation.

[0016] The Trusted Data Space Service Platform is used to store encrypted corpus data, control access for data users, schedule computing tasks, manage keys, and generate audit logs.

[0017] In one implementation, the system further includes:

[0018] The Trusted Execution Environment Cluster Management Platform is used to uniformly manage different types of Trusted Execution Environments, including: Trusted Execution Environments corresponding to data providers and Trusted Execution Environments corresponding to data requesters.

[0019] Secondly, embodiments of the present invention also provide a method for large-scale model corpus data interaction based on a trusted execution environment. The method is applied to the large-scale model corpus data interaction system based on a trusted execution environment described above, wherein the method includes:

[0020] The data provider uses encryption algorithms to encrypt the corpus data and uploads it to a trusted execution environment. The corpus data is then preprocessed, cataloged, and packaged in the data provider connector built in the trusted execution environment to obtain a data capsule.

[0021] The data requester initiates a data usage request and forwards the request to the data provider through the Trusted Data Space Service Platform. After obtaining authorization from the data provider, the data requester obtains the corpus data in the data capsule for use.

[0022] Monitor the data requester's usage of the corpus data, and delete the corpus data stored locally by the data requester after the corpus data usage period expires or the data requester completes the use of the corpus data, as well as clear the plaintext corpus data in the trusted execution environment corresponding to the data requester.

[0023] In one implementation, before the data provider encrypts the corpus data using an encryption algorithm and uploads it to the trusted execution environment, the following steps are also included:

[0024] The Trusted Data Space Service Platform deploys a public key infrastructure certification center to issue digital certificates to data providers, data requesters, and nodes in the trusted execution environment.

[0025] Data providers register and log in to user identities through the data provider connector and register and associate with the trusted data space service platform;

[0026] Data demanders register and log in to their user identities through the data demander connector and then register and associate with the trusted data space service platform.

[0027] In one implementation, the data provider encrypts the corpus data using an encryption algorithm and uploads it to a trusted execution environment. Within this trusted execution environment, a data provider connector preprocesses, catalogs, and packages the corpus data to obtain a data capsule, which includes:

[0028] The data provider deploys a trusted execution environment and a corresponding trusted execution environment cluster management platform;

[0029] The data provider encrypts the corpus data through a trusted key issued by the data provider connector, and uploads the corpus data to the data provider's trusted execution environment after remote verification of the data provider's trusted execution environment.

[0030] The data provider connector decrypts the corpus data using a private key, and performs data cataloging and data processing within the data provider connector to form a corpus data product.

[0031] The corpus data product is embedded in a digital watermark that is bound to the identity of the data provider, and the corpus data product is encrypted using the SM4 encryption algorithm and packaged into a data capsule.

[0032] The data provider uploads the data capsule to the Trusted Data Space Service Platform and puts it on the blockchain.

[0033] In one implementation, the data provider uploads the data capsule to the Trusted Data Space Service Platform, and after on-chain processing, includes:

[0034] The data requester deploys a trusted execution environment and a corresponding trusted execution environment cluster management platform.

[0035] The data requester builds a toolchain through a trusted execution environment, packages the model application, and uploads the model application to the trusted execution environment after remote verification of the data requester's trusted execution environment.

[0036] The data requester connector decrypts the model application using a private key;

[0037] The data requester encrypts the model application and generates corresponding summary information, which is then uploaded to the Trusted Data Space Service Platform for review.

[0038] In one implementation, the data requester initiates a data usage request and forwards the request to the data provider via a trusted data space service platform. After obtaining authorization from the data provider, the data requester retrieves and uses the corpus data from the data capsule, including:

[0039] When a data requester initiates a data usage request, the Trusted Data Space Service Platform sets up a digital contract for the requested corpus data and forwards the data usage request to the data provider.

[0040] The data provider authorizes the corpus data requested by the data requester, negotiates the encryption key with the data provider connector and the data requester connector, and transmits the data capsule to the trusted execution environment corresponding to the data provider.

[0041] The data requester unlocks the data capsule within the corresponding trusted execution environment and uses the corpus data within the data capsule within the authorized scope.

[0042] In one implementation, the system monitors the data requester's usage of the corpus data, and upon the expiration of the corpus data usage period or after the data requester has completed using the corpus data, deletes the corpus data stored locally by the data requester, and clears the plaintext corpus data from the trusted execution environment corresponding to the data requester, including:

[0043] Acquire the usage behavior of the data requester on the corpus data, and put the key data in the process of the corpus data being used on the blockchain and store the evidence;

[0044] The usage of the corpus data is verified and audited, and permissions are revoked when unauthorized use is detected.

[0045] After the corpus data usage period expires or the data requester completes the use of the corpus data, delete the corpus data stored locally by the data requester, and clear the plaintext corpus data in the trusted execution environment corresponding to the data requester.

[0046] Data destruction logs are uploaded to the blockchain for remote verification by data providers.

[0047] Thirdly, embodiments of the present invention also provide a terminal, wherein the terminal includes a memory, a processor, and a large model corpus data interaction program based on a trusted execution environment stored in the memory and executable on the processor. When the processor executes the large model corpus data interaction program based on a trusted execution environment, it implements the steps of the large model corpus data interaction method based on a trusted execution environment of any of the above-mentioned schemes.

[0048] Fourthly, embodiments of the present invention also provide a computer-readable storage medium, wherein the computer-readable storage medium stores a large model corpus data interaction program based on a trusted execution environment, and when the large model corpus data interaction program based on a trusted execution environment is executed by a processor, it implements the steps of the large model corpus data interaction method based on a trusted execution environment as described in any of the above schemes.

[0049] Beneficial Effects: Compared with existing technologies, this invention provides a large-scale model corpus data interaction system based on a trusted execution environment. The system includes a data provider, a data requester, and a trusted data space service platform. The data provider encrypts the corpus data using encryption algorithms and uploads it to the trusted execution environment. Within the trusted execution environment, a data provider connector preprocesses, catalogs, and packages the corpus data to obtain data capsules. The data requester submits computation tasks through the data requester connector within the trusted execution environment and, after obtaining authorization from the data provider, retrieves the corpus data from the data capsules for computation. The trusted data space service platform stores the encrypted corpus data, controls access for data users, schedules computation tasks, manages keys, and generates audit logs. This invention, through hardware isolation and end-to-end encryption within the trusted execution environment, ensures that the corpus data exists only in plaintext within the trusted execution environment throughout its entire lifecycle, from access to destruction, preventing unauthorized access and leakage. In addition, this implementation integrates trusted data space and trusted execution environment technologies, covering the entire process of corpus data from access verification, encrypted transmission, trusted processing, authorized delivery, usage auditing to secure destruction, ensuring that the operation of each link is traceable and tamper-proof. Attached Figure Description

[0050] Figure 1 The system architecture diagram of the large model corpus data interaction system based on a trusted execution environment provided in the embodiments of the present invention in the distributed peer-to-peer delivery mode.

[0051] Figure 2 The system architecture diagram of the large model corpus data interaction system based on a trusted execution environment provided in the embodiments of the present invention is shown in the centralized hosting mode of data and models.

[0052] Figure 3 This is a flowchart of a preferred embodiment of the large model corpus data interaction method based on a trusted execution environment provided by the present invention.

[0053] Figure 4 A schematic diagram of a terminal provided in an embodiment of the present invention. Detailed Implementation

[0054] To make the objectives, technical solutions, and effects of this invention clearer and more explicit, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0055] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content, operations, or steps, nor does it require execution in the described order. For example, some operations or steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0056] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0057] It should be understood that, in order to clearly describe the technical solutions of the embodiments of the present invention, the terms "first" and "second" are used in the embodiments of the present invention to distinguish identical or similar items with essentially the same function and effect. For example, "first control information" and "second control information" are only used to distinguish different control information and do not limit their order.

[0058] Those skilled in the art will understand that the words "first" and "second" do not limit the quantity or the order of execution, and that the words "first" and "second" do not necessarily imply that they are different.

[0059] It should also be understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0060] The existing technology for corpus data interaction has the following main drawbacks:

[0061] 1. High risk of privacy leakage: In traditional encrypted transmission and plaintext processing modes, the corpus data is exposed to the general computing environment after decryption. Internal personnel of the data requesting party and malicious software can obtain the plaintext corpus, and there is a lack of effective means to trace the leakage.

[0062] 2. Poor compatibility of TEE (Trusted Execution Environment) with large model corpora: Existing TEE solutions are not optimized for massive corpora. Directly processing TB-level corpora will lead to insufficient memory and low processing efficiency. For example, an experiment showed that traditional solutions require 72 hours to process 1TB of text corpora, which far exceeds the time requirements for training large models. Furthermore, it cannot be compatible with the unified processing of heterogeneous corpora such as text, images, and speech.

[0063] 3. Disruption of trustworthiness throughout the process: The existing trusted data space only covers the storage and sharing stages, while the TEE only covers the computing stage. The two are not coordinated. After the corpus is authenticated in the trusted data space, there is a lack of trusted connection during the transmission to the TEE. Furthermore, there is no subsequent auditing mechanism when the processing results (such as pre-processed corpus) in the TEE are delivered to the data requester, resulting in the problem of "trustworthy access and trustworthy processing, but untrustworthy transmission or delivery".

[0064] 4. Lack of basis for confirmation of rights and distribution of benefits: The existing scheme lacks the technology of "identifying the ownership of corpus data and tracking its use". The data provider cannot prove the ownership of the corpus data, nor can it track the actual scope of use of the data requester (such as whether the corpus data has been sublicensed to a third party), which leads to difficulties in confirming rights and sharing benefits.

[0065] 5. Performance and security imbalance: High-security cryptographic schemes suffer from excessive performance overhead and cannot meet the high efficiency requirements of large-scale model data processing.

[0066] 6. Poor cross-platform compatibility: Different TEE architectures and trusted data space protocols are incompatible with each other, which leads to repeated adaptations for cross-enterprise corpus delivery, resulting in high R&D costs and difficulty in scaling up.

[0067] To address the problems of existing technologies, this invention provides a large-scale model corpus data interaction system based on a trusted execution environment, specifically as follows: Figure 1 and Figure 2As shown, the system includes a data provider, a data requester, and a trusted data service space platform. Specifically, the data provider, as the original owner of the corpus data, encrypts the corpus data using encryption algorithms and uploads it to a trusted execution environment (TEE). Within the TEE, a data provider connector is built to preprocess, catalog, and package the corpus data into data capsules. The data requester submits computation tasks through the TEE connector and, after obtaining authorization from the data provider, acquires the corpus data from the data capsules for computation. The trusted data space service platform stores the encrypted corpus data, controls access for data users, schedules computation tasks, manages keys, and generates audit logs. Furthermore, this embodiment's system also includes a trusted execution environment cluster management platform for unified cluster management of different types of trusted execution environments, including several TEE cluster nodes. In this embodiment, this specifically includes trusted execution environments corresponding to the data provider and trusted execution environments corresponding to the data requester.

[0068] In practical applications, the Trusted Data Space Service Platform of this embodiment is deployed on a clustered management platform supporting Trusted Execution Environments (TEEs). This platform comprises three logical units: a non-secure zone, an access control and task scheduling unit, and a TEE secure zone. The non-secure zone stores encrypted massive amounts of corpus data. The access control and task scheduling unit operates outside the TEE and is responsible for receiving and processing requests from data requesters, verifying their identity and permissions, and scheduling authorized computing tasks to be executed within the TEE. The TEE secure zone is the clustered trusted execution environment, running a trusted data processing program that includes a key processing module, a secure computing engine, and an audit log generation module. The TEE clustered management platform supports various TEE solutions and pools TEEs, allowing users to deploy applications via Kubernetes, an open-source application management platform for managing containerized applications across multiple hosts in a cloud platform.

[0069] Combination Figure 1 and 2As shown, both the data provider connector and the data requester connector are deployed in a trusted execution environment (TEE), as is the trusted data space service platform. This forms a two-layer architecture of trusted data space management layer and trusted execution layer, which facilitates trusted closed-loop management of large model corpus data throughout the entire process from access to transmission to processing to delivery to destruction. This addresses the core issues of untrusted environment, uncontrollable process, and easy privacy leakage in traditional corpus delivery. Furthermore, this embodiment uses a data capsule as the core carrier for corpus data delivery. This data capsule is an integrated encapsulation generated within the trusted execution environment, deeply binding encrypted corpus data with digital contracts, aiming to achieve control upon delivery of corpus data. The core components of the data capsule in this embodiment include: a core layer, a control layer, a rights confirmation layer, and a metadata layer. The core layer consists of large model corpus encrypted using the SM4 algorithm, specifically including heterogeneous data such as text, images, and speech. This large model corpus can only be decrypted within an authorized TEE. The control layer includes digital contracts containing initial usage rules, such as the scope, frequency, duration, destruction conditions, and revenue distribution ratio of the corpus. The digital contracts support subsequent dynamic negotiation and adjustment. The ownership confirmation layer includes an invisible digital watermark bound to the data provider's identity and a corpus digest signature based on the SM3 algorithm (a commercial cryptographic hash algorithm). The invisible digital watermark is used to prevent tampering, and the corpus digest signature is used to prevent forgery. The metadata layer includes corpus type, size, ownership information, generation time, TEE node identifier, etc., to support platform directory management and blockchain traceability. Data capsules can be generated, decrypted, and executed entirely within the TEE, preventing unauthorized access. The flow records of data capsules (such as generation, authorization, delivery, and destruction) can be synchronously recorded on the blockchain for evidence storage, ensuring traceability throughout the entire lifecycle.

[0070] Based on the above embodiments, the present invention also provides a large-scale model corpus data interaction method based on a trusted execution environment. This method can be applied to the system of the fir tree embodiment. In practical applications, the method of this embodiment can be applied to a terminal, which can be an intelligent product terminal such as a computer. The terminal can be used to execute the large-scale model corpus data interaction method based on a trusted execution environment. Specifically, as... Figure 3 As shown in the figure, the large model corpus data interaction method based on a trusted execution environment in this embodiment includes the following steps:

[0071] Step S100: The data provider uses an encryption algorithm to encrypt the corpus data and uploads it to a trusted execution environment. In the data provider connector built in the trusted execution environment, the corpus data is preprocessed, cataloged, and packaged to obtain a data capsule.

[0072] First, the Trusted Data Space Service Platform deploys a Public Key Infrastructure (PKI) Certification Center to issue digital certificates to data providers, data requesters, and nodes in the Trusted Execution Environment. Next, data providers register and log in using the Data Provider Connector and associate their registration with the Trusted Data Space Service Platform. Then, data requesters register and log in using the Data Request Connector and associate their registration with the Trusted Data Space Service Platform.

[0073] Furthermore, the data provider deploys a Trusted Execution Environment (TEE) and a corresponding TEE clustered management platform. Then, the data provider encrypts the corpus data using a trusted key issued by the data provider connector. After remote verification of the data provider's TEE, the corpus data is uploaded to the TEE. Next, the data provider connector decrypts the corpus data using a private key. At this stage, only the TEE can decrypt the corpus data; malicious individuals or high-privilege administrators who steal the data cannot decrypt it. Within the data provider connector, the data provider sets up a data directory and processes the corpus data to create a corpus data product. Then, the corpus data product is embedded in a digital watermark bound to the data provider's identity and encrypted using the SM4 encryption algorithm (a symmetric encryption algorithm), encapsulating it into a data capsule. This data capsule contains encrypted corpus data, a digital contract, a digital watermark, and a corpus digest signature based on the SM3 algorithm. Finally, the data provider uploads the data capsule to the Trusted Data Space Service Platform and records it on the blockchain.

[0074] Furthermore, the data requester deploys a Trusted Execution Environment (TEE) and a corresponding TEE cluster management platform. Next, the data requester uses the TEE's Trusted Application (TEE) toolchain to package the model application. After remote verification of the data requester's TEE, the model application is uploaded to the TEE. Then, the data requester's connector decrypts the model application using its private key. At this point, only the TEE can decrypt user data; malicious individuals or high-privilege administrators cannot decrypt the data. The data requester encrypts the model application and generates a corresponding digest, which is then uploaded to the Trusted Data Space Service Platform for review. Only after passing the review can the model application be used.

[0075] Step S200: The data requester initiates a data usage request and forwards the data usage request to the data provider based on the Trusted Data Space Service Platform. After obtaining authorization from the data provider, the data requester obtains the corpus data in the data capsule for use.

[0076] A data requester discovers a corpus data product on a data space service platform and initiates a data usage request. The trusted data space service platform sets up a digital contract for the requested corpus data and forwards the data usage request to the data provider. The digital contract includes: restrictions on the data product content, restrictions on the data product delivery connector, restrictions on the data product usage connector, restrictions on the data product users, restrictions on the data product usage operations, restrictions on the number of times the data product can be used, restrictions on the data product usage duration, restrictions on the visibility range of the data product, restrictions on the availability range of the data product, notifications during data product usage, and destruction of the data product after use. Next, the data provider authorizes the requested corpus data, negotiates an encryption key with the data provider connector and the data requester connector, and transmits the data capsule to the data provider's corresponding trusted execution environment. After delivery, the data requester unlocks the data capsule within the trusted execution environment and uses the corpus data within the authorized scope.

[0077] Step S300: Monitor the data requester's usage behavior of the corpus data, and delete the corpus data stored locally by the data requester after the corpus data usage period expires or the data requester completes the use of the corpus data, as well as clear the plaintext corpus data in the trusted execution environment corresponding to the data requester.

[0078] This embodiment constructs a smart contract-based access control policy within a trusted data space using a data capsule approach to verify and audit the usage behavior of corpus data. Specifically, this embodiment acquires the usage behavior of the corpus data by the data requester, and records and notifies the key data during the usage process on the blockchain. Then, the usage behavior of the corpus data is verified and audited, and permission revocation is triggered when excessive use is detected, i.e., the TEE node stops supplying corpus data. After the corpus data usage period expires or the data requester completes the use of the corpus data, this embodiment can trigger the deletion of the corpus data stored locally by the data requester, as well as the clearing of the plaintext corpus data in the trusted execution environment corresponding to the data requester, and upload the data destruction log to the blockchain. The data requester can provide the data destruction log to the data provider for remote verification and update the digital contract.

[0079] Therefore, this embodiment proposes a two-layer architecture of a trusted data space management layer and a trusted execution layer. This facilitates a trusted closed-loop management of large-scale model corpus data throughout the entire process from access to transmission, processing, delivery, and destruction, solving the core problems of untrustworthy environments, uncontrollable processes, and easy privacy leaks in traditional corpus delivery. Furthermore, this embodiment combines digital watermarking with blockchain notarization, completing corpus watermark generation and verification within the TEE. Simultaneously, it stores corpus data delivery records (such as data provider, data requester, scope of corpus data use, and validity period of corpus data use) on the blockchain, achieving corpus data ownership confirmation and usage traceability, supporting subsequent revenue distribution. In addition, this embodiment constructs a smart contract-based access control policy within the trusted data space based on a data capsule approach, and can integrate data sandbox technology. Combined with the TEE's hardware-level identity authentication, it enables dynamic adjustment of corpus data usage permissions (such as temporary authorization and recycling after use), preventing overuse.

[0080] In practical use, the trusted data space service platform of this embodiment supports a distributed peer-to-peer delivery mode, as well as a centralized data and model hosting mode. In the distributed peer-to-peer delivery mode, such as... Figure 1 As shown. The specific process is as follows: The data provider performs step 1-1. Identity registration / authentication, and the data requester performs step 1-2. Identity registration / authentication. Next, the data provider performs step 2-1. Encrypted upload of corpus data, and the data requester performs step 2-2. Encrypted upload of model application. Further, in the trusted execution environment, the data provider connector performs step 3-1. Reporting the data product catalog and uploading it to the trusted data space service platform. In the trusted execution environment, the data provider connector performs step 3-2. Reporting model information for review. Next, the data provider connector performs step 4. Initiating a corpus data application. The trusted data space service platform performs step 5. Issuing a data application request and negotiating a digital contract. The data requester connector performs step 6. Returning the review result, and the trusted data space service center performs step 7. Returning the approval result. Then, the data provider connector and the data requester connector perform step 8. Negotiating a key to build an encrypted channel. The data provider performs step 9. Delivering the data capsule. The data requester connector and the data provider connector perform step 10. Delivery and usage monitoring, and completing the digital contract update. Therefore, in the distributed peer-to-peer interaction mode, the data demander connector and the data provider connector can deliver data capsules to each other, achieving peer-to-peer encrypted delivery.

[0081] The trusted execution environment enables functions such as identity management, access connector management, directory management, trusted data space management, digital contract management, data usage control, key management, model hosting, and data hosting. Therefore, the trusted data space service platform in this embodiment also supports a centralized data and model hosting mode. In the centralized data and model hosting mode, such as... Figure 2 As shown, the specific process is as follows: The data provider performs step 1-1. Identity registration / authentication, and the data requester performs step 1-2. Identity registration / authentication. Next, the data provider performs step 2-1. Encrypted upload of the corpus data, and the data requester performs step 2-2. Encrypted upload of the model application. Further, in the trusted execution environment, the data provider connector performs step 3-1. Reporting the data product catalog and uploading it to the trusted data space service platform. In the trusted execution environment, the data provider connector also performs step 3-2. Reporting the encrypted model application. Next, the data provider connector performs step 4. Uploading the data capsule containing the encrypted corpus data. The data requester connector performs step 5. Viewing the corpus data catalog and step 6. Initiating a corpus data request. Then, the data provider connector and the data requester connector perform step 7. Digital contract negotiation, followed by the data provider performing step 8. Reviewing the digital contract negotiation. Finally, the data requester connector and the trusted data space service platform perform step 9. Initiating the delivery of the data capsule containing the corpus data for model training, fine-tuning, and inference. This demonstrates that, under the centralized hosting model for data and models, the delivery of corpus data is accomplished by the trusted data space service platform.

[0082] Compared with the prior art, the present invention has the following significant advantages:

[0083] 1. Enhanced security: Utilizing the hardware-level security isolation of TEE, sensitive data is ensured to remain encrypted even during computation, effectively defending against attacks from the operating system and even physical hardware.

[0084] 2. Higher performance: Plaintext data is processed directly within the TEE, resulting in extremely high computational efficiency and minimal performance loss, making it ideal for processing massive large-scale model corpora.

[0085] 3. Verifiable trust: By recording access policies and audit logs through blockchain, the data usage process is made transparent and tamper-proof, establishing a trust foundation for multi-party collaboration.

[0086] 4. End-to-end protection: It achieves full-process security protection from data delivery and computation to result return, covering the entire data lifecycle.

[0087] The modules in the aforementioned large-scale model corpus data interaction system based on a trusted execution environment can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of the terminal in hardware form or independent of it, or stored in the memory of the terminal in software form, so that the processor can call and execute the corresponding operations of each module.

[0088] Based on the above embodiments, the present invention also provides a terminal, the principle block diagram of which can be as follows: Figure 4 As shown. The terminal may include one or more processors 100 ( Figure 4 (Only one is shown in the image), memory 101, and a computer program 102 stored in memory 101 and executable on one or more processors 100, such as a sleep analysis program based on multi-sensor data. When one or more processors 100 execute computer program 102, they can implement the various steps in the sleep analysis method embodiment based on multi-sensor data. Alternatively, when one or more processors 100 execute computer program 102, they can implement the functions of various modules / units in the large model corpus data interaction system embodiment based on a trusted execution environment, which is not limited here.

[0089] In one embodiment, the processor 100 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0090] In one embodiment, memory 101 may be an internal storage unit of an electronic device, such as a hard drive or RAM. Memory 101 may also be an external storage device of the electronic device, such as a plug-in hard drive, smart media card (SMC), secure digital (SD) card, flash card, etc. Furthermore, memory 101 may include both internal and external storage units. Memory 101 is used to store computer programs and other programs and data required by the terminal. Memory 101 can also be used to temporarily store data that has been output or will be output.

[0091] Those skilled in the art will understand that Figure 4 The block diagram shown is merely a partial structural diagram related to the present invention and does not constitute a limitation on the terminal to which the present invention is applied. A specific terminal may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0092] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, operational databases, or other media used in the embodiments provided by this invention can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual operating data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0093] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A trusted execution environment-based large model corpus data interaction system, characterized in that, The system comprises: a data provider, configured to encrypt corpus data using an encryption algorithm and upload the corpus data to a trusted execution environment, and preprocess, catalog and package the corpus data in a data provider connector constructed in the trusted execution environment to obtain a data capsule; a data demander, configured to submit a computing task in a data demander connector constructed in the trusted execution environment, and obtain corpus data of the data capsule for computing after being authorized by the data provider; a trusted data space service platform, configured to store the encrypted corpus data, control access of a data user, schedule a computing task, manage a key and generate an audit log; a trusted execution environment cluster management platform, configured to uniformly manage different types of trusted execution environments, including a trusted execution environment corresponding to the data provider and a trusted execution environment corresponding to the data demander; the trusted data space service platform is deployed on the trusted execution environment cluster management platform, and the trusted data space service platform comprises three logical units, namely a non-secure area, an access control and task scheduling unit and a TEE secure area; the non-secure area is configured to store massive encrypted corpus data; the access control and task scheduling unit is configured to receive and process a request of the data demander, verify an identity and an authority of the data demander, and schedule an authorized computing task to the TEE for execution; the TEE secure area is a clustered trusted execution environment, and a trusted data processing program is run in the TEE secure area; the data provider connector and the data demander connector are both deployed in the trusted execution environment, and the trusted data space service platform is also deployed in the trusted execution environment, forming a double-layer architecture of a trusted data space management layer and a trusted execution layer; a core structure of the data capsule comprises a core layer, a control layer, a right confirmation layer and a metadata layer; the core layer is a large model corpus encrypted by using an SM4 algorithm, and specifically comprises text, image and voice; the large model corpus can be decrypted only in an authorized TEE; the control layer comprises a digital contract, the digital contract comprises limiting data product content, limiting data product delivery connector, limiting data product use connector, limiting data product user, limiting data product use operation, limiting data product use times, limiting data product use time length, limiting data product visible range, limiting data product available range, data product use time notification, data product use time destruction, and the digital contract supports subsequent dynamic negotiation and adjustment; the right confirmation layer comprises an invisible digital watermark bound with an identity of the data provider and a corpus digest signature based on an SM3 algorithm; and the metadata layer comprises corpus type, size, right information, generation time and TEE node identifier, and is configured to support platform catalog management and blockchain traceability; the data capsule can be generated, decrypted and executed in the TEE throughout the whole process, unauthorized access is prevented, and flow record of the data capsule can be stored in a chain for evidence, so that the whole life cycle is traceable; the data provider is specifically configured to: deploy a trusted execution environment, and deploy a corresponding trusted execution environment cluster management platform; The corpus data encrypted by the trusted key pair dispatched by the data provider connector is uploaded to the trusted execution environment after remote verification in the trusted execution environment passes; The corpus data is decrypted by the private key, and the corpus data is set with a data directory and processed in the data provider connector to form a corpus data product; The corpus data product is embedded in a digital watermark bound to an identity, and the corpus data product is encrypted using the SM4 encryption algorithm and packaged as a data capsule; The corpus data requested by the data demander is authorized, and the encryption key is negotiated through the data provider connector and the data demander connector, and the data capsule is transmitted to the trusted execution environment; The data demander is specifically used for: Deploying a trusted execution environment and deploying a corresponding trusted execution environment cluster management platform; The model application is packaged through a trusted execution environment construction tool chain, and the model application is uploaded to the trusted execution environment after remote verification in the trusted execution environment passes; The model application is decrypted by the private key; The model application is encrypted and corresponding summary information is generated, and is uploaded to a trusted data space service platform for auditing; Initiating a data use request, the trusted data space service platform sets a digital contract for the requested corpus data, and forwards the data use request to the data provider; The data capsule is unpacked in the corresponding trusted execution environment, and the corpus data in the data capsule is used within the authorized range; The trusted data space service platform is specifically used for: Obtaining the use behavior of the data demander on the corpus data, and storing and certifying the key data in the use process of the corpus data; The use behavior of the corpus data is checked and audited, and when the use range is found to be exceeded, the authority is recovered; After the corpus data use period expires or the data demander completes the use of the corpus data, the corpus data stored locally by the data demander is deleted, and the corpus plaintext data in the corresponding trusted execution environment of the data demander is cleared; The data destruction log is chained for remote verification by the data provider.

2. The method for large model corpus data interaction based on a trusted execution environment, the method is applied to the system for large model corpus data interaction based on a trusted execution environment in claim 1, characterized in that, The method comprises: The data provider encrypts the corpus data using an encryption algorithm and uploads it to the trusted execution environment, and pre-processes, directory organizes and packages the corpus data in the data provider connector constructed in the trusted execution environment to obtain a data capsule; The data demander initiates a data use request, and forwards the data use request to the data provider based on the trusted data space service platform, and after obtaining the authorization of the data provider, obtains the corpus data in the data capsule for use; The use behavior of the data demander on the corpus data is monitored, and after the corpus data use period expires or the data demander completes the use of the corpus data, the corpus data stored locally by the data demander is deleted, and the corpus plaintext data in the corresponding trusted execution environment of the data demander is cleared.

3. The trusted execution environment-based large model corpus data interaction method according to claim 2, characterized in that, Before the data provider encrypts the corpus data using an encryption algorithm and uploads it to the trusted execution environment, it further comprises: The trusted data space service platform deploys a public key infrastructure authentication center to issue digital certificates for data providers, data demanders, and nodes of the trusted execution environment; The data provider registers and logs in the user identity through the data provider connector and registers and associates in the trusted data space service platform; The data demander registers and logs in the user identity through the data demander connector and registers and associates in the trusted data space service platform.

4. The trusted execution environment-based large model corpus data interaction method according to claim 2, characterized in that, The data provider encrypts the corpus data using an encryption algorithm and uploads it to the trusted execution environment, and pre-processes, catalogues, and packages the corpus data in the data provider connector constructed in the trusted execution environment to obtain a data capsule, including: The data provider deploys the trusted execution environment and deploys the corresponding trusted execution environment cluster management platform; The data provider encrypts the corpus data through the trusted key distributed by the data provider connector, and uploads the corpus data to the trusted execution environment of the data provider after remote verification of the trusted execution environment of the data provider; The data provider connector decrypts the corpus data through the private key, and sets up a data directory and processes data in the data provider connector to form a corpus data product; The corpus data product is embedded in a digital watermark bound to the identity of the data provider, and is encrypted using the SM4 encryption algorithm to form a data capsule; The data provider uploads the data capsule to the trusted data space service platform and performs chain-up.

5. The trusted execution environment-based large model corpus data interaction method according to claim 4, characterized in that, After the data provider uploads the data capsule to the trusted data space service platform and performs chain-up, including: The data demander deploys the trusted execution environment and deploys the corresponding trusted execution environment cluster management platform; The data demander packages the model application through the trusted execution environment construction tool chain, and uploads the model application to the trusted execution environment of the data demander after remote verification of the trusted execution environment of the data demander; The data demander connector decrypts the model application through the private key; The data demander encrypts the model application and generates corresponding digest information, and uploads it to the trusted data space service platform for review.

6. The trusted execution environment-based large model corpus data interaction method according to claim 2, characterized in that, The data demander initiates a data use request, and forwards the data use request to the data provider based on the trusted data space service platform, and after obtaining the authorization of the data provider, uses the corpus data in the data capsule, including: The data demander initiates a data use request, and the trusted data space service platform sets a digital contract for the corpus data requested by the data demander and forwards the data use request to the data provider; The data provider authorizes the corpus data requested by the data demander, negotiates the encryption key through the data provider connector and the data demander connector, and transmits the data capsule to the corresponding trusted execution environment of the data provider; The data demander opens the data capsule in the corresponding trusted execution environment and uses the corpus data in the data capsule within the authorized range.

7. The trusted execution environment-based large model corpus data interaction method according to claim 2, characterized in that, The use behavior of the corpus data by the data demander is monitored, and after the use period of the corpus data expires or the data demander completes the use of the corpus data, the corpus data stored locally by the data demander is deleted, and the corpus plaintext data in the trusted execution environment corresponding to the data demander is cleared, including: Obtaining the use behavior of the corpus data by the data demander, and storing the key data in the process of using the corpus data on a chain and notarizing the key data; Verifying and auditing the use behavior of the corpus data, and triggering authority recovery when out-of-range use is found; After the use period of the corpus data expires or the data demander completes the use of the corpus data, the corpus data stored locally by the data demander is deleted, and the corpus plaintext data in the trusted execution environment corresponding to the data demander is cleared; The data destruction log is stored on a chain for remote verification by the data provider.

8. A terminal, characterized by comprising: The terminal includes a memory, a processor, and a large model corpus data interaction program based on a trusted execution environment stored in the memory and executable on the processor. When the processor executes the large model corpus data interaction program based on the trusted execution environment, the steps of the large model corpus data interaction method based on the trusted execution environment of any one of claims 2-7 are implemented.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a large model corpus data interaction program based on a trusted execution environment, which is executed by the processor to implement the steps of the large model corpus data interaction method based on the trusted execution environment of any one of claims 2-7.

Images