Whole-process data operation and value co-creation system based on trusted data space

Abstract

A whole-process data operation and value co-creation system based on a trusted data space comprises a data acquisition and transmission module, a data aggregation and fusion module, a standard management and interoperation module, a data circulation and sharing module, a security and trusted management and control module and a system operation and monitoring module. The data acquisition and transmission module is used for data acquisition and efficient transmission, the data aggregation and fusion module is used for data integration and analysis, the standard management and interoperation module is used for standard establishment, the data circulation and sharing module is used for managing data cross-domain circulation, and the security and credibility management and control module is used for security monitoring and credibility management. And the system operation and monitoring module is used for monitoring the operation state of the system and user interaction. According to the whole-process data operation and value co-creation system based on the trusted data space, a data circulation management algorithm based on the block chain technology is proposed to manage cross-domain data circulation; and proposing a federated learning-based data privacy protection algorithm to carry out privacy protection on system data.

Description

Technical Field

[0001] The present invention relates to the fields of blockchain, federated learning, and privacy protection technology, and specifically to a full-process data operation and value co-creation system based on a trusted data space. Background Art

[0002] Blockchain technology is a decentralized data storage and management technology designed to address trust issues in data circulation. Through distributed ledgers and smart contracts, it ensures data immutability and traceability. In a full-process data operation and value co-creation system based on a trusted data space, blockchain technology is used to manage cross-domain data circulation. By establishing a transparent and trustworthy data transfer mechanism, it ensures the security and controllability of data exchanged between different entities. Furthermore, blockchain's consensus mechanism and encryption technology effectively prevent malicious tampering and unauthorized access, providing technical support for the sharing and trading of data resources.

[0003] Federated learning and privacy protection technology is a distributed machine learning method that aims to solve the problem of privacy leakage in the data sharing process. By training the model locally on each participant and exchanging only model parameters without transmitting the original data, the secure sharing of data value is achieved. In a full-process data operation and value co-creation system based on a trusted data space, federated learning combines differential privacy and secure multi-party computing technology to ensure that data participates in collaborative computing under the premise of privacy protection, avoiding data abuse and leakage. In addition, this federated learning and privacy protection technology can establish a trusted data collaboration mechanism between different data holders, ensuring data sovereignty while improving the utilization efficiency of data elements.

[0004] However, the existing full-process data operation and value co-creation system based on a trusted data space lacks a unified trust mechanism when data circulates across domains. Data may be tampered with and abused during circulation, affecting the reliability and traceability of the data. Secondly, the system relies on the traditional data centralized processing model during the collaborative calculation and analysis of data, which poses a risk of data privacy leakage and makes it difficult to meet data security and compliance requirements. Summary of the Invention

[0005] The purpose of the present invention is to provide a full-process data operation and value co-creation system based on a trusted data space, so as to solve the problems existing in the existing full-process data operation and value co-creation system based on a trusted data space proposed in the above background technology, such as the lack of a unified trust mechanism when data circulates across domains, the possibility of data being tampered with and abused during circulation, affecting the reliability and traceability of the data, and the system relying on the traditional data centralized processing mode in the process of data collaborative calculation and analysis, which poses a risk of data privacy leakage and makes it difficult to meet data security compliance requirements.

[0006] To achieve the above-mentioned objectives, the present invention provides the following technical solutions: a full-process data operation and value co-creation system based on a trusted data space, including a data acquisition and transmission module, a data aggregation and fusion module, a standard management and interoperability module, a data circulation and sharing module, a security and trusted management and control module, and a system operation and monitoring module, characterized in that: the data acquisition and transmission module is used to collect original data from multiple types of data sources and transmit it securely, ensuring that the data enters the system in real time, accurately and securely; the data aggregation and fusion module includes a multi-source data integration unit and a data fusion analysis unit, the multi-source data integration unit is used to clean, standardize and structure multivariate data to ensure data consistency and high quality, the data fusion analysis unit is used to classify, fuse and feature analyze the integrated data to ensure efficient extraction and utilization of data value; the standard management and interoperability module is used to formulate, manage and execute data standards and interface protocols to ensure the connection between modules and the connection between modules. Interconnection, interoperability and compatibility between different data systems are achieved; the data circulation and sharing module includes a data circulation management unit and a data utilization and operation unit. The data circulation management unit proposes a data circulation management algorithm based on blockchain technology to achieve tamper-proof recording, transparent supervision and full-process tracking of data during cross-domain data circulation, ensuring the security and credibility of data sharing and circulation. The data utilization and operation unit is used to commercialize and operate shared data to ensure the maximum utilization of data resources and achieve value co-creation; the security and trust management module proposes a data privacy protection algorithm based on federated learning to implement distributed authentication, dynamic access control and cross-domain data circulation privacy protection to ensure the security, credibility and privacy protection of data throughout the entire process; the system operation and monitoring module is used to monitor and manage the system operation status and data flow in real time, and interact with users for information and control instructions to ensure stable and efficient operation of the system.

[0007] Preferably, the data collection and transmission module automatically collects data from various data sources and adopts encryption and distributed transmission protocols to ensure the real-time and integrity of data collection and the security of the transmission process.

[0008] Preferably, the data aggregation and fusion module includes a multi-source data integration unit, which ensures that original data from different sources and in different formats can be accurately collected, normalized and converted into a unified data format in the first time by adopting standardized interfaces and automated data preprocessing technology, thereby providing high-quality and structured basic data for subsequent deep data fusion.

[0009] Preferably, the data aggregation and fusion module includes a data fusion analysis unit, which ensures that the multi-source data after preliminary integration can effectively remove redundancy, fill in gaps, and realize organic fusion of data in multiple dimensions by adopting data normalization, cleaning, intelligent matching and association rule analysis methods, and finally generate accurate and applicable data products, providing solid data support for decision support, data sharing and commercial value enhancement.

[0010] Preferably, the standard management and interoperability module ensures data interoperability and standard consistency between different platforms and systems by establishing unified data standards, metadata management and semantic conversion mechanisms.

[0011] Preferably, the data circulation and sharing module includes a data circulation management unit, which proposes a data circulation management algorithm based on blockchain technology. By deploying a distributed ledger and smart contract mechanism based on blockchain technology, it ensures that cross-domain data is fully recorded, verified in real time and cannot be tampered with during the circulation process, thereby ensuring the transparency and traceability of data transmission.

[0012] Preferably, the data circulation management algorithm based on blockchain technology is specifically as follows: First, the infrastructure of a trusted data space is constructed through the blockchain network to achieve tamper-proof recording and transparent supervision of data circulation, and a distributed ledger is created in the Ethereum blockchain network. The specific formula for defining the data structure model is expressed as follows:

[0013]

[0014] in, Represented as a blockchain ledger, Expressed as blocks, Represented as a block number index, Represented as each independent block, Represented as a hash value of the block, used to uniquely identify the block and ensure integrity, It is represented as a set of transactions in a block, which records the specific operations of data circulation. It is represented as the timestamp of the block, which records the time when the block was generated, ensuring the consistency of the time sequence. By initializing the distributed ledger and deploying the smart contract, it provides the underlying rule constraints and automated execution mechanism for the subsequent data flow, ensuring the auditability of the whole process of data operation. Then, the smart contract function is designed. Its core functions include data rights confirmation, access policy management and transaction verification. The contract logic is implemented through Language implementation, key functions are expressed as:

[0015]

[0016] in, Represented as data metadata, Expressed as an attribute access policy, Represented as a transaction identifier, It is expressed as a contract interface. By building the underlying ledger architecture of the trusted data space, the smart contract function realizes the automatic execution of data operation rules, providing a trusted execution environment for the value co-creation system. Secondly, it combines symmetric and asymmetric encryption technologies to ensure the confidentiality and integrity of data in circulation, and uses IPFS to achieve decentralized storage, reduce the risk of single point failure, and perform pre-processed data. Perform two-stage encryption, the specific formula is expressed as:

[0017]

[0018]

[0019] in, Represented as symmetric encrypted ciphertext, Represented as raw data, Denotes Advanced Encryption Standard, Expressed as a symmetric encryption key, Represented as a 256-bit binary key space, Indicated as belonging to, Represented as asymmetric encrypted ciphertext, Expressed as an elliptic curve integrated encryption function, Represented as the public key of the data owner, Represented as the private key of the data owner, Represented as the base point of the elliptic curve, used in the symmetric encryption phase Algorithm on raw data Encryption, using the ECIES function for the symmetric key in the asymmetric encryption stage Encrypt the ciphertext Upload to the IPFS network and obtain the content identifier. The specific formula is:

[0020]

[0021] in, Represented as an IPFS content-addressed hash function,

[0022] Through layered encryption and distributed storage, we build the security sharing requirements in the whole process of data operation, ensure the immutability of data assets in the process of value co-creation, and then implement fine-grained access control based on blockchain to ensure that data is only used by authorized parties on demand. At the same time, we support dynamic update and audit traceability of policies and define attribute sets. , using logic gate combination calculation strategy expression, the specific formula is expressed as:

[0023]

[0024] in, Expressed as attribute access control policy, Represented as the logical AND operator, Represented as the user's characteristics, Represented as the logical exclusive OR operator, Represents the specific value of the attribute. Expressed as the total number of conditions in the policy, Represented as the index of the condition in the policy, the policy With ciphertext Binding is written into the blockchain through smart contracts. The specific formula is expressed as:

[0025]

[0026] in, Represented as a policy update transaction function, it is used to write the new access control policy into the transaction record of the blockchain. Represented as a smart contract update function, It is represented as the digital signature of the data owner, which is used to verify the legitimacy of the policy update. Then, it solves the key management problem in cross-domain data sharing, records the key distribution track through the blockchain, and realizes the full traceability of the key life cycle. Initiate a key request, trigger the smart contract to verify the attribute match, It is represented as the index of the number of data users. The specific formula is:

[0027]

[0028] in, Represented as a verification function, it is used to check whether the user attribute set satisfies the access control policy. Represented as a user The attribute set includes the user's identity, role, and permission information. It is expressed as a logical judgment symbol, and the session key is negotiated through the improved ECDH protocol. The specific formula is expressed as follows:

[0029]

[0030] in, Represented as a session key, used to encrypt communications between data users and data owners, Represented as a hash-based message authentication code function, used to generate a key-derived value, Represented as a data user Temporary private key for key negotiation, Represented as a random salt value, used to enhance the security of the key and prevent replay attacks. Encrypted and written into blockchain transactions

[0031]

[0032] in, Represented as a blockchain transaction record, used to store encrypted session keys, Represented as a smart contract function that writes the encryption key to the blockchain, Indicates that the user uses the data Public key The encryption function can enhance the audit capability of the trusted data space and support the trust transfer of cross-domain value co-creation through the on-chain evidence storage mechanism of key distribution.

[0033] Preferably, the data circulation and sharing module includes a data utilization and operation unit. The data utilization and operation unit ensures that all participants can efficiently and securely access data, develop applications and realize commercial operations by building a standardized data service interface and a flexible data development platform, thereby converting the value of data after circulation into practical application results.

[0034] Preferably, the security and trusted management module proposes a data privacy protection algorithm based on federated learning. By integrating differential privacy and secure multi-party computing technology based on federated learning technology, it ensures that data is not accessed or abused without authorization during cross-subject collaboration, and ensures the privacy security, trusted authentication and dynamic management of the system during data circulation.

[0035] Preferably, the data privacy protection algorithm based on federated learning is as follows: First, a full-process data operation infrastructure based on a trusted data space is constructed to achieve secure interconnection between terminal devices, edge nodes and cloud servers, providing a decentralized collaborative environment for subsequent privacy protection algorithms. In the trusted data space, a consortium blockchain network consisting of super edge nodes is deployed. Each SEN has edge computing, model aggregation and blockchain verification functions, and initializes the global federated model parameters. ,in , Represented as an initialized high-dimensional parameter vector, Expressed as model dimensions, Represented as a model identifier, generating initial identity credentials through a digital signature protocol ,in, Expressed as the total number of participating devices, Represented as the index of the number of participating devices, Represented as a device The encrypted signature is used for authentication of subsequent communications. The specific formula is expressed as:

[0036]

[0037] in, Represented as a function for randomly initializing model parameters, Expressed as a digital signature generation function, Represented as a device A unique identifier for Represented as a device The public key of It is represented as a string concatenation operation, and the model training is completed locally on the terminal device. The user's sensitive data is protected by differential privacy technology, ensuring that the data is available but invisible in the trusted data space. The terminal device Download the latest global model from the edge node , using local datasets Conduct training, Represented as the index of federated learning iterations, calculating local gradients ,in is the loss function, and the gradient is applied Norm clipping, the specific formula is expressed as:

[0038]

[0039] in, Represented as the clipped local gradient vector, Represented as a device The local gradient vector of Expressed as norm, Represented as the gradient clipping threshold, input Laplace noise , generate privacy-preserving gradients, Represented as the Laplacian noise vector injected into the local gradient, Expressed as a Laplace distribution, It is expressed as gradient sensitivity and privacy budget to control the strength of privacy protection. Secondly, local model updates are aggregated at the edge node layer, and blockchain technology is combined to achieve trusted verification and traceability of parameters, supporting the auditability of full-process data operations. Edge nodes receive privacy gradients uploaded by terminal devices. , verify the digital signature , ensure the legality of the data source, execute the federated average algorithm, the specific formula is expressed as:

[0040]

[0041]

[0042] in, Expressed as The global model parameters of the wheel, Denoted as the learning rate, Expressed as The update amount of the global model parameters in the iteration. Secondly, the federated learning model is used to generate personalized recommendations in real time at the edge layer, combining user behavior data with global hot trends to achieve value co-creation. The edge node loads the latest model , according to the user Local behavior data , calculate the interest vector :

[0043]

[0044] in, It is expressed as a model inference function. Then, the global hotspot vector is read from the blockchain and fused to generate a hybrid recommendation score. The specific formula is expressed as:

[0045]

[0046] in, Represented as a user Recommendation score for hot information, Denoted as personalized weight, Represented as a user The interest vector, It is expressed as a global hotspot vector, which is used to describe the characteristics and trend characteristics of hot information on a global scale, and is used to describe the hot attention trends of the entire user group. Finally, the model parameters and recommendation results are permanently stored in the blockchain, supporting multi-party data value sharing and auditing, and building a sustainable operation ecosystem. The specific formula is expressed as:

[0047]

[0048] in, Expressed as The blockchain data blocks generated during the round of federated learning, Represented as the set of feedback vectors of all users, Expressed as The hash value of the blockchain data block.

[0049] Preferably, the system operation and monitoring module ensures that the platform continues to operate efficiently and stably and can detect and handle abnormal situations in a timely manner by monitoring the system operation status, data flow and log records in real time, and interacting with users on information and control instructions.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] 1. The data circulation management unit proposes a data circulation management algorithm based on blockchain technology. First, the algorithm provides the underlying trusted infrastructure for the trusted data space by constructing a distributed ledger system based on blockchain. The entire process of data circulation is recorded through the block structure, ensuring that all additions, deletions, modifications and queries of data during the circulation process can be recorded in real time and cannot be tampered with, thereby achieving a highly transparent and traceable regulatory environment. Through the smart contract mechanism deployed on the Ethereum network, the algorithm can automatically complete key operations such as data rights confirmation, access policy setting, and transaction verification on the chain, effectively improving the execution efficiency and credibility of the data operation link, and building a value co-creation basic platform with rule self-execution capabilities. Secondly, combining symmetric and asymmetric encryption technologies, double-layer encryption is implemented in the data preprocessing stage, and the ciphertext is uploaded to the decentralized IPFS system for storage, effectively reducing the risk of data leakage and single point failure, and ensuring the confidentiality and integrity of data in cross-platform and cross-subject circulation. In addition, the algorithm introduces fine-grained attribute-based policy management in access control, which can achieve precise control of data access rights and prevent data from being used without authorization. ; The access policy also supports dynamic updates and on-chain audits, which greatly enhances the flexibility and controllability of the data sharing process. In terms of cross-domain sharing, the system records the entire process of key distribution through the blockchain, solves the trust bottleneck problem in traditional key management, and realizes full-link traceability of the key life cycle. When the data user initiates an access request, the system automatically calls the smart contract to verify whether its attribute set meets the access policy. If the verification is passed, the session key is dynamically generated based on the improved key negotiation protocol and encrypted into the blockchain to ensure the security of subsequent communications. Through the integration of the above multi-dimensional mechanisms, the algorithm not only ensures the privacy, controllability and verifiability of data assets throughout the entire operation process, but also provides trust guarantees for the co-creation of data value with the participation of multiple subjects, and supports the implementation of trusted data space in multiple sensitive data fields such as government affairs, medical care, industry, and finance, with extremely high promotion value and application prospects.

[0052] 2. The security and trust management module proposes a data privacy protection algorithm based on federated learning. The algorithm first ensures the secure connection between terminal devices, edge nodes and cloud servers by building a trusted data space infrastructure, providing a solid environment support for distributed collaborative learning. On this basis, a consortium blockchain network composed of super edge nodes is deployed, so that each node has edge computing, model aggregation and blockchain verification functions, effectively enhancing the system's decentralized governance capabilities and trust mechanism. The introduction of the federated learning algorithm enables terminal devices to perform model training locally, protect user sensitive data through differential privacy technology, and ensure that data is always available and invisible, which is practical. The existing data can participate in global modeling without leaving the local area, effectively avoiding the privacy leakage risks faced by traditional data sets. At the same time, the algorithm applies L2 norm clipping and noise perturbation to the gradient during local training, which enhances the ability to resist inference attacks and provides a protected data model for edge aggregation. At the edge node layer, the algorithm verifies the terminal digital signature to ensure the integrity of data transmission and the credibility of the identity, and then uses weighted averaging to aggregate local model updates to improve model convergence efficiency. Through deep integration with the blockchain, the auditability and non-tamperability of the model parameter update process are achieved, providing technical support for subsequent data governance. During the model inference stage, the algorithm dynamically generates personalized recommendations based on local user behavior data and global hot trends, promoting the implementation of the data value co-creation mechanism. The edge nodes generate hybrid recommendation results by integrating user interest vectors and global hot vectors, which not only ensures a personalized experience, but also activates the reuse potential of data elements. Finally, the algorithm will upload the updated model parameters, user recommendation results and feedback information to the chain to form a trusted data certificate, supporting data value sharing involving multiple parties. The data privacy protection algorithm based on federated learning not only comprehensively improves the security, auditability and intelligence level of the system, but also promotes the closed-loop operation of the entire process from data collection to value transformation, and establishes a sustainable operation ecosystem that emphasizes both privacy protection and value co-creation, demonstrating strong technical foresight and industrial applicability. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a structural schematic diagram of the present invention; DETAILED DESCRIPTION

[0054] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0055] See also Figure 1The present invention provides a full-process data operation and value co-creation system based on a trusted data space, including a data acquisition and transmission module, a data aggregation and fusion module, a standard management and interoperability module, a data circulation and sharing module, a security and trusted management and control module, and a system operation and monitoring module. It is characterized in that: the data acquisition and transmission module is used to collect original data from multiple types of data sources and transmit them securely, ensuring that the data enters the system in real time, accurately and securely; the data aggregation and fusion module includes a multi-source data integration unit and a data fusion analysis unit. The multi-source data integration unit is used to clean, standardize and structure multivariate data to ensure data consistency and high quality; the data fusion analysis unit is used to classify, fuse and analyze the characteristics of the integrated data to ensure efficient extraction and utilization of data value; the standard management and interoperability module is used to formulate, manage and implement data standards and interface protocols to ensure that the modules and different data systems The data circulation and sharing module includes a data circulation management unit and a data utilization and operation unit. The data circulation management unit proposes a data circulation management algorithm based on blockchain technology to achieve tamper-proof recording, transparent supervision and full-process tracking of data during cross-domain data circulation, ensuring the security and credibility of data sharing and circulation. The data utilization and operation unit is used to commercialize and operate shared data to ensure the maximum utilization of data resources and realize value co-creation. The security and trust management module proposes a data privacy protection algorithm based on federated learning to implement distributed authentication, dynamic access control and cross-domain data circulation privacy protection to ensure the security, credibility and privacy protection of data throughout the entire process. The system operation and monitoring module is used to monitor and manage the system operation status and data flow in real time, and interact with users for information and control instructions to ensure stable and efficient operation of the system.

[0056] See Figure 1 ,Furthermore, the data collection and transmission module automatically ,collects data from various data sources and adopts encryption and ,distributed transmission protocols to ensure the real-time and integrity of data ,collection and the security of the transmission process.

[0057] See Figure 1 Furthermore, the data aggregation and fusion module includes a multi-source data integration unit. The multi-source data integration unit adopts standardized interfaces and automated data preprocessing technology to ensure that the original data from different sources and in different formats can be accurately collected, standardized and processed in the first time, and converted into a unified data format, thereby providing high-quality and structured basic data for subsequent deep data fusion.

[0058] See Figure 1Furthermore, the data aggregation and fusion module includes a data fusion analysis unit. The data fusion analysis unit ensures that the multi-source data after preliminary integration can effectively remove redundancy, fill in gaps, and realize the organic fusion of data in multiple dimensions by adopting data normalization, cleaning, intelligent matching and association rule analysis methods, and finally generates accurate and applicable data products, providing solid data support for decision support, data sharing and commercial value enhancement.

[0059] See Figure 1 Furthermore, the standard management and interoperability module ensures data interoperability and standard consistency between different platforms and systems by establishing unified data standards, metadata management and semantic conversion mechanisms.

[0060] See Figure 1 Furthermore, the data circulation and sharing module includes a data circulation management unit, which proposes a data circulation management algorithm based on blockchain technology. By deploying a distributed ledger and smart contract mechanism based on blockchain technology, it ensures that cross-domain data is fully recorded, verified in real time and cannot be tampered with during the circulation process, thereby ensuring the transparency and traceability of data transmission.

[0061] See Figure 1 Furthermore, the data circulation management algorithm based on blockchain technology is specifically as follows: First, the infrastructure of a trusted data space is built through the blockchain network to achieve tamper-proof records and transparent supervision of data circulation, create a distributed ledger in the Ethereum blockchain network, and define the data structure model. The specific formula is expressed as follows:

[0062]

[0063] in, Represented as a blockchain ledger, Expressed as blocks, Represented as a block number index, Represented as each independent block, Represented as a hash value of the block, used to uniquely identify the block and ensure integrity, It is represented as a set of transactions in a block, which records the specific operations of data circulation. It is represented as the timestamp of the block, which records the time when the block was generated, ensuring the consistency of the time sequence. By initializing the distributed ledger and deploying the smart contract, it provides the underlying rule constraints and automated execution mechanism for the subsequent data flow, ensuring the auditability of the whole process of data operation. Then, the smart contract function is designed. Its core functions include data rights confirmation, access policy management and transaction verification. The contract logic is implemented through Language implementation, key functions are expressed as:

[0064]

[0065] in, Represented as data metadata, Expressed as an attribute access policy, Represented as a transaction identifier, It is expressed as a contract interface. By building the underlying ledger architecture of the trusted data space, the smart contract function realizes the automatic execution of data operation rules, providing a trusted execution environment for the value co-creation system. Secondly, it combines symmetric and asymmetric encryption technologies to ensure the confidentiality and integrity of data in circulation, and uses IPFS to achieve decentralized storage, reduce the risk of single point failure, and perform pre-processed data. Perform two-stage encryption, the specific formula is expressed as:

[0066]

[0067]

[0068] in, Represented as symmetric encrypted ciphertext, Represented as raw data, Denotes Advanced Encryption Standard, Expressed as a symmetric encryption key, Represented as a 256-bit binary key space, Indicated as belonging to, Represented as asymmetric encrypted ciphertext, Expressed as an elliptic curve integrated encryption function, Represented as the public key of the data owner, Represented as the private key of the data owner, Represented as the base point of the elliptic curve, used in the symmetric encryption phase Algorithm on raw data Encryption, using the ECIES function for the symmetric key in the asymmetric encryption stage Encrypt the ciphertext Upload to the IPFS network and obtain the content identifier. The specific formula is:

[0069]

[0070] in, Represented as an IPFS content-addressed hash function,

[0071] Through layered encryption and distributed storage, we build the security sharing requirements in the whole process of data operation, ensure the immutability of data assets in the process of value co-creation, and then implement fine-grained access control based on blockchain to ensure that data is only used by authorized parties on demand. At the same time, we support dynamic update and audit traceability of policies and define attribute sets. , using logic gate combination calculation strategy expression, the specific formula is expressed as:

[0072]

[0073] in, Expressed as attribute access control policy, Represented as the logical AND operator, Represented as the user's characteristics, Represented as the logical exclusive OR operator, Represents the specific value of the attribute. Expressed as the total number of conditions in the policy, Represented as the index of the condition in the policy, the policy With ciphertext Binding is written into the blockchain through smart contracts. The specific formula is expressed as:

[0074]

[0075] in, Represented as a policy update transaction function, it is used to write the new access control policy into the transaction record of the blockchain. Represented as a smart contract update function, It is represented as the digital signature of the data owner, which is used to verify the legitimacy of the policy update. Then, it solves the key management problem in cross-domain data sharing, records the key distribution track through the blockchain, and realizes the full traceability of the key life cycle. Initiate a key request, trigger the smart contract to verify the attribute match, It is represented as the index of the number of data users. The specific formula is:

[0076]

[0077] in, Represented as a verification function, it is used to check whether the user attribute set satisfies the access control policy. Represented as a user The attribute set includes the user's identity, role, and permission information. It is expressed as a logical judgment symbol, and the session key is negotiated through the improved ECDH protocol. The specific formula is expressed as follows:

[0078]

[0079] in, Represented as a session key, used to encrypt communications between data users and data owners, Represented as a hash-based message authentication code function, used to generate a key-derived value, Represented as a data user Temporary private key for key negotiation, Represented as a random salt value, used to enhance the security of the key and prevent replay attacks. Encrypted and written into blockchain transactions

[0080]

[0081] in, Represented as a blockchain transaction record, used to store encrypted session keys, Represented as a smart contract function that writes the encryption key to the blockchain, Indicates that the user uses the data Public key The encryption function can enhance the audit capability of the trusted data space and support the trust transfer of cross-domain value co-creation through the on-chain evidence storage mechanism of key distribution.

[0082] See Figure 1 Furthermore, the data circulation and sharing module includes a data utilization and operation unit. The data utilization and operation unit ensures that all participants can efficiently and securely access data, develop applications and realize commercial operations by building a standardized data service interface and a flexible data development platform, thereby converting the value of data after circulation into practical application results.

[0083] See Figure 1 Furthermore, the security and trusted management module proposes a data privacy protection algorithm based on federated learning. By integrating differential privacy and secure multi-party computing technology based on federated learning technology, it ensures that data is not accessed or abused without authorization during cross-subject collaboration, and ensures the privacy security, trusted authentication and dynamic management of the system during data circulation.

[0084] See Figure 1 Furthermore, the data privacy protection algorithm based on federated learning is specifically as follows: First, a full-process data operation infrastructure based on a trusted data space is constructed to achieve secure interconnection between terminal devices, edge nodes and cloud servers, providing a decentralized collaborative environment for subsequent privacy protection algorithms. In the trusted data space, a consortium blockchain network composed of super edge nodes is deployed. Each SEN has edge computing, model aggregation and blockchain verification functions, and initializes the global federated model parameters. ,in , Represented as an initialized high-dimensional parameter vector, Expressed as model dimensions, Represented as a model identifier, generating initial identity credentials through a digital signature protocol ,in, Expressed as the total number of participating devices, Represented as the index of the number of participating devices, Represented as a device The encrypted signature is used for authentication of subsequent communications. The specific formula is expressed as:

[0085]

[0086] in, Represented as a function for randomly initializing model parameters, Expressed as a digital signature generation function, Represented as a device A unique identifier for Represented as a device The public key of It is represented as a string concatenation operation, and the model training is completed locally on the terminal device. The user's sensitive data is protected by differential privacy technology, ensuring that the data is available but invisible in the trusted data space. The terminal device Download the latest global model from the edge node , using local datasets Conduct training, Represented as the index of federated learning iterations, calculating local gradients ,in is the loss function, and the gradient is applied Norm clipping, the specific formula is expressed as:

[0087]

[0088] in, Represented as the clipped local gradient vector, Represented as a device The local gradient vector of Expressed as norm, Represented as the gradient clipping threshold, input Laplace noise , generate privacy-preserving gradients, Represented as the Laplacian noise vector injected into the local gradient, Expressed as a Laplace distribution, It is expressed as gradient sensitivity and privacy budget to control the strength of privacy protection. Secondly, local model updates are aggregated at the edge node layer, and blockchain technology is combined to achieve trusted verification and traceability of parameters, supporting the auditability of full-process data operations. Edge nodes receive privacy gradients uploaded by terminal devices. , verify the digital signature , ensure the legality of the data source, execute the federated average algorithm, the specific formula is expressed as:

[0089]

[0090]

[0091] in, Expressed as The global model parameters of the wheel, Denoted as the learning rate, Expressed as The update amount of the global model parameters in the iteration. Secondly, the federated learning model is used to generate personalized recommendations in real time at the edge layer, combining user behavior data with global hot trends to achieve value co-creation. The edge node loads the latest model , according to the user Local behavior data , calculate the interest vector :

[0092]

[0093] in, It is expressed as a model inference function. Then, the global hotspot vector is read from the blockchain and fused to generate a hybrid recommendation score. The specific formula is expressed as:

[0094]

[0095] in, Represented as a user Recommendation score for hot information, Denoted as personalized weight, Represented as a user The interest vector, It is represented as a global hotspot vector. Finally, the model parameters and recommendation results are permanently stored in the blockchain, supporting multi-party data value sharing and auditing, and building a sustainable operation ecosystem. The specific formula is expressed as:

[0096]

[0097] in, Expressed as The blockchain data blocks generated during the round of federated learning, Represented as the set of feedback vectors of all users, Expressed as The hash value of the blockchain data block.

[0098] See Figure 1 Furthermore, the system operation and monitoring module monitors the system operation status, data flow and log records in real time, and interacts with users on information and control instructions to ensure that the platform continues to operate efficiently and stably and can detect and handle abnormal situations in a timely manner.

[0099] In specific use, first, the data collection and transmission module is used to collect original data from multiple data sources and transmit it securely to ensure that the data enters the system in real time, accurately and securely; secondly, the data aggregation and fusion module includes a multi-source data integration unit and a data fusion analysis unit. The multi-source data integration unit is used to clean, standardize and structure multivariate data to ensure data consistency and high quality. The data fusion analysis unit is used to classify, fuse and analyze the characteristics of the integrated data to ensure efficient extraction and utilization of data value; then, the standard management and interoperability module is used to formulate, manage and implement data standards and interface protocols to ensure interconnection, interoperability and compatibility between modules and different data systems; secondly, the data circulation and sharing module includes a data circulation management unit and a data utilization and operation unit. The circulation management unit proposes a data circulation management algorithm based on blockchain technology, which is used to achieve tamper-proof recording, transparent supervision and full-process tracking of data during cross-domain data circulation, ensuring the security and credibility of data sharing and circulation. The data utilization and operation unit is used to commercialize and operate shared data to ensure the maximum utilization of data resources and realize value co-creation. Then, the security and trust management module proposes a data privacy protection algorithm based on federated learning, which is used to implement distributed authentication, dynamic access control and privacy protection of data cross-domain circulation, ensuring the security, credibility and privacy protection of data throughout the entire process. Finally, the system operation and monitoring module is used to monitor and manage the system operation status and data flow in real time, and interact with users for information and control instructions to ensure stable and efficient operation of the system.

[0100] Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A full-process data operation and value co-creation system based on a trusted data space, including a data acquisition and transmission module, a data aggregation and fusion module, a standard management and interoperability module, a data circulation and sharing module, a security and trust management module, and a system operation and monitoring module, characterized by: The data collection and transmission module is used to collect raw data from multiple data sources and transmit it securely; the data aggregation and fusion module is used to clean, standardize, structure and analyze multivariate data; the standard management and interoperability module is used to formulate data standards and interface protocols to ensure interconnection and interoperability; the data circulation and sharing module includes a data circulation management unit and a data utilization and operation unit. The data circulation management unit proposes a data circulation management algorithm based on blockchain technology to achieve tamper-proof recording, transparent supervision and full-process tracking of cross-domain data circulation. The data utilization and operation unit is used to commercialize and co-create value for shared data; the security and trust management module proposes a data privacy protection algorithm based on federated learning to implement distributed authentication , dynamic access control and privacy protection; the system operation and monitoring module is used to monitor the system operation status and user interaction in real time; the data circulation management algorithm based on blockchain technology builds a trusted data space according to the Ethereum blockchain, and uses distributed ledgers, smart contracts Solidity, and IPFS two-stage encryption AES256 symmetric encryption data and ECIES asymmetric encryption keys, adopts a logic gate combination calculation strategy, and generates session keys through the improved ECDH protocol to achieve full traceability of cross-domain key distribution; the data privacy protection algorithm based on federated learning builds a terminal-edge-cloud architecture, deploys a consortium blockchain network, and realizes differential privacy protection through gradient clipping, Laplace noise injection, and blockchain evidence storage, and integrates interest vectors to generate hybrid recommendation scores.

2. The full-process data operation and value co-creation system based on a trusted data space as described in claim 1 is characterized in that: The data circulation management unit proposes a data circulation management algorithm based on blockchain technology, which is used to achieve tamper-proof recording, transparent supervision and full-process tracking of data during cross-domain data circulation; the security and trust management module proposes a data privacy protection algorithm based on federated learning, which is used to implement distributed authentication, dynamic access control and cross-domain data circulation privacy protection; the system operation and monitoring module is used to monitor and manage the system operation status and data flow in real time, and interact with users for information and control instructions; the data circulation and sharing module includes a data circulation management unit and a data utilization and operation unit. The data circulation management unit proposes a data circulation management algorithm based on blockchain technology, by deploying a distributed ledger and smart contract mechanism based on blockchain technology; the data utilization and operation unit builds a standardized data service interface and a flexible data development platform.

3. The full-process data operation and value co-creation system based on a trusted data space as described in claim 2 is characterized in that: The blockchain-based data flow management algorithm builds a trusted data space based on the Ethereum blockchain. By defining a distributed ledger, each block ensures the immutability and transparent supervision of data flow records. Solidity smart contracts are deployed to implement data ownership confirmation, access policy management, and transaction verification. Two-stage encryption is used on IPFS, first using AES-256 to generate symmetric encrypted ciphertext, and then encrypting the symmetric key with ECIES to generate asymmetric encrypted ciphertext. Fine-grained access policies are calculated using logic gates based on attribute sets and written to the chain through smart contract update functions. After the data requester and the data user trigger the verification function, the improved ECDH protocol is used to generate the session key, and the evidence is stored through the smart contract function.

4. The full-process data operation and value co-creation system based on a trusted data space according to claim 3 is characterized by: The data collection and transmission module automatically collects data from various data sources and adopts encryption and distributed transmission protocols.

5. The full-process data operation and value co-creation system based on a trusted data space according to claim 4 is characterized by: The data aggregation and fusion module includes a multi-source data integration unit and a data fusion analysis unit. The multi-source data integration unit adopts standardized interfaces and automated data preprocessing technology; the data fusion analysis unit adopts data normalization, cleaning, intelligent matching and association rule analysis methods.

6. The full-process data operation and value co-creation system based on a trusted data space according to claim 5 is characterized by: The standard management and interoperability module establishes a unified data standard, metadata management and semantic conversion mechanism.

7. The full-process data operation and value co-creation system based on a trusted data space according to claim 6 is characterized by: The security and trust management module proposes a data privacy protection algorithm based on federated learning, which integrates differential privacy and secure multi-party computing technology based on federated learning technology.

8. The full-process data operation and value co-creation system based on a trusted data space according to claim 7 is characterized in that: The data privacy protection algorithm based on federated learning first builds a full-process secure interconnection architecture from terminal to edge to cloud in a trusted data space, and deploys a consortium blockchain network consisting of multiple super edge nodes (SENs). Each SEN combines edge computing, model aggregation and blockchain verification functions. The system initializes global model parameters and generates identity credentials for each device through a digital signature protocol. Subsequently, each terminal device downloads the latest global model, trains it using a local dataset, and applies norm clipping and Laplace noise injection to the calculated gradient to obtain a local gradient that satisfies differential privacy. The edge node receives and verifies the signature and privacy gradient uploaded by the device, and performs secure federated averaging to update the global parameters. On this basis, the edge layer loads the latest model in real time, generates personalized interest vectors for users, and combines them with the global hotspot vectors stored in the blockchain to achieve a hybrid recommendation score. Finally, the updated model parameters, recommendation results and user feedback are permanently uploaded to the chain, forming an auditable and traceable multi-party value sharing ecosystem.

9. The full-process data operation and value co-creation system based on a trusted data space according to claim 8 is characterized by: The system operation and monitoring module monitors the system operation status, data flow and log records in real time, and interacts with users for information and control instructions.

Images