Distributed industrial data space system, data publishing method, and data retrieval method

Abstract

The application discloses a kind of distributed industrial data space systems, data publishing method and data retrieval method, in the system, decentralized governance module is used to: through the independent parallel review of the data product submitted by the data provider node of screening audit candidate node, after approving data product release, the index abstract of data product is written into block chain;Federal search and verification module is used to: receive query request with query credentials, obtain candidate local data space set in global metadata index module after query credentials, and distribute local search request in parallel;Local data space index module is used to: after receiving local search request in each local data space, search result abstract is generated by local search through data index;Federal search and verification module is also used to: after passing through credibility verification, search result abstract is sent to data user.The application can realize decentralization and efficient search.

Description

Technical Field

[0001] The embodiments of the present invention relate to the fields of distributed and blockchain technologies, and in particular to a distributed industrial data space system, a data publishing method, and a data retrieval method. Background Technology

[0002] In typical industrial chains such as steel, chemicals, new materials, and equipment manufacturing, there is a large demand for cross-institutional and cross-regional data collaboration among upstream mining and raw material enterprises, midstream component manufacturing enterprises, and downstream equipment manufacturing enterprises. This includes data on quality testing, production process, materials research and development testing, materials performance, and supply chain status.

[0003] However, current industrial data sharing models are still dominated by centralized platforms: data publication requires manual review by the platform operator, retrieval relies on single-point indexing services, and data credibility depends on the endorsement of a central node. The overall process suffers from low review efficiency, difficulty in quickly discovering data, and high risk of unreliability at single platform points. With the development of the data space concept, the industry is gradually exploring ways to describe industrial data using standardized structures and share it in a strategy-controlled manner. However, in practical applications, problems remain, such as insufficient trust among multiple entities, reliance on centralized institutions for data product review, and difficulties in efficient data retrieval in distributed environments.

[0004] Therefore, there is a need for a new industrial data space architecture that is decentralized, does not rely on a single platform, is verifiable, auditable, traceable, and efficiently searchable.

[0005] The background section described above is merely a description made by the inventor based on his understanding, and the above content should not be regarded as evidence of prior art disclosed before the filing date of this application. Summary of the Invention

[0006] This invention provides a distributed industrial data space system to address the problems of insufficient trust among multiple entities, reliance on centralized institutions for data product review, and difficulty in efficient data retrieval in a distributed environment. The system includes:

[0007] Decentralized governance module, global metadata index module, local data space index module, federated search and verification module;

[0008] The decentralized governance module is used to: select multiple audit candidate nodes from the audit candidate node pool; independently and in parallel audit the data products submitted by the data provider nodes through the selected audit candidate nodes; and write the index digest of the data product into the blockchain after approving the release of the data product.

[0009] The federated retrieval and verification module is used to: receive query requests sent by data user nodes, the query requests carrying query credentials, obtain a set of candidate local data spaces in the global metadata index module through the query credentials, and distribute local retrieval requests in parallel to the set of candidate local data spaces;

[0010] The global metadata index module is used to: store index summaries of blockchain-de-identified data products; and determine a set of candidate local data spaces based on the received query credentials and the index summaries.

[0011] The local data space indexing module is used for: each data provider node to build a data index for the data product in the local data space; after each local data space receives a local retrieval request, it performs a local retrieval through the data index to generate a retrieval result summary and corresponding digital signature, and sends it to the federated query coordination module.

[0012] The Federated Search and Verification module is also used to send the search result summary to the data user after verifying the credibility of the digital signature of the received search result summary.

[0013] Another aspect of the present invention provides a data publishing method based on a distributed industrial data spatial system to address the problems of insufficient trust among multiple entities, reliance on centralized institutions for data product review, and difficulty in efficient data retrieval in a distributed environment. The method includes:

[0014] The decentralized governance module selects multiple audit candidate nodes from the pool of audit candidate nodes;

[0015] After receiving the data products submitted by the provider data nodes, the decentralized governance module independently and in parallel reviews the data products through the selected review candidate nodes;

[0016] After approving the release of the data product, the decentralized governance module writes the index digest of the data product into the blockchain;

[0017] The global metadata index module stores index summaries of blockchain-anonymized data products.

[0018] Another aspect of the present invention provides a data publishing method based on a distributed industrial data spatial system to address the problems of insufficient trust among multiple entities, reliance on centralized institutions for data product review, and difficulty in efficient data retrieval in a distributed environment. The method includes:

[0019] The Federated Search and Verification module receives query requests sent by data user nodes, the query requests carrying query credentials;

[0020] The global metadata indexing module determines the candidate local data space set based on the query credentials and the index summary of the data product after blockchain de-identification;

[0021] The federated search and validation module distributes local search requests in parallel to the candidate local data space set;

[0022] After receiving a local retrieval request in each local data space, a local retrieval is performed through the data index to generate a retrieval result summary and a corresponding digital signature;

[0023] After the Federated Search and Verification module verifies the credibility of the digital signature of the search result summary, it sends the search result summary to the data user.

[0024] This invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-described data publishing method or data retrieval method based on a distributed industrial data space system.

[0025] This invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the aforementioned data publishing method or data retrieval method based on a distributed industrial data space system.

[0026] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the aforementioned data publishing method or data retrieval method based on a distributed industrial data space system.

[0027] In this embodiment of the invention, the decentralized governance module is used to: select multiple audit candidate nodes from the audit candidate node pool; independently and in parallel audit the data products submitted by the data provider nodes through the selected audit candidate nodes; and write the index digest of the data product into the blockchain after approving the release of the data product; the federated retrieval and verification module is used to: receive query requests sent by data user nodes, the query requests carrying query credentials, and obtain a set of candidate local data spaces in the global metadata index module through the query credentials; and distribute local retrieval requests in parallel to the set of candidate local data spaces; the global metadata index module is used to: store the index digest of the data product after blockchain desensitization; and determine the set of candidate local data spaces based on the received query credentials and the index digest; the local data space index module is used to: have each data provider node construct a data index for the data product in its local data space; and after each local data space receives a local retrieval request, perform a local retrieval through the data index to generate a retrieval result digest and a corresponding digital signature, and send it to the federated query coordination module; the federated retrieval and verification module is also used to: after verifying the credibility of the digital signature of the received retrieval result digest, send the retrieval result digest to the data user. Compared to centralized platform-based data sharing solutions, decentralized governance modules enable decentralized parallel review of data product releases, avoiding single points of failure, low efficiency, and trust dependency issues inherent in centralized review. Furthermore, approved data product index summaries are written to the blockchain, ensuring their immutability and traceability. A global metadata index module stores anonymized index summaries, supporting rapid location of candidate local data space sets based on query credentials while protecting data privacy, enabling efficient global retrieval routing. A local data space index module builds data indexes and performs local searches on each data provider node, eliminating the need to upload or aggregate raw data, ensuring data providers' ownership and control of their data, while improving retrieval efficiency and reducing network transmission pressure. A federated retrieval and verification module uniformly receives query requests, distributes retrieval tasks in parallel, and verifies the digital signature credibility of retrieval result summaries, achieving cross-node federated retrieval and result verification, enhancing the overall system's security and credibility while maintaining retrieval efficiency. The overall system achieves decentralized governance, global index routing, local data not leaving the domain, federated parallel retrieval, and verifiable results, making it suitable for cross-entity, high-security, high-efficiency, and highly reliable data sharing and retrieval scenarios in industrial data spaces. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0029] Figure 1 This is a schematic diagram of a distributed industrial data space system in an embodiment of the present invention;

[0030] Figure 2 This is a diagram illustrating the overall architecture of the distributed industrial data space system in this embodiment of the invention.

[0031] Figure 3 This is a practical application example diagram of the distributed industrial data space system in this embodiment of the invention;

[0032] Figure 4 This is a flowchart of a data publishing method based on a distributed industrial data spatial system in an embodiment of the present invention;

[0033] Figure 5 This is an interaction diagram of data publishing based on a distributed industrial data spatial system in an embodiment of the present invention;

[0034] Figure 6 This is a flowchart of a data retrieval method based on a distributed industrial data spatial system in an embodiment of the present invention;

[0035] Figure 7 This is an interactive diagram of data retrieval based on a distributed industrial data spatial system in an embodiment of the present invention;

[0036] Figure 8 This is a schematic diagram of a computer device in an embodiment of the present invention. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0038] As mentioned earlier, existing technologies suffer from problems such as insufficient trust among multiple entities, reliance on centralized institutions for data product auditing, and difficulty in efficient data retrieval in distributed environments. To address these issues, the inventors discovered that decentralized identity (DID) can provide companies in the industry chain with verifiable identities that do not rely on a central institution. Verifiable credentials (VC) can serve as endorsements of their qualifications and capabilities, while blockchain can record key actions such as auditing, publishing, authorization, and access as tamper-proof audit credentials. Furthermore, technologies such as distributed hash table (DHT) indexes, local inverted indexes, and federated query coordination modules (FQC) can support more efficient cross-node data discovery capabilities in distributed data spaces.

[0039] Therefore, this invention proposes a new industrial trusted data space architecture that combines the standardized circulation mechanism of data space with identity authentication and the trust system of blockchain, and achieves efficient data retrieval through distributed indexing to solve the above problems.

[0040] Figure 1 This is a schematic diagram of a distributed industrial data space system in an embodiment of the present invention, such as... Figure 1 As shown, the system includes: a decentralized governance module 101, a global metadata index module 102, a local data space index module 103, and a federated retrieval and verification module 104;

[0041] The decentralized governance module 101 is used to: select multiple audit candidate nodes from the audit candidate node pool; independently and in parallel audit the data products submitted by the data provider nodes through the selected audit candidate nodes; and write the index digest of the data product into the blockchain after approving the release of the data product.

[0042] The federated retrieval and verification module 104 is used to: receive query requests sent by data user nodes, the query requests carrying query credentials, obtain a set of candidate local data spaces in the global metadata index module through the query credentials, and distribute local retrieval requests in parallel to the set of candidate local data spaces;

[0043] The global metadata index module 102 is used to: store index summaries of blockchain-de-identified data products; and determine a candidate local data space set based on the received query credentials and the index summaries.

[0044] The local data space index module 103 is used for: constructing a data index for data products in the local data space by each data provider node; and after receiving a local retrieval request in each local data space, performing a local retrieval through the data index to generate a retrieval result summary and corresponding digital signature, and sending it to the federated query coordination module.

[0045] The Federated Search and Verification Module 104 is also used to: send the search result summary to the data user after the digital signature of the received search result summary has been verified for trustworthiness.

[0046] Depend on Figure 1 As shown, compared with existing data sharing solutions based on centralized platforms, the decentralized governance module enables decentralized parallel review of data product releases, avoiding single points of failure, low review efficiency, and trust dependency issues associated with centralized review. Simultaneously, the index digest of approved data products is written to the blockchain, ensuring its immutability and traceability. The global metadata index module stores anonymized index digests, supporting rapid location of candidate local data space sets based on query credentials while protecting data privacy, achieving efficient global retrieval routing. The local data space index module builds data indexes and performs local searches on each data provider node, eliminating the need to upload or aggregate raw data, ensuring data providers' ownership and control of the data, while improving retrieval efficiency and reducing network transmission pressure. The federated retrieval and verification module uniformly receives query requests, distributes retrieval tasks in parallel, and verifies the digital signature credibility of retrieval result digests, achieving cross-node federated retrieval and result credibility verification. This ensures retrieval efficiency while enhancing the security and credibility of the entire system. The overall system achieves decentralized governance, global index routing, local data not leaving the domain, federated parallel retrieval, and verifiable results, making it suitable for cross-entity, high-security, high-efficiency, and highly reliable data sharing and retrieval scenarios in industrial data spaces.

[0047] In this embodiment, the system further includes a trust and identity infrastructure module 105, which includes:

[0048] The distributed identity module is used to authenticate the identities of all participating nodes accessing the distributed industrial data space system and obtain identity authentication identifiers. The participating nodes include audit candidate nodes, data provider nodes, and data user nodes.

[0049] The verifiable credentials module is used to generate verifiable credentials based on the qualifications submitted by participating nodes and store the hash value of the verifiable credentials on the blockchain.

[0050] In the above embodiments, a hash calculation can be performed on the query credential to generate a verifiable credential.

[0051] In this embodiment, the decentralized governance module includes:

[0052] The decentralized review committee module is used to select multiple review candidate nodes from the pool of review candidate nodes to form a decentralized review committee through smart contracts;

[0053] The smart contract review module is used to distribute data product release applications submitted by data provider nodes to each candidate node of the decentralized review committee after receiving them. This allows each candidate node to independently and in parallel review and digitally sign the data products in the data product release applications. The module also receives the digital signatures of the data products from all candidate nodes, executes a signature aggregation consensus strategy, and determines whether to approve the data product release. If approved, the module writes the index digest and digital signature set of the data product into the blockchain.

[0054] In this embodiment of the invention, the decentralized review committee (DAC) module functions to automatically and randomly select multiple nodes from the space to form a temporary review group through smart contracts.

[0055] The role of the smart contract auditing module is to execute auditing logic and collect digital signatures. Data products can only be approved for release when N-of-M consensus (e.g., 3 out of 5 auditing candidate nodes pass) is met. This eliminates the inefficiency and single-point rent-seeking risk of centralized platform manual auditing, ensuring the objectivity of data entry.

[0056] In this embodiment, the decentralized review committee module is specifically used for:

[0057] Extract the core metrics for each candidate node in the candidate node pool for review;

[0058] Based on the core indicators, the comprehensive weight of each review candidate node is calculated using a weighted normalization function, and the comprehensive weight is then globally normalized to obtain the normalized weight.

[0059] Generate a randomness seed that can be verified across the entire network based on the private key;

[0060] Each audit candidate node generates a verifiable random number based on a random seed and an identity authentication identifier;

[0061] The selection threshold is calculated based on the number of candidate nodes in the candidate node pool, the target size, and the normalized weight.

[0062] For each audit candidate node, if the verifiable random number of the audit candidate node is less than the selection threshold, the audit candidate node is added to the initial selection node set.

[0063] If the number of initially selected nodes exceeds the target size, the top target number of candidate nodes will be selected based on verifiable random numbers from smallest to largest to form a decentralized review committee.

[0064] In this embodiment, the core metrics include node reputation value, data activity, and relative network latency.

[0065] The pool of candidate nodes for review can be represented as N = {n1, n2, ..., n} M};

[0066] Each audit candidate node n i With comprehensive weight: w i =f(r i ,a i ,l i ) ;

[0067] Where, r i Let a be the node reputation value. i For data activity, l i For relative network latency, l i = RTT i (Round Trip Time) represents the time from the data provider node to the candidate review node n. i Then return the required round-trip time, RTT max This represents the maximum round-trip time for all candidate review nodes.

[0068] And f(·) is a weighted normalization function, for example: w i =α×r i +β×a i +γ×(1-l i );

[0069] And normalize it:

[0070]

[0071] in, For normalized weights;

[0072] The formula for generating random seeds is:

[0073]

[0074] in, sk For private key, t Provides network-wide verifiable randomness for timestamps or block heights, preventing node cheating.

[0075] Each candidate node i generates a verifiable random number based on a randomness seed and an identity authentication identifier. The formula is:

[0076]

[0077] Among them, nodes Based on a random seed and its identity authentication identifier DID i Generate verifiable random numbers.

[0078] node The criteria for being selected for the centralized review committee are:

[0079]

[0080] The higher the normalized weight, the larger the selectable threshold, the easier it is to meet the conditions, and the higher the probability of selection. Candidate nodes with low normalized weights can still be randomly selected, preventing them from being permanently excluded.

[0081] in:

[0082]

[0083] Where K is the target size, and each candidate node is selected based on its normalized weight ratio and the selection ratio.

[0084] The initial set of selected nodes can be represented as: .

[0085] if Then press again Take the top K from smallest to largest to obtain the decentralized review committee.

[0086] In this embodiment, the global metadata indexing module is specifically used for:

[0087] The distributed hash table (DHT) network is used to store the index digest of the data product after the index digest of the data product has been de-identified by the blockchain, and it is bound to the local data space identifier of the data provider to which the data product belongs.

[0088] In this embodiment, the local data space index module is specifically used to: after each local data space receives a local retrieval request, after the query credentials are verified in the local data space, perform a local retrieval in the local data space through the data index, generate a retrieval result summary and perform a digital signature on the local data space after the retrieval is completed, and send the retrieval result summary and the corresponding digital signature to the federated query coordination module.

[0089] The global metadata index module does not store the raw data, but only records "what data is where".

[0090] In one embodiment, the local data space indexing module is specifically used for:

[0091] The hash value of the query credential in the retrieval request is compared with the hash value of the verifiable credential stored on the blockchain by the data user node that sent the query credential.

[0092] After the consistency verification is passed, the public key of the issuer of the query credential is obtained from the blockchain. The public key is used to verify the digital signature of the proof field in the query credential to verify whether the digital signature was legally signed by the issuer's private key.

[0093] If so, check the permission content of the maximum access range field in the voucher body of the query voucher to verify whether the maximum access range matches the data product access permission policy preset locally by the data provider node;

[0094] If so, check the expiration time field value in the voucher body of the query voucher to verify whether the retrieval request was initiated after the expiration date.

[0095] If the voucher has not expired, the verification is confirmed to be successful; otherwise, the verification is confirmed to be unsuccessful.

[0096] In this embodiment, the data index is used to establish the relationship between the metadata of the data product and the index of the data product;

[0097] The local data spatial index module is specifically used for:

[0098] Parse the query credentials to obtain the search filters;

[0099] Based on the search filter, match the data index type in the local data space;

[0100] Based on the matching data index type, the search filter items are matched with the metadata of the data products in the data index corresponding to the data index type to filter the target data products;

[0101] Generate a summary of search results based on the matched target data products;

[0102] The search result digest is digitally signed using the private key of the data provider node corresponding to the local data space.

[0103] In this embodiment of the invention, the local data space indexing module ensures both global visibility of the search and prevents the transfer of ownership and storage rights of the original data products, thus guaranteeing industrial data security. When matching data index types in the local data space based on the search filters, if the search filters include keyword Bloom filters, allowed data categories, or other keyword-based conditions, the data index type is determined to be a local inverted index; if the filters include feature-based conditions such as vector summaries, the data index type is determined to be a local vector index; if both types of conditions are included, the data index type is determined to be a combination of inverted index and vector index, achieving multi-dimensional accurate matching.

[0104] For inverted index retrieval: Based on keywords, data categories, and other conditions, quickly locate the metadata of the data products associated in the data index and map it to the storage location of the corresponding target data product;

[0105] Vector index retrieval: Based on vector summaries, feature similarity is calculated, and the metadata of data products with similarity reaching a threshold is filtered out and mapped to the storage location of the corresponding target data product;

[0106] During the retrieval process, only the metadata and index of the data product are accessed, without directly retrieving the target data product itself, thus ensuring the security of the target data product.

[0107] The search result summary only contains non-sensitive core information of the data product (such as data product identifier, matching degree, data type, update time, etc.), and does not contain sensitive information in the original data product or metadata, ensuring that the core data products of the data provider node are not leaked.

[0108] After digitally signing the search result summary, the digital signature is bound to the search result summary to form a signed version of the search result summary, ensuring that the results are tamper-proof and traceable, and providing a basis for subsequent credibility verification.

[0109] In this embodiment, the local data space indexing module is also used for:

[0110] If the generation of the search result summary exceeds the first time limit, the timeout placeholder information is sent back to the federated search and verification module. If only part of the target data products are matched, the search result summary of part of the target data products and the corresponding digital signature are sent back to the federated search and verification module.

[0111] In this embodiment, the federated search and verification module includes:

[0112] The Federated Query Coordination (FQC) module receives query requests from data user nodes, parses the query credentials in the query requests to obtain core search conditions, converts them into search filter items, and sends them to the Global Metadata Index module. It then obtains the candidate local data space set returned by the Global Metadata Index module, distributes local search requests to the candidate local data space set, and sends the search result summary and corresponding digital signature to the search response verification module after receiving them.

[0113] The retrieval response verification module is used to verify whether the digital signature of the retrieval result summary is trustworthy. If it is trustworthy, the retrieval result summary is sent to the data user, and the summary of the retrieval session corresponding to the local retrieval request is written to the blockchain.

[0114] The retrieval response verification module ensures that the retrieval process is "traceable and non-repudiable," meaning that who searched for what, when, and what results were obtained are all verifiable on the blockchain.

[0115] In this embodiment, the global metadata indexing module is specifically used to: perform distributed matching between the search filter items and the index digests of the de-identified data products to obtain multiple matching index digests and corresponding matching scores; locate the local data space of the data provider node to which each matching index digest belongs as a candidate local data space and add it to the candidate local data space set; and send the candidate local data space set and the matching scores corresponding to each candidate local data space to the federated query coordination module.

[0116] The Federated Query Coordination module is also used for:

[0117] After receiving the candidate local data space set, the candidate local data spaces in the candidate local data space set are prioritized according to the matching score corresponding to the candidate local data space, and local retrieval requests are distributed to the candidate local data spaces according to the parallelism of the calculation.

[0118] In this embodiment, the federated query coordination module is specifically used for:

[0119] The size of the candidate set is calculated based on the size of the candidate local data space set and the expected hit redundancy value.

[0120] The weighted score of each candidate local data space is calculated based on the matching score of each candidate local data space, the reputation of the data provider node corresponding to the candidate local data space, and the relative network latency.

[0121] The candidate local data space set is selected from the candidate local data space set in descending order of weighted score;

[0122] The smaller value between the candidate local data space set and the maximum parallelism is determined as the parallelism.

[0123] When the size of the candidate set is greater than the parallelism, the candidate local data space set is divided into multiple batches according to the parallelism based on the weighted score from high to low, and then the local retrieval requests of the candidate local data space are distributed batch by batch.

[0124] In this embodiment of the invention, the candidate set size is m. ( ),in This represents the number of candidate local data spaces in the candidate local data space set. For the expected hit number, (e.g., 2.0) is the redundancy factor.

[0125] Weighted score of candidate local data space j The calculation formula is as follows:

[0126]

[0127] The matching score for candidate local data space j;

[0128] The node reputation value of candidate local data space j

[0129] The relative network latency (normalized value) of candidate local data space j.

[0130] , and These are the weighting coefficients, and their sum is 1.

[0131] Parallelism ( ), maximum parallelism Configurable (e.g., 16).

[0132] In this embodiment, the federated query coordination module is also used for:

[0133] For candidate local data spaces that have not received a search result summary within the second time period and whose number of retries is less than the retry threshold, the local search request is resent, and the waiting time for each retry is determined by a backoff strategy.

[0134] Once the search result summary for the fast completion threshold has been received, the sending of subsequent local search requests is terminated early.

[0135] The second duration, T_net, is for example, 800 ms, which is the network timeout duration.

[0136] In the retreat strategy, the waiting time is in To avoid the length of time, The initial retreat time, The retreat factor is r, where r is the number of rounds in which the opponent fails.

[0137] There are two ways to quickly determine the value of the threshold K_fin: one is a fixed value (e.g., the search will stop after returning 10 search result summaries), and the other is to change it dynamically according to the query type (e.g., only 1 for exact queries and 20 for fuzzy queries).

[0138] In this embodiment of the invention, a structured example of query credentials is as follows:

[0139] {

[0140] "Context": ["https: / / www.w3.org / 2018 / credentials / v1"],

[0141] "Type": ["VerifiableCredential", "QueryVC"],

[0142] "Unique identifier": "vc:query:0x93ab12ff...",

[0143] "Issuing Party": "did:mat:Authority001",

[0144] "Data User": "did:mat:RequesterA",

[0145] "Certificate Subject": {

[0146] "Query ID": "q-20251126-01823",

[0147] "Query pattern hash": "0xabc123...",

[0148] "Allowed data categories to query": ["material_properties", "steel_grade", "component_trace"],

[0149] "Keyword Bloom filter": "0x88bf...e712",

[0150] Vector digest: "0x39adfe...11",

[0151] Maximum number of results returned: 50,

[0152] "Maximum access range": "supplier-chain-level-2",

[0153] "Query Purpose": "performance-evaluation",

[0154] Expiration Date: 2025-12-31T23:59:59Z

[0155] },

[0156] "Proof domain": {

[0157] "Signature Algorithm Type": "Ed25519Signature2020",

[0158] "Issuance Time": "2025-11-26T10:12:00Z",

[0159] "Verification method": "did:mat:Authority001#key-1",

[0160] "Proof Purpose": "assertionMethod",

[0161] "Digital signature string": "eyJ..."

[0162] }

[0163] }

[0164] The following is a specific example illustrating the data flow of a distributed industrial data spatial system. Figure 2 This is a diagram illustrating the overall architecture of the distributed industrial data space system in this embodiment of the invention. Figure 3 This is a practical application example diagram of the distributed industrial data space system in this embodiment of the invention. This embodiment takes the steel industry chain as the application scenario and involves participating entities such as steel production enterprises (data users), iron ore raw material suppliers and alloy material enterprises (data providers), and industry review nodes (review candidate nodes).

[0165] Figure 2 In this distributed industrial data spatial system, the various modules can be divided into multiple layers. The local data space layer is the "source" and "territory" of the data. Each local data space is an independent data unit, and a local inverted index or vector index serves as the data index for fast local retrieval of data content. Metadata describes the attribute information of the data product (such as data type, source, format, etc.). Local data products are the raw data products for each data provider node and can be selected for use.

[0166] The blockchain layer provides a trusted infrastructure, with key functions including: DID identity authentication management, VC management, and smart contract management. Records serve as immutable evidence of critical operations (such as data access and transactions). Trustworthiness verification is based on digital signatures to verify the trustworthiness of search result summaries. Search session auditing audits and traces cross-space data retrieval sessions to ensure compliance.

[0167] The governance layer corresponds to the decentralized governance module, which implements decentralized governance, including screening and approving candidate nodes and data products, and executing signature aggregation consensus strategies. The DHT network is a distributed hash table network used to store and maintain a global index digest, allowing the entire network to quickly locate the local data space where the data is located, achieving efficient distributed indexing and routing.

[0168] The service platform layer is the core layer that provides services to the outside world, and it contains four core modules:

[0169] The Trust and Identity Foundation module includes a distributed identity module and a verifiable credential module. The global metadata index module stores global metadata information and, based on query conditions, filters a set of local data spaces that may contain the target data from the global metadata. The Federated Retrieval and Verification module coordinates query tasks across multiple local data spaces using FQC (Fulfilled Quality Control) to achieve federated retrieval and avoid data centralization. The Retrieval Response Verification module verifies the credibility of the retrieval results returned by each local data space. The Local Data Space Index module constructs a data index and performs local retrieval.

[0170] In this embodiment, corresponding Figure 3 The trust and identity foundation module is anchored by an industry authority, which completes distributed identity authentication (identity authentication identifier DID) and issues verifiable credentials (VC) for all participating nodes; the blockchain is responsible for storing the DID, VC hash value, index digest of data products, and retrieval session digest of all participating nodes.

[0171] Decentralized governance module: The pool of candidate nodes for review consists of steel industry association nodes, enterprise nodes, etc., and a decentralized review committee (DAC) is automatically formed through smart contracts to be responsible for reviewing the release of data products.

[0172] The global metadata index module is based on the DHT distributed hash table network and stores index summaries of data products such as iron ore and alloy materials after being de-identified by the blockchain.

[0173] Local data space indexing module: Each iron ore supplier (Company 1, Company 2) and alloy material supplier (Company 3, Company 4) builds an inverted index + vector index in its own local data space to manage its own raw material performance data.

[0174] Federated Search and Validation Module: Includes the Federated Query Coordination Module (FQC) and the Search Response Validation Module. It serves as a unified entry point for search requests and a hub for result verification, connecting with data users, the DHT network, and the local data spaces of various data providers.

[0175] The data user is a large steel production enterprise (client), which needs to retrieve data on iron ore and alloy materials that meet specific performance indicators to provide data support for the optimization of steelmaking processes.

[0176] The data providers include Company 1 (iron ore supplier in Region A), Company 2 (iron ore supplier in Region B), Company 3 (alloy material supplier in Region C), and Company 4 (alloy material supplier in Region A), all of whom are upstream data holders in the steel industry chain.

[0177] Step 1: Full-node identity authentication and credential issuance (Trust and identity infrastructure module)

[0178] All participating nodes first connect to the distributed industrial data space system to complete identity authentication and VC issuance, laying a trusted foundation for all subsequent operations:

[0179] Steel producers, raw material suppliers, and audit candidate nodes all submit industry qualifications and enterprise information to the trust anchor to complete DID distributed identity authentication and obtain a unique identity identifier (e.g., the identity authentication identifier DID of steel enterprise: did:steel:Producer001, and the identity authentication identifier DID of enterprise 1: did:iron:Supplier001).

[0180] Based on their own steelmaking process development and retrieval needs, steel production enterprises apply to the Trust and Identity Foundation Module for Query VC. After verifying their identity and retrieval permissions, the Trust and Identity Foundation Module issues a Query VC (generated in the structured format of the document) containing the retrieval scope, validity period and retrieval conditions, and writes the Query VC hash value into the blockchain to generate a verifiable credential.

[0181] The identity information and VC of each data provider node and audit candidate node are stored on the blockchain, enabling the full node identity to be traceable and verifiable.

[0182] Step 2: Data Product Release and Decentralized Review (Decentralized Governance Module + Global Metadata Index Module)

[0183] Each raw material supplier (data provider node) encapsulates its own raw material performance data into a data product. After decentralized review, the index summary is uploaded to the blockchain and synchronized to the DHT global metadata index network. The specific process is as follows:

[0184] Data product encapsulation: Enterprise 1 encapsulates core performance data of iron ore, such as grade, iron content, impurity content, and origin, into data product D1, builds a data index in the local data space (establishes the index relationship between the data product metadata and the original data product), and generates an index summary of the data product (de-sensitized, containing only non-sensitive identifiers, data categories, performance feature hashes, etc.).

[0185] Release Request Submission: Enterprise 1 submits a release request for data product D1 to the system, including an index summary, product description, its own DID, and VC.

[0186] Decentralized Audit Committee (DAC) Formation: The decentralized governance module extracts the core indicators (node ​​reputation value, data activity, and relative network latency) of each audit candidate node from the audit candidate node pool, calculates the comprehensive weight through a weighted normalization function, and automatically selects 3 nodes to form a decentralized audit committee (DAC) by combining the randomness seed generated by the VRF verifiable random function.

[0187] Independent Parallel Review: Each candidate review node of DAC conducts independent parallel review of Enterprise 1's data product D1, focusing on verifying the compliance, anonymization integrity, and matching with the enterprise's qualifications. After the review is passed, each node digitally signs the review result with its own private key.

[0188] Consensus determination and on-chain: The smart contract responsible for the review collects the digital signatures of all candidate nodes for review, executes the signature aggregation consensus strategy (N-of-M consensus strategy, approval is achieved when 2 or more out of 3 nodes pass), and writes the index digest of D1 + the set of review signatures into the blockchain after approval.

[0189] Global Index Synchronization: After further desensitizing the index digest of D1, the blockchain synchronizes it to the DHT global metadata index module. The DHT network stores the index digest and binds it to the local data space identifier of Enterprise 1, completing the global searchable registration of the data product.

[0190] Enterprises 2, 3, and 4 shall complete the release, review, index summary upload, and DHT synchronization of iron ore data product D2, alloy material data products D3 and D4 respectively, following the above process.

[0191] Step 3: The data user initiates a search request (Federated Search and Validation Module).

[0192] Steel production enterprises (data users) submit search requests to the federated search and verification module based on their steelmaking process optimization needs, requesting data products related to "iron ore from region A or B with an iron content ≥65% and impurity content ≤0.5%". The specific process is as follows:

[0193] Steel production enterprises construct a search request on the client side, carrying the obtained legitimate Query VC (including search filter items: allowed data category is iron_ore_performance, keyword Bloom filter is "iron content ≥65% & impurity content ≤0.5%", vector digest is iron ore performance feature vector hash, maximum access range is steel-supply-chain-level-2, etc.), and send it to the FQC of the federated search and verification module.

[0194] After receiving a search request, FQC first parses the Query VC to obtain the core search conditions, converts them into standardized search filter items that can be recognized by the DHT network, and sends them to the global metadata index module (DHT network) to request the collection of candidate local data spaces.

[0195] Step 4: Global Candidate Node Location (Global Metadata Index Module)

[0196] The DHT global metadata indexing module performs distributed matching based on search filtering items across the entire network's de-identified index digest, locating a set of candidate local data spaces that meet the criteria for data users. The specific process is as follows:

[0197] The DHT network performs distributed parallel matching of the received search filters with all the data product index summaries (D1, D2, D3, D4) stored in its own database, and selects matching index summaries: only D1 of enterprise 1 and D2 of enterprise 2 meet the search criteria of "iron ore in region A or B, iron content ≥65%, impurity content ≤0.5%".

[0198] The DHT network locates the local data space of the data provider for the matched D1 and D2 index digests, extracts the local data space identifiers (DID + access entry) of Enterprise 1 and Enterprise 2, and generates matching scores for the two candidate nodes (matching score of 0.98 for Enterprise 1 and matching score of 0.95 for Enterprise 2).

[0199] The DHT network returns the candidate local data space set (Enterprise 1, Enterprise 2) + the matching scores of each node to the FQC federated query coordinator.

[0200] Step 5: Candidate Node Ranking and Parallel Retrieval and Distribution (Federated Retrieval and Validation Module)

[0201] After receiving the candidate local data space set, FQC distributes local retrieval requests to candidate nodes in parallel according to priority and parallelism rules. The specific process is as follows:

[0202] Candidate set size calculation: FQC calculates the candidate set size, where the number of candidate nodes is 1, α=2.0, and the expected hit count is 1, resulting in a candidate set size m=min(2,2)=2, and the candidate set is determined to be Enterprise 1 and Enterprise 2.

[0203] Priority-weighted sorting: FQC extracts the matching score, node reputation value, and relative network latency of Enterprise 1 and Enterprise 2, calculates the weighted score, and determines the retrieval priority as Enterprise 1 > Enterprise 2 because Enterprise 1 has a higher score than Enterprise 2.

[0204] Parallelism calculation and request distribution: The parallelism is calculated to P=2, which supports the full parallel distribution of local retrieval requests to two candidate nodes, and the retrieval filter items and Query VC are sent synchronously to the local data space of Enterprise 1 and Enterprise 2.

[0205] Step 6: Local Data Retrieval and Result Signing (Local Data Spatial Index Module)

[0206] After receiving the local retrieval request from FQC, the local data spatial indexing modules of Enterprise 1 and Enterprise 2 complete Query VC verification, perform accurate local retrieval, and generate a signed retrieval result summary to return to FQC. The specific process is as follows:

[0207] Full-dimensional verification of Query VC: The local data space indexing module performs strict verification on the received Query VC: ① Hash consistency verification (comparing the Query VC hash with the verifiable credential stored on the blockchain); ② Issuer signature verification (using public key verification to prove the domain digital signature); ③ Access scope matching verification (confirming that the maximum access scope of the Query VC matches the local data product permission policy); ④ Validity verification (confirming that the search request has not exceeded the Query VC's expiration time). Local search is triggered after all verifications pass.

[0208] Local Precise Search: Parse the search filter items and match them with the local data index type: Since the filter items contain keyword Bloom filters, the local inverted index is called; based on the inverted index, the search filter items are matched with the local data product metadata to filter out the target data products (D1 of Enterprise 1, D2 of Enterprise 2).

[0209] The generated search result summary only includes non-sensitive information such as data product identifier, specific values ​​of iron content / impurity content, matching degree, and data update time, and does not include the original data.

[0210] Digital Signature and Result Return: Enterprise 1 and Enterprise 2 each use their own local data space private keys to digitally sign the search result digest, forming a signed version of the search result digest, and synchronously return it to FQC; if the search times out (exceeding the local processing time by 300ms), timeout placeholder information is returned; if only part of the results are matched, a partial signed digest is returned.

[0211] Step 7: Validation and Feedback of Search Results (Federated Search and Validation Module)

[0212] The federated search and verification module verifies the credibility of the received signed search result summary. Upon successful verification, the result is sent back to the data user, and the search session summary is uploaded to the blockchain for notarization. The specific process is as follows:

[0213] FQC will send the signed search result summaries from Enterprise 1 and Enterprise 2 to the search response verification module within the module.

[0214] Digital signature trustworthiness verification: The retrieval response verification module extracts the public keys of Enterprise 1 and Enterprise 2 from the blockchain, verifies the digital signature of the retrieval result digest, confirms that the signature is legal and the result has not been tampered with, and the verification is successful.

[0215] Results Feedback: The retrieval response verification module sends the retrieval result summaries (sorted by matching degree) of Enterprise 1 and Enterprise 2 to the client of the steel production enterprise (data user). The data user can view the core information of the iron ore data products that meet the criteria.

[0216] On-chain retrieval session: The retrieval response verification module extracts the core information of the entire retrieval process (data user DID, Query VC hash, candidate node set, retrieval result summary, signature verification result, etc.), generates a retrieval session summary, writes it to the blockchain to complete the notarization, and realizes the auditability, traceability, and non-repudiation of this retrieval behavior.

[0217] Step 8: On-demand point-to-point access to raw data (extended to practical industrial applications)

[0218] If a steel production enterprise needs to obtain the original data of Enterprise 1 (D1) and Enterprise 2 (D2), it can initiate an independent authorized access request to Enterprise 1 and Enterprise 2 based on the local data space access portal of the data provider in the search result summary. After Enterprise 1 and Enterprise 2 re-verify the identity and permissions of the steel production enterprise, they will transmit the original data to it in a point-to-point manner, without going through the FQC and DHT networks. This ensures that the ownership and control of the original data always belong to the data provider, which meets the security requirement of "local data not leaving the domain" for industrial data.

[0219] The core value of the system embodied in this embodiment is:

[0220] Decentralized: There is no centralized platform intervention. Data product review, global indexing, and retrieval are all executed in a distributed manner, avoiding single points of failure and trust dependencies.

[0221] Data security: The original data products are always stored in the local data space of the data provider, and only the de-sensitized index summary is synchronized to the global network. The search results are only non-sensitive summaries, thus protecting data privacy from the source.

[0222] High-efficiency retrieval: Candidate nodes are quickly located through the DHT global index, FQC supports parallel retrieval and priority sorting, and precise retrieval is achieved by combining local inverted / vector indexes, enabling millisecond-level retrieval of cross-enterprise data;

[0223] End-to-end trustworthiness: full-node identity authentication, Query VC validity verification, digital signature of search results, and end-to-end session on-chaining, constructing a full-link trustworthy system of "identity-credential-search-result";

[0224] Supply chain collaboration: Enables data sharing and retrieval across enterprises in the upstream and downstream of the steel industry chain, breaks down data silos, provides data support for the process optimization of steel production enterprises, and at the same time ensures the data asset security of upstream suppliers.

[0225] This invention also provides a data publishing method based on a distributed industrial data space system, as described in the following embodiments. Since the principle behind this method is similar to that of the distributed industrial data space system, its implementation can be found in the implementation of the distributed industrial data space system; details that are repeated will not be repeated.

[0226] Figure 4 This is a flowchart of a data publishing method based on a distributed industrial data spatial system, as described in an embodiment of the present invention. Figure 4 As shown, the method includes:

[0227] Step 401: The decentralized governance module selects multiple audit candidate nodes from the audit candidate node pool;

[0228] Step 402: After receiving the data product submitted by the provider data node, the decentralized governance module independently and in parallel reviews the data product through the selected review candidate nodes.

[0229] Step 403: After approving the release of the data product, the decentralized governance module writes the index digest of the data product into the blockchain;

[0230] Step 404: The global metadata index module stores the index summary of the data product after blockchain anonymization.

[0231] In this embodiment, the method further includes:

[0232] The distributed identity module authenticates the identities of all participating nodes accessing the distributed industrial data space system and obtains an identity authentication identifier. The participating nodes include audit candidate nodes, data provider nodes, and data user nodes.

[0233] The verifiable credentials module generates verifiable credentials based on the qualifications submitted by participating nodes and stores the hash value of the verifiable credentials on the blockchain.

[0234] In this embodiment, the decentralized governance module uses smart contracts to select multiple audit candidate nodes from the audit candidate node pool to form a decentralized audit committee.

[0235] After receiving the data product release application submitted by the data provider node, it is distributed to each review candidate node of the decentralized review committee so that each review candidate node can independently and in parallel review and digitally sign the data product in the data product release application; the digital signatures of the data product submitted by all review candidate nodes are received, and the signature aggregation consensus strategy is executed to determine whether to approve the release of the data product. If approved, the index digest and digital signature set of the data product are written into the blockchain.

[0236] In this embodiment, the decentralized governance module selects multiple audit candidate nodes from the audit candidate node pool, including:

[0237] Extract the core metrics for each candidate node in the candidate node pool for review;

[0238] Based on the core indicators, the comprehensive weight of each review candidate node is calculated using a weighted normalization function, and the comprehensive weight is then globally normalized to obtain the normalized weight.

[0239] Generate a randomness seed that can be verified across the entire network based on the private key;

[0240] Each audit candidate node generates a verifiable random number based on a random seed and an identity authentication identifier;

[0241] The selection threshold is calculated based on the number of candidate nodes in the candidate node pool, the target size, and the normalized weight.

[0242] For each audit candidate node, if the verifiable random number of the audit candidate node is less than the selection threshold, the audit candidate node is added to the initial selection node set.

[0243] If the number of initially selected nodes exceeds the target size, the top target number of candidate nodes will be selected based on verifiable random numbers from smallest to largest to form a decentralized review committee.

[0244] In this embodiment, the global metadata index module stores an index digest of the blockchain-de-identified data product, including:

[0245] The distributed hash table network is used to store the index digest of the data product after the index digest of the data product has been de-identified by the blockchain, and it is bound to the local data space identifier of the data provider to which the data product belongs.

[0246] In summary, the above embodiments, Figure 5This is an interaction diagram for data publishing based on a distributed industrial data space system in this embodiment of the invention. In the identity authentication phase, the data provider node first completes its own identity authentication, then submits industry qualifications to obtain a verifiable credential (VC). The decentralized governance module authenticates the candidate nodes for review and completes identity authentication for participating nodes. Afterwards, the decentralized governance module issues VCs to the data provider, which the data provider node can store. In the data publishing and review phase, the data provider initiates a data publishing application. The decentralized governance module selects a candidate node for review from the candidate nodes. The candidate node reviews the data publishing application and generates a digital signature set. In the index digest publishing phase, the blockchain layer executes a signature aggregation consensus strategy, verifies the data, and then uploads and registers it on the chain. The blockchain stores the index digest and digital signature set on the chain. The DHT network registers the index digest of the data product, realizing distributed index management.

[0247] This invention also provides a data retrieval method based on a distributed industrial data spatial system, as described in the following embodiments. Since the principle underlying this method is similar to that of the distributed industrial data spatial system, its implementation can be found in the implementation of the distributed industrial data spatial system; details that are repeated will not be repeated here.

[0248] Figure 6 This is a flowchart of a data retrieval method based on a distributed industrial data spatial system, as described in an embodiment of the present invention. Figure 6 As shown, the method includes:

[0249] Step 601: The federated retrieval and verification module receives a query request sent by a data user node, the query request carrying query credentials.

[0250] Step 602: The global metadata index module determines the candidate local data space set based on the query credentials and the index digest of the data product after blockchain de-identification;

[0251] Step 603: The federated search and verification module distributes local search requests in parallel to the candidate local data space set;

[0252] Step 604: After receiving a local retrieval request in each local data space, a local retrieval is performed through the data index to generate a retrieval result summary and a corresponding digital signature.

[0253] Step 605: After the Federated Search and Verification module verifies the credibility of the digital signature of the search result summary, it sends the search result summary to the data user.

[0254] In this embodiment, after each local data space receives a local retrieval request, it performs a local retrieval using the data index to generate a retrieval result summary and a corresponding digital signature, including:

[0255] After receiving a local retrieval request in each local data space, and after the query credentials are verified in the local data space, a local retrieval is performed in the local data space through the data index. After the retrieval is completed, a retrieval result summary is generated and digitally signed in the local data space. The retrieval result summary and the corresponding digital signature are then sent to the federated query coordination module.

[0256] In this embodiment, the local data space performs query credential verification, including:

[0257] The hash value of the query credential in the retrieval request is compared with the hash value of the verifiable credential stored on the blockchain by the data user node that sent the query credential.

[0258] After the consistency verification is passed, the public key of the issuer of the query credential is obtained from the blockchain. The public key is used to verify the digital signature of the proof field in the query credential to verify whether the digital signature was legally signed by the issuer's private key.

[0259] If so, check the permission content of the maximum access range field in the voucher body of the query voucher to verify whether the maximum access range matches the data product access permission policy preset locally by the data provider node;

[0260] If so, check the expiration time field value in the voucher body of the query voucher to verify whether the retrieval request was initiated after the expiration date.

[0261] If the voucher has not expired, the verification is confirmed to be successful; otherwise, the verification is confirmed to be unsuccessful.

[0262] In this embodiment, the data index is used to establish the relationship between the metadata of the data product and the index of the data product;

[0263] Local retrieval is performed in the local data space using the data index. After the retrieval is completed, a retrieval result summary is generated and digitally signed in the local data space. The retrieval result summary and the corresponding digital signature are then sent to the federated query coordination module, including:

[0264] Parse the query credentials to obtain the search filters;

[0265] Based on the search filter, match the data index type in the local data space;

[0266] Based on the matching data index type, the search filter items are matched with the metadata of the data products in the data index corresponding to the data index type to filter the target data products;

[0267] Generate a summary of search results based on the matched target data products;

[0268] The search result digest is digitally signed using the private key of the data provider node corresponding to the local data space.

[0269] In this embodiment, the method further includes:

[0270] If the generation of the search result summary exceeds the first time limit, the timeout placeholder information is sent back to the federated search and verification module. If only part of the target data products are matched, the search result summary of part of the target data products and the corresponding digital signature are sent back to the federated search and verification module.

[0271] In this embodiment, the federated search and verification module distributes local search requests in parallel to the candidate local data space set, including:

[0272] The Federated Query Coordination Module parses the query credentials in the query request to obtain the core search conditions, converts them into search filter items, and sends them to the Global Metadata Index Module. It then obtains the candidate local data space set returned by the Global Metadata Index Module and distributes the local search request to the candidate local data space set.

[0273] After the Federated Search and Verification module verifies the trustworthiness of the digital signature of the search result summary, it sends the search result summary to the data user, including:

[0274] After receiving the search result summary and its corresponding digital signature, the federated search and verification module verifies whether the digital signature of the search result summary is trustworthy. If it is trustworthy, the module sends the search result summary to the data user and writes the summary of the search session corresponding to the local search request into the blockchain.

[0275] In this embodiment, the global metadata indexing module determines a set of candidate local data spaces based on the query credentials and the index digest of the blockchain-de-identified data product, including:

[0276] The search filter items are distributedly matched with the index digests of the de-identified data products to obtain multiple matching index digests and corresponding matching scores. The local data space of the data provider node to which each matching index digest belongs is located as a candidate local data space and added to the candidate local data space set. The candidate local data space set and the matching scores corresponding to each candidate local data space are sent to the federated query coordination module.

[0277] After receiving the candidate local data space set, the federated query coordination module prioritizes the candidate local data spaces in the candidate local data space set according to the matching score corresponding to the candidate local data space, and distributes local retrieval requests to the candidate local data spaces according to the degree of parallelism of the calculation.

[0278] In this embodiment, the candidate local data spaces in the candidate local data space set are prioritized according to the matching scores corresponding to the candidate local data spaces, and local retrieval requests are distributed to the candidate local data spaces according to the degree of parallelism of the calculation, including:

[0279] The size of the candidate set is calculated based on the size of the candidate local data space set and the expected hit redundancy value.

[0280] The weighted score of each candidate local data space is calculated based on the matching score of each candidate local data space, the reputation of the data provider node corresponding to the candidate local data space, and the relative network latency.

[0281] The candidate local data space set is selected from the candidate local data space set in descending order of weighted score;

[0282] The smaller value between the candidate local data space set and the maximum parallelism is determined as the parallelism.

[0283] When the size of the candidate set is greater than the parallelism, the candidate local data space set is divided into multiple batches according to the parallelism based on the weighted score from high to low, and then the local retrieval requests of the candidate local data space are distributed batch by batch.

[0284] In this embodiment, the method further includes:

[0285] For candidate local data spaces that have not received a search result summary within the second time period and whose number of retries is less than the retry threshold, the local search request is resent, and the waiting time for each retry is determined by a backoff strategy.

[0286] Once the search result summary for the fast completion threshold has been received, the sending of subsequent local search requests is terminated early.

[0287] In summary, the above embodiments, Figure 7 This is an interaction diagram of data retrieval based on a distributed industrial data spatial system in an embodiment of the present invention, which follows... Figure 5This demonstration showcases the end-to-end operation of data in a distributed industrial data space system, from publication and retrieval to on-chain notarization. The core participants are data provider nodes and data user nodes, and the core operations are completed using the distributed industrial data space system and blockchain. In the data publication phase, after completing the preliminary review process, the data provider node executes the index digest publication operation. The published index digest is incorporated into the global metadata index module of the distributed industrial data space system. In the data retrieval phase, the data user node generates a query request carrying query credentials and sends it to the FQC (Functionally Qualified Query Center). The FQC retrieves a set of candidate local data spaces by querying the DHT (Distributed Hash Table) and distributes local retrieval requests in parallel to the candidate local data spaces, improving retrieval efficiency. After each candidate local data space completes its retrieval, it digitally signs the retrieval result digest and returns the digitally signed retrieval result digest to the system for aggregation. The system performs a credibility verification on the aggregated retrieval result digest to ensure that the data has not been tampered with and its source is legitimate. In the on-chain notarization phase, after the digital signature passes the credibility verification, the system returns the retrieval result digest to the data user node, completing the retrieval service. Meanwhile, the search result summary is synchronized to the blockchain for on-chain storage, leveraging the immutability of the blockchain to achieve full-process traceability and reliable tracking of search behavior and results.

[0288] This invention also provides a computer device. Figure 8 This is a schematic diagram of a computer device in an embodiment of the present invention. The computer device 800 includes a memory 810, a processor 820, and a computer program 830 stored in the memory 810 and executable on the processor 820. When the processor 820 executes the computer program 830, it implements a method.

[0289] This invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the aforementioned data publishing method or data retrieval method based on a distributed industrial data space system.

[0290] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the aforementioned data publishing method or data retrieval method based on a distributed industrial data space system.

[0291] In this embodiment of the invention, the decentralized governance module is used to: select multiple audit candidate nodes from the audit candidate node pool; independently and in parallel audit the data products submitted by the data provider nodes through the selected audit candidate nodes; and write the index digest of the data product into the blockchain after approving the release of the data product; the federated retrieval and verification module is used to: receive query requests sent by data user nodes, the query requests carrying query credentials, and obtain a set of candidate local data spaces in the global metadata index module through the query credentials; and distribute local retrieval requests in parallel to the set of candidate local data spaces; the global metadata index module is used to: store the index digest of the data product after blockchain desensitization; and determine the set of candidate local data spaces based on the received query credentials and the index digest; the local data space index module is used to: have each data provider node construct a data index for the data product in its local data space; and after each local data space receives a local retrieval request, perform a local retrieval through the data index to generate a retrieval result digest and a corresponding digital signature, and send it to the federated query coordination module; the federated retrieval and verification module is also used to: after verifying the credibility of the digital signature of the received retrieval result digest, send the retrieval result digest to the data user. Compared to centralized platform-based data sharing solutions, decentralized governance modules enable decentralized parallel review of data product releases, avoiding single points of failure, low efficiency, and trust dependency issues inherent in centralized review. Furthermore, approved data product index summaries are written to the blockchain, ensuring their immutability and traceability. A global metadata index module stores anonymized index summaries, supporting rapid location of candidate local data space sets based on query credentials while protecting data privacy, enabling efficient global retrieval routing. A local data space index module builds data indexes and performs local searches on each data provider node, eliminating the need to upload or aggregate raw data, ensuring data providers' ownership and control of their data, while improving retrieval efficiency and reducing network transmission pressure. A federated retrieval and verification module uniformly receives query requests, distributes retrieval tasks in parallel, and verifies the digital signature credibility of retrieval result summaries, achieving cross-node federated retrieval and result verification, enhancing the overall system's security and credibility while maintaining retrieval efficiency. The overall system achieves decentralized governance, global index routing, local data not leaving the domain, federated parallel retrieval, and verifiable results, making it suitable for cross-entity, high-security, high-efficiency, and highly reliable data sharing and retrieval scenarios in industrial data spaces.

[0292] In this embodiment of the invention, a distributed identity module is used to perform unified identity authentication on all participating nodes, establishing a trusted identity identifier to prevent identity forgery and impersonation, and enhancing the overall trust foundation of the system. A verifiable credential module generates and stores verifiable credentials on the blockchain based on node qualifications, enabling verifiable and traceable qualifications and providing a reliable basis for subsequent auditing and access control.

[0293] In this embodiment of the invention, a decentralized review committee module is used to achieve distributed screening of review nodes, avoiding single-point review and centralized power concentration, thereby improving the fairness and resistance to attacks of the review process. The review smart contract module enables automatic distribution, parallel review, and signature aggregation consensus of data publishing applications, improving review efficiency while ensuring the credibility and immutability of review results through multi-node signatures. Upon approval, the index digest and signature set are uploaded to the blockchain, achieving permanent evidence of the publishing behavior.

[0294] In this embodiment of the invention, node weights are calculated based on core indicators and a weighted normalization function. Verifiable random numbers and selection thresholds are then used to screen and review nodes, ensuring both fairness and randomness in the selection process while prioritizing high-weight, highly trustworthy nodes. The verifiable random number generation and verification mechanism ensures that the committee selection process is publicly auditable, preventing human manipulation and enhancing the transparency and credibility of decentralized governance.

[0295] In this embodiment of the invention, a distributed hash table (DHT) network is also used to store the anonymized index digest, which achieves both distributed storage and efficient retrieval of the index, while protecting the privacy of the data provider through anonymization. The index digest is bound to the local data space identifier, enabling precise positioning of the data product and its associated node, and providing a reliable routing basis for subsequent retrieval and distribution.

[0296] In this embodiment of the invention, query credentials are verified before local retrieval to ensure the legality and compliance of the retrieval request and prevent unauthorized access. After the retrieval is completed, a result digest with a local digital signature is generated to ensure the authenticity and completeness of the retrieval results and to provide a basis for subsequent trusted verification.

[0297] In this embodiment of the invention, a rigorous credential verification system is constructed through multiple layers of verification, including hash consistency, signature verification using the issuer's public key, access range matching, and expiration time verification. This effectively prevents credential forgery, unauthorized access, and expired use, ensuring the security of data access.

[0298] In this embodiment of the invention, retrieval filters are obtained by parsing query credentials, and corresponding data indexes are matched for metadata retrieval, improving the accuracy and efficiency of local retrieval. A local private key is used to digitally sign the retrieval result digest, ensuring the immutability and traceability of the retrieval results, providing a trusted foundation for federated verification.

[0299] In this embodiment of the invention, timeout preemption and partial result return mechanism are also supported to avoid overall retrieval blockage caused by slow response of individual nodes, thereby improving the robustness of the system and user experience, while ensuring the continuity of the retrieval process.

[0300] In this embodiment of the invention, the federated query coordination module implements query request parsing, candidate node location, and parallel distribution, improving the efficiency of cross-node retrieval. The retrieval response verification module performs trusted verification on the result signature to ensure the authenticity of the returned results and uploads the retrieval session summary to the blockchain, achieving traceability and immutability of retrieval behavior.

[0301] In this embodiment of the invention, distributed matching and score estimation are used to prioritize the distribution of requests to local data spaces with high matching scores, thereby improving the retrieval hit rate and response efficiency. Prioritization based on matching scores ensures that core retrieval needs are met first, optimizing system resource utilization.

[0302] In this embodiment of the invention, a weighted score is calculated based on the candidate set size, expected hit count, node reputation, and network latency to select the optimal candidate set and determine the degree of parallelism, thus achieving a balance between retrieval efficiency and resource consumption. Distributing requests in batches avoids network congestion caused by excessive single-round request volume, improving the system's concurrent processing capability and stability.

[0303] In this embodiment of the invention, a backoff and retry strategy is also adopted for timeout nodes, which improves the success rate of retrieval requests and avoids the pressure on the system caused by invalid retries.

[0304] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0305] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0306] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0307] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0308] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A distributed industrial data spatial system, characterized in that, include: Decentralized governance module, global metadata index module, local data space index module, federated search and verification module; The decentralized governance module is used to: select multiple audit candidate nodes from the audit candidate node pool; independently and in parallel audit the data products submitted by the data provider nodes through the selected audit candidate nodes; and write the index digest of the data product into the blockchain after approving the release of the data product. The federated retrieval and verification module is used to: receive query requests sent by data user nodes, the query requests carrying query credentials, obtain a set of candidate local data spaces in the global metadata index module through the query credentials; and distribute local retrieval requests in parallel to the set of candidate local data spaces. The global metadata index module is used to: store index summaries of blockchain-de-identified data products; and determine a set of candidate local data spaces based on the received query credentials and the index summaries. The local data space indexing module is used for: each data provider node to build a data index for the data product in the local data space; after each local data space receives a local retrieval request, it performs a local retrieval through the data index to generate a retrieval result summary and corresponding digital signature, and sends it to the federated query coordination module. The federated search and verification module is also used to: send the search result summary to the data user after the digital signature of the received search result summary has been verified for trustworthiness; The decentralized governance module includes: The decentralized review committee module is used to extract the core indicators of each review candidate node in the review candidate node pool; based on the core indicators, it calculates the comprehensive weight of each review candidate node through a weighted normalization function, and performs global normalization on the comprehensive weight to obtain the normalized weight; it generates a network-wide verifiable randomness seed based on the private key; it obtains verifiable random numbers for each review candidate node based on the randomness seed and identity authentication identifier; it calculates the selection threshold based on the number of review candidate nodes in the review candidate node pool, the target size, and the normalized weight; for each review candidate node, if the verifiable random number of the review candidate node is less than the selection threshold, the review candidate node is added to the preliminary selected node set; if the number of nodes in the preliminary selected node set exceeds the target size, the top target size review candidate nodes are selected according to the verifiable random numbers from smallest to largest to form the decentralized review committee. The smart contract review module is used to verify the data provider node based on verifiable credentials after receiving the data product release application submitted by the data provider node, and then distribute the data product release application to each review candidate node of the decentralized review committee. This allows each review candidate node to independently and in parallel review and digitally sign the data product in the data product release application. The module also receives the digital signatures of the data product from all review candidate nodes, executes the signature aggregation consensus strategy, and determines whether to approve the data product release. If approved, the module writes the index digest and digital signature set of the data product into the blockchain.

2. The system as described in claim 1, characterized in that, It also includes a trust and identity infrastructure module, which includes: The distributed identity module is used to authenticate the identities of all participating nodes accessing the distributed industrial data space system and obtain identity authentication identifiers. The participating nodes include audit candidate nodes, data provider nodes, and data user nodes. The verifiable credentials module is used to generate verifiable credentials based on the qualifications submitted by participating nodes and store the hash value of the verifiable credentials on the blockchain.

3. The system as described in claim 1, characterized in that, The global metadata index module is specifically used for: The distributed hash table network is used to store the index digest of the data product after the index digest of the data product has been de-identified by the blockchain, and it is bound to the local data space identifier of the data provider to which the data product belongs.

4. The system as described in claim 1, characterized in that, The local data space index module is specifically used for: after each local data space receives a local retrieval request, after the query credentials are verified in the local data space, a local retrieval is performed in the local data space through the data index; after the retrieval is completed, a retrieval result summary is generated and digitally signed by the local data space; and the retrieval result summary and the corresponding digital signature are sent to the federated query coordination module.

5. The system as described in claim 4, characterized in that, The local data spatial index module is specifically used for: The hash value of the query credential in the retrieval request is compared with the hash value of the verifiable credential stored on the blockchain by the data user node that sent the query credential. After the consistency verification is passed, the public key of the issuer of the query credential is obtained from the blockchain. The public key is used to verify the digital signature of the proof field in the query credential to verify whether the digital signature was legally signed by the issuer's private key. If so, check the permission content of the maximum access range field in the voucher body of the query voucher to verify whether the maximum access range matches the data product access permission policy preset locally by the data provider node; If so, check the expiration time field value in the voucher body of the query voucher to verify whether the retrieval request was initiated after the expiration date. If the voucher has not expired, the verification is confirmed to be successful; otherwise, the verification is confirmed to be unsuccessful.

6. The system as described in claim 4, characterized in that, The data index is used to establish the relationship between the metadata of the data product and the index of the data product. The local data spatial index module is specifically used for: Parse the query credentials to obtain the search filters; Based on the search filter, match the data index type in the local data space; Based on the matching data index type, the search filter items are matched with the metadata of the data products in the data index corresponding to the data index type to filter the target data products; Generate a summary of search results based on the matched target data products; The search result digest is digitally signed using the private key of the data provider node corresponding to the local data space.

7. The system as described in claim 6, characterized in that, The local data spatial index module is also used for: If the generation of the search result summary exceeds the first time limit, the timeout placeholder information is sent back to the federated search and verification module. If only part of the target data products are matched, the search result summary of part of the target data products and the corresponding digital signature are sent back to the federated search and verification module.

8. The system as described in claim 7, characterized in that, The federal search and verification module includes: The federated query coordination module receives query requests sent by data user nodes, parses the query credentials in the query request to obtain core search conditions, converts them into search filter items, and sends them to the global metadata index module. It then obtains the candidate local data space set returned by the global metadata index module, distributes local search requests to the candidate local data space set, and sends the search response verification module after receiving the search result summary and corresponding digital signature. The retrieval response verification module is used to verify whether the digital signature of the retrieval result summary is trustworthy. If it is trustworthy, the retrieval result summary is sent to the data user, and the summary of the retrieval session corresponding to the local retrieval request is written to the blockchain.

9. The system as described in claim 8, characterized in that, The global metadata indexing module is specifically used for: performing distributed matching between the search filter items and the index digests of the de-identified data products to obtain multiple matching index digests and corresponding matching scores; locating the local data space of the data provider node to which each matching index digest belongs as a candidate local data space and adding it to the candidate local data space set; and sending the candidate local data space set and the matching scores corresponding to each candidate local data space to the federated query coordination module. The Federated Query Coordination module is also used for: After receiving the candidate local data space set, the candidate local data spaces in the candidate local data space set are prioritized according to the matching score corresponding to the candidate local data space, and local retrieval requests are distributed to the candidate local data spaces according to the parallelism of the calculation.

10. The system as described in claim 9, characterized in that, The Federated Query Coordination module is specifically used for: The size of the candidate set is calculated based on the size of the candidate local data space set and the expected hit redundancy value. The weighted score of each candidate local data space is calculated based on the matching score of each candidate local data space, the reputation of the data provider node corresponding to the candidate local data space, and the relative network latency. The candidate local data space set is selected from the candidate local data space set in descending order of weighted score; The smaller value between the candidate local data space set and the maximum parallelism is determined as the parallelism. When the size of the candidate set is greater than the parallelism, the candidate local data space set is divided into multiple batches according to the parallelism based on the weighted score from high to low, and then the local retrieval requests of the candidate local data space are distributed batch by batch.

11. The system as claimed in claim 10, characterized in that, The Federated Query Coordination module is also used for: For candidate local data spaces that have not received a search result summary within the second time period and whose number of retries is less than the retry threshold, the local search request is resent, and the waiting time for each retry is determined by a backoff strategy. Once the search result summary for the fast completion threshold has been received, the sending of subsequent local search requests is terminated early.

12. A data publishing method based on a distributed industrial data spatial system, characterized in that, include: The decentralized governance module selects multiple audit candidate nodes from the pool of audit candidate nodes; After receiving the data products submitted by the provider data nodes, the decentralized governance module independently and in parallel reviews the data products through selected review candidate nodes; After approving the release of the data product, the decentralized governance module writes the index digest of the data product into the blockchain; The global metadata index module stores index summaries of blockchain-anonymized data products; The decentralized governance module includes: The decentralized review committee module is used to extract the core indicators of each review candidate node in the review candidate node pool; based on the core indicators, it calculates the comprehensive weight of each review candidate node through a weighted normalization function, and performs global normalization on the comprehensive weight to obtain the normalized weight; it generates a network-wide verifiable randomness seed based on the private key; it obtains verifiable random numbers for each review candidate node based on the randomness seed and identity authentication identifier; it calculates the selection threshold based on the number of review candidate nodes in the review candidate node pool, the target size, and the normalized weight; for each review candidate node, if the verifiable random number of the review candidate node is less than the selection threshold, the review candidate node is added to the preliminary selected node set; if the number of nodes in the preliminary selected node set exceeds the target size, the top target size review candidate nodes are selected according to the verifiable random numbers from smallest to largest to form the decentralized review committee. The smart contract review module is used to verify the data provider node based on verifiable credentials after receiving the data product release application submitted by the data provider node, and then distribute the data product release application to each review candidate node of the decentralized review committee. This allows each review candidate node to independently and in parallel review and digitally sign the data product in the data product release application. The module also receives the digital signatures of the data product from all review candidate nodes, executes the signature aggregation consensus strategy, and determines whether to approve the data product release. If approved, the module writes the index digest and digital signature set of the data product into the blockchain.

13. A data retrieval method based on a distributed industrial data spatial system, characterized in that, include: The Federated Search and Verification module receives query requests sent by data user nodes, the query requests carrying query credentials; The global metadata indexing module determines the candidate local data space set based on the query credentials and the index summary of the data product after blockchain de-identification; The federated search and validation module distributes local search requests in parallel to the candidate local data space set; After receiving a local retrieval request in each local data space, a local retrieval is performed through the data index to generate a retrieval result summary and a corresponding digital signature; After the Federated Search and Verification module verifies the credibility of the digital signature of the search result summary, it sends the search result summary to the data user. The index summary of the data product after blockchain anonymization is written to the blockchain by the decentralized governance module; The decentralized governance module includes: The decentralized review committee module is used to extract the core indicators of each review candidate node in the review candidate node pool; based on the core indicators, it calculates the comprehensive weight of each review candidate node through a weighted normalization function, and performs global normalization on the comprehensive weight to obtain the normalized weight; it generates a network-wide verifiable randomness seed based on the private key; it obtains verifiable random numbers for each review candidate node based on the randomness seed and identity authentication identifier; it calculates the selection threshold based on the number of review candidate nodes in the review candidate node pool, the target size, and the normalized weight; for each review candidate node, if the verifiable random number of the review candidate node is less than the selection threshold, the review candidate node is added to the preliminary selected node set; if the number of nodes in the preliminary selected node set exceeds the target size, the top target size review candidate nodes are selected according to the verifiable random numbers from smallest to largest to form the decentralized review committee. The smart contract review module is used to verify the data provider node based on verifiable credentials after receiving the data product release application submitted by the data provider node, and then distribute the data product release application to each review candidate node of the decentralized review committee. This allows each review candidate node to independently and in parallel review and digitally sign the data product in the data product release application. The module also receives the digital signatures of the data product from all review candidate nodes, executes the signature aggregation consensus strategy, and determines whether to approve the data product release. If approved, the module writes the index digest and digital signature set of the data product into the blockchain.

14. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of claim 12 or 13.

15. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method of claim 12 or 13.

16. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the method of claim 12 or 13.

Images