A Data Matching Method and System for Creditor-Debtor Relationships Based on Privacy-Preserving Computation

By implementing tamper-proof authorization records and privacy-modifying processing on debt relationship data, the problems of authorization verification and data leakage in cross-entity data matching are solved, achieving secure and reliable matching of creditor and debtor relationship data.

CN123116273APending Publication Date: 2026-11-17CHANGSHA MATCHING DIGITAL TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611318208.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-28
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In the process of cross-entity data matching, the authorized source and authorization status of debt relationship data are difficult to verify reliably, posing a risk of unauthorized data access and processing, and there is a risk of leakage of raw data during transmission, processing and correlation matching.

Method used

By anchoring unique identifiers and individual authorization credentials in an tamper-proof manner, a non-repudiable authorization record is established. Data is transmitted using independent encrypted sessions and privacy transformation is performed within the privacy computing domain to generate associated representations. Denial policies are implemented for unauthorized data, and standardized response data is generated to reduce the risk of leakage.

Benefits of technology

It enables privacy-preserving matching of creditor and debtor relationship data while avoiding unauthorized processing and data leakage, improves the security, auditability, and processing consistency of cross-entity data matching, and reduces the risk of data leakage.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123116273A_ABST
    Figure CN123116273A_ABST
Patent Text Reader

Abstract

This application discloses a method and system for matching creditor-debtor relationship data based on privacy computing, relating to the field of data privacy and security. The method includes: tamper-proof anchoring of unique identifiers and individual authorization credentials; receiving debt relationship data and verifying authorization through an independent encrypted session; importing the data into a unified privacy computing domain for privacy transformation when verification passes, and executing a rejection if verification fails; responding to privacy matching requests, completing the matching using association representations, and outputting response data standardized with externally observable features. This significantly reduces the risks of creditor data leakage, unauthorized processing, and bypass inference, improves the privacy security, auditability, and compliance of data matching between debtors and creditors, and effectively avoids the risk of disputes in debt disposal.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of data security technology, and in particular to a method and system for matching creditor and debtor relationship data based on privacy computing. Background Technology

[0002] With the increasing demand for data collaboration among financial institutions, asset management institutions, enterprises, and other data subjects, debt and creditor relationship data may be scattered across different business systems or terminals. To determine whether a relationship exists between specific debt and creditor relationship data, identity verification, data validation, and relationship matching need to be performed across different data subjects. Furthermore, debt and creditor relationship data typically includes identity information, performance status, amount range, term, and other sensitive fields. Data processing also requires constraints on authorization scope, data source, access subjects, and output results. Therefore, how to achieve cross-subject data matching while minimizing the exposure of raw data has become a problem that needs to be solved in the fields of fintech and data security.

[0003] Currently, data on debt and creditor relationships from different sources are typically uploaded to a unified server, which then performs retrieval, comparison, or rule matching based on the original fields. Some solutions also employ field encryption, hashing, or anonymization before data upload and restrict data queries through access control. For data authorization, existing systems generally establish authorization criteria by checking authorization agreements, saving authorization logs, or recording interface call information. For matching results, they typically return the hit status, the number of hits, or anonymized debt and creditor relationship information directly. In the current debt and creditor relationship data matching process, the authorized source and status of debt relationship data are difficult to reliably verify, making unauthorized data access and processing vulnerable. Furthermore, the original debt relationship data faces exposure risks during transmission, processing, and correlation matching; differences in message characteristics or response behaviors generated by different matching results may be exploited by external observers to infer the data status within the domain. Summary of the Invention

[0004] The main objective of this application is to provide a method and system for matching creditor and debtor relationship data based on privacy computing, aiming to solve the technical problem of achieving privacy matching of creditor and debtor relationship data while avoiding the risk of unauthorized processing or data leakage during the access and matching process.

[0005] To achieve the above objectives, this application provides a method for matching creditor and debtor relationship data based on privacy computing. The method is applied to a server and includes the following steps: The unique identity identifier and individual authorization certificate obtained from the target debtor's terminal are tamper-proofed and anchored to obtain tamper-proof authorization records; Debt relationship data is received through an independent encrypted session corresponding to the target debtor's terminal, and the debt relationship data is authorized and verified according to the tamper-proof authorization record; If the authorization verification of the debt relationship data passes, the debt relationship data is imported into a privacy computing domain constrained by a unified jurisdiction policy, and privacy transformation processing is performed on the debt relationship data to obtain an association representation; if the authorization verification of the debt relationship data fails, a denial policy is executed on the debt relationship data. In response to a privacy matching request sent by a target data requesting terminal, authentication is performed on the target data requesting terminal. After successful authentication, privacy matching is performed based on the association representation. Based on the privacy matching result, standardized response data with externally observable features is generated and sent to the target data requesting terminal.

[0006] Furthermore, to achieve the above objectives, this application provides a method for matching creditor and debtor relationship data based on privacy computing. This method is applied to a target debtor's terminal and includes the following steps: Obtain the target debtor's unique identifier, individual authorization certificate, and debt relationship data; Send the unique identifier and the individual authorization credential to the server so that the server can perform tamper-proof anchoring on the unique identifier and the individual authorization credential to obtain tamper-proof authorization records; Send an encrypted session request to the server to establish an independent encrypted session with the server; The debt relationship data is sent to the server through the independent encrypted session, so that the server can perform authorization verification on the debt relationship data according to the tamper-proof authorization record. When the authorization verification of the debt relationship data passes, the debt relationship data is imported into a privacy computing domain constrained by a unified governance policy for privacy transformation processing to obtain an association representation for privacy matching.

[0007] Furthermore, to achieve the above objectives, this application provides a method for matching creditor and debtor relationship data based on privacy computing. The method is applied to a target data request terminal and includes the following steps: Obtain privacy-preserving matching request data for requesting matching of creditor and debtor relationship data; Generate a privacy matching request based on the privacy matching request data; The privacy matching request is sent to the server so that the server performs privacy matching based on the privacy matching request and the pre-generated association representation, and generates standardized response data with externally observable features and standardized processing based on the privacy matching result. The association representation is obtained by the server performing privacy transformation processing on the debt relationship data that has passed the authorization verification. Receive standardized response data sent by the server; The standardized response data is parsed to obtain the matching response result corresponding to the privacy matching request.

[0008] In addition, to achieve the above objectives, this application also proposes a data matching system for creditor-debtor relationships based on privacy computing, the system comprising a server, a target debtor terminal, and a target data request terminal; The target debtor terminal is used to obtain the target debtor's unique identifier, individual authorization certificate, and debt relationship data; and to send the unique identifier and individual authorization certificate to the server. The server is used to perform tamper-proof anchoring on the unique identity identifier and individual authorization credential obtained from the target debtor's terminal to obtain tamper-proof authorization records; The target debtor terminal is also used to send an encrypted session request to the server to establish an independent encrypted session with the server; and to send debt relationship data to the server through the independent encrypted session. The server is further configured to receive debt relationship data through an independent encrypted session corresponding to the target debtor's terminal, perform authorization verification on the debt relationship data according to the tamper-proof authorization record; if the authorization verification of the debt relationship data passes, the debt relationship data is imported into a privacy computing domain constrained by a unified governance policy, and privacy transformation processing is performed on the debt relationship data to obtain an association representation; if the authorization verification of the debt relationship data fails, a denial policy is executed on the debt relationship data. The target data request terminal is configured to acquire privacy matching request data for requesting matching of creditor and debtor relationship data; generate a privacy matching request based on the privacy matching request data; and send the privacy matching request to the server. The server is also configured to respond to a privacy matching request sent by a target data request terminal, perform authentication on the target data request terminal, and after successful authentication, perform privacy matching based on the association representation, generate standardized response data with externally observable features based on the privacy matching result, and send the standardized response data to the target data request terminal. The target data request terminal is also used to receive standardized response data sent by the server; and to parse the standardized response data to obtain a matching response result corresponding to the privacy matching request.

[0009] Furthermore, to achieve the above objectives, this application also proposes a data matching device for creditor-debtor relationships based on privacy computing. The device includes: a memory, a processor, and a data matching program for creditor-debtor relationships based on privacy computing stored in the memory. The processor is used to run the data matching program for creditor-debtor relationships based on privacy computing. The computer program is configured to implement the steps of the data matching method for creditor-debtor relationships based on privacy computing as described above.

[0010] Furthermore, to achieve the above objectives, this application also proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the privacy-based computation-based debt and creditor relationship data matching method described above.

[0011] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the privacy-based computing-based debt and creditor relationship data matching method described above.

[0012] This application employs tamper-proof anchoring of unique identifiers and individual authorization credentials to create verifiable technical records of the authorizing entity, scope, and status, reducing the risk of authorization basis being replaced or modified afterward. It receives debt relationship data through independent encrypted sessions and verifies the uplink source and data fields based on the tamper-proof authorization record, linking the data transmission link with the authorization link to block data from unknown sources or exceeding the processing scope from the data access stage. By importing verified data into a privacy computing domain constrained by a unified governance policy, each privacy computing node operates according to consistent access, isolation, matching, and output rules, and uses privacy transformation to form a relational representation that can be used for computation but does not directly expose the original fields. By implementing a rejection policy on data that fails verification, unauthorized data is prevented from entering subsequent storage and computation stages. Privacy matching is performed based on the relational representation, transferring relationship judgment from the original plaintext space to a controlled representation space. By standardizing externally observable features such as the message length, response sequence, and protocol fields of the matching response, the possibility of external terminals inferring the internal matching status based on communication behavior is reduced. Therefore, while maintaining the matchability of creditor-debtor relationship data, this application establishes multi-layered constraints against original field leakage, unauthorized processing, node strategy drift, and response side-channel inference, thereby improving the security, auditability, and processing consistency of cross-entity data matching and effectively avoiding the risk of disputes in debt disposal. Furthermore, this application performs privacy matching based on field transformation data in the association representation, ensuring that the matching operation is based on data transformed by cryptography, which differs from technical solutions that directly compare the original plaintext in the privacy computing domain. This ensures that the matching stage does not rely on the original plaintext, further reducing the risk of data leakage. Attached Figure Description

[0013] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0014] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a flowchart illustrating the first embodiment of the data matching method for creditor-debtor relationships based on privacy computing in this application; Figure 2 This is a flowchart illustrating the second embodiment of the data matching method for creditor-debtor relationships based on privacy computing in this application; Figure 3This is a schematic diagram illustrating the process of performing trusted authentication of the target data request terminal based on a single root of trust in one embodiment of the privacy-based computation-based debt and creditor relationship data matching method of this application. Figure 4 This is a flowchart illustrating the third embodiment of the data matching method for creditor-debtor relationships based on privacy computing in this application; Figure 5 This is a schematic diagram of the module structure of the data matching system for creditor-debtor relationships based on privacy computing, as described in an embodiment of this application. Figure 6 This is a schematic diagram of a scenario in one embodiment of the privacy-based computation-based creditor-debtor relationship data matching system of this application; Figure 7 This is a schematic diagram of the device structure of the hardware operating environment involved in the data matching method for creditor-debtor relationships based on privacy computing in the embodiments of this application.

[0016] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0017] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0018] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0019] It should be noted that the executing entity of this application embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a computer, server, etc., or an electronic device capable of realizing the above functions. The following uses a server as an example to describe this embodiment and the following embodiments. The user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of related data comply with the relevant laws, regulations, and standards of the relevant countries and regions.

[0020] Based on this, the first embodiment of this application provides a method for matching creditor and debtor relationship data based on privacy computing, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the data matching method for creditor-debtor relationships based on privacy computing in this application.

[0021] It should be noted that in the process of matching creditor-debtor relationship data in this embodiment, under the premise of ensuring that the debtor's individual consent authorization is traceable and tamper-proof, the debtor's terminal actively and discretely uploads the debt relationship data. After encrypted transmission and authorization verification, the data is imported into a privacy computing domain constrained by a unified governance strategy. Privacy transformation and privacy matching are completed under the condition of no plaintext residue, and a standardized response that is statistically indistinguishable from externally observable features is output to the data request terminal, thereby simultaneously meeting the technical requirements of compliant authorization, privacy security and resistance to side-channel attacks.

[0022] In this embodiment, the data matching method for creditor-debtor relationships based on privacy computing includes: Step S10: Tamper-proof anchoring is performed on the unique identity identifier and individual authorization certificate obtained from the target debtor's terminal to obtain tamper-proof authorization records.

[0023] It should be noted that the target debtor terminal refers to a personal computing device or application client owned and controlled by the debtor, such as a smartphone, tablet, terminal with a debt disposal APP installed, or WeChat mini-program operating environment.

[0024] It should be noted that a unique identity identifier refers to a technical identifier that has a long-term, stable, and globally unique binding relationship with the target debtor entity. This identifier has the attributes of being tamper-proof, non-reusable, and globally uniquely bound, excluding transient identifiers such as temporary session IDs, link identifiers, and dynamic tokens, in order to support cross-domain entity traceability; for example, user primary keys generated after real-name authentication, national network identity authentication codes, or hardware binding identifiers, which are different from temporary session identifiers or dynamic tokens.

[0025] A separate authorization certificate refers to authorization proof data generated only by the target debtor after actively confirming the specific scope of data processing. Examples include digital signature authorization statements, authorization tokens issued by trusted execution environments, or blockchain authorization transactions. A separate authorization certificate can also embed a constraint statement containing the scope of use of personal sensitive information, thereby locking the processing boundaries of sensitive data from a technical perspective.

[0026] It should be noted that anti-tampering anchoring refers to the process of writing the binding relationship between a unique identifier and a separate authorization credential into a storage medium with physical or logical anti-tampering properties, and making the binding relationship unchangeable without authorization throughout its entire lifecycle; anti-tampering authorization record refers to an anti-tampering data record that records the binding relationship between a unique identifier and a separate authorization credential and can be referenced by subsequent verification.

[0027] It is understandable that this embodiment establishes an irrefutable authorization traceability root by combining debtor's active authorization with anti-tampering anchoring, ensuring that subsequent data processing has a separate consent basis, and blocking the path of platform-authorized authorization, batch silent authorization or third-party authorization.

[0028] In practice, the server pushes an independent authorization page or authorization prompt to the target debtor's terminal, which specifies the scope of data processing and authorization confirmation controls. The target debtor's terminal collects the debtor's active confirmation information in a local secure environment and packages the unique identifier and confirmation information to generate a separate authorization credential. The server or terminal-side security module writes the binding relationship between the unique identifier and the separate authorization credential into a tamper-proof storage medium.

[0029] For example, the secure storage area of ​​a trusted execution environment can be invoked to perform signature anchoring, or the binding relationship can be written as a transaction into the consortium blockchain ledger and confirmed through consensus, thereby generating a tamper-proof authorization record. This record can be read by subsequent steps to verify the data source and authorization scope.

[0030] In some embodiments, individual consent authorization is limited to operations initiated, confirmed, and modified solely by the target debtor through the target debtor's terminal, explicitly excluding any form of non-active authorization upload behavior, such as platform-authorized authorization, batch silent authorization, creditor / third-party authorization, and post-authorization supplementary authorization.

[0031] Furthermore, to enhance privacy and security during the debtor authorization process, step S10 may include: Step S101: Send independent authorization page information to the target debtor's terminal so that the target debtor's terminal can display the data processing range and authorization confirmation control based on the independent authorization page information.

[0032] It should be noted that independent authorization page information refers to page data pushed by the server to the target debtor's terminal, used to separately display the scope of data processing and authorization confirmation controls. Authorization confirmation controls are interactive elements for the debtor to express consent or refusal, such as confirmation buttons, electronic signature areas, or biometric data collection entry points.

[0033] In practice, the server generates independent authorization page information based on the security capabilities and business scenarios of the target debtor's terminal, and pushes it to the terminal through an encrypted link; the terminal renders the page locally or in a browser, displays the data processing scope to the debtor, and loads the authorization confirmation control.

[0034] Step S102: Receive the active confirmation information and unique identity identifier fed back by the target debtor terminal based on the authorization confirmation control.

[0035] It should be noted that proactive confirmation information refers to electronic records that explicitly express the debtor's consent to the processing of debt relationship data, triggered by the debtor through an authorization confirmation control, such as signature events generated by clicking the confirmation button, voice confirmation records, or facial recognition records.

[0036] In the specific implementation, after the target debtor terminal detects that the authorization confirmation control has been triggered, it collects the active confirmation information and reads the unique identity identifier from the local security module. The two are then uploaded to the server through an encrypted session. The server verifies the integrity and timestamp of the active confirmation information before proceeding to the subsequent credential generation process.

[0037] Step S103: Generate a separate authorization credential based on the active confirmation information, the data processing scope, and the unique identity identifier.

[0038] In practice, the server will actively confirm the information, data processing scope, and unique identity identifier by structuring and encoding them to generate a separate authorization credential containing the authorization scope hash, timestamp, and digital signature. For example, the above content can be encapsulated in JSON format and the credential can be signed as a whole using the SM2 algorithm.

[0039] Step S104: Write the binding relationship between the unique identifier and the individual authorization credential into the tamper-proof storage medium to obtain the tamper-proof authorization record.

[0040] It should be noted that tamper-proof storage media refers to persistent storage media with physical or logical tamper-proof properties, such as write-once storage for persistent binding relationship hash values ​​or digital signatures, controlled storage areas of hardware security modules, and secure storage areas of blockchain distributed ledgers or trusted execution environments.

[0041] In the specific implementation, the server calculates the hash of the binding relationship between the unique identity identifier and the individual authorization credential, and writes the hash along with the original binding relationship into the tamper-proof storage medium; if a blockchain is used, the binding relationship is submitted as a transaction to the consortium blockchain node, and a tamper-proof authorization record is generated after consensus confirmation.

[0042] Step S20: Receive debt relationship data through the independent encrypted session corresponding to the target debtor's terminal, and perform authorization verification on the debt relationship data according to the tamper-proof authorization record.

[0043] It should be noted that an independent encrypted session refers to an end-to-end encrypted communication channel established between the target debtor's terminal and the server, which is isolated from other terminal sessions. Examples include a two-way authentication channel based on TLS, an encrypted link based on the national cryptographic standard SM2 / SM4, or a point-to-point session based on secure sockets. The implementation methods of encrypted transmission include independent sessions and session multiplexing.

[0044] Debt relationship data refers to the set of information reported by the target debtor's terminal that describes the creditor-debtor relationship, such as creditor identification, debt amount, debt maturity date, contract number, guarantee information or repayment record; authorization verification refers to the process of verifying the upstream source, data fields and processing scope of debt relationship data based on the tamper-proof authorization record.

[0045] Understandably, the server establishes an independent encrypted session corresponding to the target debtor's terminal. By combining this independent encrypted session with authorization verification, the server ensures the confidentiality and integrity of debt relationship data during transmission and prevents unauthorized fields or forged source data from entering the privacy computing domain.

[0046] In the specific implementation, the server negotiates a session key with the target debtor's terminal and establishes an independent encrypted session, assigning an encrypted session identifier to the session; after receiving debt relationship data through the independent encrypted session, the data is associated with and stored with the session identifier; then, the tamper-proof authorization record is read, and the encrypted session identifier is mapped to a unique identity identifier to verify the upstream source, while the data fields of the debt relationship data are verified according to the data processing scope in the authorization record.

[0047] For example, if the scope of authorization is limited to "debt amount and maturity date", fields outside the scope will be truncated or marked as unauthorized; if the source or scope verification fails, a denial policy will be triggered.

[0048] Furthermore, in order to reduce the risk of data leakage during the data upload phase of the debtor, step S20 above may include: Step S201: Establish an independent encrypted session with the target debtor's terminal to obtain the encrypted session identifier corresponding to the target debtor's terminal.

[0049] It should be noted that an encrypted session identifier is a short-lived technical identifier used to uniquely identify an independent encrypted session, such as a session ID, a temporary link token, or a TLS session ticket.

[0050] In practice, the server and the target debtor's terminal establish a session key based on Diffie-Hellman key exchange or national cryptographic SM2 key negotiation, and generate an encrypted session identifier for the session; the session key resides only in the secure memory of both parties and is not permanently stored on the ground.

[0051] Step S202: Receive debt relationship data sent by the target debtor's terminal through the independent encrypted session, and associate the debt relationship data with the encrypted session identifier.

[0052] In practice, the target debtor's terminal uploads the debt relationship data in fragments or in whole within an independent encrypted session. After the server decrypts the data, it writes the encrypted session identifier into the data metadata, thus forming a record linking the data and the session.

[0053] Step S203: Verify the upstream source of the debt relationship data based on the encrypted session identifier and the unique identity identifier in the tamper-proof authorization record to obtain the source verification result.

[0054] It should be noted that the upstream source refers to the terminal identity or session source corresponding to the debt relationship data at the network transmission layer; the source verification result refers to the judgment conclusion on whether the upstream source is legal, such as "passed", "rejected" or "requires review".

[0055] In the specific implementation, the server queries the session mapping table, converts the encrypted session identifier into a unique identity identifier, and compares whether the unique identity identifier exists in the tamper-proof authorization record; if it exists and the authorization status is valid, a successful source verification result is generated; otherwise, a rejected source verification result is generated.

[0056] Step S204: Based on the data processing range in the anti-tampering authorization record, verify the data fields of the debt relationship data to obtain the range verification result.

[0057] It should be noted that a data field refers to the smallest semantically divided unit of information within the debt relationship data, such as the creditor's name, debt amount, due date, and guarantee method; the scope verification result refers to the judgment conclusion made on whether the debt relationship data exceeds the authorized scope.

[0058] In the specific implementation, the server parses the field structure of the debt relationship data and compares the identifier of each field with the whitelist of fields that are allowed to be processed within the data processing scope. If all fields to be processed are within the whitelist, a pass range verification result is generated. If there are unauthorized fields, a reject range verification result is generated and a list of unauthorized fields is recorded.

[0059] Step S205: Perform authorization verification on the debt relationship data based on the source verification result and the range verification result.

[0060] In the specific implementation, the server performs a logical AND operation on the source verification result and the range verification result; only when both pass is the debt relationship data authorization verification deemed successful; if either result is rejected, a rejection policy is triggered.

[0061] Step S31: If the authorization verification of the debt relationship data passes, the debt relationship data is imported into a privacy computing domain constrained by a unified jurisdiction policy, and privacy transformation processing is performed on the debt relationship data to obtain an association representation.

[0062] It should be noted that a privacy computing domain refers to a secure computing environment that is subject to a unified governance policy and has physical or logical isolation boundaries, such as a computing space built on a trusted execution environment, a secure multi-party computing node cluster, or an isolated sandbox.

[0063] Association representation refers to a logical data structure set up within the privacy computing domain to organize field transformation data and its session binding relationships. In some embodiments, association representation adopts a table-based mapping structure, including multiple mapping tables, each storing multiple mapping relationships, and each mapping relationship binding to an independent encrypted session. The mapping relationships are not used to establish a direct correspondence between plaintext fields and target debtors, but rather to establish a protected association between field transformation data and corresponding independent encrypted sessions.

[0064] Table partitioning refers to writing data with different field transformations into corresponding mapping tables according to at least one of the following: field type, matching constraints, target matching domain, or privacy transformation method. Multiple mapping tables can include identity mapping tables, numeric encrypted mapping tables, enumeration token mapping tables, and anchor verification mapping tables. They can also be further divided into sub-mapping tables corresponding to different target matching domains, different policy versions, or different data processing scopes according to the current unified domain management policy.

[0065] It should be noted that the unified jurisdiction strategy refers to a set of strategies that uniformly regulate authorization access, data isolation, matching constraints, and output control within the privacy computing domain; the denial strategy refers to the blocking, discarding, rollback, or auditing operations performed on debt relationship data when authorization verification fails.

[0066] In some embodiments, within the privacy computing domain, debt data undergoes a privacy transformation process with cryptographically irreversible properties, ensuring that, in the absence of a real-time authorization key or policy credentials, no computing resource can reconstruct the original plaintext debt or its identifiable fragments at a cost below a preset economic threshold within an effective time window. The privacy transformation process excludes pseudo-desensitization methods achieved solely through truncation, dimensionality reduction, or obfuscation. Subsequently, a privacy matching operation independent of the original plaintext is performed to generate an association representation that can be used for retrieval and comparison. Throughout the privacy matching operation, the original plaintext is prohibited from leaving the hardware or logical isolation boundary of the privacy computing domain, and plaintext data is prohibited from residing in storage media other than registers and caches during the operation cycle. Any logical branch jumps based on plaintext data values ​​are also prohibited.

[0067] Step S32: If the authorization verification of the debt relationship data fails, a denial policy is executed for the debt relationship data.

[0068] It is understood that this embodiment performs cryptographic transformation on debt relationship data within the privacy computing domain to generate an association representation without plaintext residue and irreversible derivation. By converting the original debt data into an irreversible association representation through privacy transformation within the domain, the matching operation does not need to rely on plaintext, eliminating the risk of plaintext leakage while preserving data availability. The denial policy blocks unauthorized data from entering the core computing layer, reducing the risk of privacy leakage.

[0069] In its implementation, when the authorization verification of debt relationship data passes, the server reads the current domain unified governance policy synchronized from the global policy source to the privacy computing domain. After the policy integrity verification passes, the debt relationship data is imported into the trusted isolation environment of the privacy computing domain. Based on the unified governance policy, the server determines the target data fields for matching from the debt relationship data. Within the privacy computing domain, based on the session cryptographic materials, a transformation algorithm that meets the requirements of cryptographic irreversibility is used to perform privacy transformation processing on the target data fields to obtain the association representation. For example, identity identifiers can be salted and hashed, amount fields can be homomorphically encrypted, and enumeration fields can be tokenized. If the authorization verification fails, the debt relationship data is imported into the isolation domain for discarding or auditing, and the subsequent matching process is terminated.

[0070] In some embodiments, when the verification fails, any equivalent denial means, including session blocking, operation prohibition, and state rollback, are automatically triggered; the data stream that fails the verification is forcibly blocked outside the core operation layer and mapped to any isolation domain with physical or logical isolation attributes to perform at least one of the following actions: caching, storage, discarding, archiving, and auditing; wherein, the isolation domain can be any equivalent carrier such as a sandbox, a trusted execution environment, a private storage cluster, and an isolated memory space.

[0071] Step S40: In response to the privacy matching request sent by the target data requesting terminal, authenticate the target data requesting terminal, and after successful authentication, perform privacy matching based on the association representation, generate standardized response data with externally observable features based on the privacy matching result, and send the standardized response data to the target data requesting terminal.

[0072] It should be noted that the target data request terminal refers to the terminal of a compliant data requester that initiates a privacy matching request to the privacy computing domain, such as the business terminal of a debt agency, asset management company, financial institution, or mediation institution; a privacy matching request refers to a service request initiated by the requesting party to the privacy computing domain based on its own claims or disposal needs, for the purpose of querying whether there is matching debt relationship data, such as an external search request.

[0073] It should be noted that externally observable features refer to response latency, message length, protocol fields, and interactive behaviors that can be observed by external entities from the network transport layer and application layer; standardized response data refers to response data generated after encapsulating privacy matching results according to preset rules, which maintains statistical consistency of externally observable features under different matching results.

[0074] It is understood that this embodiment completes the comparison of creditor-debtor relationships without disclosing plaintext through privacy matching, and eliminates observable side-channel features introduced by differences in matching results through standardized responses, thereby improving the system's resistance to inference attacks.

[0075] In some embodiments, the server responds to an external retrieval request and performs controlled output based on the presence or absence of an association representation. The output is a standardized response used to eliminate observable side-channel differences. The standardized response is implemented through deterministic protocol behavior, ensuring that regardless of whether a matching result exists, the complete control flow from request reception to response transmission is statistically indistinguishable in terms of timing distribution, message length distribution, and protocol layer feedback signal distribution, thus excluding abnormal branch jumps or differential protocol interactions caused by differences in response content.

[0076] Furthermore, in cases where multiple target data requesting terminals initiate privacy matching requests for the same debt relationship data, to reduce the risk of unauthorized access by unselected target data requesting terminals or cross-terminal reuse of access permissions, the above step S40 may further include: Step S41: Send the claim information corresponding to each target data request terminal to the target debtor terminal.

[0077] It should be noted that the claim information refers to the information generated based on the privacy matching request, trusted authentication result, and privacy matching result of the target data requesting terminal, which is used to represent the corresponding target data requesting terminal's request for subsequent processing of debt relationship data.

[0078] In one embodiment, the claim information does not directly contain plaintext debt relationship data that exceeds the data disclosure boundary. For example, the claim information may include at least one of the following: claim information identifier, de-identified terminal identifier corresponding to the target data request terminal, trust level, privacy matching request identifier, request time, matching result summary, requested access scope, and validity period.

[0079] Step S42: Receive the disposal selection instruction from the target debtor terminal based on the claim information.

[0080] It should be noted that a disposal selection instruction refers to an electronic control instruction generated by the target debtor terminal when selecting one target data request terminal from multiple target data request terminals based on the received claim information. The disposal selection instruction may include the identifier of the selected claim information, the data identifier of the debt relationship data, the unique identifier of the target debtor, the instruction generation time, a random number, and an instruction integrity certificate.

[0081] It should be understood that by binding disposal selection instructions with claim information, debt relationship data, and the identity information of the target debtor, the risk of disposal selection instructions being forged, altered, or replayed can be reduced, and servers can be prevented from granting data access permissions based on selection information from unknown sources or that has expired.

[0082] Step S43: Determine the target disposal terminal from multiple target data request terminals according to the disposal selection instruction, and grant disposal access permissions for the corresponding debt relationship data only to the target disposal terminal.

[0083] It should be noted that the target processing terminal refers to the target data request terminal indicated by the processing selection instruction. Processing access permissions refer to the technical access permissions that allow the target processing terminal to access or process corresponding debt relationship data through a controlled data interface within a preset scope and validity period. Processing access permissions can be jointly limited by the permitted data range, permitted operation types, validity period, permitted access frequency, and applicable application scenarios.

[0084] It should be understood that by binding disposal access permissions with the target disposal terminal and the corresponding debt relationship data, the standardized response data received by multiple target data request terminals can be separated from the subsequent debt relationship data access permissions, thus preventing unselected target data request terminals from continuing to access debt relationship data based solely on previously obtained standardized response data.

[0085] This embodiment combines the compliance root of debtor's active authorization with the transmission isolation root of independent encrypted sessions to first establish legitimacy and confidentiality at the data source and transmission stages. Then, it transforms plaintext debt data into associated representations through irreversible privacy transformation within the privacy computing domain, enabling matching operations to be performed without plaintext exposure. Therefore, this embodiment solves the problem of individual consent tracing through authorization anchoring, addresses the risks of transmission and field out-of-bounds access through independent encrypted sessions and authorization verification, mitigates the risk of data leakage during the computation stage through privacy transformation and in-domain matching, and addresses the risk of side-channel inference during the output stage through standardized responses. This achieves compliant authorization and privacy protection in debt data matching scenarios.

[0086] Based on the first embodiment described above, in the second embodiment of this application, the same or similar content as the first embodiment can be referred to the above description, and will not be repeated hereafter.

[0087] Based on this, please refer to Figure 2 , Figure 2 This is a flowchart illustrating the second embodiment of the data matching method for creditor-debtor relationships based on privacy computing according to this application. In this embodiment, step S31 further includes: Step S311: Read the current domain unified governance policy synchronized from the global policy source to the privacy computing domain.

[0088] It should be noted that the global policy source refers to the policy management node or policy contract responsible for maintaining and distributing the unified governance policy of the privacy computing domain; the current domain unified governance policy refers to the set of unified policies that take effect within this privacy computing domain after synchronization.

[0089] In the specific implementation, the global policy source distributes the policy set to each node of the privacy computing domain by timed push or on-demand retrieval. Before receiving the debt relationship data, the entry node of the privacy computing domain first reads the unified management policy of the current domain from the global policy source and caches it in the local trusted storage area.

[0090] Furthermore, to avoid policy drift or degraded execution, step S311 above may include: Step S3111: Obtain the authorization admission policy, data isolation policy, field transformation policy, matching constraint policy, and output control policy from the global policy source, and generate a global policy set.

[0091] It should be noted that the authorization access policy refers to the policy that specifies which subjects or data can enter the privacy computing domain; the data isolation policy refers to the policy that specifies how data is partitioned, stored hierarchically, and accessed within the domain; the matching constraint policy refers to the policy that specifies the fields, algorithms, and comparison rules that can be used for privacy matching; the output control policy refers to the policy that specifies which subjects can be output the matching results to and at what granularity; and the field transformation policy is used to characterize the correspondence between field types, matching constraints, privacy transformation methods, and privacy transformation materials.

[0092] Step S3112: Obtain the policy version identifier and policy integrity certificate corresponding to the global policy set.

[0093] It should be noted that the policy version identifier refers to the version number or timestamp used to distinguish different versions of the policy set; the policy integrity certificate refers to cryptographic evidence used to prove that the policy set has not been tampered with, such as digital signatures, hash digests or Merkle roots.

[0094] Step S3113: Obtain the local policy version identifier and local policy integrity certificate corresponding to each privacy computing node in the privacy computing domain.

[0095] In the specific implementation, each privacy computing node calculates the hash value of the local cached policy set during startup or periodic tasks and records the current local policy version identifier; after receiving the global policy set, the node reports the local policy version identifier and local policy integrity certificate to the global policy source or coordinating node.

[0096] Step S3114: Perform a consistency comparison between the policy version identifier and the policy integrity certificate and each local policy version identifier and each local policy integrity certificate to obtain the policy consistency result.

[0097] It should be noted that the policy consistency result refers to the conclusion obtained by comparing the version and integrity credentials of the global policy set with the corresponding local credentials of each node, such as "consistent" or "inconsistent".

[0098] In the specific implementation, the coordinating node compares the policy version identifier and policy integrity certificate of the global policy set with the local policy version identifier and local policy integrity certificate of each node one by one. If they are all consistent, a consistent policy consistency result is generated. If any node is inconsistent, the node is marked as a policy abnormal node. By comparing the policy version and integrity certificate across nodes, policy drift nodes are detected and located to ensure policy consistency within the domain.

[0099] Step S3115: If the policy consistency result is consistent, determine the global policy set as the unified governance policy for the current domain.

[0100] In the specific implementation, when the policy consistency result is consistent, the coordinating node sends a policy effective notification to each node and determines the global policy set as the unified management policy of the current domain; if there are inconsistent nodes, the operation circuit breaker or policy rollback is performed on the node to prevent abnormal nodes from performing operations based on outdated or tampered policies.

[0101] Step S312: Perform an integrity check on the current domain unified governance policy to obtain the policy integrity check result.

[0102] It should be noted that the policy integrity verification result refers to the conclusion obtained after verifying whether the current domain unified governance policy has been tampered with during transmission or storage, such as verification based on policy hash value, digital signature, etc.

[0103] In the specific implementation, the privacy computing domain node calculates the policy hash of the unified governance policy of the current domain and compares it with the policy integrity certificate published by the global policy source. If the hash values ​​are consistent or the signature verification is successful, a successful policy integrity verification result is generated; otherwise, a policy anomaly alarm is triggered and data import is refused.

[0104] Step S313: If the debt relationship data authorization verification passes and the policy integrity verification result passes, import the debt relationship data into the trusted isolation environment of the privacy computing domain.

[0105] It should be noted that a trusted isolated environment refers to a secure computing space within a privacy computing domain that has physical or logical isolation boundaries, such as a trusted execution environment, a secure sandbox, a container isolation environment, or a dedicated encrypted computing power pool.

[0106] In the specific implementation, the server migrates the debt relationship data from the transmission buffer to the memory space of the trusted isolation environment through a controlled isolation bus; the migration process performs secondary encryption or integrity verification on the data to prevent it from being intercepted or tampered with during the migration path.

[0107] In some embodiments, the trusted isolation environment of the privacy computing domain may include all equivalent secure execution carriers such as TEE trusted execution environment, security sandbox, container isolation, virtualization isolation, private cloud isolation, dedicated privacy node, encrypted computing power pool, and edge trusted privacy domain.

[0108] Step S314: Based on the current domain unified management strategy, determine the target data fields participating in privacy matching from the debt relationship data in the trusted isolation environment, and determine the field type corresponding to each target data field. The field type is an identity identifier, a numerical value, an enumeration, or an anchoring verification type.

[0109] It should be noted that the target data field refers to the subset of data fields selected from the debt relationship data according to the current unified jurisdiction strategy for use in privacy matching.

[0110] In the specific implementation, the server parses the matching constraint policy in the current domain's unified jurisdiction policy and extracts a whitelist of fields that are allowed to participate in the matching. The intersection of this whitelist and the field set of debt relationship data is taken to obtain the target data fields. For example, only fields such as debt amount range code and creditor industry code are selected, thereby reducing the data exposure surface involved in privacy transformation and matching, and retaining only the minimum set of fields required for matching.

[0111] Step S315: Within the privacy computing domain, determine the target privacy transformation method corresponding to each target data field based on the field type and matching constraints corresponding to each target data field.

[0112] Furthermore, after the target data field is imported into the privacy computing domain, it exists in plaintext only within the trusted isolation environment, and is not stored in plaintext on any storage medium outside the trusted isolation environment after the privacy transformation process is completed. The privacy transformation process is a cryptographic protection transformation performed on the target data field based on target privacy transformation material obtained from a trusted key source of the privacy computing domain; The association representation includes the field transformation data and a privacy check code generated based on the session cipher material corresponding to the independent encrypted session, so that there is a binding relationship between the association representation and the independent encrypted session.

[0113] Step S316: Using the target privacy transformation method corresponding to each target data field, perform field transformation on each target data field to obtain field transformation data, and construct the association representation based on the field transformation data.

[0114] In the specific implementation, the server, based on the field type and matching requirements of the target data field, calls the privacy transformation engine in a trusted isolation environment to perform format preservation encryption, token mapping, homomorphic encryption, or secret sharing processing on the target data field, and encodes the transformation result as an associated representation; for example, the debtor's ID number is tokenized, the debt amount is homomorphically encrypted, and the corresponding privacy verification code is generated and stored together.

[0115] Furthermore, to enhance the privacy of the associated representation during storage and transmission, and to prevent it from being tampered with or replaced, step S316 may include: Step S3161: When the target data field is an identity identifier field, a format-preserving encryption algorithm is used to perform format-preserving encryption processing on the identity identifier field to obtain format-preserving ciphertext.

[0116] It should be noted that identity identification fields refer to data fields with a fixed character set, field length, or data format, used to distinguish different data subjects or business objects. Identity identification fields can include personal identification identifiers, organizational identifiers, business entity numbers, or account identifiers. Format-preserving encryption means that the encrypted result has the same or compatible data format as the field before encryption.

[0117] In practical implementation, the server, within a trusted and isolated environment of the privacy computing domain, performs normalization processing on identity-related fields to obtain a sequence of characters to be encrypted. Normalization processing may include standardizing character encoding, standardizing letter case, removing delimiters without business meaning, or verifying field formats.

[0118] The server determines the radix parameters for FF1 format preservation encryption based on the character set of the character sequence to be encrypted, determines the corresponding encryption field based on the field length of the identity field, and obtains the FF1 format preservation encryption key, adjustment value, and key version identifier corresponding to the target matching field from a trusted key source. The server uses the FF1 format preservation encryption mode based on the Advanced Encryption Standard block cipher to perform multiple rounds of block transformation on the character sequence to be encrypted to obtain the format preservation ciphertext.

[0119] Within the same target matching domain, when equivalent matching is required for the same identity identifier, the same or compatible character normalization rules, FF1 format preserved encryption keys, and adjustment value generation rules are used to ensure that the same identity identifier can generate comparable format preserved ciphertext. For different target matching domains, different encryption keys or adjustment values ​​can be configured to reduce the possibility of direct association between the ciphertexts of identity identifiers in different target matching domains.

[0120] Step S3162: If the target data field is a numeric field, a homomorphic encryption algorithm is used to perform homomorphic encryption on the numeric field to obtain numeric ciphertext.

[0121] It should be noted that numeric fields refer to data fields used for performing summation, accumulation, difference calculations, or other numerical operations. Numeric fields can include debt amount, amount paid, remaining amount, installment amount, or other quantitative data.

[0122] It should be understood that by applying Paillier homomorphic encryption to numeric fields, addition homomorphic operations can be performed on the ciphertext without directly decrypting the numeric fields, reducing the risk of the plaintext being exposed during the computation process.

[0123] In the specific implementation, the server first converts the numeric field into plaintext integers within the Paillier encryption field based on its numeric type. When the numeric field contains decimals, a scaling factor can be determined based on the preset numeric precision to convert the decimals into integers; when the numeric field may contain negative numbers, signed encoding or modulo space mapping can be used to encode the negative numbers.

[0124] The server uses the same Paillier public key to encrypt the value to be matched and the requested value separately. It performs a homomorphic difference operation on the two ciphertexts to obtain the difference ciphertext representing the difference between the values. In a trusted isolation environment, or through a threshold decryption node, a zero-value detection is performed on the difference ciphertext. When the difference is zero, the value field is determined to match.

[0125] Step S3163: If the target data field is an enumeration type field, perform a hash-based message authentication code operation on the combination result of the target matching field salt value and the enumeration type field based on the message authentication code key to obtain the enumeration field token.

[0126] It should be noted that enumeration fields are data fields whose value range is limited by a preset enumeration set. Enumeration fields can include debt status, debt type, performance status, business category, or risk level.

[0127] It should be understood that by tokenizing enumeration fields based on the secret HMAC-SM3 key, enumeration fields with the same value within the same target matching domain can generate the same token, thereby supporting equal-value matching; at the same time, it is difficult for external entities that have not obtained the HMAC-SM3 key to recover the content of the enumeration field simply by enumerating all possible values.

[0128] In its implementation, the server first normalizes the enumeration fields according to the current domain's unified governance strategy, obtaining normalized enumeration values. Then, it obtains the HMAC-SM3 key, the target matching domain salt value, and the corresponding cryptographic material version identifier from a trusted key source.

[0129] The server combines the target matching field salt value and the normalized enumeration value in a preset order, and performs HMAC-SM3 operation on the combination result using the HMAC-SM3 key pair to obtain the enumeration field token.

[0130] Enumerated fields within the same target matching domain that require equality matching should use the same or repeatedly generated target matching domain salt value and a compatible HMAC-SM3 key. The target matching domain salt value cannot be randomly changed for each data entry; otherwise, the same enumeration value will generate different tokens, making direct equality matching impossible. Different target matching domains can be configured with different salt values ​​or different HMAC-SM3 keys to reduce the risk of enumeration field tokens being associated across domains.

[0131] Step S3164: If the target data field is an anchor verification field, combine the anchor salt value with the anchor verification field, and perform cryptographic hashing on the combination result to obtain the anchor digest.

[0132] It should be noted that anchored validation fields are data fields used to verify whether the data content, data version, or authorization relationship has changed. Anchored validation fields can include authorization credential validation fields, data record validation fields, record version information, or other fields that require tamper-proof verification.

[0133] It should be understood that by combining the anchor salt value with the anchor verification field and then performing SM3 hashing, an anchor digest corresponding to the field content can be generated without directly saving the plaintext of the field, so as to verify the integrity and consistency of the field later.

[0134] In its implementation, the server obtains the anchor salt value from a trusted key source or a secure random number generator. It then combines the anchor salt value with a normalized anchor verification field according to a preset encoding order and performs SM3 hashing on the result to obtain the anchor digest. The anchor salt value or its corresponding salt identifier is associated with and stored in the anchor digest. During subsequent consistency checks, the server obtains the same anchor salt value, recalculates the SM3 digest of the field to be verified according to the same normalization rules and encoding order, and compares the recalculated digest with the anchor digest in the associated representation.

[0135] Step S3165: Determine the format-preserved ciphertext, the numerical ciphertext, the enumerated field token, or the anchor digest as the field transformation data corresponding to the target data field; Step S3166: Generate a privacy check code based on the session cipher material and the field transformation data corresponding to each target data field, and associate and encode each field transformation data with the privacy check code to obtain the association representation within the privacy computing domain.

[0136] It should be noted that a privacy verification code refers to verification data generated based on session cipher materials and field transformation data, used to verify the integrity of the field transformation data and the session origin binding relationship. A privacy verification code can be a field-level privacy verification code or a session-level privacy verification code generated for multiple field transformation data.

[0137] In one embodiment, the server derives a session verification key based on the session cipher material corresponding to the independent encrypted session, and performs message authentication operations on the normalized encoding results of the field transformation data, encrypted session identifier, field type identifier, and policy version identifier to obtain a field-level privacy verification code.

[0138] In another embodiment, the server normalizes and encodes multiple field transformation data according to a preset field order, and performs message authentication operations on the encoding results based on the session verification key to obtain a session-level privacy verification code covering the multiple field transformation data. Each field transformation data is bound to the same independent encrypted session through a common session-level privacy verification code and a session binding identifier.

[0139] The server determines the target mapping table corresponding to the transformed data of each field based on the field type, target matching domain, and privacy transformation method, and establishes mapping relationships in the target mapping table. Each mapping relationship may include a mapping relationship identifier, field transformation data, privacy verification code, session binding identifier, policy version identifier, and cryptographic material version identifier. Each mapping relationship is bound to only one independent encrypted session, and one independent encrypted session can correspond to multiple mapping relationships formed by multiple target data fields.

[0140] Multiple mapping tables and their stored mapping relationships together constitute the association representation within the privacy computing domain. The association representation does not directly store the plaintext of the target data field, session cipher material, FF1 encryption key, Paillier private key, or HMAC-SM3 key.

[0141] It should be noted that the FF1 format retention encryption, Paillier homomorphic encryption, HMAC-SM3 operation and SM3 hash processing in the above embodiments are only examples. Those skilled in the art can use other format retention encryption algorithms, homomorphic encryption algorithms, hash-based message authentication code algorithms or cryptographic hash algorithms with equivalent functions.

[0142] In some embodiments, refer to Figure 3 , Figure 3 This is a flowchart illustrating the process of trusted authentication of a target data requesting terminal based on a single root of trust. For application environments with only one trusted authentication system, a privacy-based creditor-debtor relationship data matching system can perform trusted authentication of the target data requesting terminal based on a single root of trust, and then perform privacy matching after successful trusted authentication. The specific steps include: Step S301: In response to the privacy matching request sent by the target data request terminal, obtain the target trusted credential of the target data request terminal, and determine the single root of trust based on the issuer identifier carried by the target trusted credential; Step S302: Perform a trust verification on the target trusted credential based on a single root of trust to obtain the trusted credential verification result; Step S303: Compare the terminal identifier in the target trusted credential with the sending terminal identifier corresponding to the privacy matching request, and verify the request signature of the privacy matching request to obtain the request binding verification result; Step S304: If both the trusted credential verification result and the request binding verification result are passed, determine that the trusted authentication of the target data request terminal is passed, convert the privacy matching request into a request association representation, and perform a privacy comparison between the request association representation and the association representation to obtain the privacy matching result; Step S305: If the trusted credential verification result or the request binding verification result fails, determine that the trusted authentication of the target data request terminal has failed, and execute a denial policy for the privacy matching request.

[0143] It should be noted that a single root of trust refers to a root key, root certificate, hardware trust benchmark, or remote proof trust benchmark used to verify the target trusted credentials; a single root of trust corresponds to only one trusted authentication source and does not require cross-verification of the verification result through another independent root of trust. Trusted verification may include at least one of the following: signature chain verification, validity period verification, revocation status verification, authorized use verification, proof signature verification, device metric verification, and random challenge response verification.

[0144] In the specific implementation, the server calls the pre-configured single root of trust based on the issuer identifier carried by the target trusted credential to verify the signature chain, validity period, revocation status and authorized purpose of the target trusted credential. When the target trusted credential is a hardware certificate or a trusted execution environment certificate, the server can further verify the certificate signature, device metric and random challenge response to determine whether the certificate data was generated by the corresponding trusted hardware or trusted execution environment.

[0145] For example, a single root of trust could be an organization certificate root. The server verifies the device certificate submitted by the target data requesting terminal based on this organization certificate root and reads the request body type and the allowed matching service types from the extended fields of the device certificate. Alternatively, a single root of trust could be a hardware security root. The server verifies the device credentials submitted by the target data requesting terminal using this hardware security root and determines whether the device has the permission to initiate a privacy matching request based on a preset permission mapping table.

[0146] The server compares the terminal identifier in the target trusted credential with the sending terminal identifier of the privacy matching request, and verifies whether the request signature covers the request data, request random number, and encrypted session identifier; it can also query the already used request random number or request sequence number to determine whether the current privacy matching request is a duplicate request.

[0147] When the target trusted credential is verified successfully and the binding relationship between the target trusted credential and the privacy matching request is consistent, a trusted authentication success result is generated; when the target trusted credential is invalid, expired, revoked, has an incorrect authorized purpose, or has an inconsistent binding relationship, a trusted authentication failure result is generated.

[0148] When trusted authentication is successful, the server parses the matching fields and matching types in the privacy matching request, and converts the request data into a request association representation according to the same or compatible privacy transformation rules as the intra-domain association representation. The server then performs ciphertext equivalence comparison, token comparison, ciphertext set intersection, or other privacy comparisons adapted to the privacy transformation method on the request association representation and the intra-domain association representation to obtain the privacy matching result.

[0149] When trusted authentication fails, the server will not import the privacy matching request into the actual matching process, and will implement a rejection policy for the privacy matching request to prevent the request that failed trusted authentication from triggering intra-domain data queries.

[0150] It should be noted that the above-described trusted authentication based on a single root of trust is applicable to application environments that deploy only one trusted authentication system. In other embodiments where the trustworthiness of the requesting terminal is more critical, the first and second trusted credentials can be verified based on two independent roots of trust, and heterogeneous trusted authentication can be performed by combining the results of the two trusted verifications. The following description, in conjunction with the third embodiment of the debt-creditor relationship data matching method based on privacy computing of this application, further illustrates the heterogeneous trusted authentication method based on dual roots of trust.

[0151] Based on the above embodiments, please refer to Figure 4 , Figure 4 This is a flowchart illustrating the third embodiment of the data matching method for creditor-debtor relationships based on privacy computing in this application. In this third embodiment, step S40 further includes: Step S401: In response to the privacy matching request sent by the target data request terminal, obtain the first trusted credential and the second trusted credential of the target data request terminal, wherein the first trusted credential and the second trusted credential correspond to different roots of trust.

[0152] It should be noted that the first trusted credential and the second trusted credential refer to two types of credentials issued by different roots of trust, used to prove the trustworthiness of the target data requesting terminal, such as hardware trusted credentials and environment trusted credentials; the root of trust refers to the security infrastructure components that endorse trusted credentials, such as hardware security modules, trusted execution environments, or third-party authentication services.

[0153] In the specific implementation, when the target data requesting terminal sends a privacy matching request, it submits the first trusted credential and the second trusted credential together; the server extracts the signature, certificate chain or remote proof report from the credential respectively to prepare for subsequent verification.

[0154] Step S402: Perform a trust verification on the first trusted credential based on the first root of trust to obtain the first trust verification result.

[0155] It should be noted that the first root of trust refers to an independent root of trust that endorses the first trusted credential, such as a device hardware root of trust; the first trusted verification result refers to the conclusion obtained after verifying the first trusted credential based on the first root of trust.

[0156] In the specific implementation, the server calls the public key or verification interface corresponding to the first root of trust to verify the signature, certificate validity period and device identity of the first trusted credential; if the verification passes, a successful first trusted verification result is generated.

[0157] Step S403: Perform a trust verification on the second trusted credential based on the second root of trust to obtain the second trust verification result.

[0158] It should be noted that the second root of trust refers to a root of trust that endorses the second trusted credential and is physically or logically independent of the first root of trust, such as the runtime environment root of trust or the network fingerprint root of trust; the second trusted verification result refers to the conclusion obtained after verifying the second trusted credential based on the second root of trust.

[0159] In the specific implementation, the server calls the verification interface corresponding to the second root of trust to verify the environmental metric, network fingerprint or behavioral risk control score in the second trusted credential; if the verification passes, a passed second trusted verification result is generated.

[0160] Step S404: Based on the first trusted verification result and the second trusted verification result, perform heterogeneous trusted authentication on the target data request terminal to obtain heterogeneous trusted authentication result.

[0161] It should be noted that heterogeneous trusted authentication refers to the process of making a comprehensive trust determination on the target data request terminal based on the trusted verification results of two or more independent trust roots; the heterogeneous trusted authentication result refers to the conclusion of this comprehensive determination.

[0162] In the specific implementation, the server performs a weighted OR logical AND operation on the first trusted verification result and the second trusted verification result; only when all dimensions meet the preset thresholds will a passed heterogeneous trusted authentication result be generated; if any dimension fails, the access of the requesting terminal will be rejected.

[0163] Step S405: If the heterogeneous trusted authentication result is passed, the privacy matching request is converted into a request association representation, and the request association representation is compared with the association representation to obtain a privacy matching result.

[0164] In the specific implementation, the server imports the fields to be compared in the privacy matching request into a trusted isolation environment of the privacy computing domain, and generates a request association representation according to the same privacy transformation method as the target data field; then it performs secure intersection, homomorphic equivalence comparison or vector similarity calculation to obtain the privacy matching result.

[0165] Step S406: Generate standardized response data of externally observable features based on the privacy matching result, and send the standardized response data to the target data request terminal.

[0166] It should be understood that this embodiment eliminates the correlation between externally observable features and matching results by encapsulating and standardizing the matching results, thereby suppressing side-channel inference.

[0167] In the specific implementation, the server generates an encrypted response message based on the privacy matching result and the data disclosure policy, and uses fixed-length padding, constant-time algorithm and unified protocol fields to standardize the external observable features of the encrypted response message; then, according to the preset response sequence, the standardized response data is sent to the target data request terminal.

[0168] Furthermore, step S406 above may include: Step S4061: Configure the hierarchical disclosure strategy and / or the no-disclosure closed-loop strategy to obtain the data disclosure strategy set.

[0169] It should be noted that the hierarchical disclosure strategy includes configuring the debt information interaction entry point, setting the corresponding desensitization granularity and open scope based on different access permission levels, and only allowing target data request terminals with the corresponding access permission level to access the corresponding level of debt information; The non-public closed-loop strategy includes closing the public search entry, restricting privacy matching and the generation of privacy matching results to closed-loop execution within the privacy computing domain, and outputting the standardized response data generated based on the privacy matching results only to the target data request terminal that has passed the heterogeneous trusted authentication.

[0170] In some embodiments, a tiered disclosure strategy may involve establishing a debt information exchange hall, implementing gradient desensitization and tiered access control, allowing only authorized entities to access information at the corresponding level. A closed-loop, non-disclosure strategy may involve not providing a public search entry point, with matching operations and result generation executed within a closed loop within the domain, and results only directed to legitimate entities authenticated from a different source. The server adaptively reconstructs the desensitization granularity and scope of the strategy based on compliance requirements and user permissions; any equivalent substitutions to the display format, deployment architecture, or exposed boundaries are protected.

[0171] In practice, during the initialization or policy update phase, the server selects one or both of the following strategies to configure: a tiered disclosure strategy and a closed-loop strategy without disclosure, depending on the deployment scenario. For example, in an assistance scenario where public retrieval is completely prohibited, only the closed-loop strategy without disclosure is configured; in an institution matching scenario where licensed institutions are allowed to retrieve information, only the tiered disclosure strategy is configured; and in a scenario where switching is required based on business time periods, both strategies are configured and their corresponding parameter templates are bound to each strategy.

[0172] Step S4062: Determine the access permission level of the target data request terminal based on the heterogeneous trusted authentication result.

[0173] It should be noted that access permission levels refer to the level identifiers assigned to data requesting terminals based on the results of cross-source trusted authentication, which are used to limit the scope of data that the terminals can access. For example, level 1 allows viewing summary statistics, level 2 allows viewing anonymized debt information, and level 3 allows viewing complete matching results.

[0174] In the specific implementation, the server queries the permission mapping table based on the credential type, institutional qualifications, and risk control score in the heterogeneous trusted authentication results, and assigns access permission levels to the target data request terminal. For example, a terminal with only a single root trust credential is assigned the lowest level, while a terminal with two roots trust and whose qualifications have been approved can be assigned a higher level.

[0175] Step S4063: Determine the target data transparency strategy from the data transparency strategy set according to the preset application scenario parameters and the access permission level.

[0176] It should be noted that the application scenario parameters refer to the configuration parameters used to describe the business scenario in which the current privacy matching request is located, such as scenario identifier, compliance level requirements, or regulatory area code; the target data disclosure policy refers to the data disclosure policy finally selected from the set of data disclosure policies and used for the current request response.

[0177] In the specific implementation, the server reads the preset application scenario parameters and determines whether the current scenario allows public search entry. If the data disclosure strategy set only contains the non-public closed-loop strategy, it is directly determined as the target data disclosure strategy. If it only contains the hierarchical public strategy, the target data disclosure strategy is determined from the hierarchical public strategy according to the access permission level. If it contains both strategies, one is selected according to the public search entry configuration status and access permission level, and the target data disclosure strategy is determined after further filtering based on the data disclosure boundary.

[0178] Furthermore, in order to achieve dynamic adaptation and hard constraints on compliance boundaries during the data disclosure strategy selection phase, step S4063 above may include: Step S40631: Obtain the configuration status of the public search entry and the data exposure boundary according to the preset application scenario parameters.

[0179] It should be noted that the public search entry configuration status refers to the configuration identifier of whether the system provides a public search entry, such as on or off; the data exposure boundary refers to the maximum range of data that is allowed to be output to the outside in the current application scenario, such as limiting the output object, field set or network area.

[0180] In the specific implementation, the server parses the preset application scenario parameters and reads the configuration status and data exposure boundaries of the public retrieval entry. For example, in the assistance scenario, the configuration status is closed and the boundary is limited to the specified dedicated line. In the institutional matching scenario, the configuration status is open and the boundary is limited to the licensed institution.

[0181] Step S40632: When the configuration status of the public search entry is closed, the no-public closed-loop strategy is determined as the first candidate data exposure strategy.

[0182] It should be noted that the first candidate data disclosure strategy refers to the non-public closed-loop strategy that is preferred when the public search portal is closed; the non-public closed-loop strategy is preferred when the public search portal is closed to block the risk of information leakage brought about by public search.

[0183] In the specific implementation, when the public retrieval entry configuration status is closed, the server checks whether there is a closed-loop policy without public disclosure in the data disclosure policy set; if it exists, it is determined as the first candidate data disclosure policy; if it does not exist, a policy missing alarm is triggered or it falls back to the default strictest policy.

[0184] Step S40633: When the public search entry configuration status is enabled, determine the second candidate data disclosure strategy from the hierarchical disclosure strategy according to the access permission level.

[0185] It should be noted that the second candidate data disclosure strategy refers to the candidate strategy determined from the hierarchical disclosure strategy based on the access permission level when the public search portal is enabled; in scenarios where public search is allowed, hierarchical disclosure candidates are provided so that requesters with different permissions can access the corresponding level of information.

[0186] In the specific implementation, when the public retrieval entry configuration status is enabled, the server checks whether there is a hierarchical public disclosure policy in the data disclosure policy set. If it exists, the server determines the desensitization granularity and open scope corresponding to the access permission level of the target data request terminal from the hierarchical public disclosure policy as the second candidate data disclosure policy. If it does not exist, the server triggers a policy missing alarm or falls back to the default strictest policy.

[0187] Step S40634: Compare the allowed output range of the corresponding candidate data transparency strategy with the data transparency boundary to obtain the transparency boundary comparison result.

[0188] It should be noted that the allowed output range refers to the range of data objects, fields, and networks that the candidate data disclosure strategy allows to be output; the disclosure boundary comparison result refers to the conclusion obtained by comparing the allowed output range with the data disclosure boundary, such as "within the boundary" or "outside the boundary".

[0189] In the specific implementation, the server compares the allowed output range of the first or second candidate data disclosure strategy with the data disclosure boundary item by item. If the allowed output range falls entirely within the boundary, a disclosure boundary comparison result within the boundary is generated. If there are any items exceeding the boundary, an out-of-boundary result is generated and a strategy pruning or alarm is triggered. This prevents the actual output of the candidate strategy from exceeding the compliance boundary of the current scenario through boundary comparison.

[0190] Step S40635: Based on the comparison results of the transparent boundary, determine the first candidate data transparent strategy or the second candidate data transparent strategy as the target data transparent strategy.

[0191] In the specific implementation, the server determines the first or second candidate data disclosure strategy within the boundary as the target data disclosure strategy based on the boundary comparison results; if both candidates exceed the boundary, the server rejects the response or falls back to the strictest closed-loop strategy.

[0192] Step S4064: Encapsulate the privacy matching result according to the target data disclosure policy to obtain an encrypted response message.

[0193] In practical implementation, the server determines the output fields and anonymization rules based on the target data disclosure policy. It then filters, anonymizes, and encrypts the privacy matching results, generating an encrypted response message. For example, under a closed-loop policy, only a boolean value indicating whether a match exists is output; under a hierarchical disclosure policy, a gradient-anonymized debt summary is output. Through encryption and policy encapsulation, the matching results are protected by access control and confidentiality before leaving the domain.

[0194] Step S4065: Standardize the external observable features of the encrypted response message according to the preset message length, preset response sequence and preset protocol fields to obtain standardized response data.

[0195] It should be noted that the preset message length refers to the fixed or finite set of message lengths used in standardized response data; the preset response timing refers to the fixed or statistically indistinguishable timing pattern followed from request reception to response transmission; and the preset protocol fields refer to the unified field structure used by the response message at the protocol layer.

[0196] In the specific implementation, the server pads the encrypted response message to a fixed length to reach the preset message length; a constant-time algorithm is used to constrain the processing time within a preset response sequence range; and status codes, sequence numbers, and verification fields are filled according to preset protocol fields to ensure consistent message structure under different matching results. By unifying the external observable characteristics of the response, side-channel differences between matching and non-matching scenarios are eliminated.

[0197] Step S4066: Send the standardized response data to the target data request terminal based on the preset response timing; In its implementation, when the preset response time arrives, the server sends standardized response data to the target data request terminal via an encrypted channel. After transmission, an audit log is recorded, but the plaintext of the matching result or intermediate states are not recorded. Furthermore, this audit log is an auxiliary operation and maintenance audit measure and does not constitute the anti-tampering anchoring described in step S10. This allows for flexible selection of tiered compliance disclosure or fully closed-loop output, along with side-channel suppression, during the result output stage.

[0198] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the data matching method for creditor-debtor relationships based on privacy computing in this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0199] In some embodiments, the privacy-based computation-based debt-claim matching method of this application is also applied to the target debtor terminal. The target debtor terminal may be a mobile terminal, a computer terminal, a business processing terminal, or other electronic devices with data collection and communication functions.

[0200] In its implementation, the target debtor's terminal obtains the target debtor's unique identifier, individual authorization certificate, and debt relationship data. The individual authorization certificate can be generated by the server based on the proactive confirmation information, data processing scope, and unique identifier submitted by the target debtor through an independent authorization page, or it can be generated by an authorized independent authentication service. The specific content of the unique identifier, individual authorization certificate, and debt relationship data can be found in the aforementioned server-side embodiment.

[0201] The target debtor's terminal sends a unique identifier and a separate authorization credential to the server, enabling the server to perform tamper-proof anchoring on both, thus obtaining a tamper-proof authorization record. The specific process of the server generating the tamper-proof authorization record can be found in the aforementioned embodiment regarding tamper-proof anchoring.

[0202] Subsequently, the target debtor's terminal sends an encrypted session request to the server and, based on the encrypted session response returned by the server, establishes an independent encrypted session corresponding to the target debtor's terminal. The target debtor's terminal sends debt relationship data to the server through this independent encrypted session. The establishment of the independent encrypted session, the generation of session cipher materials, and the encrypted transmission method of the debt relationship data can be found in the aforementioned embodiment regarding independent encrypted sessions.

[0203] After receiving the debt relationship data, the server performs authorization verification on the data based on the tamper-proof authorization record. If the authorization verification passes, the debt relationship data is imported into a privacy computing domain constrained by a unified governance policy, and privacy transformation processing is performed on the data to obtain an association representation for privacy matching. The specific methods for authorization verification, privacy transformation processing, and association representation construction can be found in the aforementioned server-side embodiment.

[0204] It should be understood that this embodiment submits a unique identity and a separate authorization credential through the target debtor's terminal, and sends debt relationship data using an independent encrypted session, enabling the server to verify the authorization status and uplink source of the debt relationship data, and reducing the risk of exposure of the debt relationship data during transmission and access.

[0205] In some embodiments, the privacy-based computation method for matching creditor and debtor relationship data is applied to a target data request terminal. The target data request terminal may be a creditor business terminal, a financial institution terminal, a regulatory business terminal, or other electronic devices with a need for matching creditor and debtor relationship data.

[0206] In its implementation, the target data requesting terminal acquires privacy matching request data for requesting matching of creditor and debtor relationship data, and generates a privacy matching request based on the privacy matching request data. The privacy matching request data may include at least one of the following: information of the object to be matched, target matching fields, matching constraints, target matching domain identifier, and application scenario parameters.

[0207] The target data requesting terminal sends a privacy matching request to the server, enabling the server to perform privacy matching based on the privacy matching request and a pre-generated association representation. The server can convert the privacy matching request into a request association representation that uses the same or compatible privacy transformation rules as the association representation, and then perform a privacy comparison between the request association representation and the association representation to obtain a privacy matching result. The specific process of request association representation conversion, trusted authentication, and privacy comparison can be found in the aforementioned server-side embodiment.

[0208] The server generates response data based on the privacy matching results, controls the output content of the response data according to the target data disclosure policy, and standardizes externally observable characteristics such as message length, response timing, and protocol fields to obtain standardized response data. The process of generating standardized response data can be found in the aforementioned embodiments regarding the hierarchical disclosure policy, the non-disclosure closed-loop policy, and the standardization of externally observable characteristics.

[0209] The target data requesting terminal receives standardized response data sent by the server, performs integrity verification, decryption, and protocol parsing on the standardized response data, and obtains a matching response result corresponding to the privacy matching request. The content of the matching response result is determined by the target data requesting terminal's access permission level and the target data disclosure policy.

[0210] It should be understood that this embodiment generates and sends a privacy matching request with definite matching conditions through the target data request terminal, enabling the server to perform controlled privacy matching based on the association representation; by receiving and parsing the response data with externally observable features after standardization, it can obtain the matching response results within the scope of authority, while reducing the risk that external subjects will infer the authentication status or privacy matching results based on the response differences.

[0211] This application also provides a data matching system for creditor-debtor relationships based on privacy-preserving computation; please refer to [reference needed]. Figure 5 , Figure 5 This is a schematic diagram of the module structure of a data matching system for creditor-debtor relationships based on privacy computing. The system includes a server 10, a target debtor terminal 20, and a target data request terminal 30. The target debtor terminal 20 is used to obtain the target debtor's unique identifier, individual authorization certificate, and debt relationship data; and to send the unique identifier and individual authorization certificate to the server. The server 10 is used to perform anti-tampering anchoring on the unique identity identifier and individual authorization certificate obtained from the target debtor's terminal to obtain anti-tampering authorization records; The target debtor terminal 20 is also used to send an encrypted session request to the server to establish an independent encrypted session with the server; and to send debt relationship data to the server through the independent encrypted session. The server 10 is further configured to receive debt relationship data through an independent encrypted session corresponding to the target debtor's terminal, and perform authorization verification on the debt relationship data according to the tamper-proof authorization record; if the authorization verification of the debt relationship data passes, the debt relationship data is imported into a privacy computing domain constrained by a unified governance policy, and privacy transformation processing is performed on the debt relationship data to obtain an association representation; if the authorization verification of the debt relationship data fails, a denial policy is executed on the debt relationship data. The target data request terminal 30 is configured to acquire privacy matching request data for requesting matching of creditor and debtor relationship data; generate a privacy matching request based on the privacy matching request data; and send the privacy matching request to the server. The server 10 is further configured to respond to a privacy matching request sent by a target data request terminal, perform authentication on the target data request terminal, and after successful authentication, perform privacy matching based on the association representation, generate standardized response data with externally observable features based on the privacy matching result, and send the standardized response data to the target data request terminal. The target data request terminal 30 is also used to receive standardized response data sent by the server; and to parse the standardized response data to obtain a matching response result corresponding to the privacy matching request.

[0212] In some embodiments, debtors initiate debt compliance processing requests via personal terminals such as mobile apps and mini-programs. The system displays a separate authorization page in a pop-up window, requiring only the debtor's active confirmation to obtain individual authorization for sensitive information processing. The system relies on a TEE trusted execution environment or blockchain ledger to hardware-level tamper-proof anchoring of the debtor's authorization credentials to a unique global identity identifier, excluding temporary transient identifiers and soft log records, thus solidifying a traceable authorization trajectory across the entire chain.

[0213] After authorization, debtors independently enter basic debt information, and the system completes the discrete uploading of individual data points through independent encrypted sessions and independent uplink channels. The system's uplink firewall strictly blocks batch aggregation requests from creditors and institutional servers, allowing only individual terminals to proactively upload data, thus preventing centralized batch data aggregation and mitigating the risk of batch leakage at the source.

[0214] In some embodiments, after uploaded data passes authorization verification, it is incorporated into a privacy-preserving computation isolation domain. The system performs privacy transformation using an irreversible cryptographic algorithm with a security strength of 128 bits or higher, abandoning pseudo-desensitization methods such as truncation and dimensionality reduction, and generating a plaintext-free, irreversibly derivable association representation. Plaintext is strictly prohibited from leaving the isolation domain throughout the entire operation; it is only loaded momentarily in registers and caches, with no plaintext residing across media. Furthermore, the entire process adopts a branchless execution paradigm with no data-driven logical jumps.

[0215] For example, when law firms, mediation institutions, asset management companies, and other entities initiate claims, the system uses a dual heterogeneous trust root (hardware and environment) to complete multi-dimensional heterogeneous authentication, outputting anonymized matching results only to entities that pass authentication. Multiple institutions can initiate claims simultaneously, with the final disposal option entirely belonging to the debtor, achieving a reverse matching architecture and eradicating the chaos of passive debt collection. The server receives disposal selection instructions from the target debtor's terminal, determines the target disposal terminal from the target data request terminals, and only grants disposal access permissions for the corresponding debt relationship data to that target disposal terminal.

[0216] In some embodiments, under high-compliance regulatory scenarios, the system enables a closed-loop mode without public disclosure, closing all public search entry points and public disclosure halls, and completing all matching operations within the TEE privacy computing domain in a closed loop. After an external search request is received, the system performs branchless operations based on a constant-time algorithm. Regardless of whether there is a matching representation, the response latency, message length, TCP / IP protocol behavior, and application layer feedback are completely consistent, and statistical tests show that they meet the indistinguishability criterion of a p-value greater than 0.05.

[0217] This application's privacy-based creditor-debtor relationship data matching system completely abandons pseudo-protection measures such as random delays. Through full-path timing control and unified behavior across the entire protocol stack, it completely eliminates side-channel observable differences, blocks all attack paths such as member inference and traversal probing, and achieves the highest level of privacy and security protection.

[0218] Reference Figure 6 , Figure 6 This is a schematic diagram of a scenario in one embodiment of the privacy-based computation-based debt and creditor relationship data matching system of this application. The server first sends an independent authorization page to the target debtor's terminal, receives authorization confirmation information and identity identifier from the debtor, generates an authorization credential, and performs tamper-proof anchoring to obtain a tamper-proof authorization record. Subsequently, the target debtor's terminal sends debt relationship data to the server through an independent encrypted session. The server performs authorization verification on the upstream source and field range of the data based on the tamper-proof authorization record. If the authorization verification passes, the server imports the debt relationship data into the privacy computation domain, performs cryptographic protection transformation in a trusted isolated environment, and constructs an association representation through table partitioning for subsequent privacy matching. If the authorization verification fails, the server executes a denial policy. Target data requesting terminals A, B, and C send privacy matching requests to the server respectively. After completing terminal authentication and privacy matching, the server provides standardized responses and output control for the matching results. When multiple target data requesting terminals match the same debt relationship data, the server, based on the disposal selection instructions fed back by the target debtor terminal, identifies terminal A as the target disposal terminal and grants it disposal access permissions, while denying unselected terminals B and C access to the corresponding debt relationship data.

[0219] The privacy-based computation-based debt and claim relationship data matching system provided in this application employs the privacy-based computation-based debt and claim relationship data matching method described in the above embodiments, and can solve the technical problems to be addressed by the aforementioned privacy-based computation-based debt and claim relationship data matching method. Compared with the prior art, the beneficial effects of the privacy-based computation-based debt and claim relationship data matching system provided in this application are the same as those of the privacy-based computation-based debt and claim relationship data matching method provided in the above embodiments, and other technical features of the privacy-based computation-based debt and claim relationship data matching system are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0220] This application provides a data matching device for creditor-debtor relationships based on privacy computing. The data matching device for creditor-debtor relationships based on privacy computing includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the data matching method for creditor-debtor relationships based on privacy computing in any of the above embodiments.

[0221] The following is for reference. Figure 7 The diagram illustrates a structural schematic of a data matching device for creditor-debtor relationships based on privacy computing, suitable for implementing embodiments of this application. The data matching device for creditor-debtor relationships based on privacy computing in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (Personal Digital Assistants), PADs (Portable Android Devices), etc., and fixed terminals such as desktop computers. Figure 7 The privacy-based computing-based debt and creditor relationship data matching device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0222] like Figure 7As shown, the privacy-based computing-based debt and creditor relationship data matching device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in ROM 1002 (Read Only Memory) or a program loaded from storage device 1003 into RAM 1004 (Random Access Memory). RAM 1004 also stores various programs and data required for the operation of the privacy-based computing-based debt and creditor relationship data matching device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via bus 1005. I / O interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, LCDs (Liquid Crystal Displays), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the privacy-based computational debt-creditor data matching device to communicate wirelessly or wiredly with other devices to exchange data. Although the figure shows a privacy-based computational debt-creditor data matching device with various systems, it should be understood that it is not required to implement or possess all of the systems shown. More or fewer systems may be implemented alternatively.

[0223] The privacy-based computation-based debt and claim relationship data matching device provided in this application, employing the privacy-based computation-based debt and claim relationship data matching method described in the above embodiments, can solve the technical problems to be addressed by the aforementioned privacy-based computation-based debt and claim relationship data matching method. Compared with the prior art, the beneficial effects of the privacy-based computation-based debt and claim relationship data matching device provided in this application are the same as those of the privacy-based computation-based debt and claim relationship data matching method provided in the above embodiments, and other technical features in this privacy-based computation-based debt and claim relationship data matching device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0224] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0225] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0226] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the privacy-based computing-based debt and creditor relationship data matching method in the above embodiments.

[0227] The computer-readable storage medium provided in this application can be a tangible medium capable of storing computer programs, such as a hard disk, random access memory, read-only memory, flash memory, optical storage device, or magnetic storage device. When the computer program is executed by a processor, it implements the aforementioned method for matching creditor and debtor relationship data based on privacy computing.

[0228] The aforementioned computer-readable storage medium may be included in a privacy-based computing-based debt and creditor relationship data matching device; or it may exist independently and not be assembled into a privacy-based computing-based debt and creditor relationship data matching device.

[0229] The aforementioned computer-readable storage medium carries one or more programs. When the aforementioned one or more programs are executed by the privacy-based computation-based debt and claim relationship data matching device, the privacy-based computation-based debt and claim relationship data matching device performs the privacy-based computation-based debt and claim relationship data matching method as described in the above embodiments.

[0230] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including LAN (Local Area Network) or WAN (Wide Area Network)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0231] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0232] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the aforementioned privacy-based computation-based debt and claim relationship data matching method, thereby solving the technical problems to be addressed by the aforementioned privacy-based computation-based debt and claim relationship data matching method. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as the beneficial effects of the privacy-based computation-based debt and claim relationship data matching method provided in the above embodiments, and will not be repeated here.

[0233] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for matching creditor-debtor relationship data based on privacy computing, characterized in that, The method is applied to a server, and the method includes: The unique identifier and individual authorization certificate obtained from the target debtor's terminal are tamper-proofed and anchored to obtain tamper-proof authorization records; Debt relationship data is received through an independent encrypted session corresponding to the target debtor's terminal, and the debt relationship data is authorized and verified according to the tamper-proof authorization record; If the authorization verification of the debt relationship data passes, the debt relationship data is imported into a privacy computing domain constrained by a unified jurisdiction policy, and privacy transformation processing is performed on the debt relationship data to obtain an association representation; if the authorization verification of the debt relationship data fails, a denial policy is executed on the debt relationship data. In response to a privacy matching request sent by a target data requesting terminal, authentication is performed on the target data requesting terminal. After successful authentication, privacy matching is performed based on the association representation. Based on the privacy matching result, standardized response data with externally observable features is generated and sent to the target data requesting terminal.

2. The method as described in claim 1, characterized in that, The process of tamper-proof anchoring of the unique identity identifier and individual authorization certificate obtained from the target debtor's terminal to obtain tamper-proof authorization records includes: Send independent authorization page information to the target debtor's terminal so that the target debtor's terminal can display the data processing scope and authorization confirmation control based on the independent authorization page information; Receive proactive confirmation information and unique identity identifier from the target debtor's terminal based on the authorization confirmation control; A separate authorization credential is generated based on the proactive confirmation information, the data processing scope, and the unique identifier. The binding relationship between the unique identifier and the individual authorization credential is written into the tamper-proof storage medium to obtain the tamper-proof authorization record.

3. The method as described in claim 1, characterized in that, The step of receiving debt relationship data through an independent encrypted session corresponding to the target debtor's terminal, and verifying the authorization of the debt relationship data according to the tamper-proof authorization record, includes: Establish an independent encrypted session with the target debtor's terminal to obtain an encrypted session identifier corresponding to the target debtor's terminal; The debt relationship data sent by the target debtor's terminal is received through the independent encrypted session, and the debt relationship data is associated with the encrypted session identifier; Based on the encrypted session identifier and the unique identity identifier in the tamper-proof authorization record, the upstream source of the debt relationship data is verified to obtain the source verification result; Based on the data processing scope in the anti-tampering authorization record, the data fields of the debt relationship data are validated to obtain the scope validation result; Based on the source verification result and the range verification result, the debt relationship data is subject to authorization verification.

4. The method as described in claim 1, characterized in that, If the authorization verification of the debt relationship data passes, the debt relationship data is imported into a privacy computing domain constrained by a unified jurisdiction policy, and privacy transformation processing is performed on the debt relationship data to obtain an association representation, including: Read the current domain unified governance policy synchronized from the global policy source to the privacy computing domain; Perform an integrity check on the current domain unified governance policy to obtain the policy integrity check result; If the authorization verification of the debt relationship data passes and the policy integrity verification result passes, the debt relationship data will be imported into the trusted isolation environment of the privacy computing domain. According to the current domain unified jurisdiction policy, target data fields participating in privacy matching are determined from the debt relationship data in the trusted isolation environment, and the field type corresponding to each target data field is determined. The field type is identity identifier, numerical, enumeration, or anchor verification. Within the privacy computing domain, the target privacy transformation method corresponding to each target data field is determined based on the field type and matching constraints corresponding to each target data field. The target privacy transformation method corresponding to each target data field is used to transform each target data field to obtain field-transformed data, and the association representation is constructed based on the field-transformed data.

5. The method as described in claim 4, characterized in that, After being imported into the privacy computing domain, the target data field exists in plaintext only within the trusted isolation environment, and is not stored in plaintext on any storage medium outside the trusted isolation environment after the privacy transformation process is completed. The privacy transformation process is a cryptographic protection transformation performed on the target data field based on target privacy transformation material obtained from a trusted key source of the privacy computing domain; The privacy transformation process is at least one of the following: format-preserving encryption, homomorphic encryption, hash-based message authentication code operation, and cryptographic hashing. The association representation includes the field transformation data and a privacy check code generated based on the session cipher material corresponding to the independent encrypted session, so that there is a binding relationship between the association representation and the independent encrypted session.

6. The method as described in claim 4, characterized in that, The reading of the current domain unified governance policy synchronized from the global policy source to the privacy computing domain includes: The authorization admission policy, data isolation policy, field transformation policy, matching constraint policy and output control policy are obtained from the global policy source to generate a global policy set. The field transformation policy is used to represent the correspondence between field type, matching constraint condition, privacy transformation method and privacy transformation material. Obtain the policy version identifier and policy integrity credential corresponding to the global policy set; Obtain the local policy version identifier and local policy integrity certificate corresponding to each privacy computing node within the privacy computing domain; The policy version identifier and the policy integrity certificate are compared with the local policy version identifier and the local policy integrity certificate respectively to obtain the policy consistency result. If the policy consistency results are consistent, the global policy set will be determined as the unified governance policy for the current domain.

7. The method as described in claim 5, characterized in that, The step of performing field transformation on each target data field to obtain field-transformed data, and constructing the association representation based on the field-transformed data, includes: When the target data field is an identity identifier field, a format-preserving encryption algorithm is used to perform format-preserving encryption processing on the identity identifier field to obtain format-preserving ciphertext; When the target data field is a numeric field, a homomorphic encryption algorithm is used to perform homomorphic encryption on the numeric field to obtain numeric ciphertext; When the target data field is an enumeration type field, based on the message authentication code key, a hash-based message authentication code operation is performed on the combination result of the target matching field salt value and the enumeration type field to obtain the enumeration field token; When the target data field is an anchor verification field, the anchor salt value is combined with the anchor verification field, and the combination result is cryptographically hashed to obtain the anchor digest. The field transformation data that determines the format-preserved ciphertext, the numerical ciphertext, the enumerated field token, or the anchor digest as the corresponding target data field; A privacy check code is generated based on the session cryptographic material and the field transformation data corresponding to each target data field. The field transformation data is then associated with the privacy check code to obtain the association representation within the privacy computation domain.

8. The method as described in claim 1, characterized in that, The process of responding to a privacy matching request sent by a target data requesting terminal includes: authenticating the target data requesting terminal; performing privacy matching based on the association representation after successful authentication; generating standardized response data with externally observable features based on the privacy matching result; and sending the standardized response data to the target data requesting terminal. In response to a privacy matching request sent by a target data requesting terminal, a first trusted credential and a second trusted credential of the target data requesting terminal are obtained, wherein the first trusted credential and the second trusted credential correspond to different roots of trust. Based on the first root of trust, the first trusted credential is verified to obtain the first trusted verification result. Based on the second root of trust, the second trusted credential is verified to obtain the second trusted verification result. Based on the first trusted verification result and the second trusted verification result, heterogeneous trusted authentication is performed on the target data request terminal to obtain heterogeneous trusted authentication result. If the heterogeneous trusted authentication result is passed, the privacy matching request is converted into a request association representation, and the request association representation is compared with the association representation to obtain a privacy matching result; Based on the privacy matching results, standardized response data with externally observable features is generated and sent to the target data request terminal.

9. The method as described in claim 8, characterized in that, The step of generating standardized response data with externally observable features based on the privacy matching result and sending the standardized response data to the target data request terminal includes: Configure a hierarchical disclosure strategy and / or a closed-loop strategy without disclosure to obtain a set of data disclosure strategies; The access permission level of the target data requesting terminal is determined based on the heterogeneous trusted authentication result; Based on the preset application scenario parameters and the access permission level, a target data transparency strategy is determined from the data transparency strategy set; The privacy matching result is encapsulated according to the target data disclosure strategy to obtain an encrypted response message; The external observable features of the encrypted response message are standardized according to the preset message length, preset response timing and preset protocol fields to obtain standardized response data; The standardized response data is sent to the target data request terminal based on the preset response timing. The hierarchical disclosure strategy includes configuring a debt information interaction entry point, setting corresponding desensitization granularity and open scope based on different access permission levels, and only allowing target data request terminals with corresponding access permission levels to access the corresponding level of debt information. The non-public closed-loop strategy includes closing the public search entry, restricting privacy matching and the generation of privacy matching results to closed-loop execution within the privacy computing domain, and outputting the standardized response data generated based on the privacy matching results only to the target data request terminal that has passed the heterogeneous trusted authentication.

10. The method as described in claim 9, characterized in that, The step of determining the target data transparency strategy from the data transparency strategy set based on preset application scenario parameters and the access permission level includes: The configuration status of the public search entry and the data exposure boundary are obtained based on the preset application scenario parameters. When the public search entry configuration status is closed, the no-public closed-loop strategy is determined as the first candidate data exposure strategy; When the public search entry configuration status is enabled, a second candidate data disclosure strategy is determined from the hierarchical disclosure strategy according to the access permission level; The allowed output range of the corresponding candidate data leakage strategy is compared with the data leakage boundary to obtain the leakage boundary comparison result. Based on the comparison results of the transparent boundary, the first candidate data transparent strategy or the second candidate data transparent strategy is determined as the target data transparent strategy.

11. The method according to any one of claims 1 to 10, characterized in that, When there are multiple target data requesting terminals, after sending the standardized response data to the target data requesting terminals, the method further includes: Send the corresponding claim information of each target data request terminal to the target debtor's terminal; Receive disposal selection instructions from the target debtor's terminal based on the claim information; According to the disposal selection instruction, a target disposal terminal is determined from multiple target data request terminals, and disposal access permissions for the corresponding debt relationship data are only granted to the target disposal terminal.

12. A method for matching creditor-debtor relationship data based on privacy computing, characterized in that, The method is applied to the target debtor terminal, and the method includes: Obtain the target debtor's unique identifier, individual authorization certificate, and debt relationship data; Send the unique identifier and the individual authorization credential to the server so that the server can perform tamper-proof anchoring on the unique identifier and the individual authorization credential to obtain tamper-proof authorization records; Send an encrypted session request to the server to establish an independent encrypted session with the server; The debt relationship data is sent to the server through the independent encrypted session, so that the server can perform authorization verification on the debt relationship data according to the tamper-proof authorization record. When the authorization verification of the debt relationship data passes, the debt relationship data is imported into a privacy computing domain constrained by a unified governance policy for privacy transformation processing to obtain an association representation for privacy matching.

13. A method for matching creditor-debtor relationship data based on privacy computing, characterized in that, The method is applied to a target data request terminal, and the method includes: Obtain privacy-preserving matching request data for requesting matching of creditor and debtor relationship data; Generate a privacy matching request based on the privacy matching request data; The privacy matching request is sent to the server so that the server performs privacy matching based on the privacy matching request and the pre-generated association representation, and generates standardized response data with externally observable features and standardized processing based on the privacy matching result. The association representation is obtained by the server performing privacy transformation processing on the debt relationship data that has passed the authorization verification. Receive standardized response data sent by the server; The standardized response data is parsed to obtain the matching response result corresponding to the privacy matching request.

14. A data matching system for creditor-debtor relationships based on privacy-preserving computation, characterized in that, The system includes a server, a target debtor terminal, and a target data request terminal; The target debtor terminal is used to obtain the target debtor's unique identity identifier, individual authorization certificate, and debt relationship data; Send the unique identifier and the individual authorization credential to the server; The server is used to perform tamper-proof anchoring on the unique identity identifier and individual authorization credential obtained from the target debtor's terminal to obtain tamper-proof authorization records; The target debtor terminal is also used to send an encrypted session request to the server to establish an independent encrypted session with the server; and to send debt relationship data to the server through the independent encrypted session. The server is further configured to receive debt relationship data through an independent encrypted session corresponding to the target debtor's terminal, perform authorization verification on the debt relationship data according to the tamper-proof authorization record; if the authorization verification of the debt relationship data passes, the debt relationship data is imported into a privacy computing domain constrained by a unified governance policy, and privacy transformation processing is performed on the debt relationship data to obtain an association representation; if the authorization verification of the debt relationship data fails, a denial policy is executed on the debt relationship data. The target data request terminal is configured to acquire privacy matching request data for requesting matching of creditor and debtor relationship data; generate a privacy matching request based on the privacy matching request data; and send the privacy matching request to the server. The server is also configured to respond to a privacy matching request sent by a target data request terminal, perform authentication on the target data request terminal, and after successful authentication, perform privacy matching based on the association representation, generate standardized response data with externally observable features based on the privacy matching result, and send the standardized response data to the target data request terminal. The target data request terminal is also used to receive standardized response data sent by the server; and to parse the standardized response data to obtain a matching response result corresponding to the privacy matching request.

15. A data matching device for creditor-debtor relationships based on privacy computing, characterized in that, The privacy-based computation-based debt and creditor relationship data matching device includes: a memory, a processor, and a privacy-based computation-based debt and creditor relationship data matching program stored on the memory. The processor is used to run the privacy-based computation-based debt and creditor relationship data matching program, which is configured to implement the privacy-based computation-based debt and creditor relationship data matching method as described in any one of claims 1 to 11, 12, or 13.

16. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a data matching program for creditor-debtor relationships based on privacy computing, which, when executed by a processor, implements the data matching method for creditor-debtor relationships based on privacy computing as described in any one of claims 1 to 11, 12, or 13.

17. A computer program product, characterized in that, The computer program product includes a privacy-based computational data matching program for creditor-debtor relationships, which, when executed by a processor, implements the steps of the privacy-based computational data matching method for creditor-debtor relationships as described in any one of claims 1 to 11, 12, or 13.