Archives linkage management and control method based on digital and physical cross-dimensional cross verification

Abstract

The application discloses an archive linkage management and control method based on digital and physical cross-dimensional cross verification, belongs to the technical field of archive digital management and warehouse security linkage control, and aims at solving the problems that digital side access control is disconnected with physical warehouse security and lacks an executable cross-dimensional linkage disposal mechanism. The application realizes the technical effects of privacy protection, trusted authentication, unforgeable instruction and digital-physical closed-loop linkage management and control by the following steps: generating and binding access records based on verifiable credentials, performing zero-knowledge proof authentication, fusing cross-dimensional consistency detection and risk grading of physical side access control events and in-storage states, and issuing control and returning audit by a security isolation and information exchange system.

Description

Technical Field

[0001] This invention relates to the field of digital archive management and warehouse security linkage control, and in particular to an archive linkage control method based on cross-dimensional verification of digital and physical data. Background Technology

[0002] Archival management involves technologies such as digital archival management, access control and security control for storage facilities, and cross-network security isolation and information exchange. With the development of information technology in museums and cultural heritage sites, existing technologies typically deploy archival management systems on the digital side to implement archival cataloging, access control, borrowing and viewing processes, access auditing, and log retention. On the physical storage side, access control controllers, electronic locks, alarm devices, video surveillance, RFID inventory management, and inventory status data collection equipment are deployed to achieve physical protection of the storage area and controlled objects. Due to network security requirements, digital management networks and physical control networks often employ secure isolation and information exchange equipment to transmit data and issue commands, forming a layered and isolated management architecture. Some platforms also provide alarm aggregation and visualization to assist on-site personnel in handling situations.

[0003] However, the aforementioned existing technologies still have the following shortcomings:

[0004] 1. There is a lack of an executable cross-dimensional verification mechanism between digital access requests and physical access control events, warehouse location, and inventory status. Digital authorization is disconnected from physical status, and anomaly handling relies on manual verification, making it difficult to form a closed-loop linkage in a timely manner.

[0005] 2. Identity authentication and authorization verification rely heavily on centralized verification or plaintext submission of authorization attributes, resulting in high cross-system trust costs. Furthermore, they lack the capability to verify credential revocation and minimize disclosure, making it difficult to balance verifiability and privacy protection.

[0006] 3. The mechanisms for preventing counterfeiting, preventing replay, and multi-party authorization of the control commands issued across networks are not perfect. There is a risk that the commands may be forged, tampered with, or triggered by unauthorized means. Furthermore, the execution feedback and audit evidence chain is not complete, which affects traceability and security reliability.

[0007] Therefore, a method for linking and managing archives is needed to address the shortcomings of existing technologies. Summary of the Invention

[0008] One objective of this invention is to propose a file linkage management method based on cross-dimensional verification of digital and physical data. Addressing the problems in existing technologies such as the disconnect between digital file access control and physical warehouse security status, the lack of an executable cross-dimensional linkage mechanism, and the susceptibility to forgery and traceability of cross-network control commands, this invention proposes a technical solution based on zero-knowledge proof authentication that generates verifiable credentials and binds them to access request records; cross-dimensional consistency detection and risk classification decision-making that integrates physical access control events and warehouse status; and a system where a threshold number of signature nodes generate threshold signatures for control commands, which are then sent to the physical control network for verification and execution via a secure isolation and information exchange system, with feedback and auditing provided. This invention offers the technical advantages of trusted authentication under privacy protection, closed-loop linkage control between digital and physical data, unforgeable commands, and traceable auditing.

[0009] This invention provides a method for linked management and control of archives based on cross-dimensional verification of digital and physical data, including:

[0010] S1. The digital-side management system receives an access request for a target file identifier and generates an access request record, including the identity identifier, target file identifier, access operation type, and request time. S2. The identity authentication platform retrieves verifiable credentials associated with the identity identifier. The digital-side management system generates a zero-knowledge proof bound to the access request record based on the verifiable credentials and sends the access request record and zero-knowledge proof together as authorization proof data to the authentication party. S3. The authentication party verifies the zero-knowledge proof, obtains the authentication result, and sends it to the digital-side management system. The digital-side management system determines the controlled object identifier based on the target file identifier and obtains the physical-side status data corresponding to the controlled object identifier from the physical-side management system. The digital-side management system generates verification feature data based on the authentication result, access request record, and physical-side status data. S4. The cross-dimensional consistency detection model outputs a consistency judgment result based on the verification feature data, including consistency... S5. The digital management system generates linkage control decision data based on the consistency judgment result and preset risk classification rules, including risk level, controlled object identifier, and physical control action corresponding to the risk level. If the authentication result fails, the access operation corresponding to the access request is rejected and control is triggered. If the authentication result passes, control is triggered when the risk level meets the preset triggering conditions; otherwise, access operation is allowed and the audit log is recorded. S6. The digital management system generates control instructions based on the linkage control decision data. The signature nodes in the signature node set that meet the threshold number generate threshold signatures for the control instructions to obtain signature control instructions. S7. The security isolation and information exchange system sends the signature control instructions to the physical control network. When the signature verification is successful, the access controller implements physical control based on the physical control action and generates control execution feedback. The digital management system receives the control execution feedback and writes it to the audit log.

[0011] Optionally, S1 includes:

[0012] When the digital side management system receives the access request, it parses and obtains the identity identifier, the target file identifier, and the access operation type.

[0013] Collect and initiate terminal attribute data corresponding to the access request, the terminal attribute data including network address, operating system information and browser fingerprint information;

[0014] The terminal fingerprint is generated through a hash operation based on the terminal attribute data;

[0015] The requested time is generated based on a trusted time source;

[0016] The identity identifier, the target file identifier, the access operation type, the request time, and the terminal fingerprint are written into the access request record.

[0017] Optionally, S2 includes:

[0018] The identity authentication platform retrieves the verifiable credential bound to the identity from the credential repository based on the identity identifier in the access request record, and verifies the signature validity, validity period and revocation status of the verifiable credential.

[0019] When the verification is successful, the digital management system constructs a proof statement based on the authorization attribute in the verifiable credential and the access control rule corresponding to the target file identifier. The proof statement is used to indicate that the authorization attribute satisfies the access control rule.

[0020] The digital side management system uses the authorization attribute as witness data and introduces random numbers to generate a commitment value. It calculates a challenge value based on the commitment value and the access request record, generates a response value based on the challenge value, and generates a zero-knowledge proof bound to the access request record. The zero-knowledge proof includes the commitment value and the response value.

[0021] The access request record and the zero-knowledge proof are then used to generate authorization proof data.

[0022] Furthermore, the verification of the revocation status of the verifiable credential includes: performing an unforgeable revocation proof verification on the verifiable credential based on a cryptographic accumulator or based on a verifiable revocation list; and the digital side management system configures selective disclosure rules when constructing the proof statement, so that the zero-knowledge proof only proves that the authorization attribute satisfies the access control rules without disclosing the specific value of the authorization attribute.

[0023] Optionally, S3 includes:

[0024] The authentication and verification party extracts the access request record and the zero-knowledge proof based on the authorization proof data, and calls the proof verification parameters corresponding to the zero-knowledge proof to verify the zero-knowledge proof. The proof verification parameters include a proof verification key and a digest calculation rule for binding the zero-knowledge proof with the access request record. The authentication and verification party calculates a request digest for the access request record based on the digest calculation rule, and generates a pass flag when the zero-knowledge proof verification passes and the request digest satisfies the binding verification. A failure flag is generated when the zero-knowledge proof verification fails or the binding verification fails, so as to obtain the authentication result, and sends the authentication result to the digital side management system. The digital side management system retrieves the controlled object identifier corresponding to the target file identifier in the file and controlled object association table based on the target file identifier in the access request record. The physical side management system obtains physical side status data based on the controlled object identifier. The physical side status data includes warehouse location data, access control event data, and in-stock status data corresponding to the controlled object identifier.

[0025] The physical-side management system encapsulates the physical-side status data to generate a status upload message. The status upload message includes a controlled object identifier, physical-side status data, a status timestamp, and a random value. The security isolation and information exchange system receives the status upload message and performs format verification and replay verification on it. If the verification passes, the system forwards the status upload message to the digital-side management system. The digital-side management system performs integrity verification on the received status upload message, parses it to obtain the physical-side status data, performs time alignment and field standardization on the physical-side status data, and concatenates the authentication result, the access request record, and the processed physical-side status data to generate verification feature data.

[0026] Optionally, S4 includes:

[0027] The cross-dimensional consistency detection model performs feature encoding on the verification feature data, generating a digital side feature vector corresponding to the access request record and a physical side feature vector corresponding to the physical side status data, respectively.

[0028] The cross-dimensional consistency detection model performs cross-attention calculation on the digital side feature vector and the physical side feature vector to obtain the cross-dimensional association weight;

[0029] The cross-dimensional consistency detection model fuses the digital-side feature vector and the physical-side feature vector based on the cross-dimensional correlation weights to obtain a fused feature vector;

[0030] The cross-dimensional consistency detection model calculates the consistency score based on the fused feature vector and generates a consistency label or inconsistency label based on a preset consistency threshold;

[0031] When generating the inconsistency marker, the cross-dimensional consistency detection model extracts the field combination that causes the consistency score to be lower than the consistency threshold from the verification feature data based on the cross-dimensional association weight as the evidence of inconsistency.

[0032] The consistency score, the consistency marker or the inconsistency marker, and the inconsistency evidence constitute the consistency determination result.

[0033] Furthermore, when the cross-dimensional consistency detection model performs feature encoding on the verification feature data, it also includes extracting the event sequence of the access control event data based on a preset time window and performing position encoding; the cross-dimensional consistency detection model weights the physical side feature vector according to the data credibility weight to reduce the impact of low credibility data on the consistency score.

[0034] Optionally, S5 includes:

[0035] The digital management system extracts consistency scores and inconsistency evidence based on the consistency determination results; maps the inconsistency evidence to evidence weights according to preset risk grading rules, and calculates risk scores based on the consistency scores and evidence weights; determines the risk level according to the correspondence between risk scores and risk levels in the preset risk grading rules; determines the physical lockdown action corresponding to the risk level according to the correspondence between risk levels and physical lockdown actions in the preset risk grading rules, and generates coordinated control decision data, which includes the risk level, the controlled object identifier, and the physical lockdown action; the preset risk grading rules also include risk level trigger conditions, wherein when the authentication result is a failure flag, the digital management system rejects the access operation corresponding to the access request, determines the risk level as a preset high-risk level, generates a lockdown trigger flag, and proceeds to the next step;

[0036] When the authentication result is marked as passed, the digital management system generates a lockdown trigger mark and proceeds to the next step when the risk level meets the risk level trigger condition. When the risk level does not meet the risk level trigger condition, the access operation corresponding to the access request is allowed, and the access request record, the consistency judgment result, and the linkage control decision data are written into the audit log.

[0037] Optionally, S6 includes:

[0038] The digital management system generates control instructions based on the linkage control decision data. The control instructions include the controlled object identifier, physical containment action, instruction timestamp, and random value. The system also calculates an instruction digest for the control instructions based on a preset digest algorithm.

[0039] The digital side management system determines the signing nodes to participate in the signing from the set of signing nodes according to preset selection rules, and sends the instruction digest to the signing nodes;

[0040] Each of the aforementioned signature nodes calls the threshold signature key fragment corresponding to its own signature node to generate a partial signature for the instruction digest, and returns the partial signature to the digital side management system;

[0041] The digital side management system verifies the validity of the received partial signatures, and when the number of valid partial signatures reaches a threshold number, it aggregates the partial signatures to generate a threshold signature, thereby obtaining a signature control instruction. The signature control instruction includes the control instruction, the threshold signature, and a set of identifiers used to identify the signature nodes participating in the signing.

[0042] Optionally, the S7 includes:

[0043] The security isolation and information exchange system receives signature control instructions and performs format verification and replay verification on the signature control instructions. The replay verification includes determining whether the signature control instructions are repeated based on the instruction timestamp and random value in the control instructions.

[0044] When the format verification and the replay verification pass, the security isolation and information exchange system forwards the signature control instruction to the physical control network;

[0045] The physical control network verifies the threshold signature in the signature control instruction. The verification includes obtaining the corresponding threshold signature verification parameters based on the signature node identifier set and verifying the instruction digest of the control instruction.

[0046] When the signature is verified, the physical control network determines the warehouse area corresponding to the controlled object identifier based on the controlled object identifier, and sends an execution command matching the physical sealing action to the access controller;

[0047] After executing the execution command, the access control controller generates a lockdown execution feedback, which includes the controlled object identifier, physical lockdown action, execution result, and execution time.

[0048] The security isolation and information exchange system transmits the lockdown execution feedback back to the digital management system, which then associates the lockdown execution feedback with the access request record and writes it into the audit log.

[0049] The beneficial effects of this invention are:

[0050] 1. Achieve an executable closed-loop linkage between digital access control and physical warehouse security: By performing cross-dimensional consistency checks on access request records and physical access control events, in-stock status, and warehouse location data, and outputting consistency scores and inconsistency evidence, and combining risk classification rules to automatically generate lockdown decisions, physical lockdowns can be triggered in a timely manner and a closed-loop handling system can be formed when there are abnormal accesses or digital-physical inconsistencies, reducing manual verification costs and delay risks.

[0051] 2. Enhance authentication credibility and compliance without disclosing authorization attributes: Based on zero-knowledge proofs that are generated and bound to access request records and verified by the authentication and verification party, it can prove that the authorization meets access control rules while avoiding the exposure of attributes in plaintext. Combined with the verification of credential validity period and revocation status, it reduces the risk of credential misuse, expiration, or continued use after revocation.

[0052] 3. Enhance the anti-counterfeiting capabilities and traceability of cross-network linkage commands: Introduce a threshold signature mechanism for the control commands and verify the signature on the physical control network side. At the same time, perform format verification and replay verification in the security isolation and information exchange links, and send back control execution feedback to the audit log. This can effectively prevent command forgery, tampering and replay, forming a complete evidence chain of "decision, issuance, execution and receipt", and enhancing the ability to trace responsibility. Attached Figure Description

[0053] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0054] Figure 1 This is a flowchart of a file linkage management method based on cross-dimensional verification of digital and physical data, as proposed in this invention. Detailed Implementation

[0055] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0056] refer to Figure 1 A method for coordinated management and control of archives based on cross-dimensional verification of digital and physical data includes:

[0057] S1. The digital-side management system receives an access request for a target file identifier and generates an access request record, including the identity identifier, target file identifier, access operation type, and request time. S2. The identity authentication platform retrieves verifiable credentials associated with the identity identifier. The digital-side management system generates a zero-knowledge proof bound to the access request record based on the verifiable credentials and sends the access request record and zero-knowledge proof together as authorization proof data to the authentication party. S3. The authentication party verifies the zero-knowledge proof, obtains the authentication result, and sends it to the digital-side management system. The digital-side management system determines the controlled object identifier based on the target file identifier and obtains the physical-side status data corresponding to the controlled object identifier from the physical-side management system. The digital-side management system generates verification feature data based on the authentication result, access request record, and physical-side status data. S4. The cross-dimensional consistency detection model outputs a consistency judgment result based on the verification feature data, including consistency... S5. The digital management system generates linkage control decision data based on the consistency judgment result and preset risk classification rules, including risk level, controlled object identifier, and physical control action corresponding to the risk level. If the authentication result fails, the access operation corresponding to the access request is rejected and control is triggered. If the authentication result passes, control is triggered when the risk level meets the preset triggering conditions; otherwise, access operation is allowed and the audit log is recorded. S6. The digital management system generates control instructions based on the linkage control decision data. The signature nodes in the signature node set that meet the threshold number generate threshold signatures for the control instructions to obtain signature control instructions. S7. The security isolation and information exchange system sends the signature control instructions to the physical control network. When the signature verification is successful, the access controller implements physical control based on the physical control action and generates control execution feedback. The digital management system receives the control execution feedback and writes it to the audit log.

[0058] In this specific embodiment, S1 includes:

[0059] The digital management system executes the access request upon receiving an access request for the target file identifier. The access request is submitted via an HTTPS interface and the request body is a JSON structure encoded in UTF-8. The digital management system performs syntax parsing on the request body and extracts the identity identifier, target file identifier, and access operation type according to preset field names. The identity identifier is a string identifier that can uniquely point to the user subject and is consistent with the binding relationship of the verifiable credential of the identity authentication platform. The target file identifier is a string identifier used to uniquely locate the file entry in the file management database. The access operation type is a preset enumeration code and its value is limited to VIEW, DOWNLOAD, EXPORT, and BORROW to represent four types of access operations: viewing, downloading, exporting, and borrowing registration, respectively.

[0060] After parsing, the digital side management system collects terminal attribute data corresponding to the access request. This terminal attribute data consists of network address, operating system information, and browser fingerprint information. The network address is taken from the source IP address of this HTTPS connection and is carried by the reverse proxy through a trusted intranet link when the system deploys a reverse proxy. The digital side management system verifies the reverse proxy device identifier before using this data. The operating system information is obtained by parsing the User-Agent field of the access request and generated according to the standardized format "OSNAME / MAJOR.MINOR" to ensure that the same operating system version corresponds to the same string. The browser fingerprint information is generated by constructing and standardizing a fixed set of fields. This fixed set of fields includes the User-Agent plaintext, Accept-Language plaintext, time zone offset in minutes, screen width in pixels, screen height in pixels, color depth in bits, Canvas rendering digest, and WebGL vendor and renderer strings. The Canvas rendering digest is generated by the browser in... After rendering on the canvas with a preset font, font size and fixed text "ARCHIVE-AUTH", the RGBA byte sequence is exported and SHA-256 is performed on the byte sequence. Field standardization includes removing leading and trailing whitespace, converting field names to lowercase, converting numeric fields to decimal strings and concatenating fields in a fixed order.

[0061] The digital side management system generates terminal fingerprints based on terminal attribute data through hash operations. The terminal fingerprints are calculated as follows:

[0062] ;

[0063] in Indicates terminal fingerprint, This represents a hash function using the SHA-256 algorithm. This represents the canonical string corresponding to a network address. This represents a normalized string corresponding to operating system information. This represents the normalized concatenated string corresponding to the browser fingerprint information. This indicates a string concatenation operation performed using the fixed delimiter "|".

[0064] The digital side management system generates request times based on a trusted time source. The trusted time source is implemented by a time synchronization module and establishes an NTS authentication channel with an authorized NTP server and maintains system clock synchronization with UTC time scale. The time synchronization module reads the current UTC time verified by NTS each time a request time is generated and outputs it with millisecond precision and monotonicity verification to ensure that the request time generated within the same digital side management system does not roll back.

[0065] The digital management system writes the identity identifier, target file identifier, access operation type, request time, and terminal fingerprint into the access request record and assigns a globally unique record identifier to the access request record for subsequent binding with zero-knowledge proofs and association with audit logs. The access request record is stored in the audit database in an append-only manner and is simultaneously written to a log file that only increases and never decreases to form a traceable access record.

[0066] In this specific embodiment, S2 includes:

[0067] The identity authentication platform and the digital management system collaborate within the same trusted management domain. The identity authentication platform receives access request records from the digital management system and reads the identity identifier ID and request time from them. Subsequently, the verifiable credential (VC) bound to the ID is retrieved from the credential repository. The verifiable credential (VC) is a structured signature data object that includes a credential identifier (CID), an issuer identifier (IID), and a valid start time. Valid deadline Authorization attribute set And the issuer's digital signature for the VC The set of authorized attributes Includes character attributes With license level attributes Both use a preset code table for encoding. This preset code table remains consistent and unchanged across the digital management system, the identity authentication platform, and the authentication and verification party. The identity authentication platform performs credential verification on the VC and generates a verification result. The credential verification includes signature validity verification, validity period verification, and revocation status verification. The signature validity verification is performed by reading the issuer's public key corresponding to the IID from the issuer's public key directory. And on The signature verification was completed. The hash algorithm used for signature verification was SHA-256, and the hash of the normalized serialized bytes of VC was calculated to avoid signature verification failure due to differences in field order. The validity period verification passed the comparison. and , Completed and or If the process fails, the cancellation status verification is performed using a verifiable cancellation list mechanism. This cancellation list is carried out using a Merkle tree and is periodically published by the issuing party. The common parameters of the Merkle tree include tree depth. And the SHA-256 hash algorithm, with each leaf node corresponding to a credential index. Furthermore, the leaf value is a revocation flag, with 0 indicating no revocation and 1 indicating revocation. The identity authentication platform reads the Merkel path corresponding to the CID from the credential repository. And read the signed revocation list root from the revocation list publisher. and its signature In the Use after verification right Perform path recalculation and confirm that the leaf value corresponding to CID is 0 to complete the verification of the unforgeable revocation proof;

[0068] When all three types of verifications pass, the identity authentication platform sends the VC and credential verification pass mark back to the digital side management system and establishes a session identifier SID associated with the access request record for subsequent audit association. When any verification fails, the platform sends back a credential verification failure mark and the digital side management system generates authorization proof data containing only the access request record and the failure mark and proceeds to the next step.

[0069] After receiving the credential verification pass flag, the digital side management system determines the access control rule corresponding to the file from the access control rule table based on the target file identifier (DID) and access operation type (OP). The access control rules A deterministic strategy object containing a set of allowed role codes. With the set of license level codes And construct a proof statement based on this. To express "existence" In and Make and ,and "For a valid signature from a VC that has not been revoked and is still valid," and simultaneously configure selective disclosure rules. To enable zero-knowledge proofs to disclose only the CID, IID, and credential structure version number VER, and not the credential structure version number VER. and The specific value to be taken;

[0070] Digital side management system will and Mapped to a scalar witness vector according to a preset code table. It also calls a cryptographically secure random number generator to generate random vectors. Subsequently in the elliptic curve group The above uses the Pedersen commitment algorithm to respectively... Generate commitment components and match them with the The signature proof preprocessing commitment components together constitute the commitment value. ,in Pick The curve group and its order is The access request record is processed according to the normalized serialization rules to obtain a byte string, and then a binding digest is calculated. To achieve the binding of zero-knowledge proofs with access request records, challenge values ​​and response values ​​are then generated based on the Fiat-Shamir transformation, specifically as follows:

[0071] ;

[0072] in Indicates the challenge value. This refers to a hash function that performs SHA-256 on the input byte string and maps it to an integer. This represents a byte string representing the commitment value, formed by concatenating commitment components in a fixed order. This represents the binding summary of the access request record. This indicates a byte string concatenation operation. Indicates to Modulo operation Representing the elliptic curve group The rank, Represents the response vector. Represents a random vector. Represents the witness vector;

[0073] The digital side management system assembles zero-knowledge proof ZKP into a package containing com, The system uses a data structure containing CID, IID, VER, PID (Verification Statement Identifier), and SID (Session Identifier). It then encapsulates the access request record with ZKP to generate authorization proof data, which is sent to the authentication party via a TLS connection for subsequent verification steps.

[0074] In this specific embodiment, S3 includes:

[0075] It is accomplished collaboratively by the authentication and verification party, the digital management system, the physical management system, and the security isolation and information exchange system;

[0076] The authentication and verification party receives the authorization certificate data from the digital management system and parses it to obtain the Access Request Record (ARR) and the Zero-Knowledge Proof (ZKP). The ARR is the access request record data object written in step S1 and normalized and serialized, which includes the identity identifier (ID), target file identifier (DID), access operation type (OP), and request time. With terminal fingerprint ZKP is the zero-knowledge proof data object generated in step S2, which includes the commitment value com and the response vector. The document identifier (CID), issuer identifier (IID), document structure version number (VER), proof statement identifier (PID), and session identifier (SID) are all present.

[0077] The authentication and verification party retrieves the proof verification parameters that match ZKP from the proof verification parameter library by PID. The proof verification parameters A deterministic configuration set containing the proof verification key VK and group parameters. Its class Abstract Calculation Rules and challenge value mapping rules VK is the verification key that matches the commitment structure in step S2 and is stored in the hardware key module of the authenticator in read-only mode. Take the elliptic group secp256r1 and For the order of the group, This is used to normalize and serialize Access Request Records (ARRs) by fixing the field order, lowercaseing field names, converting numeric fields to decimal, and fixing the delimiter, in order to generate a unique byte string. Used to map hash outputs to integer challenge values ​​and press Take the mold;

[0078] Authentication and verification basis Calculate the request digest for ARR and use it to complete the binding verification of zero-knowledge proof and access request record, while also based on... The challenge value is recalculated and VK is invoked to perform verification on ZKP to obtain the authentication result. The request digest and challenge value are calculated as follows:

[0079] ;

[0080] in Indicates a request summary. SER(ARR) represents a hash function that takes SHA-256 encoding and interprets the output as non-negative integers in big-endian order. The byte string obtained by performing normalized serialization on the Access Request Record (ARR) Indicates the challenge value. This represents the byte string representing the commitment value in zero-knowledge proof ZKP. This indicates a byte string concatenation operation. This represents the modulo operation. Representing the elliptic curve group The order;

[0081] The authentication and verification party passes the zero-knowledge proof verification and uses the stated... and The recalculated result When the challenge value matches the one used in the verification process, a pass flag is generated as the authentication result. When the zero-knowledge proof verification fails or the challenge value does not match, a failure flag is generated as the authentication result. The authentication result message containing the SID and the authentication result is sent to the digital management system through the TLS channel.

[0082] After receiving the authentication result, the digital side management system performs an equivalent search in the file and controlled object association table based on the target file identifier DID in the ARR to obtain the controlled object identifier COID that uniquely corresponds to the DID. The controlled object identifier COID is a string identifier used by the physical side management system to uniquely identify the cultural relic entity or its packaging carrier and is consistent with the control object table of the access controller.

[0083] The digital management system sends a status acquisition request to the physical management system, carrying the COID and the alignment time base. The physical-side management system reads physical-side status data from the physical status database based on the COID and generates a status upload message SUM. This physical-side status data includes warehouse location data (LOC), access control event data (EVE), and inventory status data (INV). LOC is a fixed set of fields including the warehouse area code, warehouse door number, and storage location number. EVE is a sequence of access control events arranged in ascending time order, with a time range limited to [specific time range missing]. It includes the event time, event type code and access control point number. INV is the inventory status code and its value is limited to IN or OUT, which is determined by the fusion result of RF inventory and inbound / outbound registration.

[0084] The status transmission message SUM includes the controlled object identifier COID, the physical side status data (LOC, EVE, INV), and the status timestamp. and random values ,in The physical side management system reads its NTS-certified UTC clock, which is generated with millisecond precision. Generated by a cryptographically secure random number generator of the physical management system, with a length of 128 bits;

[0085] After receiving SUM, the security isolation and information exchange system performs format verification and replay verification on it. The format verification includes field existence, field type, and length boundary checks. The replay verification checks (COID, If the triplet has been completed within the preset sliding window of 1800s, the message is discarded if duplicates are found. If the verification passes, the SUM is forwarded to the digital management system.

[0086] The digital-side management system performs an integrity check on the received SUM and, upon successful verification, parses out LOC, EVE, and INV. This integrity check is performed by verifying the HMAC field carried in the SUM. The HMAC uses the HMAC-SHA-256 algorithm and a shared key pre-injected and periodically rotated between the digital-side management system and the security isolation and information exchange system. calculate;

[0087] The digital-side management system then performs time alignment and field standardization on the physical-side status data, where time alignment will... Event times in EVE are uniformly converted to UTC milliseconds and then... The relative time offset field is calculated based on the benchmark. Field standardization maps the warehouse area code, event type code and in-stock status code to normalized enumeration values ​​consistent with the digital side code table, and fills missing fields according to preset null value rules.

[0088] The digital side management system will eventually concatenate the authentication results, access request records (ARR), and processed physical side status data (LOC, EVE, INV) in a fixed field order to generate verification feature data (CFD) and write the CFD into the feature cache.

[0089] In this specific embodiment, S4 includes:

[0090] The digital side management system calls the cross-dimensional consistency detection model to perform consistency determination on the verification feature data CFD and outputs the consistency determination result.

[0091] The verification feature data CFD consists of authentication results, access request records (ARR), and processed physical side status data (LOC, EVE, INV), with the field order being fixed.

[0092] The cross-dimensional consistency detection model employs a deterministic network structure that combines two-way encoding with cross-path attention. Model parameters are fixed at deployment time and do not update automatically during runtime. The model's embedding dimension is fixed. The number of cross-attention heads is fixed at 1 The number of cross-attention layers is fixed. The hidden layer dimension of the feedforward network is fixed at 1. The activation function is GELU and the dropout is 0;

[0093] When encoding features for CFD, the model maps ARR to a sequence of digit-side tokens and generates digit-side feature vectors. The digital token sequence includes, in a fixed order, authentication result token, target file identifier token, access operation type token, request time token, and terminal fingerprint token. The authentication result token is obtained by marking or failing and mapping it to two types of trainable embedding vectors. The target file identifier token and terminal fingerprint token are first SHA-256 encoded, and the lower 64 bits are truncated as an index before being mapped to trainable embedding vectors to control the vocabulary size. The access operation type token is mapped to four types: VIEW, DOWNLOAD, EXPORT, and BORROW, and the request time token is... and The relative offset millisecond value is used as a numerical feature, and after being truncated and normalized, it is mapped to a 128-dimensional vector through two fully connected layers;

[0094] The model simultaneously maps (LOC, EVE, INV) to a sequence of physical tokens and generates a physical feature vector. The physical-side token sequence includes warehouse location tokens, in-stock status tokens, and access control event token sequences in a fixed order. The warehouse location token is mapped to a trainable embedding vector by concatenating the warehouse area code, warehouse door number, and storage location number. The in-stock status token is mapped to a trainable embedding vector according to two categories: IN and OUT. The access control event token sequence is generated from access control event data EVE within a preset time window. The intercepted data is obtained and sorted in ascending order by event time, with a fixed upper limit for the event sequence length. And when the number of events exceeds the limit, the closest one is retained. 32 events were recorded and the missing parts were padded with masks;

[0095] The access control event token is composed of the event type code, the access control point number, and the relative time offset. The relative time offset is rounded down to the second and mapped to a discrete bucket and a position code is introduced. The position code adopts a fixed sine and cosine position code and uses the event's sequence number as the position index to form a reproducible temporal representation.

[0096] The model generates physical-side feature vectors. A data credibility weight is applied to the physical-side token to reduce the impact of low-credibility data. This data credibility weight follows a deterministic calculation rule and has an output range of [missing information]. The warehouse location token has a fixed weight of 1.0. The access control event token has a weight of 1.0 when the event source is the access control controller and its message passes integrity verification, and a weight of 0.8 when the event source is the upper-level aggregation device. The in-warehouse status token has a weight of... Take 0.9 and in The value is set to 0.6, and the embedding vector of each token is multiplied by its corresponding weight before being input into the cross-attention module;

[0097] The cross-attention module uses the digital-side token sequence as the query and the physical-side token sequence as the key and value to calculate the cross-dimensional association weight matrix and complete two-layer stacked updates. Then, it performs average pooling on the updated digital-side token sequence to obtain... The updated physical token sequence is obtained by performing average pooling. and will and The fused feature vector is obtained by concatenating the elements in a fixed order. ;

[0098] The model is based on fused feature vectors Calculate the consistency score and generate a consistency label or inconsistency label, wherein the consistency score is calculated as follows:

[0099] ;

[0100] in This represents the consistency score and its value range is [value range missing]. This represents the Sigmoid function. Represents the classification weight vector and its dimension is 1. Consistent and solidified after model training is completed. Indicates by and The fused feature vector obtained by concatenation It represents a classification bias scalar and is fixed after the model training is completed;

[0101] The consistency threshold is fixed at ,when Output a consistent flag when Output inconsistency flags and extract evidence of inconsistency in real time;

[0102] The extraction of inconsistency evidence is based on the interpretability output of the cross-dimensional association weight matrix. When outputting inconsistency markers, the model aggregates the cross-dimensional association weight matrix by query token to obtain the contribution score of each physical token and selects the tokens with the highest contribution scores. Each physical token is recorded, along with the digital token with the highest associated weight. The "digital field name, physical field name, associated weight value, and corresponding original field value" are then assembled into a field combination as evidence of inconsistency. This combination, along with the consistency score and the consistency mark or inconsistency mark, constitutes the consistency judgment result and is returned to the digital management system.

[0103] In this specific embodiment, S5 includes:

[0104] The digital side management system executes this action after receiving the authentication result and the consistency determination result. The digital side management system reads the consistency score from the consistency determination result. Set of Inconsistent Evidence ,in For the range of values The consistency score indicates that the digital and physical sides are more consistent, and the higher the score, the more consistent the digital and physical sides are. It is a set of inconsistent evidence consisting of a combination of fields, and each piece of evidence contains a digital field name, a physical field name, an association weight value, and the corresponding original field value;

[0105] The digital side management system will classify risks according to preset risk classification rules. Each piece of evidence is mapped to an evidence weight, forming a set of evidence weights. The mapping of evidence weights is accomplished using a fixed rule table, which is fixed during system deployment. The rule table categorizes evidence into three types based on field combinations: warehouse location conflict, in-warehouse status conflict, and access control event conflict, assigning weights of 0.40, 0.35, and 0.25 respectively. A warehouse location conflict is defined as evidence containing a warehouse area code or warehouse door number field whose physical value does not match the warehouse location whitelist bound to the target file identifier (DID) in the file-controlled object association table. An in-warehouse status conflict is defined as evidence containing an in-warehouse status data INV field with a value of OUT. An access control event conflict is defined as evidence containing access control event data EVE field within a time window. There is a forced door opening event code or an illegal card swipe event code that matches the warehouse door number corresponding to the controlled object identifier COID;

[0106] After completing the evidence weight mapping, the digital side management system uses consistency scoring. With evidence weight set Calculate risk score And cut it to The calculation satisfies:

[0107] ;

[0108] in Indicates the risk score. This represents the function that takes the minimum value. This represents the consistency score reduction factor and takes... Indicates consistency score, This represents the summation operator. Indicates the number of evidence weights and Represents the set of evidence weights The first in Each piece of evidence has a weight;

[0109] The digital management system determines the risk level according to the correspondence between risk scores and risk levels in a preset risk grading rule. The risk levels are fixed at three levels: low risk, medium risk, and high risk. When determined to be low risk, When it was determined to be medium risk, It was identified as high risk at the time;

[0110] The digital management system determines physical control actions according to the correspondence between risk levels and physical control actions in the preset risk classification rules. The physical control actions are fixed as three categories: no control, target warehouse door locking, and warehouse area control. Low risk corresponds to no control, medium risk corresponds to target warehouse door locking, and high risk corresponds to warehouse area control. Target warehouse door locking is defined as issuing a lock execution command to the physical control network to lock the warehouse access control point corresponding to the controlled object identifier COID, and simultaneously switching the access policy of the access control point to allow only security personnel to pass. Warehouse area control is defined as issuing a lock execution command to the physical control network to lock all access control points in the warehouse area where the COID is located, and setting the area alarm linkage to the on state.

[0111] The digital management system generates coordinated control decision data, which includes risk level, controlled object identification (COID), and physical containment actions.

[0112] The preset risk grading rules also include risk level triggering conditions, and the lockdown triggering logic is jointly determined by the authentication result and the risk level. When the authentication result is marked as a failure, the digital side management system rejects the access operation corresponding to the access request and directly determines the risk level as high risk, while generating a lockdown triggering mark and entering S6. When the authentication result is marked as a pass, the digital side management system generates a lockdown triggering mark and enters S6 when the risk level is high risk, generates a lockdown triggering mark and enters S6 when the risk level is medium risk and the access operation type OP is DOWNLOAD or EXPORT, and enters S6 when the risk level is medium risk and the evidence set is inconsistent. When evidence of a conflict in the database status is included, a blockade trigger flag is generated and the process enters S6. If the risk level is low or the above-mentioned medium-risk trigger conditions are not met, the access operation corresponding to the access request is allowed, and the access request record, consistency judgment result, and linkage control decision data are associated with the session identifier SID and written into the audit log to form a traceable handling record.

[0113] In this specific embodiment, S6 includes:

[0114] After the digital side management system generates a blockade trigger flag in step S5, it executes the command to form a signature control instruction that can be issued across networks and cannot be forged.

[0115] The digital management system constructs control instructions (CMDs) based on the coordinated control decision data. The control instructions (CMDs) are deterministic data structures with fixed field order and include at least the controlled object identifier (COID), physical containment action (ACT), and instruction timestamp. and random values Where COID is the same as in step S5, and ACT is taken from the physical containment action enumeration value determined in step S5. The digital-side management system reads an NTS-certified UTC clock to generate a clock with millisecond precision and is required to meet the following conditions. , The random number is generated by the cryptographically secure random number generator of the digital side management system and is 128 bits long and written as a hexadecimal string;

[0116] The digital side management system performs normalized serialization on the CMD according to fixed rules to obtain byte strings and calculates instruction digests. The normalized serialization employs rules such as lowercase field names, decimalization of numeric fields, encoding of enumerated fields according to a code table, and using a fixed separator "|" between fields while prohibiting the omission of empty fields to ensure the uniqueness of the digest. The instruction digest is obtained as follows:

[0117] ;

[0118] in Indicates the instruction summary. The hash function is represented by SHA-256, and SER(CMD) represents the byte string generated by the control command CMD according to the normalized serialization rules.

[0119] The digital side management system maintains a set of signature nodes. The distributed generation and fragmented distribution of threshold signature keys are completed during the system initialization phase, with the threshold number fixed. And the total number of nodes is fixed. Key sharding employs a distributed key generation process based on Shamir secret sharing, generating keys only once and not updating them online after generation. The signature algorithm uses BLS threshold signing with curve parameters set to BLS12-381, and uses deterministic hash-to-curve mapping. Mapping to signature group elements eliminates signature inconsistencies caused by differences in message format. The digital side management system stores the master verification public key (PK) and the shard verification public keys corresponding to each signature node. And the PK and All configurations are permanently written to the read-only system during the initialization phase.

[0120] The digital side management system selects from the preset rules. The set of signature node identifiers that participate in the signing is determined in the middle. The preset selection rule is a deterministic rule and includes two parts: node health status filtering and a minimum recently used strategy. Node health status is determined by the bidirectional TLS heartbeat connection maintained with each signing node, and a node is deemed unavailable if it does not respond for three consecutive heartbeat cycles. The minimum recently used strategy selects three nodes from the available nodes in ascending order based on the node signature counter and the most recent signature timestamp maintained by the digital side management system. ;

[0121] The digital management system communicates with the digital side via a bidirectional TLS connection. Each signature node in the process sends a signature request message, the signature request message containing an instruction digest. Session ID (SID) and command timestamp With random values And the signing node, after receiving it, Perform time offset verification with the local NTS-certified UTC clock and reject the signature if the offset exceeds 5 seconds;

[0122] Each signing node calls the threshold signature key fragment stored in its hardware key module when the time deviation verification passes. right Generate partial signature and will contain and Partially signed response messages are returned to the digital management system;

[0123] The digital side management system performs validity checks on the received partial signatures. These validity checks include verifying the length and encoding format of the partial signatures and verifying the public key based on fragmentation. The cryptographic school verifies the signature and discards the part of the signature if the verification fails, and selects other available nodes according to the same preset selection rules to make up for the threshold number;

[0124] The digital side management system has reached a threshold number of valid partial signatures. At that time, an aggregation operation is performed on the corresponding partial signatures to generate a threshold signature. The aggregation operation is based on the set of participating node identifiers. Lagrange coefficient interpolation was performed on the BLS12-381 group to output a unique aggregate signature result. The length is fixed at 96 bytes;

[0125] The digital side management system will include CMD and threshold signature. and a set of identifiers used to identify the signing nodes participating in the signing. The signature control instruction (SCMD) is encapsulated and generated. The SCMD and the linkage control decision data of step S5 are associated by SID and written into the audit log to ensure the traceability of the evidence chain for subsequent cross-network distribution, signature verification execution and receipt audit.

[0126] In this specific embodiment, S7 includes:

[0127] It is accomplished collaboratively by a security isolation and information exchange system, a physical control network, access control controllers, and a digital management system;

[0128] The security isolation and information exchange system receives the signature control command (SCMD) and parses it to obtain the control command (CMD) and threshold signature. and signature node identifier set The CMD includes the Controlled Object Identifier (COID), the Physical Control Action (ACT), and the Instruction Timestamp. With random values For CMD threshold signatures with a fixed length of 96 bytes, It is a set of signature node identifiers that participate in the signature process, and the number of elements is fixed at 3.

[0129] The security isolation and information exchange system performs format verification and replay verification on SCMD. Format verification includes field existence verification, field type verification, field length boundary verification, and enumeration value verification. The field length boundary verification limits the COID to printable ASCII strings of 1 to 64 characters. For 13-bit UTC millisecond timestamps The signature is a 32-character hexadecimal string corresponding to a 128-bit random number; ACT represents one of three conditions: no lockdown, target warehouse door locked, or warehouse area locked; each signature node in I is identified as a printable ASCII string of 1 to 32 characters. Duplicates are not allowed within the internal code.

[0130] The replay verification includes based on and The repetition and timeliness assessments are performed, with the repetition assessment done by checking the triplet (COID). Whether the replay cache already exists in the security isolation and information exchange system and whether the sliding window of the replay cache is fixed at 1800s are used to determine its timeliness. The system uses an NTS-certified UTC clock that is isolated from and exchanges information with the security system, and determines an invalid command if the time deviation exceeds 30 seconds.

[0131] When both format verification and replay verification pass, the security isolation and information exchange system forwards the SCMD to the physical control network and keeps the SCMD byte sequence unchanged during the forwarding process to avoid signature discrepancies.

[0132] After receiving the SCMD, the physical control network verifies the threshold signature and only enters the blockade execution process upon successful signature verification. The signature verification includes two parts: obtaining threshold signature verification parameters and digest verification. The threshold signature verification parameters are determined by the physical control network based on… Read the corresponding fragmented verification public key from the local read-only key directory. And follow the steps Constructing aggregated verification public keys using consistent Lagrange coefficient interpolation rules The result is that the digest verification is performed on the CMD by performing normalized serialization consistent with step S6 and recalculating the instruction digest. Completed, the instruction summary is as follows:

[0133] ;

[0134] in Indicates the instruction summary. This indicates a hash function and uses SHA-256. SER(CMD) represents a byte string generated by applying the control command CMD according to the following rules: lowercase field names, decimalization of numeric fields, encoding of enumerated fields according to a code table, and using a fixed separator "|" between fields, with empty fields prohibited. CMD indicates that it contains COID, ACT, etc. and Control commands;

[0135] When the signature verification is successful, the physical control network uses the COID to query the mapping table from controlled objects to warehouse areas within the physical control network to determine the warehouse area identifier (RID) and the set of access control point identifiers associated with the RID. Based on the ACT, a set of execution commands to be sent to the access control controller is generated. When the ACT is to lock the target warehouse door, the set of execution commands only contains the target access control point locking command bound to the COID and simultaneously switches the access control point's access policy to allow only security roles to pass. When the ACT is to lock the warehouse area, the set of execution commands includes the lock command for the target access control point bound to the COID. The command to lock all restricted areas within the premises is executed, and the alarm linkage in the warehouse area is simultaneously activated.

[0136] The physical control network sends execution commands through the control link with the access controller and waits for the access controller to return execution confirmation. After completing the locking and policy switching, the access controller generates a lockdown execution feedback FB, which includes the controlled object identifier (COID), physical lockdown action (ACT), execution result (RES), and execution time. RES takes the value SUCCESS or FAIL, and in the case of FAIL, it carries an error code to indicate one of the three failure reasons: latch failure, policy switching failure, or device offline. The access control controller reads its local NTS-certified UTC clock, which is generated with millisecond precision.

[0137] After receiving the FB, the security isolation and information exchange system performs format verification and replay verification on it. After passing the verification, it sends the FB back to the digital management system. After receiving the FB, the digital management system associates the FB with the access request record (ARR), the linkage control decision data, and the signature control instruction (SCMD) according to the session identifier (SID) and writes them into the audit log to form a closed-loop evidence chain of "decision, issuance, execution, and receipt".

[0138] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

[0139] This invention addresses the technical problem of "disconnection between digital systems and physical warehouse security, lacking an executable cross-dimensional linkage mechanism." It uses verifiable credentials and zero-knowledge proofs for trusted authentication of digital access requests. The authentication results are then cross-validated with physical access control events, inventory status, and warehouse location data to generate consistency scores and inconsistency evidence. Based on risk grading rules, physical control actions are automatically generated and executed across the network, ensuring that digital access behavior is constrained by physical status and triggers executable actions. Simultaneously, control commands employ threshold signatures and undergo anti-replay verification and acknowledgment auditing in isolated exchange links to guarantee the authenticity, integrity, and traceability of commands during cross-network transmission and execution. Ultimately, this achieves a closed-loop linkage control effect from digital to physical systems.

[0140] In terms of algorithm structure, this invention addresses the aforementioned technical problems by improving upon "executable linkage" and "cross-domain trust": First, it binds and verifies zero-knowledge proofs and access request records through digest binding and challenge response mechanisms, reducing the risk of proof misuse and replay, and ensuring that digital authorization results have a trustworthy basis for linkage processing; Second, it introduces cross-attention fusion of digital and physical features into the cross-dimensional consistency detection model, performs time window extraction and location encoding on access control event sequences, applies credibility weights to physical data to reduce the impact of low-credibility data on consistency scores, and outputs field-level evidence to support risk mapping and lockdown decisions when inconsistencies occur; Third, it introduces an instruction structure containing timestamps and random values, threshold signature aggregation, and signature verification execution mechanism on the linkage instruction side, avoiding single-point unauthorized issuance, improving the security and stability of lockdown action triggering, and thus more effectively achieving the technical effect of cross-dimensional linkage control.

Claims

1. A method for coordinated management and control of archives based on cross-dimensional verification of digital and physical data, characterized in that, include: S1. The digital side management system receives access requests for the target file identifier and generates access request records, including the identity identifier, target file identifier, access operation type, and request time. S2. The identity authentication platform retrieves verifiable credentials associated with the identity identifier. The digital side management system generates a zero-knowledge proof bound to the access request record based on the verifiable credentials. The access request record and the zero-knowledge proof are combined to form authorization proof data and sent to the authentication and verification party. S3. The authentication and verification party verifies the zero-knowledge proof to obtain the authentication result and sends it to the digital side management system. The digital side management system determines the controlled object identifier based on the target file identifier and obtains the physical side status data corresponding to the controlled object identifier from the physical side management system. The digital side management system generates verification feature data based on the authentication result, access request record and physical side status data. S4. The cross-dimensional consistency detection model performs feature encoding on the verification feature data, generating a digital side feature vector corresponding to the access request record and a physical side feature vector corresponding to the physical side status data, respectively. The cross-dimensional consistency detection model performs cross-attention calculation on the digital side feature vector and the physical side feature vector to obtain the cross-dimensional association weight; The cross-dimensional consistency detection model fuses the digital-side feature vector and the physical-side feature vector based on the cross-dimensional correlation weights to obtain a fused feature vector; The cross-dimensional consistency detection model calculates a consistency score based on the fused feature vector and generates a consistency label or an inconsistency label based on a preset consistency threshold. When generating the consistency marker, the cross-dimensional consistency detection model uses the consistency score and the consistency marker to form a consistency determination result. When generating the inconsistency marker, the cross-dimensional consistency detection model extracts the field combination that causes the consistency score to be lower than the consistency threshold from the verification feature data based on the cross-dimensional association weight as inconsistency evidence, and uses the consistency score, the inconsistency marker, and the inconsistency evidence to constitute the consistency judgment result; S5. The digital management system generates linkage control decision data based on the consistency judgment result and the preset risk classification rules, including risk level, controlled object identifier and physical blocking action corresponding to the risk level. When the authentication result is failed, the access operation corresponding to the access request is rejected and the blocking is triggered. When the authentication result is passed, the blocking is triggered when the risk level meets the preset triggering conditions. Otherwise, the access operation is allowed and the audit log is recorded. S6. The digital side management system generates control instructions based on the linkage control decision data. Signature nodes in the signature node set that meet the threshold number generate threshold signatures for the control instructions to obtain signature control instructions. S7, the security isolation and information exchange system sends signature control instructions to the physical control network. When the signature is verified, the access controller is controlled to implement physical sealing based on the physical sealing action and generate sealing execution feedback. The digital side management system receives the sealing execution feedback and writes it into the audit log.

2. The method for linked management and control of archives based on cross-dimensional verification of digital and physical data as described in claim 1, characterized in that, S1 includes: When the digital side management system receives the access request, it parses and obtains the identity identifier, the target file identifier, and the access operation type. Collect and initiate terminal attribute data corresponding to the access request, the terminal attribute data including network address, operating system information and browser fingerprint information; A terminal fingerprint is generated through a hash operation based on the terminal attribute data; The requested time is generated based on a trusted time source; The identity identifier, the target file identifier, the access operation type, the request time, and the terminal fingerprint are written into the access request record.

3. The method for file linkage management based on digital and physical cross-dimensional verification according to claim 1, characterized in that, S2 include: The identity authentication platform retrieves the verifiable credential bound to the identity from the credential repository based on the identity identifier in the access request record, and verifies the signature validity, validity period and revocation status of the verifiable credential. When the verification is successful, the digital management system constructs a proof statement based on the authorization attribute in the verifiable credential and the access control rule corresponding to the target file identifier. The proof statement is used to indicate that the authorization attribute satisfies the access control rule. The digital side management system uses the authorization attribute as witness data and introduces random numbers to generate a commitment value. It calculates a challenge value based on the commitment value and the access request record, generates a response value based on the challenge value, and generates a zero-knowledge proof bound to the access request record. The zero-knowledge proof includes the commitment value and the response value. The access request record and the zero-knowledge proof are then used to generate authorization proof data.

4. The method for linked management and control of archives based on cross-dimensional verification of digital and physical data as described in claim 1, characterized in that, S3 includes: The authentication and verification party extracts the access request record and the zero-knowledge proof based on the authorization proof data, and calls the proof verification parameters corresponding to the zero-knowledge proof to verify the zero-knowledge proof. The proof verification parameters include the proof verification key and the digest calculation rules used to bind the zero-knowledge proof to the access request record. The authentication and verification party calculates a request digest for the access request record according to the digest calculation rules, and generates a pass flag when the zero-knowledge proof verification is passed and the request digest satisfies the binding verification, and generates a failure flag when the zero-knowledge proof verification fails or the binding verification fails, so as to obtain the authentication result and send the authentication result to the digital side management system. The digital side management system retrieves the controlled object identifier corresponding to the target file identifier in the file and controlled object association table based on the target file identifier in the access request record; The physical side management system obtains physical side status data based on the controlled object identifier. The physical side status data includes warehouse location data, access control event data, and in-stock status data corresponding to the controlled object identifier. The physical side management system encapsulates the physical side status data to generate a status uploading message, which includes the controlled object identifier, physical side status data, status timestamp, and random value. The security isolation and information exchange system receives the status transmission message and performs format verification and replay verification on the status transmission message. When the verification passes, the system forwards the status transmission message to the digital side management system. The digital side management system performs integrity verification on the received status transmission message, parses it to obtain the physical side status data, performs time alignment and field standardization processing on the physical side status data, and concatenates the authentication result, the access request record and the processed physical side status data to generate verification feature data.

5. The method for file linkage management based on digital and physical cross-dimensional verification according to claim 1, characterized in that, S5 include: When generating the inconsistency marker, the digital management system extracts the consistency score and inconsistency evidence based on the consistency judgment result, maps the inconsistency evidence to evidence weights according to the preset risk grading rules, calculates the risk score based on the consistency score and the evidence weights, determines the risk level according to the correspondence between the risk score and the risk level in the preset risk grading rules, determines the physical lockdown action corresponding to the risk level according to the correspondence between the risk level and the physical lockdown action in the preset risk grading rules, and generates linkage control decision data, which includes the risk level, the controlled object identifier, and the physical lockdown action. The preset risk classification rule also includes risk level triggering conditions. When the authentication result is a failure mark, the digital side management system rejects the access operation corresponding to the access request, determines the risk level as a preset high-risk level, generates a blockade trigger mark, and proceeds to the next step. When the authentication result is marked as passed, the digital management system generates a lockdown trigger mark and proceeds to the next step when the risk level meets the risk level trigger condition. When the risk level does not meet the risk level trigger condition, the access operation corresponding to the access request is allowed, and the access request record, the consistency judgment result, and the linkage control decision data are written into the audit log.

6. The method for file linkage management based on digital and physical cross-dimensional verification according to claim 1, characterized in that, S6 include: The digital management system generates control instructions based on the linkage control decision data. The control instructions include the controlled object identifier, physical containment action, instruction timestamp, and random value. The system also calculates an instruction digest for the control instructions based on a preset digest algorithm. The digital side management system determines the signing nodes to participate in the signing from the set of signing nodes according to preset selection rules, and sends the instruction digest to the signing nodes; Each of the aforementioned signature nodes calls the threshold signature key fragment corresponding to its own signature node to generate a partial signature for the instruction digest, and returns the partial signature to the digital side management system; The digital side management system verifies the validity of the received partial signatures, and when the number of valid partial signatures reaches a threshold number, it aggregates the partial signatures to generate a threshold signature, thereby obtaining a signature control instruction. The signature control instruction includes the control instruction, the threshold signature, and a set of identifiers used to identify the signature nodes participating in the signing.

7. The method for linked management and control of archives based on cross-dimensional verification of digital and physical data as described in claim 1, characterized in that, S7 includes: The security isolation and information exchange system receives signature control instructions and performs format verification and replay verification on the signature control instructions. The replay verification includes determining whether the signature control instructions are repeated based on the instruction timestamp and random value in the control instructions. When the format verification and the replay verification pass, the security isolation and information exchange system forwards the signature control instruction to the physical control network; The physical control network verifies the threshold signature in the signature control instruction. The verification includes obtaining the corresponding threshold signature verification parameters based on the signature node identifier set and verifying the instruction digest of the control instruction. When the signature is verified, the physical control network determines the warehouse area corresponding to the controlled object identifier based on the controlled object identifier, and sends an execution command matching the physical sealing action to the access controller; After executing the execution command, the access control controller generates a lockdown execution feedback, which includes the controlled object identifier, physical lockdown action, execution result, and execution time. The security isolation and information exchange system transmits the lockdown execution feedback back to the digital management system, which then associates the lockdown execution feedback with the access request record and writes it into the audit log.

8. The method for file linkage management based on cross-dimensional verification of digital and physical data as described in claim 3, characterized in that, The verification of the revocation status of the verifiable credential includes: performing an unforgeable revocation proof verification on the verifiable credential based on a cryptographic accumulator or based on a verifiable revocation list; Furthermore, the digital side management system configures selective disclosure rules when constructing the proof statement, so that the zero-knowledge proof only proves that the authorization attribute satisfies the access control rules without revealing the specific value of the authorization attribute.

9. The method for linked management and control of archives based on cross-dimensional verification of digital and physical data as described in claim 1, characterized in that, The physical-side status data includes warehouse location data, access control event data, and in-stock status data corresponding to the controlled object identifier; When the cross-dimensional consistency detection model performs feature encoding on the verification feature data, it also includes extracting the event sequence of the access control event data based on a preset time window and performing position encoding. The cross-dimensional consistency detection model weights the physical side feature vectors based on data credibility weights to reduce the impact of low-credibility data on the consistency score.

Images