A data element encapsulation model for cross-domain interaction security

By separating the data element encapsulation model, the problem of binding the encapsulation structure with business logic in cross-domain interaction is solved, realizing flexible encapsulation and secure flow of data in different scenarios, and improving the efficiency and security of cross-domain interaction.

CN123116349APending Publication Date: 2026-11-17ECONOMIC TECH RES INST OF STATE GRID HENAN ELECTRIC POWER +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611102953.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-07-23
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing data encapsulation technologies are difficult to reuse flexibly across scenarios and systems. The encapsulation structure is tightly bound to specific business logic, resulting in poor portability and low efficiency in cross-domain interaction.

Method used

This paper proposes a data element encapsulation model for cross-domain interaction security, which includes an element plane and a control plane. Through the separate design of data layer, metadata layer, security layer, capability layer and application layer, it realizes unified data representation and dynamic binding of security policies, and supports flexible encapsulation and secure circulation across domains.

Benefits of technology

It enables flexible encapsulation and secure circulation of data in different scenarios, improves the efficiency of cross-domain interaction, and adapts to the security needs of different application scenarios through dynamic security policies, thereby enhancing the flexibility and security of data interaction.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123116349A_ABST
    Figure CN123116349A_ABST
Patent Text Reader

Abstract

This invention provides a data element encapsulation model for cross-domain interaction security, comprising: an element plane and a control plane; the element plane includes a data layer and a metadata layer. The data layer is used for preprocessing collected and stored multi-source heterogeneous data; the metadata layer is used to abstract the data in the data layer to form metadata, realizing a unified representation of multi-source heterogeneous data; the control plane includes a security layer, a capability layer, and an application layer; the security layer is configured with an atomic security capability library, a dynamic policy encapsulation module, and an audit trail module; the capability layer encapsulates standardized data operation interfaces; the application layer is used to perform layered release of data value based on the data and capabilities provided by the data layer, metadata layer, and security layer. The element plane and the control plane are decoupled from each other. This invention, through the decoupling of the element plane and the control plane, achieves the decoupling of data entities and data applications, enabling data to be flexibly encapsulated and securely circulated in different scenarios.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of information processing technology, and in particular to a data element encapsulation model for cross-domain interaction security. Background Technology

[0002] With the rapid development of information technology and the widespread adoption of technologies such as the Internet, cloud computing, and artificial intelligence, the speed and scale of data generation are constantly expanding. Data has gradually become an important national strategic resource, with its global and fluid attributes increasingly enhanced. Circulation is essential for realizing the value of data, and cross-domain interaction is an inherent requirement of data circulation. However, existing warehouse-style data resources have low efficiency in secure external interaction, and data models only emphasize the structure of data objects themselves, making it difficult to meet the element-based requirements for data processing and configuration during cross-domain interaction. Therefore, defining a standardized data encapsulation model that can be processed and configured has become a critical issue that urgently needs to be addressed.

[0003] Existing data encapsulation technologies include solutions based on data elements. However, while these data element solutions aim to provide functional services for specific businesses by packaging data resources with specific algorithms and rule engines, their encapsulation structure tightly binds data to specific business logic, making it difficult to achieve flexible reuse across scenarios and systems, resulting in poor portability. Summary of the Invention

[0004] This invention provides a data element encapsulation model for cross-domain interaction security, which addresses the shortcomings of existing data encapsulation models in achieving flexible reuse across scenarios and systems.

[0005] A data element encapsulation model for cross-domain interaction security includes: an element plane and a control plane; The feature plane includes a data layer and a metadata layer. The data layer is used to preprocess the collected and stored multi-source heterogeneous data; the metadata layer is used to abstract the data in the data layer to form metadata, thereby realizing a unified representation of multi-source heterogeneous data. The control plane includes a security layer, a capability layer, and an application layer; The security layer is configured with an atomic security capability library, a dynamic policy encapsulation module, and an audit trail module; The atomic security capability library contains several independently callable atomic security capabilities; the atomic security capabilities include at least one of encryption, desensitization, watermarking, or digital signature; The dynamic policy encapsulation module is used to dynamically combine the atomic security capabilities with the root access scenario risk value to generate a security policy, and bind the security policy to the data entities in the data layer, the metadata in the metadata layer, and the capability interfaces encapsulated in the capability layer; The audit trail module is used to record the data element package identifier, calling entity, calling time, calling interface, output summary, and policy version; The capability layer encapsulates standardized data operation interfaces; The application layer is located at the top of the control plane and is used to release data value in layers based on the data and capabilities provided by the data layer, metadata layer, and security layer. The element plane and the control plane are decoupled from each other, and a controlled association is established through the globally unique data element encapsulation identifier built into the model, the standardized metadata list output by the metadata layer, the security policy file generated by the policy encapsulation module, and the capability interface description file output by the capability layer.

[0006] Furthermore, in the data element encapsulation model for cross-domain interaction security described above, the data layer preprocesses the collected and stored multi-source heterogeneous data by: firstly, using data cleaning tools to deduplicate, denoise, detect outliers, and fill in missing values; then, eliminating differences in data dimensions and standardizing the data; finally, for multi-source data with business relationships, establishing associations based on unified subject identifiers, device identifiers, or spatiotemporal attributes, and integrating and storing them according to application requirements; data that does not need to be integrated is stored independently after standardization.

[0007] Furthermore, in the data element encapsulation model for cross-domain interaction security described above, the metadata layer includes: descriptive metadata, structural metadata, management metadata, and service metadata; The descriptive metadata is used to abstractly describe the data content; The structural metadata is used to abstractly describe the structure of the data; The management metadata is used to abstractly describe the data's lifecycle, security attributes, and ownership claims; The service metadata is used to define the upper-level functional interfaces supported by the data.

[0008] Furthermore, in the data element encapsulation model for cross-domain interaction security described above, the generation of the security policy includes: collection Context information The context information Includes: main context Context of the object Context and behavioral context ; Wherein, the main context Obtain from the authentication information of the current accessing entity; The object context Obtained from the management metadata output by the metadata layer; The environment context This includes location information, network information, device information, and time information; The location information is obtained by the server based on the source IP address of the access request; the network information is determined by the client through the network type reported in the request header; the device information is obtained by the server based on the device characteristics carried in the access request; and the time information is determined by the server's system time. The behavioral context The frequency of calls to the capability interfaces in the capability layer is obtained by the server. Based on the collected context information, the risks of each entity are quantified. Object risk Environmental risks and behavioral risks And normalize each risk factor; ,in, The level of trust in the authenticated entity; ; ,in This represents the number of visits in the last 5 minutes. in, For location credibility, For network stability, For equipment safety, The time sequence compliance is determined based on whether the current access time falls within an allowed access period. The overall risk value for the current access scenario is obtained by weighted summation of the normalized subject risk, object risk, environmental risk, and behavioral risk. ;

[0009] in, , , , As corresponding weights; Based on the overall risk value Determine the risk level of the current session, and set a periodic check interval based on the risk level to continuously monitor the session and re-collect context information to update the situational risk; Based on the overall risk value, the access scenario is divided into multiple risk levels. For each risk level, the corresponding atomic security capabilities are dynamically combined from the atomic security capability library and matched with auxiliary policy constraints to generate a dynamic security policy. The dynamic security policy is bound to data entities, metadata, and capability interfaces to perform secure processing on the data output of the data layer.

[0010] Furthermore, in the data element encapsulation model for cross-domain interaction security described above, the capability layer includes: The basic operation capability module is designed to meet general data processing needs and encapsulates corresponding data operation interfaces, including at least one of sorting, selection, and filtering interfaces. The advanced application capability module is used to generate customized data intermediate states and operation interfaces by abstracting typical data application patterns. It encapsulates at least one of the following: statistical analysis interface, machine learning interface, and large model training and inference interface.

[0011] Furthermore, based on the data element encapsulation model for cross-domain interaction security described above, the application layer provides four value release modes: The associated release mode is used to fuse and analyze multi-source heterogeneous data, transforming the raw data into reusable knowledge assets; The fusion release model is used to deeply fuse multimodal data to generate predictive models and semantic networks; The multiplier release model is used to couple data with other production factors to promote the integration of the digital economy and the real economy. The circulation release model is used to achieve the social distribution of data value through sharing and trading.

[0012] Furthermore, the data element encapsulation model for cross-domain interaction security, as described above, includes the following encapsulation: Descriptive metadata, structural metadata, management metadata, and service metadata are extracted from the data layer to form the metadata description basis of the data element encapsulation; Based on the interfaces encapsulated in the capability layer and the policies generated in the security layer, the security policies and operational capability interfaces are encapsulated to generate dynamic access policies. The dynamic access policies support dynamic updates of authorization rules according to changes in the access context without changing the main structure of the data element encapsulation component. Data is bound to the hash value of the security policy to prevent policy tampering through hash value verification. The encapsulated data element package is stored in the database, and the registration information is submitted to the registration center. The data element package is encapsulated in JSON format and includes at least four dimensions: identifier, metadata, capability interface, and security policy.

[0013] Furthermore, the data element encapsulation model for cross-domain interaction security described above also includes a data element encapsulation protocol suite. This protocol suite extends the data element encapsulation from a static encapsulation object into a discoverable, parseable, controllable, and assemblable interactive unit, thereby supporting the standardized circulation and secure use of data elements in different application environments. The data element encapsulation protocol suite includes: an identifier resolution protocol and an access acquisition protocol.

[0014] Furthermore, in the data element encapsulation model for cross-domain interaction security described above, the identifier resolution protocol takes the data element encapsulation identifier as input and is used to establish a three-level mapping relationship between the data element encapsulation identifier, the identifier resolution record, and the data element encapsulation object, so as to decouple the encapsulation identifier from the underlying storage location and support external subjects to perform cross-domain discovery and location of data element encapsulation through stable and persistent identifiers. The identifier resolution protocol defines registration, resolution, update, and revocation operations; The registration operation is used to establish a binding relationship between the data element package identifier, the identifier resolution record and the data element package object, and store it as a package registry; The parsing operation is used to query the identifier parsing record corresponding to the identifier of the data element package; The update operation is used to modify the mapping relationship when the package storage is migrated or the server endpoint is changed; The undo operation is used to update the operational status of a package when it fails, expires, or is no longer providing services. The identifier resolution protocol adopts a hierarchical resolution mode, consisting of three levels: root resolution node, domain resolution node, and end resolution node. The root resolution node records the service address and running status of each domain resolution node. When a resolution request is received, the domain prefix is ​​extracted from the data element encapsulation identifier, and the corresponding domain resolution node is queried based on the domain prefix. The domain resolution node manages the namespace in a specific domain and its subordinate end resolution nodes. When it receives a resolution request, it determines the end resolution node responsible for the data element encapsulation based on the authoritative prefix in the data element encapsulation identifier and directs the resolution request to the end resolution node. The endpoint parsing node stores the identifier parsing record corresponding to the data element encapsulation identifier, maintains the mapping relationship between the data element encapsulation identifier and the data element encapsulation object service, and acts as the direct executor of identifier registration, parsing, updating and revocation; The identifier resolution record includes at least the data element encapsulation object type, encapsulation model version, lifecycle status, and object service address.

[0015] Furthermore, in the data element encapsulation model for cross-domain interaction security described above, the access acquisition protocol uses the unique identifier of the data element encapsulation as the access entry point, the encapsulation security policy as the access constraint, and the capability layer interface as the call exit point, supporting secure calls to the encapsulation under multiple subjects, multiple purposes, and multiple scenarios, and transforming traditional static data resources into dynamically operable structured data encapsulations. The access acquisition protocol includes six execution phases: initialization, identifier resolution, request construction, request processing, data processing, and response. (1) Initialization phase: The raw data is processed into standard data element encapsulations and registered to the data element encapsulation registration center to generate a list of data element encapsulations; the list of data element encapsulations includes data element encapsulation identifier, metadata digest, security policy hash value, data access address, version, and data element encapsulation status. (2) Identifier resolution stage: The data user initiates a resolution request to the identifier resolution service based on the unique identifier of the encapsulated component; the encapsulated component retrieval, field parsing, lifecycle management and service endpoint location are realized by relying on the four types of metadata in the metadata layer; at the same time, the policy summary and capability interface list in the list are extracted as the basis for subsequent access control and capability invocation. (3) Request construction stage: After obtaining the controlled access entry, an access request is generated based on the parsed list of capability interfaces and business scenario; the access request includes at least: data element encapsulation identifier, access credential, request capability, request time, random number, and initial context; wherein the request capability is determined based on the list of capability interfaces returned by the capability layer to determine the data operation called in this request; the access credential, request time, random number, and context information serve as inputs for the security layer to perform identity authentication, anti-replay verification, and risk assessment; (4) Request processing phase: This phase is executed collaboratively by the policy execution point (PEP), policy decision point (PDP), and policy information point (PIP) within the trusted third-party organization. After receiving an access request, the Policy Enforcement Point (PEP) retrieves the manifest file based on the data element encapsulation identifier and verifies whether the security policy hash, metadata digest, capability interface digest, and data element encapsulation version carried in the request are consistent with the information registered in the manifest. If the verification fails, the request is rejected directly and an audit record is retained. After the verification is successful, the policy execution point (PEP) verifies the validity of the access credentials, timestamp, and random number, and checks whether the target capability identifier is within the scope of the publicized interface list. If the verification is correct, the request is converted into an XACML request and forwarded to the policy decision point (PDP). The Policy Information Point (PIP) collects three types of access contexts: subject, object, and environment. The Policy Decision Point (PDP) conducts situational risk assessment based on the context, calculates the global risk value by comprehensively considering subject risk, object risk, environmental risk, and behavioral risk, and classifies the risk into three levels: low, medium, and high, and denies the risk. The strategy decision point (PDP) combines the strategy summary bound to the list to retrieve the atomic security capability library and risk-policy mapping table, matches the current risk level to generate a security capability combination, and distributes it to the data element package management center. (5) Data processing stage: The data element package management center dynamically assembles the atomic capabilities defined in the security layer based on the decision results of the policy decision point (PDP), performs security processing on the data in the data layer, and then sends the data back to the policy execution point (PEP). (6) Response phase: The policy execution point (PEP) returns the securely processed data along with the authorized available capability interfaces to the data user.

[0016] Beneficial effects: The data element encapsulation model for cross-domain interaction security provided by this invention organizes a five-layer protocol cluster—data layer, metadata layer, security layer, capability layer, and application layer—into a two-layer architecture: an element layer and a control layer. By constructing this two-layer architecture, data entities and data applications are separated. In the element layer, four types of standardized metadata—description, structure, management, and service—are defined to uniformly abstract and represent the underlying heterogeneous data resources. In the control layer, processing, operation, and security control are implemented during cross-domain data interaction. This two-layer architecture decouples data entities and data applications, enabling flexible encapsulation and secure data flow in different scenarios. Data entities maintain a unified description and stable storage in the element layer, while data applications are configured on demand in the control layer using atomic security policies and capability interfaces. The two establish a controlled association through data element encapsulation identifiers, metadata, policy files, and capability description files. By changing the policy encapsulation and capability interfaces, data element encapsulations adapted to different scenarios can be generated.

[0017] The data element encapsulation model for cross-domain interaction security provided by this invention, on the one hand, utilizes standardized data access and operation interfaces pre-embedded in the control plane capability layer to complete the binding of basic operation capabilities and advanced application capabilities during the encapsulation stage. After receiving the data encapsulation, the user can directly call the capability layer interface to complete data processing and use, improving the interaction efficiency of existing warehouse-style solutions. On the other hand, a dynamic policy binding mechanism is adopted in the control plane to embed atomic security capabilities such as encryption and desensitization into the data encapsulation. At the same time, the context-adaptive policy encapsulation technology is used to support the dynamic loading of different security control policies according to the application scenario, realizing the transformation from passive defense to intrinsic security. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 This application provides a schematic diagram of the data element encapsulation model structure for cross-domain interaction security. Figure 2 A schematic diagram of the security layer protection strategy provided in this application; Figure 3 A schematic diagram of the data element encapsulation access and acquisition protocol provided in this application. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0021] This application provides a data element encapsulation model for cross-domain interaction security, aiming to solve problems such as low efficiency of secure interaction and lack of standardized and configurable data encapsulation models in the process of cross-domain data sharing.

[0022] Figure 1 This is a schematic diagram of the data element encapsulation model structure for cross-domain interaction security provided in this application, as shown below. Figure 1 As shown, the model includes: an element plane and a control plane; the element plane includes a data layer and a metadata layer, the data layer is used to preprocess the collected and stored multi-source heterogeneous data; the metadata layer is used to abstract the data in the data layer to form metadata, thereby realizing a unified representation of multi-source heterogeneous data. The control plane includes a security layer, a capability layer, and an application layer; The security layer is configured with an atomic security capability library, a policy encapsulation module, and an audit trail module. The atomic security capability library contains several independently callable atomic security capabilities. The atomic security capabilities include at least one of encryption, desensitization, watermarking, or digital signature. The policy encapsulation module is used to dynamically combine the atomic security capabilities according to the access context risk value to generate a security policy, and bind the security policy to the data entities in the data layer, the metadata in the metadata layer, and the capability interfaces encapsulated by the capability layer. The audit trail module is used to record the data element encapsulation identifier, calling subject, calling time, calling interface, output summary, and policy version. The capability layer encapsulates standardized data access interfaces and operation interfaces. The application layer is located at the top of the control plane and is used to perform layered release of data value based on the data and capabilities provided by the data layer, metadata layer, and security layer. The element plane and the control plane are decoupled from each other and a controlled association is established through the globally unique data element encapsulation identifier built into the model, the standardized metadata list output by the metadata layer, the security policy file generated by the policy encapsulation module, and the capability interface description file output by the capability layer.

[0023] Specifically, the data element encapsulation model for cross-domain interaction security provided in this application consists of, from bottom to top, a data layer, a metadata layer, a security layer, a capability layer, and an application layer.

[0024] The data layer and metadata layer constitute the feature dimension, which achieves standardized abstraction and representation of heterogeneous data features through a unified modeling method. The feature dimension is responsible for data generation, updating, and maintenance, while the generation and use of metadata are subject to the strategies of the capability layer, management layer, and security layer.

[0025] The security layer, capability layer, and application layer form a control plane, which is responsible for the processing, operation, and security control of data during cross-domain interactions, ultimately realizing the value release of data in different scenarios.

[0026] The data layer is responsible for collecting and storing multi-source heterogeneous energy data and preprocessing the collected data. The data preprocessing steps include: first, using data cleaning tools to deduplicate, denoise, detect outliers, and fill in missing values; then, eliminating differences in data units and standardizing the data; finally, for multi-source data with business relationships, establishing associations based on unified subject identifiers, equipment identifiers, or spatiotemporal attributes, and integrating them according to application requirements; data that does not need to be integrated is stored independently after standardization.

[0027] The metadata layer is responsible for initially abstracting the raw data from the data layer to form metadata, thereby achieving a unified representation of multi-source heterogeneous data. Specifically, it is divided into descriptive metadata, structural metadata, management metadata, and service metadata.

[0028] Metadata descriptions are used to abstractly describe data content. Table 1 shows the design content of the metadata descriptions.

[0029] Table 1 describes the metadata.

[0030] Structure metadata is used to abstractly describe the structure of data. Table 2 shows the design content of structure metadata.

[0031] Table 2 Structure Metadata

[0032] Management metadata is used to abstractly describe the lifecycle, security attributes, and ownership claims of data. Table 3 shows the design content of management metadata.

[0033] Table 3 Management Metadata

[0034] Service metadata is used to define the upper-level functional interfaces supported by the data. Table 4 shows the design content of service metadata.

[0035] Table 4 Service Metadata Design

[0036] Figure 2 A schematic diagram of the security layer protection strategy provided for this application is shown below. Figure 2 As shown, the security layer employs a multi-layered protection mechanism, coupled with a context-adaptive strategy encapsulation mechanism, covering the entire lifecycle of data acquisition, transmission, storage, consumption, and destruction, achieving end-to-end security and controllability. The security layer breaks down various security functions such as encryption, de-identification, watermarking, and digital signatures into several independently callable atomic security capabilities; combined with... Figure 2 The data flow protection requirements at each stage are illustrated. Access scenario risks are dynamically calculated based on data sensitivity levels, user role levels, environmental risks, and behavioral risks. Corresponding atomic security capabilities are matched and combined to generate a security policy file. This security policy file is then bound and encapsulated with the data payload, metadata file, and capability interface file to form a data element encapsulation with security control capabilities. The security policy carried by this encapsulation flows synchronously with the data. During the acquisition stage, dynamic desensitization and quality assessment are performed; during the transmission stage, encrypted transmission and digital signatures are performed; during the storage stage, static encryption and backup storage are performed; during the consumption stage, authentication and authorization and data metering are performed; and during the destruction stage, dynamic overwriting and hard data destruction are performed. The security layer relies on a phased and differentiated combination of atomic security capabilities to ensure trusted data flow, secure computation, and controllable use throughout the entire process.

[0037] The security layer provided by this invention includes an atomic security capability library, a policy binding module, and an audit trail module. The atomic security capability library provides basic security capabilities such as encryption, data masking, and watermarking. The dynamic policy encapsulation module dynamically combines atomic security capabilities based on access context risk values ​​to generate security policies, and binds the generated security policies to data entities in the data layer, metadata in the metadata layer, and capability interfaces encapsulated in the capability layer, enabling the security policies to flow synchronously with the data. The audit trail module fully records the data element encapsulation identifier, calling entity, calling time, calling interface, output summary, and policy version information.

[0038] To address the need for dynamic security policy encapsulation, this invention designs a context-adaptive policy encapsulation mechanism to achieve dynamic data security protection. First, various security control logics are decomposed into standardized atomic security capabilities. Then, based on access context risks, these atomic security capabilities are dynamically assembled to generate an adaptive security policy. Finally, the security policy is bound to the data. Specifically, the generation of this context-adaptive policy includes the following steps: Access Context Collection: The system collects four types of context information. , including the main context Context of the object Context and behavioral context The environmental context includes device, network, time, and location information.

[0039] Contextual risk quantification: This application assesses the risk of a single access at time [time]. The state retains four types of normalization factors that can be obtained from the package metadata or the runtime environment: Main risk ,in, The level of trust in the authenticated entity.

[0040] The subject's trust level is mapped from role levels L1-L5 to a trust score of [0,1]. This score is obtained as follows: After the subject completes identity authentication, the system determines the subject's identity based on the access token and reads its role level (…). ) and basic trust level ( This is used as the main context for access decisions in the security layer, and the basic level of trust is assigned by the system according to different roles.

[0041] Object risk Data sensitivity levels L1-L5 are mapped to [0,1].

[0042] Environmental risk is determined by location credibility. Network stability Equipment safety and timing compliance jointly determined, , , , It is a value in the interval [0,1].

[0043] Environmental risks ; Behavioral risk ,in, This refers to the number of accesses in the last 5 minutes, specifically the number of times the same certified entity initiates access to and access to the same data element package and calls to capabilities (capability interfaces) within the current 5-minute statistical window.

[0044] When a visitor locates a data feature encapsulation (DWID) according to the protocol and requests the creation of an access session, the counter increments by 1. After the session is established, the counter increments by 1 for each successful invocation of a data feature encapsulation capability. Frequency thresholds configured for the deployment side.

[0045] Subject level and trust level are obtained from authenticated user information; data sensitivity level is obtained from the metadata of data element encapsulation; and behavior frequency is maintained by a server-side counter. The environmental context is resolved by the server based on the client IP, network evidence, device characteristics, and server time.

[0046] Location trustworthiness indicates the degree of trustworthiness of the source network region. The client IP is obtained directly from the network connection by the server, and the trustworthiness is determined by the source IP of the access request.

[0047] Network stability describes the degree to which the current network connection type supports continuous and reliable data usage. It is reported by the client via networkType or X-Dataware-Network-Type. The current rating is 1.0 for Ethernet; 0.6 for Wi-Fi / 4G; 0.4 for 3G; 0.1 for 2G; and 0.5 for unknown.

[0048] Device security level indicates whether the security status of the access terminal meets the usage requirements of the data element encapsulation component.

[0049]

[0050] Timing compliance indicates whether the current access time meets the allowed usage time specified by the data element encapsulation security policy.

[0051] The current system uses server time as the standard, and the permitted access period is from 8:00 to 22:00.

[0052] in, This represents the normalized distance between the current time and the boundary of the most recent allowed access period. The further away from the allowed period, the lower the time sequence compliance.

[0053] Contextual risk assessment: The overall risk value of the current access context is obtained by weighted summation of the four types of risks obtained after quantification. The weights sum to 1. Simultaneously, based on the current session risk level, the session is continuously monitored periodically at 1-second (High), 2-second (Medium), and 3-second (Low) intervals to collect contextual information and reassess the situational risk.

[0054]

[0055] Risk Classification and Policy Encapsulation: The system generates dynamic security policies and auxiliary policy constraints (including rate limiting, field filtering, row count display, and audit rules) based on the access context risk. Dynamic security policies are generated by dynamically combining encryption, desensitization, watermarking, and digital signature capabilities from the atomic security capability library. Each capability in the atomic security capability library has an independent and standardized calling interface. When the access context is at the LOW level, the security layer performs digital signatures on the output results and implements basic rate limiting and full audit auxiliary policy constraints; at the MEDIUM level, the security layer combines partial desensitization, metadata watermarking, and digital signature atoms, and implements field selection and row count limit auxiliary policy constraints; at the HIGH level, the security layer combines strong desensitization, traceable watermarking, encryption, and digital signature atoms, outputting only securely processed public or low-sensitivity fields; at the DENY level, no data operation interfaces are opened, no data output processing chain is generated, and existing sessions are revoked.

[0056] The capability layer encapsulates standardized data access and operation interfaces, including basic operational capabilities and advanced application capabilities. The capability layer comprises a basic operational capability module and an advanced application capability module.

[0057] Basic operational capabilities module: Basic operational capabilities cater to general data processing needs, providing corresponding data operation interfaces, including sorting, selection, and filtering.

[0058] Sorting: Sort data in ascending or descending order by a specified field to quickly retrieve the data you need.

[0059] Selection: Filter target data subsets according to requirements, and support multi-field combination query selection.

[0060] Filtering: The rules engine retains only the core data required for the business and filters out irrelevant data.

[0061] Mapping: Provides data format conversion and field mapping functions.

[0062] Advanced application capability module: Advanced application capabilities abstract typical data application patterns to generate customized data intermediate states and operation interfaces, enabling plug-and-play functionality at the application layer. Core modules include statistical analysis, machine learning, and large-scale model training and inference.

[0063] As the top layer of the model, the application layer is business-oriented and ultimately realizes the value of data. The application layer provides four value release modes: Associated Release Model: Explore the implicit relationships and synergistic effects between multi-source heterogeneous data, break down data silos, establish a cross-industry, cross-organization, and cross-domain data interconnection mechanism, and use big data analysis, artificial intelligence and other technologies to integrate and analyze scattered large-scale data, transforming raw data into reusable knowledge assets.

[0064] Fusion release mode: Deeply integrate multimodal data to break through the value boundaries of single data and generate systematic knowledge or higher-order value that goes beyond the simple summation of the original data, similar to a nuclear fusion reaction - integrating scattered fragmented data elements into knowledge bodies such as predictive models and semantic networks under specific technical conditions.

[0065] Multiplier release model: Deeply couples data with other production factors to stimulate multiplier effects, promotes the integration of the digital economy and the real economy, and improves production efficiency and the level of industrial digitalization.

[0066] Circulation and Release Model: Focusing on the market-based allocation of data elements, this model achieves the social distribution of data value through sharing and trading. Based on blockchain and smart contract technologies, a decentralized data trading platform is built to ensure clear data ownership and transparent pricing.

[0067] Existing technologies struggle to effectively decouple data entities from application logic for cross-domain data element circulation, and they also exhibit significant shortcomings in dynamic binding and full lifecycle management of security policies. Furthermore, when applied to multi-entity cross-domain data sharing scenarios, these approaches either employ a warehouse-style storage model leading to low interaction efficiency, or rely on functional encapsulation solutions to bind data to specific algorithms with fixed encapsulation granularity, failing to flexibly adapt to the needs of different business scenarios and security levels, and thus hindering the large-scale market circulation of data elements. In contrast, the five-layer component-two-layer plane data element encapsulation model proposed in this invention organizes a five-layer protocol cluster (data layer, metadata layer, security layer, capability layer, and application layer) into a two-layer plane structure: an element plane and a control plane. By constructing this two-layer plane architecture, data entities and data applications are separated. The element plane defines four types of standardized metadata—description, structure, management, and service—to uniformly abstract and represent the underlying heterogeneous data resources. The control plane implements processing, operation, and security control during cross-domain data interaction. By constructing a two-layer architecture of elements and control, data entities and data applications are decoupled, enabling data to be flexibly encapsulated and securely circulated in different scenarios. Data entities maintain a unified description and stable storage in the element layer, while data applications are configured on demand in the control layer in the form of atomic security policies and capability interfaces. The two establish a controlled association through data element encapsulation identifiers, metadata, policy files, and capability description files. By changing policy packages and capability interfaces, data element encapsulations adapted to different scenarios can be generated.

[0068] Flexible encapsulation of data elements is a secure organizational model in the data supply phase. Based on differentiated business needs and security requirements, it flexibly binds and integrates data, access control policies, and metadata. By adjusting the encapsulation granularity and embedding adaptive security policies, it achieves refined data management, flexible authorization, and efficient sharing, enhancing the security and controllability of the entire cross-domain data flow process.

[0069] The encapsulation of the data element encapsulation model in this application specifically includes the following steps: Step 1: Extract descriptive metadata, structural metadata, management metadata, and service metadata to form the metadata description basis of the data element encapsulation, which is used to provide a searchable and parsable external interface during cross-domain transfer. Step 2: Based on the design of the capability layer and security layer, encapsulate security policies and operational capability interfaces to generate dynamic access policies. These dynamic access policies support dynamic updates of authorization rules based on changes in access scenarios without altering the main structure of the encapsulated components. Data is bound to the hash value of the security policy to prevent policy tampering through hash value verification, ensuring that data access control remains effective even after crossing organizational boundaries. Step 3: Store the encapsulated data element package in the database and submit the registration information to the registration center; the data element package is encapsulated in JSON format and includes at least four dimensions: identifier, metadata, capability interface and security policy, to support addressability, accessibility and controllability.

[0070] Specifically, based on the five-layer model architecture proposed in this application, the complete encapsulation process of the data element encapsulation component is divided into three steps: encapsulation information definition extraction, dynamic security policy encapsulation, and data encapsulation component storage. The detailed descriptions of each step are as follows: Encapsulation information definition extraction: Extracting the metadata required to construct the encapsulation from the original resources. First, the attributes of the feature plane are extracted. Based on the four types of metadata standards designed in the metadata layer, description, structure, management and service metadata are extracted from the data layer to form the identification base of the encapsulation, providing a searchable and parsable external call interface for cross-domain flow.

[0071] Dynamic security policy encapsulation: Embeds adaptive dynamic access policies to establish a relationship between data, security policies, and operational capabilities; authorization rules can be dynamically updated as access context risks change without modifying the main architecture of the encapsulation component. Binding data to security policy hash digests ensures that data holders' customized data access permissions remain effective after cross-organizational transfers; relying on hash digest verification mechanisms, it prevents security policies from being tampered with.

[0072] Data encapsulation storage: The encapsulated data element components are stored in the database, and a complete set of registration information is submitted to the data element encapsulation registry center. The standardized data element encapsulation uses JSON format for structural encapsulation, covering four core modules: unique identifier, metadata, capability interface, and security policy, and possesses addressable, accessible, and manageable characteristics.

[0073] Furthermore, the data element encapsulation model for cross-domain interaction security provided in this application also includes a data element encapsulation protocol suite. The data element encapsulation protocol suite is used to extend the data element encapsulation from a static encapsulation object into an interactive unit that can be discovered, parsed, controlled, and assembled, so as to support the standardized circulation and secure use of data elements in different application environments. The data element encapsulation protocol suite includes: an identifier resolution protocol and an access acquisition protocol.

[0074] Specifically, to support the trusted circulation and controlled use of data element encapsulation components across subjects, systems, and scenarios, this application further designs a data element encapsulation component protocol suite based on the five-layer componentization-two-layer bit-plane model. The goal of the protocol design is to extend the data element encapsulation components from static encapsulation objects into discoverable, parseable, controllable, and assemblable interactive units, thereby supporting the standardized circulation and secure use of data elements in different application environments.

[0075] (1) Identifier Resolution Protocol Design To support stable discovery and resolution of data feature encapsulated documents (DWIDs) in multi-subject, cross-platform, and cross-security domain environments, this application designs a data feature encapsulated document identifier resolution protocol. This protocol takes the data feature encapsulated document identifier (DWID) as input, and determines the current state of the data feature encapsulated document and its associated data feature encapsulated document object service by querying the record information associated with the identifier. The data feature encapsulated document object service then locates the specific data feature encapsulated document object based on the identifier (i.e., the requester uses the DWID to query the associated data feature encapsulated document resolution record (DWR), which contains information such as the data feature encapsulated document status and object service address. When the data feature encapsulated document is available, the requester further accesses the corresponding data feature encapsulated document object service; the object service locates the specific data feature encapsulated document object from its internal object directory or storage system based on the DWID). Protocol establishment... A three-tiered mapping relationship exists, where DWID represents the data feature encapsulation identifier, DWR (Dataware Resolution Record) represents the data feature encapsulation identifier record associated with this identifier, and DW (Dataware) represents the identified data feature encapsulation object. This hierarchical mapping method decouples the encapsulation identifier from the underlying storage location, allowing external entities to still use stable and persistent identifiers for discovery and location even if the encapsulation is migrated or the server endpoint changes.

[0076] The protocol defines four basic operations: registration, resolution, update, and revocation. The registration operation is used to establish the binding relationship between DWID, identifier record, and data element encapsulation object and store it as an encapsulation registry; the resolution operation is used to query the identifier record of the data element encapsulation corresponding to the DWID; the update operation is used to modify the mapping relationship when the encapsulation storage is migrated or the service endpoint is changed; and the revocation operation is used to update the status when the encapsulation becomes invalid, expires, or no longer provides services.

[0077] The data element encapsulation identifier resolution adopts a hierarchical resolution model, consisting of three levels: root resolver, domain resolver, and endpoint resolver. The root resolver, located at the highest level of the identifier resolution system, records information such as the service address and operating status of each domain resolver. When a DWID resolution request is received, the root resolver extracts the prefix from the identifier and queries the corresponding domain resolver's service information based on the prefix. Domain resolvers, located in the middle layer of the system, are responsible for managing the namespaces within a specific domain and their subordinate endpoint resolvers. Upon receiving a request, a domain resolver determines the endpoint resolver responsible for that data element encapsulation based on the DWID and directs the resolution request to the appropriate endpoint. End endpoint resolvers, located at the leaf layer of the identifier resolution system, are the direct executors of data element encapsulation identifier registration, resolution, update, and revocation. End endpoint resolvers store the data element encapsulation resolution record (DWR) corresponding to the DWID and maintain the mapping relationship between the DWID and the data element encapsulation object service. The data element encapsulation resolution record mainly includes the data element encapsulation object type, encapsulation model version, lifecycle status, and object service address.

[0078] (2) Access Acquisition Protocol After the data element encapsulation completes its identifier resolution, external entities can obtain access to the encapsulation. To further standardize the access, invocation, and result return process of the encapsulation in a cross-domain environment, this application designs a data element encapsulation access and acquisition protocol. Traditional invocation methods rely on services provided by the platform. The access and acquisition protocol proposed in this application uniquely identifies the data element encapsulation as the entry point, uses the encapsulation security policy as a constraint, and uses the capability interface as the invocation exit point. This is used to support the secure use of the data element encapsulation under different entities, different purposes, and different scenarios, transforming data from a traditional static resource into a dynamically operable structured data encapsulation.

[0079] Figure 3 A schematic diagram of the data element encapsulation access and acquisition protocol provided in this application is shown below. Figure 3 As shown, the access acquisition protocol consists of six phases: initialization phase, identifier resolution phase, request construction phase, request processing phase, data processing phase, and response phase. Each phase is described in detail below.

[0080] Initialization phase: Based on the aforementioned flexible encapsulation process, the system processes the original data resources into standardized data element encapsulations and submits them to the data element encapsulation registration center to complete registration, generating a list of data element encapsulations; the list information includes a unique identifier for the data element encapsulation, metadata hash digest, security policy hash digest, data access address, version information, and data element encapsulation status.

[0081] This phase predefines the risk-atomic security capability mapping relationship and establishes mapping rules from risk level to atomic security capability set: risk level corresponds to atomic security capability combination; during system operation, the matching atomic security capability combination can be retrieved based on the risk level of the real-time access context. .

[0082] The system stores the security policy ontology and security policy hash digest as pairs in the policy database; when subsequent policy decision points perform access determination, they directly retrieve the matching security policy from this database. This stage corresponds to... Figure 3 Step 1: Package List Registration Process.

[0083] Identifier Resolution Phase: The data usage direction identifier resolution service initiates a resolution request based on the globally unique identifier of the data element encapsulation component. The identifier resolution service first verifies the format of this unique identifier and distinguishes its resolution domain based on the identifier prefix. Then, it searches the data element encapsulation component registry center to verify whether the corresponding data element encapsulation component has completed registration and whether its current running status is valid. This phase corresponds to... Figure 3 Step ② Access and obtain, Step ③ Package identifier parsing.

[0084] If the data element encapsulation is registered and valid, the parsing service returns a manifest file summary to the data user. This summary includes a metadata summary, a security policy summary, a list of capability interfaces, and controlled access points. If the data element encapsulation is expired or revoked, controlled access points will not be issued. In this stage of the process, the four types of metadata in the metadata layer play different roles: descriptive metadata is used to realize the retrieval and discovery of data element encapsulations; structural metadata is used to parse data fields and data formats; management metadata is used to determine the running status of data element encapsulations and manage their entire lifecycle; and service metadata is used to locate service endpoints and retrieve available operational capabilities. Meanwhile, the security policy summary and the list of capability interfaces returned in this stage will serve as the basis for subsequent access control and data capability invocation.

[0085] Request Construction Phase: After obtaining a controlled access point, the data user constructs an access request based on the list of capability interfaces and application scenario descriptions returned during the parsing phase. The access request must carry at least the following information: a unique identifier for the data element encapsulation, access credentials, requested capability, request time, random number, and initial context information.

[0086] The request capability determines the data operation invoked in this request based on the list of capability interfaces returned by the capability layer; the access credentials, request time, random number, and initial context information serve as inputs for the security layer to conduct identity authentication risk assessment.

[0087] Request processing phase: This phase is completed collaboratively by the Policy Enforcement Point (PEP), Policy Decision Point (PDP), and Policy Information Point (PIP) within the trusted third-party organization. This phase corresponds to... Figure 3 The entire interaction process from steps ② to ⑩ is as follows. The specific execution logic is as follows: 1. Integrity summary verification After receiving an access request from a data user, the Policy Enforcement Point (PEP) first retrieves the corresponding manifest file based on the unique identifier (DID) of the data element encapsulation. It then verifies whether the policy digest hash(P) carried in the request matches the policy digest registered in the manifest file. Simultaneously, it verifies whether the metadata digest, capability interface digest, and data element encapsulation version information match. If the digest verification fails, it indicates a risk of policy, metadata, or capability interface being tampered with, replaced, or bypassed. The PEP directly rejects the request and triggers an audit log. If the digest verification passes, the PEP continues to verify the validity of the access credentials, the legality of the request timestamp and random number, and checks whether the requested capability belongs to the capability interface list declared in the manifest file.

[0088] 2. Request forwarding and context collection After all pre-verifications are completed, the user's access request is converted into an XACML request and forwarded to the Policy Decision Point (PDP). At this point, the Policy Information Point (PIP) collects the context information of the current access request, including the subject context, object context, environment context, and behavior context.

[0089] 3. Quantitative classification of situational risks The Policy Decision Point (PDP) performs a situational risk assessment based on the contextual information provided by the Policy Information Point (PIP). It comprehensively calculates the risks of the subject, object, environment, and behavior to form the risk value of the current access situation and maps it to low, medium, and high risk levels and deny.

[0090] 4. Security capability matching and distribution The Policy Decision Point (PDP) locates the atomic security capability library and risk policy mapping table in the security layer based on the policy summary bound to the manifest file. It then generates the policy security capabilities for this access based on the current risk level and forwards the assembled security capabilities to the Data Element Encapsulation Management Center. This policy instance not only provides an access decision (allow or deny) but also determines the allowed capability interfaces and the combination of atomic security capabilities to be executed. For example, in a low-risk scenario, the PDP can allow query capabilities and only perform summary verification and access measurement; in a medium-risk scenario, the PDP can restrict the output granularity to aggregated results and require audit logs; in a high-risk scenario, the PDP can further perform field desensitization, watermark embedding, encrypted transmission, and strong auditing, or directly deny access when the risk exceeds a threshold.

[0091] Data processing stage: This stage corresponds to Figure 3 intermediate steps Data security processing: The Data Element Encapsulation Management Center dynamically assembles atomic capabilities defined in the security layer based on the decision results of the Policy Decision Point (PDP), processes the data in the data layer, and then forwards the processed data to the PEP.

[0092] Response phase: This phase corresponds to Figure 3 intermediate steps , The policy execution point (PEP) returns the securely processed data to the data user.

[0093] Existing data encapsulation technologies based on standardized product forms offer solutions that provide measurable, standardized delivery carriers for data products, addressing static measurement and compliance registration issues. However, they focus on the static description and packaging of data products, lacking the ability to dynamically bind security policies. Security constraints typically exist independently as platform rules, management system configurations, or contractual terms, posing risks of unauthorized data retention, unauthorized secondary forwarding, and misuse in dynamic cross-domain data transfer scenarios, resulting in insufficient security controllability. This application designs a context-adaptive strategy encapsulation mechanism at the security layer, abstracting security capabilities such as encryption, de-identification, and hashing into independently callable atomic security capabilities. During the access phase, the system combines atomic security capabilities based on the access context risk, determining the currently requested accessible fields, encryption status, output granularity, de-identification method, and audit requirements.

[0094] Existing data encapsulation technologies based on data element encapsulation components propose this novel data abstraction as the basic unit for the circulation and use of data elements. By standardizing the encapsulation of heterogeneous and multi-dimensional data, it possesses three basic capabilities: controllability, measurability, and assemblability. While this lays a theoretical foundation for the standardized construction of data elements, related work is still in its early stages. This application, building upon the basic abstraction of data elements, further proposes a five-layer component-two-dimensional data element encapsulation component model, along with corresponding identifier resolution and access acquisition protocols. This extends the data element encapsulation component from a static encapsulation object into an interactive unit that is discoverable, parsable, callable, and auditable. On the one hand, through the mapping mechanism of encapsulation component identifier—registration list—data encapsulation component object, the metadata, storage location, service endpoint, and capability interface of the encapsulation component are associated, enabling external entities to complete cross-domain discovery and location based on stable identifiers. On the other hand, through the context-adaptive strategy encapsulation mechanism of the security layer, atomic security capabilities such as encryption, desensitization, and watermark embedding are dynamically assembled according to the access context risk. During the execution of the access acquisition protocol, the request subject, calling capabilities, accessible fields, and output granularity are verified and controlled, extending the data element encapsulation component from a static encapsulation object to a controllable and assemblable interactive unit, supporting the collaboration of data element encapsulation components in cross-platform and cross-system environments.

[0095] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A data element encapsulation model for cross-domain interaction security, characterized in that, include: Element planes and control planes; The feature plane includes a data layer and a metadata layer. The data layer is used to preprocess the collected and stored multi-source heterogeneous data. The metadata layer is used to abstract the data in the data layer to form metadata, thereby achieving a unified representation of multi-source heterogeneous data. The control plane includes a security layer, a capability layer, and an application layer; The security layer is configured with an atomic security capability library, a dynamic policy encapsulation module, and an audit trail module; The atomic security capability library contains several independently callable atomic security capabilities; The atomic security capabilities include at least one of encryption, desensitization, watermarking, or digital signature; The dynamic policy encapsulation module is used to dynamically combine the atomic security capabilities with the root access scenario risk value to generate a security policy, and bind the security policy to the data entities in the data layer, the metadata in the metadata layer, and the capability interfaces encapsulated in the capability layer; The audit trail module is used to record the data element package identifier, calling entity, calling time, calling interface, output summary, and policy version; The capability layer encapsulates standardized data operation interfaces; The application layer is located at the top of the control plane and is used to release data value in layers based on the data and capabilities provided by the data layer, metadata layer, and security layer. The element plane and the control plane are decoupled from each other, and a controlled association is established through the globally unique data element encapsulation identifier built into the model, the standardized metadata list output by the metadata layer, the security policy file generated by the policy encapsulation module, and the capability interface description file output by the capability layer.

2. The data element encapsulation model for cross-domain interaction security according to claim 1, characterized in that, The data layer preprocesses the collected and stored multi-source heterogeneous data by: firstly, using data cleaning tools to deduplicate, denoise, detect outliers, and fill in missing values; then, eliminating differences in the dimensions of the data and standardizing the data; finally, for multi-source data with business relationships, establishing associations based on unified subject identifiers, device identifiers, or spatiotemporal attributes, and integrating and storing them according to application requirements; and storing data that does not need to be integrated independently after standardization.

3. The data element encapsulation model for cross-domain interaction security according to claim 1, characterized in that, The metadata layer includes: descriptive metadata, structural metadata, management metadata, and service metadata; The descriptive metadata is used to abstractly describe the data content; The structural metadata is used to abstractly describe the structure of the data; The management metadata is used to abstractly describe the data's lifecycle, security attributes, and ownership claims; The service metadata is used to define the upper-level functional interfaces supported by the data.

4. The data element encapsulation model for cross-domain interaction security according to claim 1, characterized in that, The generation of the security policy includes: collection Context information The context information Includes: main context Context of the object Context and behavioral context ; Wherein, the main context Obtain from the authentication information of the current accessing entity; The object context Obtained from the management metadata output by the metadata layer; The environment context This includes location information, network information, device information, and time information; The location information is obtained by the server based on the source IP address of the access request; the network information is determined by the client through the network type reported in the request header; the device information is obtained by the server based on the device characteristics carried in the access request; and the time information is determined by the server's system time. The behavioral context The frequency of calls to the capability interfaces in the capability layer is obtained by the server. Based on the collected context information, the risks of each entity are quantified. Object risk Environmental risks and behavioral risks And normalize each risk factor; ,in, The level of trust in the authenticated entity; ; ,in This represents the number of visits in the last 5 minutes. in, For location credibility, For network stability, For equipment safety, The time sequence compliance is determined based on whether the current access time falls within an allowed access period. The overall risk value for the current access scenario is obtained by weighted summation of the normalized subject risk, object risk, environmental risk, and behavioral risk. ; in, , , , As corresponding weights; Based on the overall risk value Determine the risk level of the current session, and set a periodic check interval based on the risk level to continuously monitor the session and re-collect context information to update the situational risk; Based on the overall risk value, the access scenario is divided into multiple risk levels. For each risk level, the corresponding atomic security capabilities are dynamically combined from the atomic security capability library and matched with auxiliary policy constraints to generate a dynamic security policy. The dynamic security policy is bound to data entities, metadata, and capability interfaces to perform secure processing on the data output of the data layer.

5. The data element encapsulation model for cross-domain interaction security according to claim 1, characterized in that, The capability layer includes: The basic operation capability module is designed to meet general data processing needs and encapsulates corresponding data operation interfaces, which include at least one of sorting, selection, and filtering interfaces. The advanced application capability module is used to generate customized data intermediate states and operation interfaces by abstracting typical data application patterns. It encapsulates at least one of the following: statistical analysis interface, machine learning interface, and large model training and inference interface.

6. The data element encapsulation model for cross-domain interaction security according to claim 1, characterized in that, The application layer provides four value release modes: The associated release mode is used to fuse and analyze multi-source heterogeneous data, transforming the raw data into reusable knowledge assets; The fusion release model is used to deeply fuse multimodal data to generate predictive models and semantic networks; The multiplier release model is used to couple data with other production factors to promote the integration of the digital economy and the real economy. The circulation release model is used to achieve the social distribution of data value through sharing and trading.

7. The data element encapsulation model for cross-domain interaction security according to claim 1, characterized in that, The encapsulation of the data element encapsulation model includes: Descriptive metadata, structural metadata, management metadata, and service metadata are extracted from the data layer to form the metadata description basis of the data element encapsulation; Based on the interfaces encapsulated in the capability layer and the policies generated in the security layer, the security policies and operational capability interfaces are encapsulated to generate dynamic access policies. The dynamic access policies support dynamic updates of authorization rules according to changes in the access context without changing the main structure of the data element encapsulation component. Data is bound to the hash value of the security policy to prevent policy tampering through hash value verification. The encapsulated data element package is stored in the database, and the registration information is submitted to the registration center. The data element package is encapsulated in JSON format and includes at least four dimensions: identifier, metadata, capability interface, and security policy.

8. The data element encapsulation model for cross-domain interaction security according to claim 1, characterized in that, It also includes a data element encapsulation protocol suite, which is used to extend data element encapsulations from static encapsulation objects into discoverable, parseable, controllable, and assemblable interactive units to support the standardized circulation and secure use of data elements in different application environments; The data element encapsulation protocol suite includes: an identifier resolution protocol and an access acquisition protocol.

9. The data element encapsulation model for cross-domain interaction security according to claim 8, characterized in that, The identifier resolution protocol takes the data element encapsulation component identifier as input and is used to establish a three-level mapping relationship between the data element encapsulation component identifier, the identifier resolution record and the data element encapsulation component object, so as to decouple the encapsulation component identifier from the underlying storage location and support external entities to perform cross-domain discovery and location of data element encapsulation components through stable and persistent identifiers. The identifier resolution protocol defines registration, resolution, update, and revocation operations; The registration operation is used to establish a binding relationship between the data element package identifier, the identifier resolution record and the data element package object, and store it as a package registry; The parsing operation is used to query the identifier parsing record corresponding to the identifier of the data element package; The update operation is used to modify the mapping relationship when the package storage is migrated or the server endpoint is changed; The undo operation is used to update the operational status of a package when it fails, expires, or is no longer providing services. The identifier resolution protocol adopts a hierarchical resolution mode, consisting of three levels: root resolution node, domain resolution node, and end resolution node. The root resolution node records the service address and running status of each domain resolution node. When a resolution request is received, the domain prefix is ​​extracted from the data element encapsulation identifier, and the corresponding domain resolution node is queried based on the domain prefix. The domain resolution node manages the namespace in a specific domain and its subordinate end resolution nodes. When it receives a resolution request, it determines the end resolution node responsible for the data element encapsulation based on the authoritative prefix in the data element encapsulation identifier and directs the resolution request to the end resolution node. The endpoint parsing node stores the identifier parsing record corresponding to the data element encapsulation identifier, maintains the mapping relationship between the data element encapsulation identifier and the data element encapsulation object service, and acts as the direct executor of identifier registration, parsing, updating and revocation; The identifier resolution record includes at least the data element encapsulation object type, encapsulation model version, lifecycle status, and object service address.

10. The data element encapsulation model for cross-domain interaction security according to claim 8, characterized in that, The access and acquisition protocol uses the unique identifier of the data element encapsulation component as the access entry point, the encapsulation component security policy as the access constraint, and the capability layer interface as the call exit point. It supports secure calls to encapsulation components under multiple subjects, multiple purposes, and multiple scenarios, and transforms traditional static data resources into dynamically operable structured data encapsulation components. The access acquisition protocol includes six execution phases: initialization, identifier resolution, request construction, request processing, data processing, and response. (1) Initialization phase: The raw data is processed into standard data element encapsulations and registered to the data element encapsulation registration center to generate a list of data element encapsulations; the list of data element encapsulations includes data element encapsulation identifier, metadata digest, security policy hash value, data access address, version, and data element encapsulation status. (2) Identifier resolution stage: The data user initiates a resolution request to the identifier resolution service based on the unique identifier of the encapsulated component; the encapsulated component retrieval, field parsing, lifecycle management and service endpoint location are realized by relying on the four types of metadata in the metadata layer; at the same time, the policy summary and capability interface list in the list are extracted as the basis for subsequent access control and capability invocation. (3) Request construction stage: After obtaining the controlled access entry, an access request is generated based on the parsed list of capability interfaces and business scenario; the access request includes at least: data element encapsulation identifier, access credential, request capability, request time, random number, and initial context; wherein the request capability is determined based on the list of capability interfaces returned by the capability layer to determine the data operation called in this request; the access credential, request time, random number, and context information serve as inputs for the security layer to perform identity authentication, anti-replay verification, and risk assessment; (4) Request processing phase: This phase is executed collaboratively by the policy execution point (PEP), policy decision point (PDP), and policy information point (PIP) within the trusted third-party organization. After receiving an access request, the Policy Enforcement Point (PEP) retrieves the manifest file based on the data element encapsulation identifier and verifies whether the security policy hash, metadata digest, capability interface digest, and data element encapsulation version carried in the request are consistent with the information registered in the manifest. If the verification fails, the request is rejected directly and an audit record is retained. After the verification is successful, the policy execution point (PEP) verifies the validity of the access credentials, timestamp, and random number, and checks whether the target capability identifier is within the scope of the publicized interface list. If the verification is correct, the request is converted into an XACML request and forwarded to the policy decision point (PDP). The Policy Information Point (PIP) collects three types of access contexts: subject, object, and environment. The Policy Decision Point (PDP) conducts situational risk assessment based on the context, calculates the global risk value by comprehensively considering subject risk, object risk, environmental risk, and behavioral risk, and classifies the risk into three levels: low, medium, and high, and denies the risk. The strategy decision point (PDP) combines the strategy summary bound to the list to retrieve the atomic security capability library and risk-policy mapping table, matches the current risk level to generate a security capability combination, and distributes it to the data element package management center. (5) Data processing stage: The data element package management center dynamically assembles the atomic capabilities defined in the security layer based on the decision results of the policy decision point (PDP), performs security processing on the data in the data layer, and then sends the data back to the policy execution point (PEP). (6) Response phase: The policy execution point (PEP) returns the securely processed data along with the authorized available capability interfaces to the data user.