An access control system and method based on ABAC model
Through an access control system based on the ABAC model, combined with XACML and NLP technology, RBAC cannot meet the needs of complex distributed systems, and flexible resource management and simplified permission strategy management are realized, improving user experience and business visibility.
Patent Information
- Application Number
- CN202211633151.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-12-19
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2042-12-19
AI Technical Summary
Role-based access control model (RBAC) cannot meet the business needs of complex distributed systems. Centralized access control permission management is complex and user operation experience is poor, making it difficult to meet the permission management needs of the ABAC model.
An access control system based on the ABAC model was designed, using authorization service center, policy management service center and authentication service center, combining XACML description permission policy and NLP algorithm engine, and simplifying the use of heterogeneous microservice architecture and natural language processing technology. Users write permission policies, generate XACML permission policies and implement access control.
It realizes a more flexible resource management granularity, simplifies XACML policy description, reduces learning costs, improves business visibility and readability, and is suitable for collaborative management of access control policies for complex distributed systems.
Smart Images

Figure CN116032579B_ABST
Abstract
Description
Technical Field
[0001] The present application belongs to the field of network and information security technology, and specifically relates to an access control system and method based on an ABAC model. Background Art
[0002] Access control technology is an important technical means to protect network information resources. Its goal is to manage user access rights on the basis of users sharing system resources to the greatest extent. According to the system security policy, the authorized services or resources are provided to legitimate users, and the service requests of legitimate users that exceed their authority and the unauthorized access requests of illegal users are rejected, so as to prevent illegal tampering and abuse of system resources and ensure the security of information system. The role-based access control (RBAC) model is a common access control model. In practical applications, it is mainly used for account authorization. With the development of computer technology, business scenarios are becoming more and more complex, and software cloudification and distributed architecture have become the mainstream trend. RBAC can no longer meet the current business needs. Based on it, the attribute-based access control (ABAC) model has been developed. According to the "Attribute-based Access Control Definition Guide" published by the National Institute of Standards and Technology of the United States, ABAC is an access control method that grants or denies the subject's request to perform operations on the object based on the specified attributes of the subject, the specified attributes of the object, environmental conditions, and a set of policies specified by these attributes and conditions. The security policy of the ABAC model supports fine-grained access control by combining different types of attributes. Its flexible, dynamic, and rich policy expressions make it suitable for complex distributed systems and can cope with complex authorization scenarios.
[0003] For access control of large-scale distributed application systems that apply the ABAC model, centralized access control permission management cannot meet the needs. To implement permission management of the ABAC model, a unified permission policy language model is required. The user operation complexity and learning cost brought by the permission policy language model seriously affect the user operation experience and popularity. Summary of the invention
[0004] In view of the above problems, the first aspect of the present application proposes an access control system based on the ABAC model, which includes:
[0005] The authorization service center receives the authorization information written by the user and generates the permission policy based on the authorization information. The permission policy is based on the XACML description.
[0006] The policy management service center uniformly stores and configures permission policies. Permission policies are submitted to the policy management service center by the authorization service center.
[0007] The authentication service center receives the authentication request, matches the permission policy according to the access information described in the authentication request, executes the permission policy to complete the authentication and returns the authentication result, thereby realizing access control. The permission policy is provided to the authentication service center by the policy management service center.
[0008] Furthermore, the authorized service centers include:
[0009] The resource warehouse stores basic data; the knowledge base stores basic feature data; the permission policy management workbench obtains basic data and assigns attribute tags of the XACML model to the data to generate basic feature data; the NLP algorithm engine uses the NLP algorithm to extract the data of the authorization information and obtain the authorization feature data.
[0010] Furthermore, the access control system adopts a heterogeneous microservice architecture, and the NLP algorithm engine is developed based on StanfordCoreNLP.
[0011] Furthermore, the NLP algorithm engine performs feature matching on the authorization feature data and the basic feature data, obtains the attribute tag data of the XACML model in the authorization information and generates a permission policy syntax tree. The permission policy management workbench is also used to extract semantic information from the permission policy syntax tree, and generates an XACML permission policy by mapping the semantic information and the XACML permission policy elements through preset mapping rules.
[0012] Furthermore, the authorization service center includes: an authorization workbench, which guides the user to write authorization information, and the guidance method includes at least one of grammar guidance, error prompt, and automatic association.
[0013] Furthermore, the guiding method also includes recommendation completion, and marking the recommended terms according to the user's selection, and extracting the marked information and storing it in the knowledge base.
[0014] The second aspect of the present application proposes an access control method based on the ABAC model, including:
[0015] Receive authorization information written by the user, generate permission policies based on the authorization information, and the permission policies are based on XACML descriptions;
[0016] Unified storage and configuration permission policies;
[0017] Receive an authentication request, match the permission policy according to the access information described in the authentication request, execute the permission policy to complete the authentication and return the authentication result, thereby achieving access control.
[0018] Furthermore, a feature database based on XACML attributes is constructed, authorization information is extracted through the NLP algorithm and feature matching is performed with the data features in the feature database to generate a permission policy syntax tree, semantic information is extracted from the permission policy syntax tree, and the XACML permission policy is generated through rule mapping of semantic information and XACML permission policy elements.
[0019] Furthermore, it also includes: guiding the user to write the authorization information, and the guiding method includes at least one of grammar guidance, error prompt, and automatic association.
[0020] Furthermore, the guiding method also includes recommendation completion, and marking the recommended terms according to the user's selection, and extracting the marked information and storing it in the knowledge base.
[0021] The present invention establishes an access control method applicable to large-scale distributed application systems based on the ABAC model, and the ABAC model enables it to have a more flexible resource management granularity; uses XACML to describe access control policies, which is conducive to the collaborative management of access control policies of distributed systems; and guides users to generate permission policies through the NLP algorithm, so that users can formulate sophisticated and complex permission control policies according to the grammatical structure of natural language, which solves the problems of complex syntax and high learning cost of XACML policy description, and increases the visibility and readability of the business. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] The accompanying drawings help to further understand the present application. The elements of the accompanying drawings are not necessarily in proportion to each other. For ease of description, only the parts related to the relevant invention are shown in the accompanying drawings.
[0023] Figure 1 It is a schematic diagram of the structure of an access control system based on the ABAC model in one embodiment of the present invention;
[0024] Figure 2 This is the architecture diagram of the XACML policy language model;
[0025] Figure 3 It is a structural diagram of an authorization service center in another embodiment of the present invention;
[0026] Figure 4 It is a schematic diagram of the authorization process of the authorization service center in the access control system based on the ABAC model in another embodiment of the present invention;
[0027] Figure 5 The figure is a schematic diagram of the authentication process of the authentication service center in the access control system based on the ABAC model in another embodiment of the present invention. DETAILED DESCRIPTION
[0028] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention.
[0029] Figure 1 The structure diagram of an access control system based on the ABAC model in one embodiment of the present invention. The system includes:
[0030] The authorization service center 100 receives the authorization information written by the user, and generates a permission policy based on the authorization information. The permission policy is based on the XACML description.
[0031] eXtensible Access Control Markup Language (XACML) is an access policy control language approved by OASIS, an organization for the promotion of structured information standards. In this embodiment, XACML is used to implement the authorization framework of the ABAC access control model. According to the XACML standard 3.0 released by OASIS, the policy language model architecture of XACML is as follows: Figure 2 As shown, the definition rules <rule>,Strategy <policy>, Policy Set <policyset>Three main modules, <policy>Used to describe an access control security policy, which is one or more rules <rule>A collection of; <policyset>It is used to describe multiple access control security policies, which can include <policy>or <policyset>In addition, XACML also defines a set of policy combination algorithms to coordinate policy decisions specified by multiple management entities, thereby ultimately providing authorization results.
[0032] rule <rule>It’s a strategy <policy>The basic element is the smallest unit in the XACML policy language model.
[0033] refer to Figure 2 ,rule <rule>include:
[0034] Target, which is used to constrain the attribute set to be restricted by the rule, including four basic attributes, namely resource, subject, action and environment;
[0035] Effect, which indicates the result after the rule is applied. The result includes two values, namely, Permit and Deny. The meaning of effect is that if the basic attribute elements (i.e., resources, subjects, actions, and environments) in a decision request match the basic attributes of the target, the corresponding decision result is given.
[0036] Conditions, which are predicate structures that represent other conditions that assist decision making in addition to the restrictions in the Target (such as contextual information such as the time limit requirements for operations);
[0037] Obligation Expressions;
[0038] Advice Expressions.
[0039] refer to Figure 2 ,Strategy <policy>It includes target, rule, rule combining algorithm, obligation expression and advice expression. <rule>Similar to the target in .
[0040] refer to Figure 2 , policy set <policyset>It is the top-level element of XACML, which contains the target (Target), several policies (Policy) or other policy sets (PolicySet), policy merging algorithm (Policy Combining Algorithm), obligation expressions (Obligation Expressions) and advice expressions (Advice Expressions).
[0041] In XACML, there may be computational conflicts between policies and / or rules. For example, a policy may contain multiple rules whose effects may be the same or mutually exclusive. In this case, the final decision of the policy / policy set is determined by the rule / policy combining algorithm according to the applicable level.
[0042] Common merging algorithms include:
[0043] Deny-overrides algorithm, the basic idea is that once a rule or policy application gets a deny result, the return result is Deny;
[0044] The basic idea of the Permit-overrides algorithm is that as long as the application result of a rule or policy is a permit, the return result is a permit.
[0045] The basic idea of the first-applicable algorithm is that in the process of applying a set of rules or policies, if there is a rule or policy that can be applied (its result can be approved or rejected), the algorithm stops immediately and returns the application result of this rule or policy;
[0046] Only-one-applicable algorithm is only applicable to the strategy combination algorithm. When there is only one rule that can be applied in the strategy, the decision result of this rule is taken as the final decision result.
[0047] Figure 3 This is a schematic diagram of the structure of the authorization service center 100 in another preferred embodiment, which specifically includes:
[0048] The resource warehouse 110 provides a data basis for the generation of the permission policy, and at least stores information such as XACML attribute data, such as the basic attributes of the target of the aforementioned rules and policies, namely, resources, subjects, actions and environments. The data in the resource warehouse can be obtained through database scanning, file import, interface call, etc., and is usually structured data.
[0049] The privilege policy management workstation 120 pre-reads the basic data in the resource repository and performs tagged preprocessing, assigns attribute tags of the XACML model to the basic data. Specifically, it extracts the key data in the resource repository, such as the attribute information of XACML, the keywords describing the privilege policy, the attribute dictionary, and the operation dictionary, extracts the information participating in authorization through pre-established data recognition rules and manual annotation, etc., and performs annotation on this information. For example, the position is the subject attribute, which is stored in the knowledge base, and a caching mechanism is added to improve performance.
[0050] In a further embodiment, the privilege policy management workstation also provides an authorization workbench for users to write authorization information, and provides functions such as guidance, prompts for privilege policy description, automatic association and recommended completion of authorization information for users. Specifically, the guidance and prompts for privilege policy description include syntax guidance for privilege policy description and self-check of policy description statements, and prompt users for abnormalities such as expression errors, authorization information errors, and ambiguities during writing;联想推荐 is performed on the policy content through a syntax guidance template and analysis of the part-of-speech of the input words. For example, when the keyword "at" is input, environmental attribute data, etc. are recommended; functions such as associative completion are performed when the user inputs the subject and resources to assist the user in completing the writing quickly and accurately; at the same time, corresponding annotation is performed on the entry when the user selects the recommended word, and this annotation information will also enter the knowledge base as basic feature data to provide support for subsequent natural language processing.
[0051] The knowledge base 130 stores the basic feature data processed by the privilege policy management workstation with tags.
[0052] The natural language processing (NLP) algorithm engine 140 provides algorithmic capability support for the generation of the privilege policy. In this embodiment, the access control system based on the ABAC model adopts a heterogeneous microservice architecture, which is compatible with the development of capability modules with different responsibilities using different technology stacks. Among them, the NLP algorithm engine can be developed using libraries such as python, jieba, and Stanford CoreNLP, while other services are developed using java. The heterogeneous capability module services are made serviceable through the microservice framework and registered in the registry for RPC remote calls. Further, the NLP algorithm engine can also provide algorithm support for the process of writing authorization information on the authorization workbench.
[0053] In the specific implementation process, the NLP algorithm engine extracts the authorization information written by the user through semantic analysis, keyword extraction, relationship extraction, etc., and generates a permission policy syntax tree through Stanford CoreNLP. After the policy description text entered by the user is processed by NLP, the basic feature data stored in the knowledge base can be used to match the authorization feature data extracted from the authorization information, obtain the attribute features such as resources, subjects, actions, and environments, and identify information such as utility and conditions, and then generate a permission policy syntax tree.
[0054] Furthermore, the permission policy management workstation 120 is preset with mapping rules between the permission policy syntax tree and the XACML model. After obtaining semantic information from the permission policy syntax tree, the XACML permission policy is generated by mapping the semantic information with the permission policy elements of the XACML model, such as filling in the subject and object by part of speech, and judging the effectiveness of the policy as permission or rejection by part of speech and word meaning. The XACML permission policy elements at least include the rules in the aforementioned XACML. <rule>The annotation terms can also be matched by rules. The construction of mapping rules can be done with the help of a template engine, preferably FreeMarker.
[0055] The policy management service center 200 uniformly stores and configures the authority policy, and the authority policy is submitted to the policy management service center by the authorization service center;
[0056] The authentication service center 300 receives the authentication request, matches the authority policy according to the access information described in the authentication request, and returns the authentication result, thereby implementing access control. The authority policy is provided to the authentication service center by the policy management service center.
[0057] Figure 4 The following is a schematic diagram of the authorization process of the authorization service center in the access control system based on the ABAC model in another embodiment of the present invention, which includes:
[0058] S101, obtaining data information, which can be obtained by database scanning, file importing, interface calling, etc.;
[0059] S102, data information preprocessing, including assigning attribute labels to obtain basic feature data and generate a knowledge base;
[0060] S103, receiving the authorization information written by the user. In this process, the authorization workbench assists the user in writing by means of association recommendation and the like, and marks the user's recommended terms;
[0061] S104, extract authorization information, and generate a permission policy syntax tree through an NLP algorithm. The data basis of the NLP algorithm comes from the data information in S101 and the knowledge base in S102. The specific NLP algorithm can be Stanford CoreNLP;
[0062] S105, extracting semantic information of the permission policy syntax tree, and generating an XACML permission policy through rule mapping;
[0063] S106: Save the permission policy and report it to the policy management service center.
[0064] Figure 5 The following is a schematic diagram of an authentication process of an authentication service center in an access control system based on an ABAC model in another embodiment of the present invention, which includes:
[0065] S301, loading the authority policy of the policy management service center in real time, and further, parsing the basic attribute data and authority policy file of XACML and putting them into the cache to improve the authentication performance;
[0066] S302, receiving an authentication request from an application;
[0067] S303, parsing and accessing the context in the authentication request to obtain relevant information such as XACML attributes;
[0068] S304, according to the information obtained in S303, the permission policy is hit;
[0069] S305, making a policy decision through the authority policy to generate an authentication result;
[0070] S306: Return the authentication result to the application.
[0071] The access control system and method based on the ABAC model proposed in the above embodiments solve the problem of distributed system policy collaborative management through the XACML policy language model on the basis of the ABAC model; and solve the problem of complex syntax and high learning cost of XACML policy description through natural language processing. It allows users to formulate sophisticated and complex permission control policies according to the grammatical structure of natural language, increasing the visibility and readability of the business. The method and system are particularly suitable for building access control for zero-trust systems, supporting identity-based and application-based dynamic access control, and can be dynamically adjusted to respond to evolving risk factors, so that companies or organizations that have been hacked can isolate relevant employees and data resources to control damage.
[0072] Although the contents of the present application are specifically demonstrated and described in conjunction with the preferred embodiments, those skilled in the art should understand that various changes made to the present application in form and detail without departing from the spirit and scope of the present application as defined by the appended claims and without making any creative work are within the scope of protection of the present application.< / rule> < / policyset> < / rule> < / policy> < / rule> < / policy> < / rule> < / policyset> < / policy> < / policyset> < / rule> < / policy> < / policyset> < / policy> < / rule>
Claims
1. An access control system based on the ABAC model, characterized in that: include: The authorization service center receives the authorization information written by the user and generates a permission policy based on the authorization information. The permission policy is based on the XACML description. The authorization service center includes: Resource warehouse, storing basic data, the basic data is obtained by database scanning, file import and interface calling, and stored as structured data; The permission policy management workbench obtains the basic data and assigns attribute tags of the XACML model to the basic data to generate basic feature data; A knowledge base storing basic feature data that has been labeled by the permission policy management workbench; An NLP algorithm engine, which uses an NLP algorithm to extract data of the authorization information and obtain authorization feature data. The NLP algorithm engine is developed based on Stanford CoreNLP, wherein the NLP algorithm engine performs feature matching on the authorization feature data and the basic feature data, obtains attribute tag data of the XACML model in the authorization information and generates a permission policy syntax tree. The permission policy management workbench is also used to extract semantic information from the permission policy syntax tree, and perform rule mapping between the semantic information and XACML permission policy elements through preset mapping rules to generate an XACML permission policy; The policy management service center uniformly stores and configures the authority policies, which are submitted to the policy management service center by the authorization service center; The authentication service center receives the authentication request, matches the permission policy according to the access information described in the authentication request, and returns the authentication result to implement access control. The permission policy is provided by the policy management service center to the authentication service center. The access control adopts a heterogeneous microservice architecture, and the heterogeneous capability modules are serviced through the microservice framework and registered in the registration center for RPC remote calls.
2. The access control system based on the ABAC model according to claim 1 is characterized in that: The authorized service centers include: The authorization workbench guides the user to write authorization information, and the guiding method includes at least one of grammar guidance, error prompt, and automatic association.
3. The access control system based on the ABAC model according to claim 2 is characterized in that: The guiding method also includes recommendation completion, and marking the recommended terms according to the user's selection, and extracting the marked information and storing it in the knowledge base.
4. An access control method based on the ABAC model, characterized in that: include: Receive authorization information written by a user, and generate a permission policy based on the authorization information. The permission policy is based on an XACML description and specifically includes: Resource warehouse, storing basic data, the basic data is obtained by database scanning, file import and interface calling, and stored as structured data; The permission policy management workbench obtains the basic data and assigns attribute tags of the XACML model to the basic data to generate basic feature data; A knowledge base storing basic feature data that has been labeled by the permission policy management workbench; An NLP algorithm engine, which uses an NLP algorithm to extract data of the authorization information and obtain authorization feature data. The NLP algorithm engine is developed based on Stanford CoreNLP, wherein the NLP algorithm engine performs feature matching on the authorization feature data and the basic feature data, obtains attribute tag data of the XACML model in the authorization information and generates a permission policy syntax tree. The permission policy management workbench is also used to extract semantic information from the permission policy syntax tree, and perform rule mapping between the semantic information and XACML permission policy elements through preset mapping rules to generate an XACML permission policy; Unified storage and configuration permission policies; Receive an authentication request, match the permission policy according to the access information described in the authentication request, execute the permission policy to complete the authentication and return the authentication result, so as to implement access control. The access control adopts a heterogeneous microservice architecture, and the heterogeneous capability modules are serviced through the microservice framework and registered in the registration center for RPC remote calls.
5. The access control method based on the ABAC model according to claim 4 is characterized in that: Also includes: The user is guided to write the authorization information, wherein the guiding method includes at least one of grammar guidance, error prompt, and automatic association.
6. The access control method based on the ABAC model according to claim 5 is characterized in that: The guiding method also includes recommendation completion, and marking the recommended terms according to the user's selection, and extracting the marked information and storing it in the knowledge base.
Citation Information
Cited By
Zero-trust gateway access control method and system
CN117375934A
