A Formal Verification-Based Compliance Verification Method for Office Network Access Control Policies

Abstract

This invention discloses a formal verification-based compliance verification method for office network access control policies, relating to the field of network service technology. The method includes: S1: collecting raw configuration data and auxiliary support data, preprocessing them, storing them, and constructing a policy experience database; S2: decoupling multi-attribute composite judgment conditions of the policy and constructing a hierarchical policy, implementing hierarchical verification units and performing layered verification, and constructing a condition dependency graph to optimize the verification path; S3: performing incremental verification and parallel accelerated verification based on the quantitative analysis results of the impact of changes, and outputting full verification process data; S4: fusing the optimization process data to construct an adaptive verification scheduling model and performing steady-state analysis. This method solves the problems of path explosion, high computational complexity, and low verification efficiency due to frequent policy changes in office network access under multi-attribute conditions.

Description

Technical Field

[0001] This invention relates to the field of network service technology, specifically to a method for compliance verification of office network access control policies based on formal verification. Background Technology

[0002] In office network access control, network security issues become increasingly complex as network scale continues to expand. Existing technologies typically employ formal verification methods based on symbolic execution and constraint solving to verify compliance with multi-attribute conditions in access control policies. These methods rely on the logical relationships between conditions to verify each policy condition one by one and enumerate possible execution paths.

[0003] For example, the invention patent with announcement number CN119583152B discloses a method and system for access control of network services, applied to network service providers. It deploys an auxiliary server, a target server, and a firewall. The target server runs a target service. The method includes: the auxiliary server receiving a user's request; the request being adapted to a predetermined message format of the auxiliary server, the content including combined information and a checksum; the combined information including an access identifier and firewall configuration commands; the auxiliary server finding the associated target server based on the access identifier and forwarding the combined information and checksum to the target server; the target server verifying the combined information and checksum; if the verification result is successful, the target server, based on the configuration commands, notifies the firewall to adjust firewall settings regarding user access to the target service.

[0004] For example, the invention patent announcement CN112788025B discloses a network access control method and display device, which enables the exception access of certain network connection data when the end user has not given permission, thereby avoiding the risk of application process privilege escalation and reducing workload. The method includes: when the end user license agreement is closed, if the network data connection in the network access data is identified to contain preset tag parameters, the exception access function is executed on the network access data.

[0005] However, as the number of condition dimensions increases, the computational complexity of symbolic execution and constraint solving also rises, leading to verification timeouts or memory overflows, making it impossible to reach compliance conclusions within a limited time. Furthermore, in scenarios with large-scale policy rules that require frequent changes or batch verification, traditional methods often cannot guarantee verification throughput and latency, thus affecting the availability and stability of compliance checks. Especially in environments with large-scale and frequently changing network access control policies, existing technologies face severe performance bottlenecks and struggle to meet the requirements of real-time verification.

[0006] Therefore, to address the above issues, there is an urgent need for a compliance verification method for office network access control policies based on formal verification. Summary of the Invention

[0007] Technical problems to be solved

[0008] To address the shortcomings of existing technologies, this invention provides a formal verification-based method for compliance verification of office network access control policies. This method solves the problems of path explosion, excessive computational complexity, and low verification efficiency due to frequent policy changes in office network access under multi-attribute conditions.

[0009] Technical solution

[0010] To achieve the above objectives, this invention provides the following technical solution: a formal verification-based compliance verification method for office network access control policies, comprising: S1: collecting original configuration data and auxiliary support data for office network access control, preprocessing the original configuration data and auxiliary support data, storing the data, and constructing a policy experience database; S2: decoupling multi-attribute composite judgment conditions of the policy and constructing a hierarchical policy, implementing hierarchical verification units and performing layered verification, constructing a condition dependency graph to optimize the verification path and update the verification status in real time; S3: performing quantitative analysis of the impact of policy changes based on the verification results before and after the policy change, the verification duration, and the number of affected devices, performing incremental verification and parallel accelerated verification based on the quantitative analysis results, and outputting full verification process data; S4: fusing the optimization process data to construct an adaptive verification scheduling model and performing steady-state analysis, and generating diagnostic reports, switching policy versions, and implementing closed-loop optimization evolution based on the steady-state analysis results.

[0011] Further, the specific steps for collecting the original configuration data and auxiliary support data for office network access control are as follows: Collect office network access control policy verification data, including original policy configuration data and auxiliary support data; the original policy configuration data includes policy rule conditions, policy rule actions, and policy rule priorities; the auxiliary support data includes user credential data, terminal device information, time policy configuration data, and network partition information data; collect original policy rules through the firewall configuration interface, directly obtaining the original condition attributes, network access conditions, rule actions, and rule priorities of each rule; collect user credential data through the Lightweight Directory Access Protocol interface, including user identifiers and assigned roles; collect terminal device information through the terminal management platform interface, including unique device identifiers and device types; collect time period data through time policy configuration, including the start and end points of the time range; collect network partition data through the network configuration management interface, including subnet addresses and virtual LAN identifiers; access permissions are determined by combining the network access conditions defined in the policy rules with the subnet addresses and virtual LAN identifiers in the network partition information to determine which users and terminal devices can access specific network areas.

[0012] Further, the specific steps for preprocessing and storing the original configuration data and auxiliary support data to construct the strategy experience database are as follows: The role codes in user identity information, type labels in terminal device information, time expressions in time period definitions, and area identifiers in network partition information are uniformly standardized and encoded, converted into an internally unified representation format to eliminate ambiguity; the minimum-maximum normalization method is used to scale the numerical attributes in the multi-source strategy data to a unified range; the standardized and normalized strategy data are then appended with unique strategy rule identifiers and collection timestamps before being stored and used to construct the strategy experience database.

[0013] Furthermore, the specific steps for decoupling the multi-attribute composite judgment conditions of the strategy and constructing a hierarchical strategy are as follows: Dimensional decomposition of strategy attributes: User identity, terminal hardware identifier, access initiation time, and target network area identifier are identified as discrete dimensions, and independent constraint sets are extracted to construct independently verifiable strategy sub-units; Independent description of each attribute constraint is achieved using a formula based on temporal logic: Linear temporal logic is used to represent the logical assertion of role membership, where the clock variable T1 represents the access initiation time, T2 represents the access end time, and the constraint of the access period is expressed in the form that T1 is less than T2; multi-temporal logic is used to represent the linear temporal constraints on the clock variable; CTL computational tree temporal logic is used to describe the reachability proposition in the condition dependency graph, and the relationships between conditions are represented by nodes and edges in the condition dependency graph; In the combination stage, based on the original logical relationship of the strategy, logical connectors are used to merge the independent sub-units into a complete strategy, realizing the transformation of the original strategy into a hierarchical structure composed of multi-formal sub-units.

[0014] Furthermore, the specific steps for implementing hierarchical verification units and performing hierarchical verification are as follows: Basic verification is achieved by verifying the legitimacy of the user and device by checking the user credentials and the device's unique identifier; hierarchical verification is achieved by comparing the current time with the specified time period to verify whether the access time period is legal, and checking whether the IP address requested by the user conforms to the access permissions of the target network area; the hierarchical division follows the rules of no cross-layer pre-dependencies and aggregation of fields of the same category, and the verification order is determined by sorting historical verification time from shortest to longest and failure probability from highest to lowest to minimize the expected verification time; the input of each level is standardized auxiliary support data and policy rule condition fragments under the corresponding category, where the input of the basic verification layer is user credential data and device unique identifier, and the input of the advanced verification layer is time policy configuration data and network partition information data; the legality verification process decides whether to continue to verify the next level of conditions based on the verification results of each level of conditions. If the verification of a certain level of conditions fails, the verification of the subsequent levels is skipped; if it passes, the verification result is passed to the next level; the compliance judgment of all conditions is completed through hierarchical verification, and the judgment result and the corresponding complete verification path are encapsulated into the compliance verification result of the current policy rule.

[0015] Furthermore, the specific steps for constructing a condition dependency graph to optimize the verification path and update the verification status in real time are as follows: A condition dependency graph is constructed by treating each condition in the strategy as a node in the graph and the logical dependencies between conditions as edges. In the condition dependency graph, the direction of the edges indicates that the result of one condition verification is a prerequisite for another condition. In the valid verification process, the original verification duration and the updated verification duration of each rule are recorded as edge weights on the path, used to weigh the priorities of different paths in the graph. The number of devices affected by the changes is tracked and recorded, the total number of changed rules and the total number of affected devices are counted, path selection is optimized using the A* search algorithm, and the dependency graph is updated in real time. The verification status is updated promptly by monitoring the path verification progress in real time.

[0016] Furthermore, the specific steps for quantitative analysis of the impact of the change through verification results, verification duration, and the number of affected devices before and after the strategy change are as follows: Obtain the compliance verification results before and after the change of the i-th change rule, the number of change rules, the number of devices affected by the change, the total number of affected devices, and the updated verification duration and the original verification duration of the i-th rule; calculate the ratio of the original verification duration to the updated verification duration of the i-th change strategy rule, and then multiply it by the time impact weighting factor to obtain the time impact item; calculate the ratio of the number of devices affected by the change to the total number of affected devices of the i-th change strategy rule, and then multiply it by the impact degree weighting factor to obtain the scope impact item; calculate the absolute value of the compliance verification results before and after the change of the i-th change rule to obtain the compliance difference item; add the compliance difference item, the time impact item, and the scope impact item to obtain the individual impact value of the i-th change strategy rule; sum the individual impact values ​​of all change strategy rules and divide by the number of change strategy rules to obtain the change impact value.

[0017] Furthermore, the specific steps for performing incremental verification, parallel accelerated verification, and outputting full verification process data based on the quantitative analysis results of the change impact are as follows: By comparing the change impact value and the impact threshold in real time, when the change impact value is less than the impact threshold, the verified results are retained and archived in the strategy experience database, and incremental verification is only carried out on the changed sub-units; when the change impact value is greater than or equal to the impact threshold, all change impact values ​​are sorted in descending order to obtain the impact queue, and a verification time window is added to the first k% of the impact queue and more computing nodes are allocated for parallel processing; the time complexity, path reachability, and computing resource consumption indicators generated during the full verification process are output; if the change impact value obtained after recalculation is less than the impact threshold, the strategy is cached, and the impact queue is updated in real time to dynamically adjust the verification order and improve the verification throughput.

[0018] Furthermore, the specific steps for constructing an adaptive verification scheduling model and performing steady-state analysis by integrating full-scale verification process data are as follows: The time complexity records, path reachability records, and computational resource consumption indicators generated during the full-scale verification process are used as inputs. Based on the real-time generated impact queue and verification time window allocation records, verification task priority records, node load balancing records, and path dependency topology records are generated. These are combined with intermediate state cache hit records, historical verification result confidence levels, and change impact value sequences to form a modeling input set. The impact queue batch number and verification task identifier are written as version traceability fields. The data sampling range and data contribution ranking for training are determined by combining key verification throughput indicators, forming dynamic update constraints. The new constraints are implemented through incremental verification triggering rules, elastic allocation rules for parallel computing nodes, and cache invalidation and warm-up rules. Based on a multi-objective feedback control algorithm, online training is conducted using a modeling input set. In a continuous decision space, the algorithm learns the weight parameter adjustment strategy of the verification heuristic function, the computing node scheduling strategy, and the time window allocation strategy. This ensures that the overall throughput of the verification allocation decision is greater than the throughput threshold and the verification duration of a single verification allocation decision is less than the latency threshold. Path reachability feedback drives the recursive pruning of the verification state space and the iterative update of the caching strategy to improve the reuse rate of high-value intermediate results. An adaptive verification scheduling model is constructed, and the verification scheduling strategy version identifier and scheduling adaptation value are output. The scheduling adaptation value is then written into the strategy dynamic optimization record table.

[0019] Furthermore, the specific steps for generating diagnostic reports, switching strategy versions, and implementing closed-loop optimization evolution based on steady-state analysis results are as follows: When the scheduling adaptation value is greater than or equal to the adaptation threshold, the optimization strategy version is maintained and the version parameters are fixed. At the same time, the current round's score value, performance indicator logs, and the corresponding verification scheduling strategy version identifier are associated and archived in the strategy experience database to enrich historical experience data. By comparing the scheduling adaptation value and the adaptation threshold in real time, when the scheduling adaptation value is less than the adaptation threshold, based on the real-time updated dependency graph and associated path verification data, the condition combination explosion point in the verification path where the number of nodes is greater than the node threshold is located, and the abnormal resource consumption area where the verification duration of a single verification allocation decision is greater than the latency threshold is identified, generating a bottleneck mode report containing the above quantitative characteristics. Based on the bottleneck mode report, historical verification scheduling strategy versions with scheduling adaptation values ​​greater than or equal to the adaptation threshold are matched from the strategy experience database and loaded, and the verification throughput is reconstructed by loading historical experience data.

[0020] Beneficial effects

[0021] The present invention has the following beneficial effects:

[0022] (1) This invention decouples the multi-attribute composite judgment conditions of the strategy and performs formal abstraction, makes the strategy conditions independent and performs hierarchical verification, thereby achieving high efficiency and controllability of the full verification process, and effectively solves the problems of path explosion and computational infeasibility in the prior art.

[0023] (2) This invention optimizes the selection of verification paths and improves verification throughput by constructing a dependency graph and using path optimization strategies. By using the A* search algorithm to optimize path selection and updating the dependency graph in real time, redundant path verification is effectively reduced, ensuring optimal resource utilization and thus achieving the effect of path optimization, solving the problem of low path traversal efficiency in traditional methods.

[0024] (3) This invention significantly improves the verification efficiency in scenarios with frequent changes by using quantitative analysis of the impact value of changes and an incremental verification strategy. The method dynamically adjusts the verification process according to the impact value of changes, performs incremental verification only on the changed part, avoids repeated verification of the entire strategy, and thus achieves the effect of accelerating the verification of change strategies, solving the problem of low verification efficiency caused by large-scale changes in traditional methods.

[0025] (4) This invention improves resource utilization in the full verification process through an adaptive verification scheduling model and steady-state analysis. Based on the optimized process data and steady-state analysis results, the allocation of computing resources is dynamically adjusted, and the execution order of verification tasks is optimized to ensure the reuse of high-value intermediate results, thereby improving the verification throughput and solving the problems of uneven resource utilization and excessive verification latency in the prior art.

[0026] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0027] Figure 1 The flowchart of the office network access control policy compliance verification method based on formal verification of the present invention is shown below.

[0028] Figure 2 This is a dependency graph showing the conditions for compliance verification of the network access control policy in this invention.

[0029] Figure 3 This is a multi-dimensional dynamic impact analysis diagram for verifying the compliance of the strategy of this invention. Detailed Implementation

[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0031] Please see Figures 1-3 This invention provides a technical solution: a formal verification-based compliance verification method for office network access control policies, comprising: S1: collecting original configuration data and auxiliary support data for office network access control, preprocessing the original configuration data and auxiliary support data, storing the data, and constructing a policy experience database; S2: decoupling multi-attribute composite judgment conditions of the policy and constructing a hierarchical policy, implementing hierarchical verification units and performing layered verification, constructing a condition dependency graph to optimize the verification path and update the verification status in real time; S3: performing quantitative analysis of the impact of policy changes by analyzing the verification results before and after the policy change, the verification duration, and the number of affected devices, performing incremental verification and parallel accelerated verification based on the quantitative analysis results, and outputting full verification process data; S4: fusing the optimization process data to construct an adaptive verification scheduling model and performing steady-state analysis, and generating diagnostic reports, switching policy versions, and performing closed-loop optimization evolution based on the steady-state analysis results.

[0032] Specifically, the steps for collecting the original configuration data and auxiliary support data for office network access control are as follows: Collect office network access control policy verification data, including original policy configuration data and auxiliary support data. Original policy configuration data includes policy rule conditions, policy rule actions, and policy rule priorities; auxiliary support data includes user credential data, terminal device information, time policy configuration data, and network partition information data. Collect original policy rules through the firewall configuration interface, directly obtaining the original condition attributes, network access conditions, rule actions, and rule priorities of each rule, and mapping these fields to the policy rule conditions, rule actions, and priority variables in the subsequent formal model to ensure accurate expression and deduction during the full verification process. Collect user credential data through the Lightweight Directory Access Protocol interface. User credential data includes user identifiers and assigned roles, where user identifiers are mapped to user identity variables in the subsequent formal model, and assigned roles are mapped to role variables. The legality of identity and role is verified in subsequent verification. Terminal device information is collected through the terminal management platform interface. This information includes a unique device identifier and device type. The unique device identifier maps to a device identity variable, and the device type maps to a device category variable. Subsequent verification uses these variables to confirm the legality and compliance of the devices. Time period data is collected through the time policy configuration interface. This data includes the start and end points of the time range, which map to time window variables in the verification model. Subsequent verification will use these time variables to confirm whether access complies with the specified time period. Network partition data is collected through the network configuration management interface. This data includes subnet addresses and virtual LAN identifiers. Subnet addresses map to network area identifier variables, and virtual LAN identifiers map to network area permission variables. Subsequent verification uses these variables to determine whether users or devices have the corresponding network access permissions. Access permissions are determined by combining the network access conditions defined in the policy rules with the subnet addresses and virtual LAN identifiers in the network partition information. The formal model will then use these mapped variables, combined with rule actions, to verify the legality of users and terminal devices, thereby determining which users and terminal devices can access specific network areas.

[0033] In this implementation plan, by collecting and standardizing the raw configuration data and supporting data required for office network access control policies, each data field is mapped to a corresponding variable in the subsequent formal model, ensuring consistency and accuracy throughout the full verification process. Specifically, raw policy configuration data is obtained through the firewall configuration interface and mapped to policy rule conditions, rule actions, and priority variables; user identity data is collected through the Lightweight Directory Access Protocol interface and mapped to user identity variables; terminal device information is collected through the terminal management platform interface and mapped to device identity and device type variables; time policy data is collected through the time policy configuration interface and mapped to time window variables; and network partition information is collected through the network configuration management interface and mapped to network zone identifiers and permission variables. This process ensures data consistency, eliminates ambiguity, and provides a reliable data foundation for subsequent compliance verification, effectively improving verification efficiency and accuracy.

[0034] Specifically, the steps for preprocessing and storing the original configuration data and auxiliary support data to build a strategy experience database are as follows: The role codes in user identity information, type labels in terminal device information, time expressions in time period definitions, and area identifiers in network partition information are uniformly standardized and encoded, converted into an internally unified representation format to eliminate ambiguity, ensure format compatibility across different data sources, and eliminate interpretive ambiguity caused by inconsistent data representation. For numerical attributes, the minimum-maximum normalization method is used to scale the numerical attributes in multi-source strategy data to a unified range, achieving dimensionless processing and avoiding the impact of inconsistent data scales on subsequent analysis. The standardized and normalized strategy data is then appended with unique strategy rule identifiers and collection timestamps before being stored and used to build the strategy experience database. This ensures the integrity and timeliness of the strategy data, provides a high-quality data foundation for subsequent strategy analysis and optimization, supports historical data traceability and version management, and enhances the accuracy and executability of strategy optimization and verification.

[0035] In this implementation plan, standardized coding and minimum-maximum normalization ensure the consistency of strategy data from different data sources in terms of format and scale, eliminating problems caused by ambiguity in expression and inconsistent data scales. This process allows strategy data to be processed and analyzed within a unified framework, avoiding the adverse effects of data inconsistency on subsequent verification and optimization. At the same time, the addition of unique identifiers for strategy rules and collection timestamps ensures the timeliness and integrity of the data, laying the foundation for the construction of a strategy experience database and improving the accuracy and reliability of subsequent strategy analysis, optimization, and verification.

[0036] Specifically, the steps for decoupling multi-attribute composite decision conditions and constructing a hierarchical strategy are as follows: When decomposing the strategy attributes into dimensions, user identity, terminal hardware identifier, access initiation time, and target network area identifier are first identified as independent discrete dimensions, and a set of independent constraints is extracted for each dimension. Each constraint is independently described using a formula based on temporal logic to ensure its verifiability. In the logical assertion of role membership, linear temporal logic is used, defining clock variables T1 to represent the access initiation time and T2 to represent the access end time, and expressing the access time limit through the constraint that T1 is less than T2. ​​Furthermore, multi-temporal logic is used to model the linear temporal constraints on the clock variables to capture the complex relationships in the time sequence. To further describe the interdependencies between conditions, CTL computation tree temporal logic is used to represent the reachability propositions in the condition dependency graph, using nodes and edges in the graph to clarify the sequential relationships and dependencies between conditions. Based on the logical relationships of the original strategy, the independent sub-units are merged using logical connectors. For example, if the user's identity is "administrator" and the access time is "9:00 to 18:00 on weekdays," then these two sub-units are combined into a complete policy using AND logic. Ultimately, through this logical connection, the independent policy sub-units are effectively integrated, realizing the transformation of the original policy into a hierarchical structure composed of multi-formal sub-units, ensuring that both the complexity and verifiability of the policy are improved simultaneously.

[0037] In this implementation scheme, independent verification of each policy sub-unit is achieved by decomposing policy attributes into dimensions and extracting independent sets of constraints. Temporal logic formulas are used to describe each attribute constraint in detail, ensuring the precise expression of access time periods and conditional dependencies. Linear and multi-temporal logic are used to model the constraints on access times and time order, and CTL computation tree temporal logic is used to express the reachability propositions between conditions, thus clearly defining the relationships and dependencies between policy sub-units. In the final combination stage, based on the logical relationships of the original policy, the independent sub-units are merged through logical connectors, successfully transforming the policy from its original structure to a hierarchical structure of multi-formal sub-units, improving the policy's verification capability and operability.

[0038] Specifically, the steps for implementing hierarchical verification units and performing layered verification are as follows: By comparing the current time with a specified time period, the access time period is verified to be legal, and the IP address requested by the user is checked to ensure it meets the access permissions of the target network area, thus achieving hierarchical verification. The hierarchical division follows the rules of no cross-layer pre-dependencies and aggregation of fields of the same category, ensuring that each layer of verification depends only on the input data of its own layer. Through analysis of historical verification time and failure probability, verification priorities are determined in ascending order of time and frequency to minimize the expected verification time. The input for each layer consists of standardized auxiliary support data and policy rule condition fragments under the corresponding category. The input data for the basic verification layer includes user credential data and device unique identifiers, while the input data for the advanced verification layer includes time policy configuration data and network partition information data. The legal verification process determines whether to continue verifying the next level of conditions based on the verification results of each layer. If the verification of a certain layer fails, the verification of subsequent layers is skipped to avoid invalid calculations; if it passes, the verification result is passed to the next layer, proceeding step by step. Figure 2 This is a dependency graph of network access control policy compliance verification conditions in this embodiment. Finally, through layered verification, the compliance determination of all conditions is completed, and the determination result of each layer and the corresponding complete verification path are encapsulated into the compliance verification result of the current policy rule, ensuring the clarity, accuracy and efficiency of the policy layered verification process.

[0039] This implementation scheme employs a hierarchical verification method to ensure efficient verification of complex policy rules. The hierarchical verification process is tiered based on access time periods and IP address matching, ensuring that each layer of verification relies only on its own input data. Analysis of historical verification time and failure probabilities optimizes the verification order, prioritizing conditions with high failure probabilities and low time consumption, thereby minimizing verification latency. Input data for each layer is standardized by category, ensuring data consistency and accuracy. The verification process dynamically determines whether to proceed to the next layer based on the verification results of each layer, avoiding unnecessary calculations and improving efficiency. Ultimately, through hierarchical verification and step-by-step progression, compliance determination for all conditions is completed, ensuring the completeness and accuracy of policy verification while effectively reducing redundant calculations in the hierarchical verification process.

[0040] Specifically, the steps for constructing a condition dependency graph to optimize the verification path and update the verification status in real time are as follows: A condition dependency graph is constructed by treating each condition in the strategy as a node in the graph and the logical dependencies between conditions as edges. In the condition dependency graph, the direction of the edges indicates that the result of one condition verification is a prerequisite for another condition, ensuring that verification proceeds in the order of the dependencies between conditions. In the valid verification process, the original verification time and the updated verification time of each rule are recorded as edge weights on the path, used to weigh the priority of different paths in the graph, helping to determine whether paths with longer verification times are worth prioritizing. The system tracks and records the number of devices affected by changes, and counts the total number of change rules and the total number of affected devices. It optimizes path selection using the A* search algorithm, where the state represents the current verification progress, the cost function represents the total duration of verified paths, the heuristic function represents the estimated time from the current path to the target state, and the target condition is the state where all rules have been verified. The system selects the optimal path using the A* algorithm, ensuring priority verification of paths with a significant impact on compliance, and updates the dependency graph in real time to optimize computational resource allocation. By monitoring path verification progress in real time and updating verification status promptly, the system ensures path optimization and the rational use of computational resources, further improving efficiency and accuracy.

[0041] In this implementation scheme, efficient management of verification paths was successfully achieved by constructing a condition dependency graph and applying the A* search algorithm to optimize path selection. Logical dependencies between conditions are connected by edges in the graph, ensuring verification proceeds in the order of dependency and avoiding redundant computation. The original and updated verification times for each rule are recorded as edge weights, allowing for a reasonable evaluation of the priority of each path in the graph. Using the A* algorithm, based on a comprehensive consideration of verification time and path impact, paths with a greater impact on compliance are prioritized for verification, significantly improving verification efficiency. Real-time monitoring of path verification progress and updating of the dependency graph effectively controls the allocation of computational resources, improving the accuracy and throughput of policy verification.

[0042] Specifically, the steps for quantitatively analyzing the impact of a policy change by analyzing the verification results, verification duration, and number of affected devices before and after the change are as follows: Obtain the compliance verification results before and after the change of the i-th rule, the number of changed rules, the number of devices affected by the change, the total number of affected devices, and the updated and original verification durations of the i-th rule. Perform statistical analysis on the verification duration under different change scenarios using regression analysis. Use historical verification data for training to fit the relationship between duration and change impact, and obtain the time impact weighting factor based on the regression results. Calculate the device impact weighting factor by statistically analyzing the ratio of the number of affected devices to the total number of devices in the system, combined with historical verification results and the scope of impact. Calculate the ratio of the original verification duration to the updated verification duration of the i-th policy change rule, and multiply it by the time impact weighting factor to obtain the time impact term, which quantifies the impact of the change on verification time. Calculate the ratio of the number of affected devices to the total number of affected devices for the i-th policy change rule, and multiply it by the degree of impact weighting factor to obtain the scope impact term, reflecting the size of the change's impact range. Subsequently, by calculating the absolute value of the compliance verification results before and after the change of the i-th change rule, a compliance difference term is obtained, representing the degree of impact of the change on the compliance verification results. The compliance difference term, time impact term, and scope impact term are added together to obtain the individual impact value of the i-th change strategy rule. By summing the individual impact values ​​of all change strategy rules and dividing by the number of change strategy rules, the overall change impact value is obtained, used to quantify the impact of the overall change on the strategy verification process. This process provides data support for subsequent incremental and differentiated verification, ensuring the computational efficiency and accuracy of the optimized verification.

[0043] The specific formula for calculating the impact value of the change is as follows:

[0044] ;

[0045] In the formula, This indicates the impact value of the change, reflecting the overall impact of the policy and rule changes on the compliance verification results; This indicates the number of rules that have been changed, reflecting the number of policy rules that have been affected. Indicates the first The compliance verification results before and after the rule change reflect the degree of impact of the change on the verification results; This indicates the time-related weighting factor, which adjusts the impact of the post-change verification time on the change's impact value. This indicates the verification time after the update, reflecting the computational complexity of the rule verification after the change; This indicates the original verification time, reflecting the computational complexity of the original rule verification; This indicates the equipment impact weighting factor, which adjusts the impact of the change scope on the change impact value. Indicates the first The number of device users affected by the rule change reflects the actual scope of impact of the rule change; This indicates the total number of affected devices, reflecting the overall scale of the system.

[0046] In this embodiment, the compliance verification result for the first group is 0.375, the verification time ratio is 2.299, the equipment ratio is 0.120, the time influence weighting factor is 0.65, the equipment influence weighting factor is 0.35, and the individual influence value is 1.911; the compliance verification result for the second group is 0.951, the verification time ratio is 1.386, the equipment ratio is 0.550, the time influence weighting factor is 0.55, the equipment influence weighting factor is 0.45, and the individual influence value is 1.961; the compliance result for the third group is... The verification result for Group 1 was 0.732, the verification time ratio was 2.009, the equipment ratio was 0.320, the time impact weighting factor was 0.60, the equipment impact weighting factor was 0.40, and the individual impact value was 2.065. The compliance verification result for Group 4 was 0.599, the verification time ratio was 2.047, the equipment ratio was 0.460, the time impact weighting factor was 0.58, the equipment impact weighting factor was 0.42, and the individual impact value was 1.979. The calculated change impact value (CIV) was 1.979.

[0047] Table 1. Multi-dimensional quantitative analysis of the impact parameters of strategy rule changes

[0048]

[0049] like Figure 3 As shown, this is a multi-dimensional dynamic impact analysis diagram for strategy compliance verification provided in an embodiment of this application. Combined with the data in Table 1... Figure 3It is evident that the impact of different strategies and rules varies significantly under different conditions. Differences in compliance verification results, verification time ratios, and equipment ratios all influence the change impact value. For example, under Rule 3, when the verification time ratio is as high as 2.009 and the time impact weighting factor is 0.60, the individual impact value exceeds 2.065, surpassing the impact threshold and indicating a high-impact area. This triggers a coordinated adjustment involving parallel verification and comprehensive review by the system. Under Rule 1, when the verification time ratio is 2.299 but the equipment ratio is as low as 0.120, the individual impact value is only 1.911, below the impact threshold, indicating a low-impact area. Incremental verification continues normally, demonstrating the good controllability of daily changes. Under Rule 2, when the difference in compliance verification results is as high as 0.951 and the equipment impact weighting factor is 0.45, the individual impact value rises to 1.961. Although it does not exceed the threshold, it is close to the critical point. The system initiates incremental verification and strengthens monitoring, reflecting the ability to prevent potential impacts. This indicates that multidimensional factor analysis can effectively capture the dynamic changes during the change process, and the fluctuations in the change impact value are highly correlated with the actual change status. The impact of changes increases significantly with the increase in verification time, consistent with the trend of verification complexity. While the proportion of devices has some impact, the impact remains within a controllable range due to the low difference in compliance. This distribution pattern provides a reliable basis for real-time monitoring of strategy changes and allocation of verification resources. It allows for dynamic adjustment of verification strategies and actions based on different change types and impact levels. It also provides effective data support for change quality trend analysis, anomaly warning, and intelligent decision-making, contributing to a comprehensive improvement in verification efficiency, system stability, and overall compliance.

[0050] This implementation plan, through comprehensive analysis of the compliance verification results, verification duration, and number of affected devices of the changed rules, accurately quantifies the impact of each changed policy rule on the verification process. The calculation of time impact, scope impact, and compliance difference items combines time consumption, impact scope, and compliance changes, effectively reflecting the comprehensive impact of the change on policy verification. Finally, by calculating the change impact value, the overall impact of the change on the verification process can be assessed, providing a scientific basis for subsequent incremental and differentiated verification. This ensures optimized verification processes, improved computational efficiency, and reduced unnecessary computational overhead in scenarios with frequent policy changes.

[0051] Specifically, the steps for performing incremental verification, parallel accelerated verification, and outputting full verification process data based on the quantitative analysis results of change impact are as follows: During real-time comparison of change impact values ​​and impact thresholds, when the change impact value is less than the impact threshold, the system directly retains the verified results and archives them to the strategy experience database. Simultaneously, incremental verification is performed only on the changed sub-units to avoid unnecessary computational overhead. When the change impact value is greater than or equal to the impact threshold, the system sorts all change impact values ​​in descending order, generates an impact queue, and prioritizes more detailed verification of the top k% of changes based on the queue's sorting. For the top k% portion, the system increases the verification time window and allocates more computing nodes for parallel processing, thereby accelerating the verification process. In specific implementation, k% represents 15%, used to control the priority verification of the top k% of changes, typically between 0 and 100, with the specific value depending on the strategy complexity and the allocation of computing resources. The unit of the time window is usually milliseconds, and the upper limit of the time window is usually set to the maximum tolerable computational latency to ensure the timeliness of the full verification process. The verification time window allocation strategy depends on the complexity and importance of the change. The system dynamically adjusts the verification time allocated to each change unit based on priority. Compute nodes can originate from local clusters, containers, or thread pools, depending on the system architecture. When allocating compute nodes, the system selects appropriate resources from a pre-defined parallel execution environment based on the task's parallelism requirements, ensuring efficient parallel execution of tasks. Containers and thread pools are common choices in parallel execution environments, suitable for lightweight tasks and large-scale computing tasks respectively, guaranteeing efficient utilization of computing resources. Through this mechanism, the system can adjust the verification order in real time and dynamically adjust the verification priority based on the impact of the change, improving verification throughput and ensuring efficient system operation.

[0052] In this implementation plan, the full policy verification process is effectively optimized by comparing the impact value of a change with the impact threshold in real time. When the impact value of a change is lower than the impact threshold, the system only performs incremental verification on the changed sub-units and archives the verified results to the policy experience database, avoiding unnecessary duplicate calculations. If the impact value of a change exceeds the threshold, the system sorts all change impact values ​​and prioritizes verifying the changes with the greatest impact. By allocating more verification time windows and computing resources to the top k% of changes for parallel processing, the system accelerates the full verification process and improves the throughput and efficiency of verification. During the calculation process, the order of the impact queue is adjusted in real time, and the policy cache is optimized in subsequent calculations, making the full verification process more dynamic and efficient, thereby maximizing resource utilization and optimizing the scheduling of verification tasks.

[0053] Specifically, the steps for constructing an adaptive verification scheduling model and performing steady-state analysis by integrating full-scale verification process data are as follows: Time complexity records, path reachability records, and computational resource consumption indicators generated during the full-scale verification process are used as inputs. Combined with real-time generated impact queues and verification time window allocation records, verification task priority records, node load balancing records, and path dependency topology records are generated to ensure that the execution order and allocation of computational resources for each verification task are efficient and targeted. Simultaneously, intermediate state cache hit records, historical verification result confidence levels, and change impact value sequences are combined to form a modeling input set, enabling optimization and scheduling of verification resources from multiple dimensions. Impact queue batch numbers and verification task identifiers are written to ensure that the version of each verification task is traceable and to maintain the consistency of historical data. Based on key verification throughput indicators, the data to be used for training is determined. The system prioritizes the verification of key data points by sampling range and data contribution ranking, while dynamically updating constraints and optimizing the execution of incremental verification triggering rules, parallel computing node elastic allocation rules, and cache invalidation and preheating rules. Based on a multi-objective feedback control algorithm, online training is conducted on the modeled input set to learn how to adjust the weight parameters of the verification heuristic function, optimize the computing node scheduling strategy and time window allocation strategy, and ensure that the overall throughput of verification allocation decisions is greater than the throughput threshold and the verification duration of a single verification allocation decision is less than the latency threshold. By driving the recursive pruning of the verification state space and the iterative update of the caching strategy through path reachability feedback, the reuse rate of high-value intermediate results is further improved. An adaptive verification scheduling model is constructed, and the optimized verification scheduling strategy version identifier and scheduling adaptation value are output. The scheduling adaptation value is written into the strategy dynamic optimization record table to support subsequent optimization.

[0054] In this implementation scheme, by comprehensively analyzing time complexity, path reachability, and computational resource consumption, the priority of verification tasks and node load balancing are adjusted in real time to ensure efficient utilization of computational resources. The dynamically adjusted verification strategy optimizes the allocation of verification tasks based on throughput and latency thresholds, improving overall throughput and single-pass verification efficiency. Simultaneously, through a multi-objective feedback control algorithm and an adaptive verification scheduling model, the scheduling strategy and time window allocation strategy of computing nodes are optimized, ensuring the reuse rate of high-value intermediate results. This effectively improves the efficiency and accuracy of the full verification process, reduces redundant computation, and increases verification throughput.

[0055] Specifically, the steps for generating diagnostic reports, switching strategy versions, and implementing closed-loop optimization based on steady-state analysis results are as follows: When the scheduling fit value is greater than or equal to the fit threshold, the system maintains the current optimization strategy version and solidifies the version parameters to ensure the stability of the strategy version. Simultaneously, the current round's score, performance indicator logs, and corresponding verification scheduling strategy version identifiers are associated and archived in the strategy experience database, thereby enriching historical experience data. This process provides valuable historical data support for subsequent strategy optimization and adjustments. By comparing the scheduling fit value with the fit threshold in real time, when the scheduling fit value is less than the fit threshold, the system will conduct in-depth analysis of potential bottleneck areas in the verification path based on the real-time updated dependency graph and associated path verification data. In particular, the system will locate the explosion points of conditional combinations where the number of nodes exceeds the node threshold and identify abnormal resource consumption areas where the verification duration of a single verification allocation decision exceeds the latency threshold. A bottleneck pattern report is generated, specifying the above quantitative characteristics and detailing the performance bottlenecks and abnormal resource consumption that may occur during the full verification process. Based on the bottleneck pattern report, the system matches and loads historical verification scheduling policy versions with scheduling fit values ​​greater than or equal to the fit threshold from the policy experience database. By loading this historical experience data, the system can reconstruct the verification throughput and optimize it during the new full verification process. This cyclical process helps the system continuously improve verification efficiency and ensures that each verification is performed within a more efficient policy framework, thereby further improving the accuracy and execution efficiency of policy optimization.

[0056] In this implementation, the policy verification and resource allocation process is optimized by comparing the scheduling adaptation value and the adaptation threshold in real time. When the scheduling adaptation value meets the threshold requirement, the system fixes the optimization policy version and archives relevant data to enrich historical experience data and support subsequent decision-making. When the scheduling adaptation value is lower than the threshold, the system analyzes the dependency graph and path verification data to accurately locate areas where performance bottlenecks may occur, generates bottleneck pattern reports, and identifies areas of abnormal resource consumption. By matching and loading historical experience data, the system can reconstruct the verification throughput, optimize the verification policy, and improve verification efficiency, thereby ensuring the efficient execution and continuous optimization of the policy.

[0057] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0058] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for compliance verification of office network access control policies based on formal verification, characterized in that, Includes the following steps: S1: Collect the raw configuration data and auxiliary support data for office network access control, preprocess the raw configuration data and auxiliary support data, store them, and build a policy experience database; S2: Decoupling strategy for multi-attribute composite judgment conditions and constructing a hierarchical strategy, implementing hierarchical verification units and performing layered verification, constructing a condition dependency graph to optimize the verification path and update the verification status in real time; S3: Quantitatively analyze the impact of the change by using the verification results, verification time and number of affected devices before and after the strategy change. Based on the quantitative analysis results of the change impact, perform incremental verification and parallel accelerated verification and output full verification process data. S4: Integrate optimization process data to construct an adaptive verification scheduling model and perform steady-state analysis. Based on the steady-state analysis results, generate diagnostic reports, switch strategy versions, and implement closed-loop optimization evolution.

2. The method for compliance verification of office network access control policies based on formal verification according to claim 1, characterized in that: The specific steps for collecting the original configuration data and auxiliary support data for office network access control are as follows: The system collects verification data for office network access control policies. This data includes original policy configuration data and supporting data. Original policy configuration data includes policy rule conditions, policy rule actions, and policy rule priorities. Supporting data includes user credential data, terminal device information, time policy configuration data, and network partition information. Original policy rules are collected via the firewall configuration interface, directly obtaining the original condition attributes, network access conditions, rule actions, and rule priorities for each rule. User credential data is collected via the Lightweight Directory Access Protocol (LightCOP) interface, including user identifiers and assigned roles. Terminal device information is collected via the terminal management platform interface, including unique device identifiers and device types. Time period data is collected via time policy configuration, including the start and end points of the time range. Network partition data is collected via the network configuration management interface, including subnet addresses and virtual LAN identifiers. Access permissions are determined by combining the network access conditions defined in the policy rules with the subnet address and virtual LAN identifier in the network partition information to determine which users and terminal devices can access a specific network area.

3. The method for compliance verification of office network access control policies based on formal verification according to claim 1, characterized in that: The specific steps for preprocessing and storing the original configuration data and auxiliary support data to construct a strategy experience database are as follows: The role codes in user identity information, type labels in terminal device information, time expressions in time period definitions, and area identifiers in network partition information are uniformly standardized and encoded, and converted into an internal unified representation format to eliminate ambiguity. The minimum-maximum normalization method is used to scale the numerical attributes in multi-source policy data to a unified range. The standardized and normalized policy data are then appended with unique policy rule identifiers and collection timestamps before being stored and used to build a policy experience database.

4. The method for compliance verification of office network access control policies based on formal verification according to claim 1, characterized in that: The specific steps for constructing a hierarchical strategy based on multi-attribute composite decision conditions for the decoupling strategy are as follows: The policy attributes are decomposed dimensionally: user identity, terminal hardware identifier, access initiation time, and target network area identifier are identified as discrete dimensions, and independent sets of constraints are extracted to construct independently verifiable policy sub-units. Each attribute constraint is independently described using a formula based on temporal logic: linear temporal logic is used to represent the logical assertion of role membership, where the clock variable T1 represents the access initiation time and T2 represents the access end time, and the constraint of the access period is expressed in the form that T1 is less than T2. ​​Multi-temporal logic is used to represent the linear temporal constraints on the clock variable. CTL computation tree temporal logic is used to describe the reachability proposition in the condition dependency graph, and the nodes and edges in the condition dependency graph represent the sequential relationship between conditions. In the combination stage, based on the original logical relationship of the policy, logical connectors are used to merge the independent sub-units into a complete policy, realizing the transformation of the original policy into a hierarchical structure composed of multi-formal sub-units.

5. The method for compliance verification of office network access control policies based on formal verification according to claim 1, characterized in that: The specific steps for implementing the hierarchical verification unit and performing hierarchical verification are as follows: Basic verification is achieved by verifying the legitimacy of users and devices by checking user credentials and device unique identifiers. Hierarchical verification is achieved by comparing the current time with a specified time period to verify the legality of the access time period and checking whether the IP address requested by the user meets the access permissions of the target network area. The hierarchical division follows the rules of no cross-layer pre-dependencies and aggregation of fields of the same category. The verification order is determined by sorting historical verification time from shortest to longest and failure probability from highest to lowest to minimize the expected verification time. The input of each level is standardized auxiliary support data and policy rule condition fragments under the corresponding category. The input of the basic verification layer is user credential data and device unique identifier, and the input of the advanced verification layer is time policy configuration data and network partition information data. The compliance verification process determines whether to continue verifying the next level of conditions based on the verification results of each level. If the verification of a certain level of conditions fails, the verification of subsequent levels is skipped; if it passes, the verification result is passed to the next level. The compliance determination of all conditions is completed through layered verification, and the determination result and the corresponding complete verification path are encapsulated as the compliance verification result of the current policy rule.

6. The method for compliance verification of office network access control policies based on formal verification according to claim 1, characterized in that: The specific steps for constructing the condition dependency graph to optimize the verification path and update the verification status in real time are as follows: A condition dependency graph is constructed by treating each condition in the strategy as a node in the graph and the logical dependencies between conditions as edges. In the condition dependency graph, the direction of the edge indicates that the result of one condition verification is a prerequisite for another condition. In the valid verification process, the original verification time and the updated verification time of each rule are recorded as edge weights on the path, which are used to weigh the priority of different paths in the graph. The number of devices affected by the change is tracked and recorded, the total number of all changed rules and the total number of affected devices are counted, the path selection is optimized by the A* search algorithm, and the dependency graph is updated in real time. The verification status is updated in a timely manner by monitoring the path verification progress in real time.

7. The method for compliance verification of office network access control policies based on formal verification according to claim 1, characterized in that: The specific steps for quantitatively analyzing the impact of the strategy change by using verification results, verification time, and the number of affected devices before and after the change are as follows: Obtain the compliance verification results before and after the change of the i-th rule, the number of changed rules, the number of devices affected by the change, the total number of affected devices, and the verification time after the update and the original verification time of the i-th rule; Calculate the ratio of the original verification time to the updated verification time of the i-th change strategy rule, and then multiply it by the time impact weighting factor to obtain the time impact term; Calculate the ratio of the number of devices affected by the change to the total number of affected devices for the i-th change policy rule, and then multiply it by the impact degree weighting factor to obtain the scope impact term; calculate the absolute value of the compliance verification results before and after the change of the i-th change policy rule to obtain the compliance difference term; add the compliance difference term, the time impact term, and the scope impact term to obtain the individual impact value of the i-th change policy rule; sum the individual impact values ​​of all change policy rules and divide them by the number of change policy rules to obtain the change impact value.

8. The method for compliance verification of office network access control policies based on formal verification according to claim 1, characterized in that: The specific steps for performing incremental verification, parallel accelerated verification, and outputting full verification process data based on the quantitative analysis results of the change impact are as follows: By comparing the impact value of the change with the impact threshold in real time, when the impact value of the change is less than the impact threshold, the verified results are archived to the strategy experience database, and incremental verification is carried out only for the changed sub-unit. When the impact value of a change is greater than or equal to the impact threshold, all impact values ​​are sorted in descending order to obtain an impact queue. A verification time window is added to the first k% of the impact queue, and more computing nodes are allocated for parallel processing. The time complexity, path reachability, and computing resource consumption indicators generated during the full verification process are output. If the impact value of the change is less than the impact threshold after recalculation, the strategy is cached, and the impact queue is updated in real time to dynamically adjust the verification order and improve the verification throughput.

9. The method for compliance verification of office network access control policies based on formal verification according to claim 1, characterized in that: The specific steps for constructing an adaptive verification scheduling model and performing steady-state analysis by integrating the full verification process data are as follows: The time complexity record, path reachability record, and computing resource consumption index generated during the full verification process are used as inputs. Based on the real-time generated impact queue and verification time window allocation record, verification task priority record, node load balancing record, and path dependency topology record are generated. The intermediate state cache hit record, historical verification result confidence, and change impact value sequence are combined to form the modeling input set. The impact queue batch number and verification task identifier are written as version traceability fields. The data sampling range and data contribution ranking for participating in training are determined by combining the key indicators of verification throughput and forming dynamic update constraints. The update constraints are implemented with incremental verification trigger rules, parallel computing node elastic allocation rules, cache invalidation and preheating rules. Based on a multi-objective feedback control algorithm, online training is conducted using a modeling input set. In a continuous decision space, the algorithm learns the weight parameter adjustment strategy, computing node scheduling strategy, and time window allocation strategy for the verification heuristic function. This ensures that the overall throughput of the verification allocation decision is greater than the throughput threshold and the verification duration of a single verification allocation decision is less than the latency threshold. Path reachability feedback drives the recursive pruning of the verification state space and the iterative update of the caching strategy to improve the reuse rate of high-value intermediate results. An adaptive verification scheduling model is constructed, and the verification scheduling strategy version identifier and scheduling adaptation value are output. The scheduling adaptation value is then written into the strategy dynamic optimization record table.

10. The method for compliance verification of office network access control policies based on formal verification according to claim 1, characterized in that: The specific steps for generating diagnostic reports, switching strategy versions, and implementing closed-loop optimization evolution based on steady-state analysis results are as follows: When the scheduling adaptation value is greater than or equal to the adaptation threshold, the optimization strategy version is maintained and the version parameters are fixed. At the same time, the current round score value, performance indicator logs and the corresponding verification scheduling strategy version identifier are associated and archived to the strategy experience database to enrich historical experience data. By comparing the scheduling adaptation value and the adaptation threshold in real time, when the scheduling adaptation value is less than the adaptation threshold, based on the real-time updated dependency graph and associated path verification data, the system locates the conditional explosion point where the number of nodes in the verification path exceeds the node threshold, and identifies abnormal resource consumption areas where the verification duration of a single verification allocation decision exceeds the latency threshold, generating a bottleneck pattern report containing the above quantitative characteristics. Based on the bottleneck pattern report, the system matches and loads historical verification scheduling policy versions with scheduling adaptation values ​​greater than or equal to the adaptation threshold from the policy experience database, and reconstructs the verification throughput by loading historical experience data.

Images