Firewall policy auditing method and system based on IP asset profiling and network topology

By adopting a firewall policy auditing method based on IP asset profiling and network topology, the problems of low efficiency and poor accuracy in firewall policy auditing are solved, realizing automated policy auditing and risk scoring, and improving the efficiency and security of policy management.

CN123123027APending Publication Date: 2026-11-17INFORMATION & TELECOMM COMPANY SICHUAN ELECTRIC POWER
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611609701.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-19
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing technologies suffer from low efficiency and poor accuracy in firewall policy review. Manual review is difficult to handle issues such as incomplete work order information, inconsistencies between asset registration information and actual operating status, and complex network paths, which lead to low policy activation efficiency and security risks.

Method used

The firewall policy auditing method based on IP asset profiling and network topology obtains firewall policy application work orders, performs field parsing and integrity verification, establishes IP asset profiles, matches source and destination addresses, derives access paths, and matches policies with the existing firewall policy library to generate audit opinions and recommended policies.

Benefits of technology

It improves the efficiency and accuracy of firewall policy auditing, reduces repeated manual verification, identifies abnormal asset situations, reduces security risks, and enhances the security and intelligence level of policy management.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123123027A_ABST
    Figure CN123123027A_ABST
Patent Text Reader

Abstract

This invention relates to the field of network security management technology, and provides a firewall policy review method and system based on IP asset profiling and network topology. The method includes: obtaining firewall policy application work orders and performing field parsing and integrity verification; associating and fusing IP asset data based on IP addresses to establish IP asset profiles; matching source and destination addresses with the IP asset profiles respectively to generate asset consistency verification results; deriving the access path from the source address to the destination address based on the network topology; matching the access requirements corresponding to the firewall policy application work orders with existing firewall policy libraries; and generating review opinions and recommended policies based on the firewall policy application work orders when no existing policy meets the access requirements. By implementing the linkage between firewall policy application work order review, asset consistency verification, and access path derivation, the efficiency and accuracy of firewall policy review are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of network security management technology, specifically to a firewall policy auditing method and system based on IP asset profiling and network topology. Background Technology

[0002] In enterprise network security management, firewall policy auditing is a crucial step in ensuring business interconnection, access control, and perimeter protection.

[0003] Currently, firewall policy requests are typically submitted as work orders by business or operations personnel, and security administrators then manually review them based on asset ledgers, business descriptions, network topology, and existing policies. Because work order information comes from diverse sources, details such as source address, destination address, port, protocol, business system affiliation, and validity period are often incomplete, inconsistent, or contradictory, requiring reviewers to repeatedly verify information. Furthermore, the lack of a unified and dynamic management mechanism for the relationship between IP addresses and business systems, organizational affiliation, and operational status makes it easy for discrepancies to arise between asset registration information and actual operational status. This makes it difficult for reviewers to promptly identify whether addresses are truly usable, belong to existing assets, or have been duplicated or incorrectly affiliated to business systems.

[0004] Furthermore, enterprise internal networks typically contain multiple security domains, routing nodes, and firewall nodes, resulting in complex access paths. Manual intervention makes it difficult to accurately determine the actual path between the source and destination addresses, and to promptly identify whether existing rules in the policy library already meet the requirements, or whether there are redundancies, duplications, or conflicts. This easily leads to problems such as low policy activation efficiency, inconsistent review standards, overly open policies, and the accumulation of security risks. Therefore, there is an urgent need for a firewall policy review technology capable of correlating and analyzing work order information, IP asset status, network paths, and existing policies to improve review accuracy and management efficiency, and reduce the risks associated with manual reliance. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a firewall policy auditing method and system based on IP asset profiling and network topology, thereby solving the problems of low efficiency and poor accuracy in firewall policy auditing in existing technologies.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A firewall policy auditing method based on IP asset profiling and network topology includes the following steps: Step 1: Obtain the firewall policy application work order, and perform field parsing and field integrity verification on the source address, destination address, port, protocol, access direction, responsible person, business system, application reason and validity period in the firewall policy application work order; Step 2: Associate and merge IP asset data based on IP addresses to create IP asset profiles; Step 3: For firewall policies that pass field integrity verification, submit a work order, match the source address and destination address with the IP asset profile respectively, and generate asset consistency verification results; Step 4: For the source and destination addresses that have passed the asset consistency verification, based on the network topology and in combination with security domains, routing devices, firewall nodes and business systems, deduce the access path from the source address to the destination address, and determine the security domains and firewall nodes that the source address passes through to the destination address. Step 5: Match the access requirements corresponding to the firewall policy request work order with the existing firewall policy library to determine whether there are existing policies that meet the access requirements. Step 6: When there is no existing policy that meets the access requirements, generate review comments and recommended policies based on the firewall policy by submitting a work order.

[0007] Furthermore, in step 2, the IP asset data includes at least two of the following data sources: CMDB, IP address management, DNS, DHCP, traffic session records, security device logs, vulnerability scanning, business system ledgers, and historical work orders.

[0008] Furthermore, in step 2, the IP asset profile includes at least two of the following fields: Organization affiliation, asset type, activity status, open ports, historical communication records, business systems, data centers, environment, risk labels, and recent alarm information.

[0009] Furthermore, in step 3, when the asset consistency verification result is inconsistent, an audit opinion requiring manual review is output, and a risk work order is generated, specifically including: Using the source address and the destination address as query conditions, respectively, query the corresponding asset records in the IP asset profile; If the source address or destination address is not recorded in the IP asset profile, and the address is determined to be active based on traffic session records, security device logs, or other operational data, then the corresponding address will be determined to be unregistered and already in use. If the IP asset profile contains a record of the source address or the destination address, but there is no corresponding traffic session record within a preset time window, then the corresponding address is determined to be registered but not used. If the same source address or the same destination address corresponds to multiple asset records in the IP asset profile, and the asset type field or organization affiliation field in the multiple asset records is inconsistent, then the corresponding address will be determined as duplicate registration. If the source address or the destination address is inconsistent with the business system field in the IP asset profile and the business system field in the firewall policy application work order, the corresponding address will be determined as having an inconsistent business system. When any of the above situations occur, the corresponding source address or destination address will be marked as a risk address, and a risk work order will be generated according to the corresponding situation.

[0010] Furthermore, in step 4, based on the network topology and considering security domains, routing devices, firewall nodes, and business systems, the access path from the source address to the destination address is derived, specifically including: Obtain network security domain segmentation data, firewall node connection locations and interface configuration information in the network topology, topological connection relationships of routing devices, and the security domain to which the business system belongs; construct a directed graph by treating security domains, routing devices, firewall nodes, and business systems as nodes and the connectivity relationships between nodes as edges. Using the security domain to which the source address belongs as the starting node and the security domain to which the destination address belongs as the ending node, search all access paths from the starting node to the ending node in the directed graph. If multiple access paths are found, one path is selected as the recommended configuration path based on historical traffic data, routing priority, and security domain policy. Based on the recommended configuration path, determine the security domains and firewall nodes that the source address and destination address will pass through.

[0011] Further, in step 5, the access requirement corresponding to the firewall policy request work order is matched with the existing firewall policy library, specifically including: Read the source address, destination address, port and protocol of each existing policy from the existing firewall policy library, and match the source address, destination address, port and protocol corresponding to the firewall policy application work order with each existing policy; Generate policy hit results, which include duplicate policies, redundant policies, or conflicting policies; When the source address, destination address, port, and protocol of the firewall policy request ticket are exactly the same as those of the existing policy, the policy hit result will be determined as a duplicate policy. When the source address, destination address, port, and protocol of the existing policy cover the firewall policy application ticket, the policy hit result is determined as a redundant policy; When the firewall policy request ticket overlaps with an existing denial policy in terms of source address, destination address, port, and protocol, the policy hit result is determined to be a conflicting policy. When the strategy hit result is a duplicate strategy, it is determined that there is an existing strategy that satisfies the access requirement.

[0012] Furthermore, in step 6, generating review comments and recommended policies based on the firewall policy application work order specifically includes: Extract risk assessment factors, which include at least port risk, vulnerability risk, security domain risk, firewall path risk, validity period risk, and policy hit result; calculate a risk score based on the risk assessment factors; If the risk score falls within the low-risk score range, then an approval opinion will be output, and the risk level will be determined as low risk. If the risk score falls within the medium risk range, then the manual review comments and individual scores for each risk assessment factor will be output, and the risk level will be determined as medium risk. If the risk score falls within the high-risk score range, a rejection review opinion and individual scores for each risk assessment factor will be output, and the risk level will be determined as high risk. When the review opinion is low-risk and approved or medium-risk and manually reviewed, a recommended policy is generated according to the principle of least privilege. The source address of the recommended policy is the specific source host address corresponding to the firewall policy application work order, the destination address is the specific destination service address corresponding to the work order, the port and protocol are the ports and protocols used in the actual business, and the validity period is the period required by the business.

[0013] Further, the risk score is calculated, specifically including: Each risk assessment factor is assigned a weight coefficient and a value range; the corresponding individual score is determined based on the actual value of each risk assessment factor; the product of each individual score and its corresponding weight coefficient is summed to obtain the risk score.

[0014] Furthermore, following step 6, the process also includes online verification of the approved recommendation strategy, specifically including: The generated recommended policy is then sent to the corresponding firewall device and made effective. Collect traffic hit data of the recommended policy on the firewall device, and determine whether the recommended policy is hit by actual access traffic based on the traffic hit data; If the recommended strategy is not hit within the preset verification period, a strategy optimization suggestion is generated and recorded in the strategy optimization list.

[0015] A firewall policy auditing system based on IP asset profiling and network topology includes: The work order parsing module is used to obtain firewall policy application work orders and to parse and verify the integrity of the source address, destination address, port, protocol, access direction, responsible person, business system, application reason and validity period in the firewall policy application work order. The IP asset profile building module is used to associate and merge IP asset data based on IP addresses in order to pre-build IP asset profiles. The asset verification module is used to match the source address and the destination address with the IP asset profile respectively, and generate an asset consistency verification result; The network topology path deduction module is used to deduce the access path from the source address to the destination address based on the network topology, which includes security domains, routing devices, firewall nodes, and business systems, and to determine the security domains and firewall nodes that the source address passes through to the destination address. The policy matching module is used to match the access requirements corresponding to the firewall policy application work order with the existing firewall policy library. When there is an existing policy that meets the access requirements, the module outputs an audit opinion that no new policy needs to be added. The risk review module is used to calculate a risk score based on risk assessment factors when there is no existing strategy that meets the access requirements, generate a review opinion based on the risk score, and generate a recommended strategy according to the principle of least privilege when the review opinion is low-risk and approved or medium-risk and manually reviewed.

[0016] Compared with the prior art, the beneficial effects of the present invention are: By linking and analyzing firewall policy application tickets, IP asset profiles, network topology, and the existing firewall policy library, the system can automatically complete field validation, asset ownership verification, access path deduction, policy hit results, and risk scoring during the submission and review stages of firewall policy application tickets. This significantly improves the efficiency and accuracy of firewall policy review, reducing repeated manual verification and ticket returns for resubmission. Simultaneously, by identifying asset anomalies such as unregistered but used, registered but not used, duplicate registration, and inconsistencies between business systems, the system can improve the consistency between asset ledgers and actual operational status, reducing the risk of shadow assets and bot IPs being mistakenly allowed. Furthermore, by identifying duplicate, redundant, or conflicting policies and generating recommended policies based on the principle of least privilege, the system can effectively reduce security risks such as excessive policy openness, cross-domain violations, port amplification, and long-term openness. A closed-loop optimization mechanism is formed through policy deployment verification and traffic hit feedback, thereby improving the overall security, standardization, and intelligence of firewall policy management. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a firewall policy auditing method based on IP asset profiling and network topology.

[0019] Figure 2 This is a flowchart for the strategy-based intelligent review process.

[0020] Figure 3 This is a block diagram of the overall system architecture. Detailed Implementation

[0021] The embodiments of the present invention / invention are described in detail below with reference to the accompanying drawings.

[0022] Example 1

[0023] A firewall policy auditing method based on IP asset profiling and network topology, the process of which is as follows: Figure 1 As shown, it includes the following steps: Step 1: Obtain the firewall policy application work order, and perform field parsing and field integrity verification on the source address, destination address, port, protocol, access direction, responsible person, business system, application reason and validity period in the firewall policy application work order; Step 2: Associate and merge IP asset data based on IP addresses to create IP asset profiles; Step 3: For firewall policies that pass field integrity verification, submit a work order, match the source address and destination address with the IP asset profile respectively, and generate asset consistency verification results; Step 4: For the source and destination addresses that have passed the asset consistency verification, based on the network topology and in combination with security domains, routing devices, firewall nodes and business systems, deduce the access path from the source address to the destination address, and determine the security domains and firewall nodes that the source address passes through to the destination address. Step 5: Match the access requirements corresponding to the firewall policy request work order with the existing firewall policy library to determine whether there are existing policies that meet the access requirements. Step 6: When there is no existing policy that meets the access requirements, generate review comments and recommended policies based on the firewall policy by submitting a work order.

[0024] In practice, firewall policy application tickets are retrieved from the ITSM system or ticketing system. These tickets must contain at least the following fields: source address, destination address, port, protocol, access direction, responsible person, business system, reason for application, and validity period. The fields in the firewall policy application ticket are parsed. After parsing, the completeness of the fields is verified, including but not limited to: checking for missing fields in the source address, destination address, port, protocol, access direction, responsible person, business system, reason for application, and validity period; checking if the address format conforms to the preset IP address format, network segment format, or host format; checking if the port value exceeds the legal range; checking if the protocol belongs to the preset protocol enumeration; and checking if the validity period is empty. If any required field is missing, formatted incorrectly, or has an illegal value, a "supplement missing fields" review comment is output, and the firewall policy application ticket is returned to the applicant for supplementation. This approach blocks low-quality firewall policy application tickets at the entry point, preventing subsequent IP asset profiling, access path deduction, and policy matching from being based on incorrect input, thereby reducing invalid calculations and repeated returns.

[0025] Using the source and destination addresses from firewall policy application work orders as query criteria, corresponding records are retrieved from IP asset data. IP asset data originates from multiple different systems; records containing the same IP address are associated with the same IP asset profile. During association and fusion, the IP addresses from each data source are first standardized in format, and then a master index is created based on the IP addresses. For multiple records corresponding to the same IP address, fields such as organization affiliation, asset type, activity status, open ports, historical communication records, business system, data center, environment, risk label, and recent alarm information are written into the corresponding IP asset profile. The IP asset profile transforms scattered IP asset data into a unified data object centered on IP addresses.

[0026] For firewall policy application work orders that pass field integrity verification, the IP asset profile is queried using the source address and destination address as matching keys. The corresponding asset record is then located within the IP asset profile, and an asset consistency verification result is generated. Asset consistency verification can be performed as follows: If the IP asset profile does not contain an asset record for the source or destination address, but the address is active based on traffic session records, security device logs, or other operational data, it is determined to be unregistered and already in use. If the IP asset profile contains an asset record for the source or destination address, but no corresponding traffic session record exists within a preset time window, it is determined to be registered but not used. If the same source or destination address corresponds to multiple asset records in the IP asset profile, and the asset type or organization affiliation fields in these records are inconsistent, it is determined to be duplicate registration. If the business system field of the source or destination address in the IP asset profile is inconsistent with the business system field in the firewall policy application work order, it is determined to be a business system inconsistency. Other inconsistencies, such as missing responsible persons, abnormal activity status, or inconsistent environments, can also be used as supplementary conditions for determining inconsistency. When any of the above situations occur, the corresponding source address or destination address will be marked as a risk address, and an audit opinion of "manual review, risk warning work order" will be output. At the same time, the reasons for the inconsistency, related evidence and verification results will be recorded.

[0027] Based on the IP asset profile corresponding to the source address, the security domain to which the source address belongs is determined. Based on the IP asset profile corresponding to the destination address, the security domain to which the destination address belongs is determined. The connection relationships between security domains, routing devices, firewall nodes, and business systems are then retrieved from the network topology. Starting from the security domain of the source address and ending at the security domain of the destination address, an access path is searched within the network topology. The searched access path includes the source security domain, the destination security domain, and the routing devices and firewall nodes between adjacent security domains. This step transforms abstract access requirements into specific network configuration locations. For example, if the source address is located in the production application security domain and the destination address is located in the database security domain, the access path might pass through the first and second firewall nodes sequentially. This step is performed based on asset consistency verification. If the actual security domains corresponding to the source and destination addresses are not determined first, the starting and ending positions of the path derivation may be incorrect, leading to policy matching on the wrong firewall nodes.

[0028] The system reads existing policies from the existing firewall policy library for the corresponding firewall node and matches the access requirements corresponding to the firewall policy request with these existing policies. Matching dimensions include source address, destination address, port, and protocol. It then determines whether the existing policies satisfy the access requirements of the firewall policy request. If the access path passes through multiple firewall nodes, the existing policies on each firewall node are checked separately. Only when all existing policies on the firewall nodes satisfy the access requirements can it be determined that the access requirement from the source address to the destination address has been met. If the existing policies satisfy the access requirements, then... Figure 2 As shown, the output provides review comments without requiring the addition of new strategies, thus avoiding redundant configuration.

[0029] When no existing policy meets the access requirements, an audit opinion and a recommended policy are generated based on the firewall policy application ticket, IP asset profile, asset consistency verification results, access path, and policy hit results. The audit opinion can include approval, manual review, or rejection. The recommended policy is generated according to the principle of least privilege. The source address of the recommended policy is the specific source host address corresponding to the firewall policy application ticket, the destination address is the specific destination service address corresponding to the firewall policy application ticket, the port and protocol are the ports and protocols actually used in the business, and the validity period is the period required by the business. In one implementation, if some firewall nodes on the access path already have existing policies that meet the access requirements, while other firewall nodes do not, then a recommended policy is generated only for the firewall nodes that do not have existing policies that meet the access requirements. This approach further avoids repeatedly generating policies for all firewall nodes along the entire path.

[0030] Furthermore, in step 2, the IP asset data includes at least two of the following data sources: CMDB, IP address management, DNS, DHCP, traffic session records, security device logs, vulnerability scanning, business system ledgers, and historical work orders.

[0031] In practice, the IP asset profile is constructed by retrieving IP asset data from CMDB, IP address management, DNS, DHCP, traffic session records, security device logs, vulnerability scans, business system ledgers, and historical work orders according to a preset cycle. Specifically, CMDB provides asset name, asset type, organizational affiliation, responsible person, and asset status; IP address management provides IP address allocation, network segment, and usage status; DNS provides the resolution relationship between IP addresses and hostnames; DHCP provides IP address lease records and lease duration; traffic session records provide the active status of IP addresses, historical communication records, communication ports, and protocols; security device logs provide access records, policy hit records, and recent alarm information; vulnerability scans provide open ports and vulnerability risks; business system ledgers provide the correspondence between IP addresses and business systems; and historical work orders provide historical application information for the same source address, destination address, port, and protocol.

[0032] The reason for using at least two data sources is that a single data source typically only reflects one aspect of IP assets. For example, a CMDB (Content Management Database) can reflect registration information, but cannot independently prove whether an IP address is currently in actual use; traffic session records can reflect the actual use of an IP address, but cannot independently prove the organizational affiliation and business system of the IP address. By associating the CMDB with traffic session records, both "registered" and "used" can be determined simultaneously, thereby identifying unregistered but used and registered but unused assets. Therefore, multi-source IP asset data further improves the reliability of asset consistency verification. A single data source can only solve the asset record query problem, while multi-source data fusion further solves the problem of contradictions between registration information and actual operational information.

[0033] Furthermore, in step 2, the IP asset profile includes at least two of the following fields: Organization affiliation, asset type, activity status, open ports, historical communication records, business systems, data centers, environment, risk labels, and recent alarm information.

[0034] In practice, an IP asset profile is created for each IP address. The IP address serves as the index for the IP asset profile record. The meanings of each field are as follows: Organization Affiliation: Indicates the organization to which the asset corresponding to the IP address belongs; Asset Type: Indicates the asset category, such as server, terminal, or network device; Activity Status: Indicates whether the IP address has traffic session records or security device logs within a preset time window; Open Ports: Records actual open ports identified by vulnerability scans or operational data; Historical Communication Records: Records the source address, destination address, port, protocol, and communication time; Business System: Indicates the business system to which the asset corresponding to the IP address belongs; Data Center: Indicates the data center where the asset is located; Environment: Indicates the production environment, testing environment, or other operating environment; Risk Label: Indicates whether the asset has vulnerability risks, alarm risks, or attribution risks; Recent Alarm Information: Records the most recent security alarms related to this IP address. Through this step, subsequent asset consistency verification of source and destination addresses is no longer based solely on firewall policy application work order fields, but can be based on the actual asset status. This improves the accuracy and interpretability of firewall policy auditing.

[0035] Furthermore, in step 3, when the asset consistency verification result is inconsistent, an audit opinion requiring manual review is output, and a risk work order is generated, specifically including: Using the source address and the destination address as query conditions, respectively, query the corresponding asset records in the IP asset profile; If the source address or destination address is not recorded in the IP asset profile, and the address is determined to be active based on traffic session records, security device logs, or other operational data, then the corresponding address will be determined to be unregistered and already in use. If the IP asset profile contains a record of the source address or the destination address, but there is no corresponding traffic session record within a preset time window, then the corresponding address is determined to be registered but not used. If the same source address or the same destination address corresponds to multiple asset records in the IP asset profile, and the asset type field or organization affiliation field in the multiple asset records is inconsistent, then the corresponding address will be determined as duplicate registration. If the source address or the destination address is inconsistent with the business system field in the IP asset profile and the business system field in the firewall policy application work order, the corresponding address will be determined as having an inconsistent business system. When any of the above situations occur, the corresponding source address or destination address will be marked as a risk address, and a risk work order will be generated according to the corresponding situation.

[0036] In specific implementation, such as Figure 2As shown, when the firewall policy request ticket fields are complete, the process enters the stage of matching IP asset profiles and determining whether the matching results are consistent. Specifically, the source address and destination address are used as query conditions to read the corresponding asset records from the IP asset profiles, and the following judgments are performed.

[0037] If the IP asset profile does not contain a record for the corresponding address, but there are activity records for that address in the traffic session logs or security device logs within a preset time window, then the address is determined to be unregistered and in use. For example, if a source address does not exist in the CMDB and IP address management, but that source address has continuously initiated accesses to multiple destination addresses within the last seven days, it indicates that the address is in use in the actual network. This address is marked as a risk address, and a risk ticket is generated. The risk ticket records the address, judgment type, traffic session logs, and relevant security device logs; through this process, shadow assets that cannot be identified by the CMDB alone can be identified, avoiding granting access permissions to active assets that have not completed registration. If the IP asset profile contains an asset record for the corresponding address, but there are no traffic session records within a preset time window, then the address is determined to be registered but not used. The preset time window can be set according to the asset type or business system. For example, a 30-day time window is used for production environment assets, and a 7-day time window is used for temporary assets. The specific length of the preset time window does not constitute a limitation of this invention. A registered but unused status may indicate that the asset is offline, the address is idle, or the business system ledger has not been updated in a timely manner. A risk ticket is generated to prevent continued access to zombie IPs or deactivated assets. If the same IP address corresponds to multiple asset records, and the asset type or organizational affiliation of these records is inconsistent, the address is considered a duplicate registration. For example, the same IP address corresponds to a production server and belongs to organization one in the first record, but to a test terminal and belongs to organization two in the second record; since two records cannot simultaneously serve as reliable evidence of asset affiliation, the address is marked as a risk address, and the conflicting fields and corresponding records are listed in the risk ticket. The business system fields in the IP asset profile are compared with the business system fields in the firewall policy application ticket. If they are inconsistent, the corresponding address is determined to be from an inconsistent business system. For example, the firewall policy application ticket declares the destination address belongs to business system A, but the destination address in the IP asset profile belongs to business system B. This situation may indicate an error by the applicant, and instead of automatic approval, a review opinion requiring manual verification is generated.

[0038] For any of the above inconsistencies, the corresponding source or destination address will be marked as a risk address, and a risk work order will be generated. This work order, along with the reason for the inconsistency, related evidence, and verification results, will be output to the manual review process. If the asset consistency verification fails, the process will proceed to manual review instead of directly entering network topology path deduction. This is because when there are asset ownership anomalies, the security domain to which the source or destination address belongs may be unreliable. If the access path continues to be automatically deduced, incorrect firewall nodes may be identified. By adding an asset consistency verification before path deduction, erroneous path deduction can be blocked at the data source.

[0039] Furthermore, in step 4, based on the network topology and considering security domains, routing devices, firewall nodes, and business systems, the access path from the source address to the destination address is derived, specifically including: Obtain network security domain segmentation data, firewall node connection locations and interface configuration information in the network topology, topological connection relationships of routing devices, and the security domain to which the business system belongs; construct a directed graph by treating security domains, routing devices, firewall nodes, and business systems as nodes and the connectivity relationships between nodes as edges. Using the security domain to which the source address belongs as the starting node and the security domain to which the destination address belongs as the ending node, search all access paths from the starting node to the ending node in the directed graph. If multiple access paths are found, one path is selected as the recommended configuration path based on historical traffic data, routing priority, and security domain policy. Based on the recommended configuration path, determine the security domains and firewall nodes that the source address and destination address will pass through.

[0040] In practice, security domains, routing devices, firewall nodes, and business systems are converted into nodes in a directed graph, and network connectivity is converted into directed edges. Each edge can record attributes such as connection direction, routing priority, and security domain policy. The security domains of the source and destination addresses are obtained from the IP asset profile and used as the start and end nodes, respectively. An access path from the start node to the end node is searched in the directed graph.

[0041] If only one access path that satisfies the security domain policy is found, that path is designated as the recommended configuration path. If multiple access paths are found, they are filtered based on the following factors: whether there are records of source addresses or communication between the same business system along the corresponding path in historical traffic data; whether the routing priority is higher than other candidate paths; whether the candidate paths conform to the security domain policy; and whether the firewall nodes in the candidate paths are available. In one implementation, access paths that do not conform to the security domain policy can be eliminated first, and then sorted according to routing priority; when multiple paths have the same routing priority, the path with historical traffic records is selected as the recommended configuration path. Based on the recommended configuration path, the starting node, ending node, traversed security domains and firewall nodes, the interface of each firewall node, and the access direction corresponding to each firewall node are output.

[0042] Further, in step 5, the access requirement corresponding to the firewall policy request work order is matched with the existing firewall policy library, specifically including: Read the source address, destination address, port and protocol of each existing policy from the existing firewall policy library, and match the source address, destination address, port and protocol corresponding to the firewall policy application work order with each existing policy; Generate policy hit results, which include duplicate policies, redundant policies, or conflicting policies; When the source address, destination address, port, and protocol of the firewall policy request ticket are exactly the same as those of the existing policy, the policy hit result will be determined as a duplicate policy. When the source address, destination address, port, and protocol of the existing policy cover the firewall policy application ticket, the policy hit result is determined as a redundant policy; When the firewall policy request ticket overlaps with an existing denial policy in terms of source address, destination address, port, and protocol, the policy hit result is determined to be a conflicting policy. When the strategy hit result is a duplicate strategy, it is determined that there is an existing strategy that satisfies the access requirement.

[0043] In practice, after parsing firewall policy request ticket fields, constructing IP asset profiles, and deriving network topology paths, the system receives the access request corresponding to the firewall policy request ticket and reads the source address, destination address, port, and protocol of existing policies from the existing firewall policy library one by one. The system matches the source address, destination address, port, and protocol in the firewall policy request ticket with existing policies to determine if a policy already exists that satisfies the access request. Specifically, when the source address, destination address, port, and protocol of the firewall policy request ticket are completely identical to the corresponding fields of an existing policy, the system marks the existing policy as a duplicate policy and thus determines that an existing policy already exists that satisfies the access request; when the source address, destination address, port, and protocol range of an existing policy covers the access request of the firewall policy request ticket, the system marks it as a redundant policy, indicating that the current request does not require adding a new policy or only needs to converge the existing policy; when the access request of the firewall policy request ticket overlaps with an existing denial policy in terms of source address, destination address, port, or protocol, the system marks it as a conflicting policy. Through the above matching method, the system can automatically identify problems such as duplicate activation, broad coverage and access conflicts before policy approval, thereby reducing the workload of manually checking the policy library item by item, avoiding firewall rule bloat, and providing a clear basis for subsequent risk scoring and least privilege principle recommended policy generation.

[0044] Furthermore, in step 6, generating review comments and recommended policies based on the firewall policy application work order specifically includes: Extract risk assessment factors, which include at least port risk, vulnerability risk, security domain risk, firewall path risk, validity period risk, and policy hit result; calculate a risk score based on the risk assessment factors; If the risk score falls within the low-risk score range, then an approval opinion will be output, and the risk level will be determined as low risk. If the risk score falls within the medium risk range, then the manual review comments and individual scores for each risk assessment factor will be output, and the risk level will be determined as medium risk. If the risk score falls within the high-risk score range, a rejection review opinion and individual scores for each risk assessment factor will be output, and the risk level will be determined as high risk. When the review opinion is low-risk and approved or medium-risk and manually reviewed, a recommended policy is generated according to the principle of least privilege. The source address of the recommended policy is the specific source host address corresponding to the firewall policy application work order, the destination address is the specific destination service address corresponding to the work order, the port and protocol are the ports and protocols used in the actual business, and the validity period is the period required by the business.

[0045] In practice, when no existing policy meets the access requirements, the system generates review opinions and recommended policies based on the firewall policy application work order. The system first extracts risk assessment factors. These factors include at least port risk, vulnerability risk, security domain risk, firewall path risk, validity period risk, and policy hit result. Specifically, port risk indicates whether the requested port has an excessively large range, or whether low-risk or high-risk ports are open; vulnerability risk indicates whether the destination address or related assets have medium- or high-risk vulnerabilities; security domain risk indicates whether the access relationship between the source and destination addresses' security domains involves cross-domain access, access from a low-security domain to a high-security domain, or vice versa; firewall path risk indicates the number of firewall nodes traversed in the access path and the cross-domain level; validity period risk indicates whether the duration of the requested policy exceeds the preset period; and policy hit result indicates whether the current firewall policy application work order contains duplicate, redundant, or conflicting policies with existing policies. The system calculates a risk score based on the risk assessment factors.

[0046] When the risk score falls within the low-risk range, the system outputs an approval opinion and classifies the risk level as low. When the risk score falls within the medium-risk range, the system outputs a manual review opinion and individual scores for each risk assessment factor, classifying the risk level as medium. When the risk score falls within the high-risk range, the system outputs a rejection opinion and individual scores for each risk assessment factor, classifying the risk level as high. When the review opinion is low-risk and approved, or medium-risk and requires manual review, the system generates a recommended policy according to the principle of least privilege. The source address of the recommended policy is the specific source host address corresponding to the firewall policy application ticket, the destination address is the specific destination service address corresponding to the firewall policy application ticket, the port and protocol are the ports and protocols actually used in the business, and the validity period is the period required by the business.

[0047] In one implementation, the specific source host address can be determined based on the IP asset profile and business system affiliation corresponding to the source address; the specific destination service address can be determined based on the IP asset profile, open ports, and business system corresponding to the destination address; the ports and protocols actually used by the business can be determined based on traffic session records, security device logs, and business system ledgers; and the required service duration can be determined based on the validity period of the firewall policy application ticket and the service assurance cycle. When the source address requested by the firewall policy application ticket is a large network segment, but historical communication records only show that a few specific hosts communicate with the destination service, the system will converge the source address of the recommended policy to the specific host address, instead of directly using the entire network segment. When the port requested by the firewall policy application ticket is a port range, but the destination service actually only uses one or more of these ports, the system will converge the port of the recommended policy to the port actually used by the business. When relevant fields cannot be determined based on IP asset profiles, historical communication records, or business system ledgers, the system will not directly set the uncertain fields to arbitrary addresses or arbitrary ports, but will output supplementary material requirements.

[0048] Further, the risk score is calculated, specifically including: Each risk assessment factor is assigned a weight coefficient and a value range; the corresponding individual score is determined based on the actual value of each risk assessment factor; the product of each individual score and its corresponding weight coefficient is summed to obtain the risk score.

[0049] In practical implementation, risk scoring is used to quantify the overall risk of firewall policy application tickets, addressing the problems in existing technologies where review opinions mainly rely on human experience, different reviewers have inconsistent judgment standards for similar firewall policy application tickets, and it is difficult to uniformly compare multiple risk assessment factors. Specifically, after completing firewall policy application ticket parsing, IP asset profile matching, asset consistency verification, and policy hit analysis, the system extracts risk assessment factors and calculates risk scores for firewall policy application tickets that are not satisfied by existing firewall policies. These risk assessment factors include at least port risk, vulnerability risk, security domain risk, firewall path risk, validity period risk, and policy hit result. In this embodiment, the risk score is set to an integer value from 1 to 100, with higher scores indicating higher risk. Specifically, 1-39 points represent the low-risk score range, 40-69 points the medium-risk score range, and 70-100 points the high-risk score range.

[0050] In the specific calculation, the system can first set the actual value x for each risk assessment factor. i With weighting coefficient w i Where i = 1, 2, ..., n, n is the number of risk assessment factors, and satisfies: ,

[0051] Where R represents the final risk score; i represents the sequence number of the risk assessment factor; n represents the total number of risk assessment factors involved in this risk score calculation; w i This represents the weight coefficient corresponding to the i-th risk assessment factor. The actual value of each risk assessment factor can be mapped to a standardized value between 0 and 1 according to preset rules. For example, in port risk, if only a single low-risk port is applied for, a lower score can be obtained; if multiple ports or high-risk ports are applied for, a higher score is obtained; in vulnerability risk, if the asset corresponding to the destination address has a high-risk vulnerability, the score is higher than that of the case with a medium-risk vulnerability; in security domain risk, access within the same security domain has the lowest score, access across security domains has a higher score, and if access from a low security domain to a high security domain is involved, the score increases further; in firewall path risk, the more firewall nodes passed through and the more security domains crossed, the higher the score; in validity period risk, short-term validity period has a lower score, and long-term validity period has a higher score; in policy hit results, if an existing policy that meets the access requirements already exists, there is no risk; if a redundant policy exists, the risk is low; if a conflicting policy exists, the risk is high. The specific value range of each of the above risk assessment factors can be pre-configured by the system. It is not limited to linear mapping, but can also adopt a segmented mapping method to meet the requirements of enterprise security strategy management.

[0052] After obtaining the actual values ​​of each risk assessment factor, the system calculates the comprehensive risk score using a weighted summation method, which can be expressed mathematically as follows: ,

[0053] Where R is the final risk score; i represents the index of the risk assessment factor; n represents the total number of risk assessment factors involved in this risk score calculation; w i This represents the weight coefficient corresponding to the i-th risk assessment factor. The risk score ranges from 1 to 100. If the calculated result is a decimal, it can be rounded up or down according to preset rules. The system can compare the risk score with preset thresholds: when 1≤R≤39, it is determined to be in the low-risk score range; when 40≤R≤69, it is determined to be in the medium-risk score range; and when 70≤R≤100, it is determined to be in the high-risk score range. Therefore, the risk score not only reflects the magnitude of a single risk assessment factor, but also reflects the comprehensive impact of multiple risk assessment factors, thus avoiding a one-sided judgment based on only one indicator.

[0054] For example, in one optional implementation, if a firewall policy application has a low port risk, no known vulnerabilities at the destination address, access within the same security domain, a path passing through only one firewall, and a short validity period, and there are no conflicting existing policies, then the overall scores for each risk assessment factor will be low, typically falling within the 1-39 point range. The system can automatically determine this as low-risk and approve it. If the firewall policy application involves cross-security domain access and has a longer validity period, but the port and vulnerability risks are average, then the risk score may fall within the 40-69 point range, and the system will output medium-risk for manual review. If the firewall policy application simultaneously has high-risk open ports, a high-risk vulnerability at the destination address, crosses multiple security domains, and has a long validity period, then the superposition of multiple high-weight risk assessment factors will result in a score between 70-100, and the system will output high-risk rejection accordingly. Through the above calculation method, this invention transforms the difficult-to-quantify review criteria into a calculable, interpretable, and verifiable risk score, thereby making the boundaries between low-risk, medium-risk, and high-risk clearly defined.

[0055] Furthermore, following step 6, the process also includes online verification of the approved recommendation strategy, specifically including: The generated recommended policy is then sent to the corresponding firewall device and made effective. Collect traffic hit data of the recommended policy on the firewall device, and determine whether the recommended policy is hit by actual access traffic based on the traffic hit data; If the recommended strategy is not hit within the preset verification period, a strategy optimization suggestion is generated and recorded in the strategy optimization list.

[0056] In practice, after generating the recommended policy, the system does not immediately end the review process. Instead, it further executes the policy deployment verification step to form a closed loop of review, distribution, monitoring, and optimization. Specifically, after outputting a review opinion indicating low-risk approval or medium-risk manual review, the system distributes the recommended policy to the corresponding firewall device to make it effective. The distribution location corresponds to the determined recommended configuration path to ensure that the recommended policy is on the correct firewall node in the actual access path. After the policy takes effect, the system continuously collects traffic hit data generated by the firewall device within a preset verification period and performs correlation analysis between the traffic hit data and the recommended policy to determine whether the recommended policy is actually hit by access traffic. If no hit record is detected within the preset verification period, the system determines that the recommended policy has not been actually used, or that there are situations such as the policy scope being too narrow, the configuration location being inaccurate, or the service not yet triggering access. The system then generates policy optimization suggestions, which include at least narrowing or adjusting one or more of the following: source address, destination address, port, protocol, or validity period. The system records the policy optimization suggestions in the policy optimization list for subsequent review and adjustment. Through this step, the present invention can not only identify risks and provide recommended policies during the policy application stage, but also further verify the effectiveness of the policies after they are launched, thereby avoiding redundant policies that have no hits for a long time from occupying policy library resources and improving the accuracy, traceability and continuous optimization capabilities of firewall policy management.

[0057] A firewall policy auditing system based on IP asset profiling and network topology includes: The work order parsing module is used to obtain firewall policy application work orders and to parse and verify the integrity of the source address, destination address, port, protocol, access direction, responsible person, business system, application reason and validity period in the firewall policy application work order. The IP asset profile building module is used to associate and merge IP asset data based on IP addresses in order to pre-build IP asset profiles. The asset verification module is used to match the source address and the destination address with the IP asset profile respectively, and generate an asset consistency verification result; The network topology path deduction module is used to deduce the access path from the source address to the destination address based on the network topology, which includes security domains, routing devices, firewall nodes, and business systems, and to determine the security domains and firewall nodes that the source address passes through to the destination address. The policy matching module is used to match the access requirements corresponding to the firewall policy application work order with the existing firewall policy library. When there is an existing policy that meets the access requirements, the module outputs an audit opinion that no new policy needs to be added. The risk review module is used to calculate a risk score based on risk assessment factors when there is no existing strategy that meets the access requirements, generate a review opinion based on the risk score, and generate a recommended strategy according to the principle of least privilege when the review opinion is low-risk and approved or medium-risk and manually reviewed.

[0058] In specific implementation, such as Figure 3 As shown, the firewall policy review system based on IP asset profiling and network topology includes a work order parsing module, an IP asset profiling construction module, an asset verification module, a network topology path deduction module, a policy matching module, and a risk review module. Each module can be deployed on the same application server or deployed on different computing nodes in a service-oriented manner and interact with data through interfaces.

[0059] After receiving a firewall policy application work order, the work order parsing module performs structured parsing on the source address, destination address, port, protocol, access direction, responsible person, business system, application reason, and validity period. It also verifies the completeness of required fields. If any are missing or abnormal, it directly outputs a "supplement missing fields" audit result, blocking invalid firewall policy application work orders at the entry point. For firewall policy application work orders with complete fields, the IP asset profiling module uses the source and destination addresses as indexes, integrating relevant data from CMDB, IP address management, DNS, DHCP, traffic session records, security device logs, vulnerability scans, business system ledgers, and historical work orders to form an IP asset profile containing organizational affiliation, asset type, active status, open ports, historical communication records, business system, data center, environment, risk label, and recent alarm information. The asset verification module then matches the source and destination addresses in the firewall policy application work order with the above profile to identify situations such as unregistered but used, registered but unused, duplicate registration, and inconsistencies with business systems. When inconsistencies are found, a risk work order is generated, prompting manual review. Subsequently, the network topology path deduction module constructs a directed graph with security domains, routing devices, firewall nodes, and business systems as nodes, based on network security domain segmentation data, firewall node deployment locations and interface information, routing device topology connections, and the security domain to which the business system belongs. It then searches for access paths from the security domain of the source address to the security domain of the destination address. If multiple paths exist, it selects the recommended configuration path by combining historical traffic data, routing priorities, and security domain policies, thereby determining the security domains and firewalls traversed from the source address to the destination. The policy matching module then compares the access requirements corresponding to the firewall policy application with the existing firewall policy library, identifying duplicate, redundant, or conflicting policies. If an existing policy exists that meets the access requirements, it directly outputs a review opinion stating that no new policy needs to be added. If no matching policy exists, the risk assessment module extracts factors such as port risk, vulnerability risk, security domain risk, firewall path risk, validity period risk, and policy hit results. It calculates a risk score according to preset weights and outputs an assessment opinion: low-risk approved, medium-risk subject to manual review, or high-risk rejected. Simultaneously, in cases of low-risk approval or medium-risk subject to manual review, a recommended policy is generated according to the principle of least privilege. Through the collaboration of these modules, this invention transforms the firewall policy assessment process, which originally relied on human experience, into an automated process based on asset facts, path deduction, and risk classification. This reduces policy conflicts and the risk of accidental approval, while improving policy assessment efficiency and consistency of results.

[0060] Example 2 In a specific implementation scenario, a company's production system A plans to access the database service of production system B within the same data center. The applicant submits a firewall policy request ticket in the system, including the source address, destination address, port, protocol, access direction, responsible person, business system, reason for request, and validity period. The source address is the specific IP address corresponding to the application server of production system A, the destination address is the specific IP address corresponding to the database server of production system B, the port is TCP 1521, the protocol is TCP, the validity period is 30 days, and the reason for request is "System A has added a new interface call, which requires access to the database service of system B."

[0061] Upon receiving the firewall policy request ticket, the ticket parsing module first performs structured parsing of the ticket fields and verifies the integrity of the source address, destination address, port, protocol, and validity period. Since the ticket is complete, the system proceeds to the next steps. Subsequently, the IP asset profiling module integrates relevant data from CMDB, IP address management, traffic session records, security device logs, and business system ledgers to generate an IP asset profile. This profile identifies the source address as an application server, the business system as production system A, and its activity status as active. The destination address is identified as a database server, the business system as production system B, and the destination address shows recent database access traffic. This step reconciles the static request information in the firewall policy request ticket with the actual operational asset facts, thus avoiding misjudgments based solely on similar names or manual registration errors.

[0062] Next, the asset verification module matches the source and destination addresses with the IP asset profiles, confirming that both are registered assets consistent with the business system ledger, and that there are no instances of unregistered but used, registered but unused, or inconsistent with the business system. The system then proceeds to match the access path and policy. Subsequently, the network topology path deduction module constructs a directed graph based on security domain divisions, firewall node deployment locations, and routing device topology relationships, and searches for an access path from the security domain of the source address to the security domain of the destination address. The system discovers that the access request only needs to pass through a single boundary firewall between the production business area and the database area, and the access direction is clearly from the production business area to the database area. Through this crucial step, the system transforms the "source address to destination address" access requirement into a specific endpoint: "which firewall and which security domain will control it?"

[0063] Subsequently, the policy matching module compares the source address, destination address, port, and protocol of the firewall policy request with the existing firewall policy database entry by entry. The comparison reveals that an existing firewall policy already exists that allows the address range of production system A to access the TCP 1521 port of the database server of production system B. This policy is identical to the one in the current firewall policy request. Based on this, the system determines that the access requirement corresponding to the firewall policy request has been duplicated by an existing policy, and therefore constitutes a duplicate policy, rather than requiring a new policy. This prevents the same business requirement from requiring the same firewall rule to be enabled repeatedly, avoids the continuous expansion of the policy database, and prevents increased operational complexity due to excessive duplicate policies.

[0064] If the existing policies are insufficient to meet the access requirements in this scenario, the system further enters the risk review module. The risk review module calculates a risk score by combining port risk, vulnerability risk, security domain risk, firewall path risk, validity period risk, and policy hit results. For example, if the target database server has a medium-risk vulnerability, and the access crosses a security domain with a long validity period, the overall risk score will increase accordingly; if the port is a necessary port for database operations, and the path is clear and the scope is convergent, the risk score will decrease. The system categorizes the results into low risk (1-39 points), medium risk (40-69 points), and high risk (70-100 points) according to the risk scoring rules, and outputs the corresponding conclusions. If it is low risk, the review is approved directly; if it is medium risk, manual review is required; if it is high risk, access is rejected.

[0065] In this implementation scenario, since the system identifies the application as a duplicate policy, it ultimately outputs a review opinion stating that no new policy needs to be added. Compared with existing technologies, this invention can not only determine "whether it can be enabled," but also further determine "whether there is already an available policy, whether there are risks, and the level of risk," thereby achieving a closed-loop review from firewall policy application work order review, asset consistency verification, access path deduction to policy matching. Therefore, this invention effectively solves the problems of strong reliance on manual review, inconsistent judgment standards, numerous duplicate activations, many policy conflicts, and low review efficiency in existing firewall policy review processes, and improves review accuracy and network security protection capabilities.

[0066] The above embodiments are merely preferred embodiments of the present invention and are not intended to limit the technical solutions of the present invention. Any technical solution that can be implemented based on the above embodiments without creative effort should be considered to fall within the scope of protection of the patent of the present invention.

Claims

1. A firewall policy auditing method based on IP asset profiling and network topology, characterized in that, Includes the following steps: Step 1: Obtain the firewall policy application work order, and perform field parsing and field integrity verification on the source address, destination address, port, protocol, access direction, responsible person, business system, application reason and validity period in the firewall policy application work order; Step 2: Associate and merge IP asset data based on IP addresses to create IP asset profiles; Step 3: For firewall policy application work orders that pass field integrity verification, match the source address and destination address with the IP asset profile respectively to generate asset consistency verification results; Step 4: For the source and destination addresses that have passed the asset consistency verification, based on the network topology and in combination with security domains, routing devices, firewall nodes and business systems, deduce the access path from the source address to the destination address, and determine the security domains and firewall nodes that the source address passes through to the destination address. Step 5: Match the access requirements corresponding to the firewall policy request ticket with the existing firewall policy library to determine if there are any existing policies that meet the access requirements. Step 6: When there is no existing policy that meets the access requirements, generate review comments and recommended policies based on the firewall policy by submitting a work order.

2. The firewall policy auditing method based on IP asset profiling and network topology according to claim 1, characterized in that, In step 2, the IP asset data includes at least two of the following data sources: CMDB, IP address management, DNS, DHCP, traffic session records, security device logs, vulnerability scanning, business system ledgers, and historical work orders.

3. The firewall policy auditing method based on IP asset profiling and network topology according to claim 1, characterized in that, In step 2, the IP asset profile includes at least two of the following fields: Organization affiliation, asset type, activity status, open ports, historical communication records, business systems, data centers, environment, risk labels, and recent alarm information.

4. The firewall policy auditing method based on IP asset profiling and network topology according to claim 1, characterized in that, In step 3, when the asset consistency verification result is inconsistent, an audit opinion requiring manual review is output, and a risk work order is generated, specifically including: Using the source address and the destination address as query conditions, respectively, query the corresponding asset records in the IP asset profile; If the source address or destination address is not recorded in the IP asset profile, and the address is determined to be active based on traffic session records, security device logs, or other operational data, then the corresponding address will be determined to be unregistered and already in use. If the IP asset profile contains a record of the source address or the destination address, but there is no corresponding traffic session record within a preset time window, then the corresponding address is determined to be registered but not used. If the same source address or the same destination address corresponds to multiple asset records in the IP asset profile, and the asset type field or organization affiliation field in the multiple asset records is inconsistent, then the corresponding address will be determined as duplicate registration. If the source address or the destination address is inconsistent with the business system field in the IP asset profile and the business system field in the firewall policy application work order, the corresponding address will be determined as having an inconsistent business system. When any of the above situations occur, the corresponding source address or destination address will be marked as a risk address, and a risk work order will be generated according to the corresponding situation.

5. The firewall policy auditing method based on IP asset profiling and network topology according to claim 1, characterized in that, In step 4, based on the network topology and considering security domains, routing devices, firewall nodes, and business systems, the access path from the source address to the destination address is derived, specifically including: Obtain network security domain segmentation data, firewall node connection locations and interface configuration information in the network topology, topological connection relationships of routing devices, and the security domain to which the business system belongs; construct a directed graph by treating security domains, routing devices, firewall nodes, and business systems as nodes and the connectivity relationships between nodes as edges. Using the security domain to which the source address belongs as the starting node and the security domain to which the destination address belongs as the ending node, search all access paths from the starting node to the ending node in the directed graph. If multiple access paths are found, one path is selected as the recommended configuration path based on historical traffic data, routing priority, and security domain policy. Based on the recommended configuration path, determine the security domains and firewall nodes that the source address and destination address will pass through.

6. The firewall policy auditing method based on IP asset profiling and network topology according to claim 1, characterized in that, In step 5, the access requirements corresponding to the firewall policy request work order are matched with existing firewall policy libraries, specifically including: Read the source address, destination address, port and protocol of each existing policy from the existing firewall policy library, and match the source address, destination address, port and protocol corresponding to the firewall policy application work order with each existing policy; Generate policy hit results, which include duplicate policies, redundant policies, or conflicting policies; When the source address, destination address, port, and protocol of the firewall policy request ticket are exactly the same as those of the existing policy, the policy hit result will be determined as a duplicate policy. When the source address, destination address, port, and protocol of the existing policy cover the firewall policy application ticket, the policy hit result is determined as a redundant policy; When the firewall policy request ticket overlaps with an existing denial policy in terms of source address, destination address, port, and protocol, the policy hit result is determined to be a conflicting policy. When the strategy hit result is a duplicate strategy, it is determined that there is an existing strategy that satisfies the access requirement.

7. The firewall policy auditing method based on IP asset profiling and network topology according to claim 6, characterized in that, In step 6, the review comments and recommended policies are generated based on the firewall policy application work order, specifically including: Extract risk assessment factors, which include at least port risk, vulnerability risk, security domain risk, firewall path risk, validity period risk, and policy hit result; calculate a risk score based on the risk assessment factors; If the risk score falls within the low-risk score range, then an approval opinion will be output, and the risk level will be determined as low risk. If the risk score falls within the medium risk range, then the manual review comments and individual scores for each risk assessment factor will be output, and the risk level will be determined as medium risk. If the risk score falls within the high-risk score range, a rejection review opinion and individual scores for each risk assessment factor will be output, and the risk level will be determined as high risk. When the review opinion is low-risk and approved or medium-risk and manually reviewed, a recommended policy is generated according to the principle of least privilege. The source address of the recommended policy is the specific source host address corresponding to the firewall policy application work order, the destination address is the specific destination service address corresponding to the work order, the port and protocol are the ports and protocols used in the actual business, and the validity period is the period required by the business.

8. The firewall policy auditing method based on IP asset profiling and network topology according to claim 7, characterized in that, The calculation of the risk score specifically includes: Each risk assessment factor is assigned a weight coefficient and a value range; the corresponding individual score is determined based on the actual value of each risk assessment factor; the product of each individual score and its corresponding weight coefficient is summed to obtain the risk score.

9. The firewall policy auditing method based on IP asset profiling and network topology according to claim 1, characterized in that, Following step 6, the process also includes online verification of the approved recommendation strategy, specifically including: The generated recommended policy is then sent to the corresponding firewall device and made effective. Collect traffic hit data of the recommended policy on the firewall device, and determine whether the recommended policy is hit by actual access traffic based on the traffic hit data; If the recommended strategy is not hit within the preset verification period, a strategy optimization suggestion is generated and recorded in the strategy optimization list.

10. A firewall policy auditing system based on IP asset profiling and network topology, characterized in that, include: The work order parsing module is used to obtain firewall policy application work orders and to parse and verify the integrity of the source address, destination address, port, protocol, access direction, responsible person, business system, application reason and validity period in the firewall policy application work order. The IP asset profile building module is used to associate and merge IP asset data based on IP addresses in order to pre-build IP asset profiles. The asset verification module is used to match the source address and the destination address with the IP asset profile respectively, and generate an asset consistency verification result; The network topology path deduction module is used to deduce the access path from the source address to the destination address based on the network topology, which includes security domains, routing devices, firewall nodes, and business systems, and to determine the security domains and firewall nodes that the source address passes through to the destination address. The policy matching module is used to match the access requirements corresponding to the firewall policy application work order with the existing firewall policy library. When there is an existing policy that meets the access requirements, the module outputs an audit opinion that no new policy needs to be added. The risk review module is used to calculate a risk score based on risk assessment factors when there is no existing strategy that meets the access requirements, generate a review opinion based on the risk score, and generate a recommended strategy according to the principle of least privilege when the review opinion is low-risk and approved or medium-risk and manually reviewed.