Intelligent route selection method, system, medium and equipment

Through the security assessment of routing nodes and the use of multiple attack detection models, combined with static routing priority scores, the problem of insufficient security of static routing in complex attack environments is solved, and efficient and secure Routing path selection improves the overall security and stability of the network.

CN118921313AActive Publication Date: 2024-11-08MING YANG POINT TIME TECH (SHENYANG) CO LTD
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202411408479.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-10
Publication Date
2024-11-08
Estimated Expiration
2044-10-10

AI Technical Summary

Technical Problem

The existing static routing method cannot effectively ensure routing security when facing complex network attacks, and dynamic routing protocols increase system complexity and operation and maintenance costs, and cannot adapt to the increasingly complex threat situation in modern network environments.

Method used

By conducting security assessment of routing nodes, combining priority scores in static routing, a comprehensive routing priority score is generated, and a variety of attack detection models are used to monitor and evaluate the security of each routing node, and calculate the 0 to 1 Security scores and couples them with the priority score of static routes for routing path selection.

Benefits of technology

It significantly improves the security and robustness of static routing, avoids the selection of unsafe nodes, improves the overall security and stability of the network, reduces the risk of misjudgment, and adapts to complex threats to modern network environments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118921313A_ABST
    Figure CN118921313A_ABST
Patent Text Reader

Abstract

The invention discloses an intelligent route selection method and system, a medium and equipment, the intelligent route selection system comprises a security scoring unit, a comprehensive scoring unit and a routing table updating unit, the security scoring unit is in communication connection with the comprehensive scoring unit, and the comprehensive scoring unit is in communication connection with the routing table updating unit. According to the method and the device, a comprehensive routing priority score is generated by performing security evaluation on the routing node and combining the priority score in the static routing, so that the communication efficiency of the node is considered and the security of the node is fully evaluated in the routing selection process, and finally, an efficient and secure routing path is selected.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of routing technology, in particular to an intelligent routing selection method, system, medium and equipment. Background Art

[0002] In today's network environment, with the continuous evolution of various complex attack methods, network security issues have become particularly prominent. Static routing, as a foundational technology for network communications, while offering excellent stability and efficiency, suffers from significant security deficiencies. Traditional static routing methods typically determine routing paths based solely on preset priority scores, neglecting to assess the security of network nodes. Consequently, they cannot effectively guarantee routing security when the network is subjected to various attacks. This issue is particularly acute when facing complex network threats such as distributed denial of service (DDoS) attacks, address resolution protocol (ARP) spoofing, password attacks, and IP address hijacking attacks.

[0003] Currently, some technical solutions attempt to enhance network security by introducing dynamic routing or adding additional security mechanisms. However, these solutions often rely on dynamic network topology adjustments or complex security protocol implementations, which increases system complexity and operational costs, and performs poorly in dealing with specific attack scenarios in static network environments. However, these dynamic routing protocols require frequent exchange of routing information, which may introduce additional network overhead and latency in large-scale networks. In addition, although some solutions improve network security by adding simple security checks to routing, these methods often rely on static rules and cannot effectively respond to increasingly complex attack methods, especially when attack behaviors are dynamic and diverse.

[0004] Existing static routing methods primarily rely on preset priority scores to determine routing paths. While this approach offers excellent stability and efficiency, it suffers from significant shortcomings when addressing network security threats. First, traditional static routing lacks an assessment of the security status of network nodes, making routing security insecure in the face of complex attacks such as DDoS attacks, ARP spoofing, password attacks, and IP address hijacking. Due to the diverse and dynamic nature of these attack vectors, the security status of network nodes is constantly changing, and existing technologies are unable to reflect these changes in real time. Consequently, static routing may, in some cases, select nodes that are under attack, increasing the security risk of the entire network.

[0005] Furthermore, while some existing technologies attempt to improve security by introducing dynamic routing or adding simple security mechanisms, these approaches either require frequent network topology adjustments, increasing system complexity and maintenance costs, or rely solely on static security rules, failing to effectively address dynamically changing attack behaviors. This demonstrates that existing static routing methods have inherent flaws in security assessment and are unable to adapt to the increasingly complex threat landscape in modern network environments. Summary of the Invention

[0006] To this end, the technical problem to be solved by the present invention is to provide an intelligent routing selection method, system, medium and device, which generates a comprehensive routing priority score by performing a security assessment on the routing nodes and combining the priority score in the static routing to ensure that in the routing selection process, not only the communication efficiency of the node is considered, but also its security is fully evaluated, and ultimately a routing path that is both efficient and secure is selected.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: An intelligent routing selection method comprises the following steps: S100) Perform security assessment on each route in the routing network. The specific steps are as follows: S101) Collecting traffic data on each routing node; S102) classifying and preprocessing the traffic data collected in step S101), wherein the preprocessing includes data cleaning, feature extraction, and category labeling; S103) Inputting the pre-processed traffic data into different analysis and detection models by category for analysis and detection; wherein the analysis and detection models include a DDoS attack analysis and detection model, an ARP spoofing attack analysis and detection model, a password attack analysis and detection model, and an IP address hijacking attack analysis and detection model; S104) normalizing the analysis and detection results obtained in step S103) and outputting a probability value, calculating an average of the probability values ​​outputted cumulatively n times and sorting the average probability values ​​by size, wherein the normalization method adopted is a Min-Max normalization method; wherein n is an integer greater than or equal to 50; S105) Calculate the security score of each routing node. The specific formula is as follows: Where, is the probability value of DDoS attack; is the probability value of ARP spoofing attack; is the probability value of password attack; The probability value of IP address hijacking attack; Index j, index k, index l, and index m are all positive numbers greater than 0, and their values ​​are determined according to the following strategy: According to the result of sorting the average values ​​of the probability values ​​obtained in step S104), the value range of the index related to the average value of the first-ranked probability value is (0.25, 0.5], the value range of the index related to the average value of the second-ranked probability value is (0.5, 1], the value range of the index related to the average value of the third-ranked probability value is (1, 2], and the value range of the index related to the average value of the fourth-ranked probability value is (2, 3]; S200) couples the security score obtained in step S100) with the priority score of each routing node in the static route to calculate a comprehensive score for each routing node, and selects a routing path based on the comprehensive score; wherein the comprehensive score is calculated using the following formula: Where e is a natural constant; Pref is the priority score of a routing node in the routing table of other routing nodes.

[0008] In the above intelligent routing selection method, the probability value output by the normalization process ranges from 0 to 1.

[0009] In the above intelligent routing selection method, in step S104), the average value of the probability value is a square average.

[0010] In the above intelligent routing selection method, the values ​​of index j, index k, index l, and index m are determined according to the following strategy: According to the result of sorting the average values ​​of the probability values ​​obtained in step S104), the value range of the index related to the average value of the probability value ranked first is (0.25, 0.3], the value range of the index related to the average value of the probability value ranked second is [0.8, 1], the value range of the index related to the average value of the probability value ranked third is [1.5, 2], and the value range of the index related to the average value of the probability value ranked fourth is [2.5, 3].

[0011] In the above intelligent routing selection method, in step S101), when collecting traffic data on each routing node, data collection is performed according to the traffic data type required by each analysis and detection model.

[0012] In the above intelligent routing selection method, in step S200), when selecting a routing path based on the comprehensive score, the routing table in each routing node is first updated based on the comprehensive score, and then data is transmitted based on the updated routing table.

[0013] A system for performing route selection using the above-mentioned intelligent route selection method comprises: A security scoring unit is used to perform a security scoring on each routing node; the security scoring unit includes a data acquisition module, a data processing module, an analysis and detection model module, a standardization processing module, and a security score calculation module. The data acquisition module is in communication with the data processing module, the data processing module is in communication with the analysis and detection model module, the analysis and detection model module is in communication with the standardization processing module, and the standardization processing module is in communication with the security scoring module. A comprehensive scoring unit, configured to comprehensively score each routing node using the security score and priority score of the routing node; A routing table updating unit, configured to update the routing table in each routing node according to the comprehensive score; The security scoring unit is communicatively connected to the comprehensive scoring unit, and the comprehensive scoring unit is communicatively connected to the routing table updating unit.

[0014] The above system, the analysis and detection model module includes a DDoS attack analysis and detection model module, an ARP spoofing attack analysis and detection model module, a password attack analysis and detection model module and an IP address hijacking attack analysis and detection model module.

[0015] A computer-readable storage medium stores a computer program, which implements the above-mentioned intelligent routing method when executed by a processor.

[0016] The computer device includes a readable storage medium, a processor, and a computer program stored in the readable storage medium and executable on the processor. When the computer program is executed by the processor, the intelligent routing selection method is implemented.

[0017] The technical solution of the present invention achieves the following beneficial technical effects: 1. The present invention significantly improves the security of static routing, making it more flexible and effective in protecting the overall security of the network in the face of various network attacks.

[0018] 2. The present invention can effectively consider security factors in the static routing selection process, avoiding the selection of unsafe nodes when facing potential network attacks, thereby improving the overall security of the network.

[0019] 3. The present invention can significantly improve the security and robustness of routing without affecting network performance.

[0020] 4. This invention introduces multiple attack detection models to monitor and evaluate the security of each routing node, generating a security score between 0 and 1. This score is then coupled with the priority score in static routing to form a comprehensive score. This comprehensive score not only considers the node's topological location and traditional priority factors, but also incorporates its security into the decision-making process. As a result, when making route selections, it can effectively avoid directing traffic to nodes with potential security risks, greatly improving the overall security and stability of the network.

[0021] 5. The design of this invention also balances system efficiency and practicality. The model accurately identifies and evaluates various attack types, ensuring the reliability of the security score. The innovative approach of combining this score with a priority score not only simplifies the complexity of routing selection but also ensures the algorithm's applicability in real-world network environments. Therefore, while improving network security, this invention maintains excellent performance and practical application value, fully demonstrating its superiority.

[0022] 6. This invention enhances the intelligent routing system's sensitivity to potential threats by introducing differentiated treatment of different attack probabilities. This ensures that when vulnerable nodes are identified, the probability of the associated attack plays a greater role in the security score. This ensures that the system can promptly reflect changes in the network environment, optimize routing strategies, and reduce the risk of misjudgment, significantly improving network security and adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a working principle diagram of the intelligent routing system of the present invention; Figure 2 A flow chart of intelligent routing selection in the present invention; Figure 3 Schematic diagram of a computer device capable of intelligent routing. DETAILED DESCRIPTION

[0024] The present invention is further described below with reference to examples.

[0025] like Figure 1 As shown, the intelligent routing selection system of the present invention includes a security scoring unit, a comprehensive scoring unit and a routing table updating unit. The security scoring unit is communicatively connected to the comprehensive scoring unit, and the comprehensive scoring unit is communicatively connected to the routing table updating unit.

[0026] Among them, the security scoring unit is used to perform a security score on each routing node; the security scoring unit includes a data acquisition module, a data processing module, an analysis and detection model module, a standardization processing module and a security score calculation module, the data acquisition module is communicatively connected to the data processing module, the data processing module is communicatively connected to the analysis and detection model module, the analysis and detection model module is communicatively connected to the standardization processing module, and the standardization processing module is communicatively connected to the security scoring module. In this embodiment, the analysis and detection model module includes a DDoS attack analysis and detection model module, an ARP spoofing attack analysis and detection model module, a password attack analysis and detection model module and an IP address hijacking attack analysis and detection model module; the comprehensive scoring unit is used to perform a comprehensive score on each routing node using the security score and priority score of the routing node; the routing table update unit is used to update the routing table in each routing node according to the comprehensive score.

[0027] like Figure 2 As shown, the intelligent routing system is used for routing selection. The specific steps are as follows: S100) Perform security assessment on each route in the routing network. The specific steps are as follows: S101) Collecting traffic data on each routing node; S102) classifying and preprocessing the traffic data collected in step S101), wherein the preprocessing includes data cleaning, feature extraction, and category labeling; S103) inputting the pre-processed traffic data into different analysis and detection models according to categories for analysis and detection; S104) normalizing the analysis and detection results obtained in step S103) and outputting a probability value, calculating the average of the probability values ​​outputted cumulatively n times and sorting the average probability values ​​by size, wherein the normalization method adopted is the Min-Max normalization method; wherein n is an integer greater than or equal to 50, and in the present invention, preferably 100; S105) Calculate the security score of each routing node. The specific formula is as follows: Where, is the probability value of DDoS attack; is the probability value of ARP spoofing attack; is the probability value of password attack; The probability value of IP address hijacking attack; Index j, index k, index l, and index m are all positive numbers greater than 0, and their values ​​are determined according to the following strategy: According to the result of sorting the average values ​​of the probability values ​​obtained in step S104), the value range of the index related to the average value of the first-ranked probability value is (0.25, 0.5], the value range of the index related to the average value of the second-ranked probability value is (0.5, 1], the value range of the index related to the average value of the third-ranked probability value is (1, 2], and the value range of the index related to the average value of the fourth-ranked probability value is (2, 3]; S200) couples the security score obtained in step S100) with the priority score of each routing node in the static route, calculates a comprehensive score of each routing node, and selects a routing path based on the comprehensive score.

[0028] The priority score used in the comprehensive scoring of a routing node is the priority score of the routing node in the routing table of the previous hop routing node.

[0029] In scenarios where a more nuanced response to a particular attack is required, the values ​​of index j, index k, index l, and index m are determined according to the following strategy: According to the result of sorting the average values ​​of the probability values ​​obtained in step S104), the value range of the index related to the average value of the probability value ranked first is (0.25, 0.3], the value range of the index related to the average value of the probability value ranked second is [0.8, 1], the value range of the index related to the average value of the probability value ranked third is [1.5, 2], and the value range of the index related to the average value of the probability value ranked fourth is [2.5, 3].

[0030] The purpose of increasing the difference between two adjacent numbers of index j, index k, index l and index m is to make the security score more subtle and accurate in responding to the impact of a certain attack, thereby further ensuring the security of the routing path.

[0031] For the usage scenario of the intelligent routing selection system in the first use invention, the historical traffic data of the scenario or scenarios similar to the scenario can be used to train the DDoS attack analysis and detection model, the ARP spoofing attack analysis and detection model, the password attack analysis and detection model, and the IP address hijacking attack analysis and detection model, and at the same time, it can be known which type of attack accounts for a higher proportion in the scenario, thereby facilitating more emphasis on this type of attack when scoring security.

[0032] After the intelligent routing system of the present invention has been used for a period of time, the intelligent routing system can adjust the values ​​of index j, index k, index l, and index m according to the average value changes of the probability values ​​of various attacks in the historical traffic data, or the user can adjust them by himself.

[0033] In this embodiment, the analysis and detection models used include the DDoS attack analysis and detection model in the DDoS attack analysis and detection model module, the ARP spoofing attack analysis and detection model in the ARP spoofing attack analysis and detection model module, the password attack analysis and detection model in the password attack analysis and detection model module, and the IP address hijacking attack analysis and detection model in the IP address hijacking attack analysis and detection model module. The machine learning model used by the DDoS attack analysis and detection model is a random forest model, the machine learning model used by the ARP spoofing attack analysis and detection model is a K-nearest neighbor algorithm model, the machine learning model used by the password attack analysis and detection model is a logistic regression model, and the machine learning model used by the IP address hijacking attack analysis and detection model is an anomaly detection algorithm model.

[0034] And in step S101), when collecting the flow data on each routing node, data collection is performed according to the flow data type required by each analysis and detection model.

[0035] In step S200), the comprehensive score is calculated using the following formula: Where e is a natural constant; Pref is the priority score of a routing node in the routing table of other routing nodes.

[0036] After calculating the comprehensive score of each routing node, when selecting a routing path based on the comprehensive score, the routing table in each routing node is first updated based on the comprehensive score, and then data is transmitted based on the updated routing table.

[0037] Specifically, the present invention first uses the traffic capture tool Wireshark to obtain traffic data from each routing node in a static routing network. The traffic data is then classified and preprocessed. The preprocessed traffic data is then input into a DDoS attack analysis and detection model, an ARP spoofing attack analysis and detection model, a password attack analysis and detection model, and an IP address hijacking attack analysis and detection model to distinguish four different router attacks. The different types of traffic data captured by Wireshark for each analysis and detection model are as follows: (1) DDoS attack: traffic rate, packet size, and number of connections; (2) ARP spoofing attack: number of ARP packets, ARP packet frequency, source and destination MAC address distribution; (3) Password attacks: number of login attempts per hour, login success rate, and IP address distribution; (4) IP address hijacking attack: IP address usage, IP address allocation change records, and source records.

[0038] Four analysis and detection models perform timely analysis and detection on traffic data captured by Wireshark. The results are normalized after each model, outputting a probability value between 0 and 1. A probability value closer to 0 indicates that the routing node is less vulnerable to a particular attack, while a probability value lower than 0 indicates that it is more susceptible to a particular attack. The greater the vulnerability to a particular attack, the greater the weight of the probability value of that attack in the security score. This requires the intelligent routing selection system of the present invention to be more sensitive to that attack, meaning it has a more subtle perception of it. This, in turn, requires that the probability value of that attack be amplified to a certain degree during the security score calculation process. Therefore, the present invention utilizes an exponential method to differentiate the probability values ​​of different attacks, making the probability values ​​of attacks to which a route is vulnerable more prominent. This can reduce the possibility that the intelligent routing selection system of the present invention misjudges the vulnerability of a route to an attack when the probability value of that route is vulnerable decreases briefly.

[0039] because Indicates the priority score of the routing node in the routing table of other routing nodes. According to the static routing rules, The smaller the value, the higher the priority of the router. Generally speaking, the smaller the score, the higher the priority of the router, and vice versa.

[0040] Finally, after calculating the comprehensive score of each routing node in the static routing network, the comprehensive score is used to update the routing table to guide the transmission of data in the routing nodes and realize intelligent routing selection.

[0041] Based on the above-mentioned intelligent routing selection method, accordingly, this example also provides a computer-readable storage medium storing a computer program, which implements the following steps when executed by a processor: collecting traffic data of each routing node in the static routing network, calculating the security score of each routing node, and then calculating the comprehensive score of the routing node based on the security score and priority score of each routing node, and then updating the routing table based on the comprehensive score of each routing node, and finally using the updated routing table to perform routing selection and data transmission.

[0042] like Figure 3As shown, based on the above-mentioned intelligent routing selection method and computer-readable storage medium, in this embodiment, a computer device is also provided, which includes a readable storage medium, a processor, and a computer program stored on the readable storage medium and capable of running on the processor, wherein the readable storage medium and the processor are both arranged on a bus, and when the processor executes the computer program, the following steps are implemented: collecting traffic data of each routing node in the static routing network, calculating the security score of each routing node, and then calculating the comprehensive score of the routing node based on the security score and priority score of each routing node, and then updating the routing table based on the comprehensive score of each routing node, and finally using the updated routing table to perform routing selection and data transmission.

[0043] Obviously, the above embodiments are merely examples for clarity of explanation and are not intended to limit the implementation methods. Those skilled in the art will readily appreciate that other variations or modifications based on the above descriptions are possible. It is not necessary and impossible to enumerate all implementation methods here. Obvious variations or modifications arising therefrom remain within the scope of protection of the claims of this patent application.

Claims

1. An intelligent routing selection method, characterized in that: The steps include: S100) Perform security assessment on each route in the routing network. The specific steps are as follows: S101) Collecting traffic data on each routing node; S102) classifying and preprocessing the traffic data collected in step S101), wherein the preprocessing includes data cleaning, feature extraction and category labeling; S103) inputting the pre-processed traffic data into different analysis and detection models according to categories for analysis and detection; wherein the analysis and detection models include DDoS attack analysis and detection models, ARP spoofing attack analysis and detection models, password attack analysis and detection models and IP address hijacking attack analysis and detection models; S104) performing standardization processing on the analysis and detection results obtained in step S103) and outputting a probability value, calculating the average value of the probability values ​​outputted cumulatively n times and sorting the average values ​​of the probability values ​​according to their size, and the method adopted for the standardization processing is the Min-Max standardization method; wherein n is an integer greater than or equal to 50; S105) Calculate the security score of each routing node. The specific formula is as follows: In the formula, is the probability value of DDoS attack; is the probability value of ARP spoofing attack; is the probability value of password attack; is the probability value of IP address hijacking attack; Index j, index k, index l, and index m are all positive numbers greater than 0, and their values ​​are determined according to the following strategy: According to the result of sorting the average values ​​of the probability values ​​obtained in step S104), the value range of the index associated with the average value of the first-ranked probability value is (0.25, 0.5], the value range of the index associated with the average value of the second-ranked probability value is (0.5, 1], the value range of the index associated with the average value of the third-ranked probability value is (1, 2], and the value range of the index associated with the average value of the fourth-ranked probability value is (2, 3]; S200) couples the security score obtained in step S100) with the priority score of each routing node in the static route, calculates the comprehensive score of each routing node, and selects the routing path based on the comprehensive score; wherein the comprehensive score is calculated by the following formula: Where e is a natural constant; Pref is the priority score of a routing node in the routing table of other routing nodes.

2. The intelligent routing selection method according to claim 1, characterized in that: In step S104), the probability value output by the normalization process ranges from 0 to 1.

3. The intelligent routing selection method according to claim 1, characterized in that: In step S104), the average value of the probability value is a square mean.

4. The intelligent routing selection method according to claim 1, characterized in that: The values ​​of index j, index k, index l and index m are determined according to the following strategy: According to the result of sorting the average values ​​of the probability values ​​obtained in step S104), the value range of the index related to the average value of the first-ranked probability value is (0.25, 0.3], the value range of the index related to the average value of the second-ranked probability value is [0.8, 1], the value range of the index related to the average value of the third-ranked probability value is [1.5, 2], and the value range of the index related to the average value of the fourth-ranked probability value is [2.5, 3].

5. The intelligent routing selection method according to claim 1, characterized in that: In step S101), when collecting traffic data on each routing node, data collection is performed according to the type of traffic data required by each analysis and detection model.

6. The intelligent routing selection method according to claim 1, characterized in that: In step S200), when selecting a routing path based on the comprehensive score, the routing table in each routing node is first updated based on the comprehensive score, and then data is transmitted based on the updated routing table.

7. A system for performing route selection using the intelligent route selection method according to claim 1, characterized in that: include: A security scoring unit, used to perform security scoring on each routing node; The safety scoring unit includes a data acquisition module, a data processing module, an analysis and detection model module, a standardization processing module and a safety scoring calculation module. The data acquisition module is connected to the data processing module in communication, the data processing module is connected to the analysis and detection model module in communication, the analysis and detection model module is connected to the standardization processing module in communication, and the standardization processing module is connected to the safety scoring module in communication; A comprehensive scoring unit, used to comprehensively score each routing node using the security score and priority score of the routing node; A routing table updating unit, used for updating the routing table in each routing node according to the comprehensive score; The security scoring unit is communicatively connected to the comprehensive scoring unit, and the comprehensive scoring unit is communicatively connected to the routing table updating unit.

8. The system according to claim 7, characterized in that The analysis and detection model modules include DDoS attack analysis and detection model modules, ARP spoofing attack analysis and detection model modules, password attack analysis and detection model modules and IP address hijacking attack analysis and detection model modules.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the intelligent routing selection method according to claims 1 to 6 is implemented.

10. A computer device comprising a readable storage medium, a processor, and a computer program stored on the readable storage medium and executable on the processor, characterized in that: When the computer program is executed by a processor, the intelligent routing selection method according to claims 1 to 6 is implemented.

Citation Information

Patent Citations

  • Security based network access selection

    CN101690097A

  • Route selection method, network device, system, computer program product and medium

    CN117354234A

  • Switch data scheduling method and switch

    CN118337728A

  • Threat evaluation system and method

    US20130347116A1

  • Methods for detecting and mitigating malicious network activity based on dynamic application context and devices thereof

    US20200186563A1