Network security systematic defense and supervision platform in gateway field

Through the systematic defense supervision platform of network security, data flow classification, feature extraction and desensitization risk analysis, the problem of insufficient risk warning and intelligence sharing of key information infrastructure is solved, and timely response and efficient protection of new attacks are achieved.

CN118944950AInactive Publication Date: 2024-11-12WUHAN ANYU INFORMATION SECURITY TECH CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202411169845.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-25
Publication Date
2024-11-12
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Critical information infrastructure has shortcomings in risk warning and intelligence sharing, and it is difficult to detect and respond to potential security threats in a timely manner. Traditional security supervision methods are unable to do so when facing new attacks.

Method used

A systematic defense supervision platform for network security was designed, including data flow classification module, feature extraction identification module and desensitization risk analysis module. Through data flow classification, feature extraction, normalization processing and pentagonal radar graph analysis, in real time Collect and evaluate the risk level of network data packets, carry out automated attack interference and suppression and intelligent security operations.

Benefits of technology

The security protection level of critical information infrastructure has been improved, and through accurate attack tracing and continuous security assessment and improvement, the probability of misjudgment is reduced, and the judgment and protection capabilities of network security situations are enhanced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118944950A_ABST
    Figure CN118944950A_ABST
Patent Text Reader

Abstract

The invention, which relates to the technical field of network security supervision, discloses a network security systematic defense supervision platform in the gateway field, comprising a data stream classification module, a feature extraction and identification module, and a desensitization risk analysis module. The data stream classification module is used for dividing different types of data streams transmitted from one node to another node into a webpage data stream, a text data stream, an audio and video data stream, a database access data stream and a sensor data stream; the feature extraction and identification module is also used for collecting secondary conventional data information and secondary conventional bidding information of the database access data stream and the sensor data stream transmitted in unit time, and sending the secondary conventional data information and the secondary conventional bidding information to the desensitization risk analysis module; according to the invention, comprehensive evaluation can be carried out according to the first-level desensitization risk factor and the second-level desensitization risk factor, and the network security situation can be accurately judged.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of network security supervision, and in particular to a network security systematized defense supervision platform in a key infrastructure field. Background Art

[0002] With the continuous development of technology and the popularization of the Internet, network information security faces increasingly complex and diverse threats. Common attack methods include computer viruses, malware, denial of service attacks, etc. In addition, there are more covert and advanced attack methods, such as APT (Advanced Persistent Threat) attacks, zero-day vulnerability exploits, etc. The security of critical information infrastructure is related to the national economy and people's livelihood. It is the cornerstone for ensuring the normal operation of the social economy and the orderly production and life. With the continuous advancement of informatization and digital transformation, critical information infrastructure has assumed the basic architecture responsibility of the superstructure.

[0003] At present, critical information infrastructure has deficiencies in risk warning and intelligence sharing, making it difficult to detect and respond to potential security threats in a timely manner. Traditional security supervision methods are powerless in the face of new attacks, and more advanced and effective technical means are needed to improve security supervision capabilities. Therefore, a network security systematization defense supervision platform in the key infrastructure field is proposed to cope with the current complex network information security environment and ensure the security of critical information infrastructure; by building comprehensive security perception, accurate attack tracing, automated attack interference and suppression, intelligent security operation and response, and continuous security assessment and improvement capabilities, the security protection level of critical information infrastructure is improved. Summary of the invention

[0004] The present invention aims at the technical problems existing in the prior art and provides a network security systematic defense and supervision platform in the key infrastructure field.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] A network security systematized defense and supervision platform in key infrastructure fields, including:

[0007] Data stream classification module, feature extraction and recognition module, desensitization risk analysis module;

[0008] The data stream classification module is used to classify different types of data streams transmitted from one node to another into web page data stream, text data stream, audio and video data stream, database access data stream and sensor data stream;

[0009] The feature extraction and recognition module is used to collect the first-level conventional data information and the first-level conventional sub-label information of the web page data stream, text data stream and audio and video data stream transmitted in a unit time, and send it to the desensitization risk analysis module; it is also used to collect the second-level conventional data information and the second-level conventional sub-label information of the database access data stream and sensor data stream transmitted in a unit time, and send it to the desensitization risk analysis module;

[0010] The first-level conventional data information includes the data packet length YLi, the maximum data packet transmission unit YDi and the data block size YKi of the web page data stream, text data stream and audio and video data stream transmitted per unit time;

[0011] The data packet length, the maximum transmission unit of the data packet and the data block size in the first-level conventional data information are normalized, and the normalization formula is:

[0012] Normalization

[0013] Among them, YLi represents the current data packet length, YLi max Indicates the maximum length of the data packet, YLi min Indicates the minimum value of the data packet length;

[0014] Normalization

[0015] Among them, YDi represents the maximum transmission unit of the current data packet, YDi max Indicates the maximum value of the maximum transmission unit of the data packet, YDi min Indicates the minimum value of the maximum transmission unit of a data packet;

[0016] Normalization

[0017] Among them, YKi represents the current data block size, YKi max Indicates the maximum size of the data block, YKi min Indicates the minimum value of the data block size;

[0018] Different weight coefficients are assigned to the packet length, maximum transmission unit of the packet and data block size in the normalized first-level conventional data information. The weight of the packet length is preset to be w L , the weight of the maximum transmission unit of a data packet is w D , the weight of the data block size is w K , the normalized first-level conventional data information is YJi, and the calculation formula is:

[0019] YJi=w l ×Normalized YLi+w D ×Normalized YDi+wK ×Normalized YKi;

[0020] In the formula, w L +w D +w K =1, and each weight coefficient is adjusted according to actual needs.

[0021] The desensitization risk analysis module is used to perform a first-level desensitization risk analysis on the normalized first-level conventional data information and first-level conventional sub-label information of the web page data stream, text data stream and audio and video data stream transmitted within a unit time to generate a first-level desensitization risk factor;

[0022] The desensitizing risk analysis module is also used to perform secondary desensitizing risk analysis on the secondary conventional data information and secondary conventional sub-label information of the database access data stream and sensor data stream transmitted within a unit time to generate secondary desensitizing risk factors.

[0023] The feature extraction and identification module collects in real time the data packet length YLi, the maximum data packet transmission unit YDi, the data block size YKi, the data packet transmission rate YCi and the data packet transmission delay YSi of the data packets in the web page data stream, text data stream and audio and video data stream transmitted per unit time, and forms a pentagonal radar chart by simulating and calculating the five sets of data to obtain the desensitization risk coefficient and risk level.

[0024] The five sets of collected data are preprocessed, which specifically includes the following steps:

[0025] Remove outliers, missing values, and duplicate data from the five sets of data.

[0026] Construct a pentagonal radar chart and calculate the area of ​​the pentagon, which includes the following steps:

[0027] S101: construct a pentagonal radar chart, which contains five data axes, and map the five sets of preprocessed data onto the five axes of the radar chart;

[0028] S102: Based on the specific values ​​in the five sets of data, draw points on the corresponding axes, and connect the five points on the five axes to form a pentagon;

[0029] S103: Calculate the area of ​​the connected pentagons using a geometric formula. For a convex pentagon, divide it into triangles, calculate the areas of the triangles respectively, and sum them up.

[0030] The relationship between the pentagon area and the desensitization risk value is calculated to judge the risk level, which specifically includes the following steps:

[0031] S104: Establish a mapping model between the pentagon area and the desensitization risk coefficient through the historical data of the five groups of data, and use the regression analysis method to train the model;

[0032] S105: According to the mapping model, different risk levels are set, and thresholds of the pentagonal area are set; for example, the risk levels are divided into three levels: high, medium, and low, and corresponding area thresholds are set;

[0033] S106: when collecting and processing five sets of data in real time, calculating the pentagonal area of ​​each data packet, and judging the risk level according to the set threshold value;

[0034] S107: Set a time window T, perform batch processing on the data packets in each time window, and at the end of the window, calculate the average area of ​​all data packets in the current window to judge the risk level of the current window.

[0035] The mapping model between the pentagon area and the desensitization risk coefficient is established, and a regression analysis method is used to train the model. The specific method is:

[0036] S201: Take the pentagon area as the independent variable X and the desensitization risk coefficient as the dependent variable Y;

[0037] S202: Setting the corresponding relationship between the area of ​​a part of the pentagon obtained through the pentagon radar chart and the desensitization risk coefficient;

[0038] S203: Construct a linear regression model, and the calculation formula is:

[0039] Y=β0+β1X+∈;

[0040] Among them, Y represents the desensitization risk coefficient, β0 represents the intercept, β1 represents the slope, X represents the area of ​​the pentagon, and ∈ represents the error term, which is the difference between the model prediction value and the actual observation value;

[0041] S204: Use the least squares formula to estimate the regression coefficients β0 and β1 by minimizing the residual sum of squares. The calculation formula is:

[0042]

[0043] Among them, RSS means minimizing the residual sum of squares, y i represents the actual observed desensitization risk coefficient value, x i represents the actual observed pentagon area value, n represents the number of observations, β0 represents the intercept, and β1 represents the slope;

[0044] S205: Solve the equation group to obtain the values ​​of β0 and β1. The calculation formula is:

[0045]

[0046] In the formula, represents the mean value of the desensitization risk coefficient actually observed, represents the actual observed mean area of ​​the pentagon, y i represents the actual observed desensitization risk coefficient value, x i represents the actual observed pentagon area value, n represents the number of observations, β0 represents the intercept, and β1 represents the slope;

[0047] S206: Calculate the above regression coefficient using statistical software to obtain the regression coefficient result. The regression equation is:

[0048] Y = -0.01 + 0.004 × X;

[0049] Among them, Y represents the desensitization risk coefficient and X represents the area of ​​the pentagon.

[0050] The first-level conventional data information collected by the feature extraction and recognition module includes five groups of data of the data packet. The interval risk value is obtained by dividing the above data packets according to time windows and calculating the pentagon area in a single time window, which specifically includes the following steps:

[0051] 301: Set the time window T1, count the number of all data packets transmitted in a single time window, and record the size of each data packet;

[0052] S302: according to the size of the data packet, set a series of size intervals, and allocate each data packet to a corresponding interval according to the size of the data packet;

[0053] If the number of packets in a certain interval exceeds the preset limit, the interval is further subdivided. If the number of packets in a certain interval is insufficient, they are extracted from the remaining time windows;

[0054] S303: Calculate the sum of the pentagonal areas of the data packets in each time window, and calculate the interval volume according to the sum of the pentagonal areas and the interval length, where the interval volume is the product of the sum of the pentagonal areas and the interval length, and set a risk threshold for each interval;

[0055] 304: For data packets in the interval exceeding the risk threshold, a separate desensitization risk assessment is performed using a linear regression model;

[0056] S305: Desensitizing the high-risk data packets according to the evaluation results.

[0057] Dynamically adjusting the number of data packets collected by the feature extraction and recognition module and issuing medium- and high-risk warnings specifically includes the following steps:

[0058] S401: dividing the data packets according to step S302 so that each interval contains a maximum number of data packets;

[0059] S402: Perform risk assessment on each interval, calculate the interval risk value, set the risk warning threshold, and when the risk value of a certain interval exceeds the warning threshold, mark the interval as a medium-high risk interval;

[0060] S403: When a certain interval is marked as medium or high risk, further group the data packets in the interval, divide the data packets in the interval into multiple sub-intervals according to the specific attributes of the data packets, and re-evaluate the risk of each sub-interval;

[0061] S404: If there are still medium- and high-risk areas in the sub-interval, the sub-interval is further grouped into finer granularity and the risk assessment process is repeated;

[0062] Continue iterating the grouping until a sufficiently small risk area or single data packet is found;

[0063] S405: within the final risk sub-interval, each data packet is evaluated and calculated individually to determine medium and high risk data packets;

[0064] Desensitize the identified medium and high risk data packets.

[0065] The beneficial effects of the present invention are:

[0066] Normalization processing can eliminate the dimensional differences between different parameters, so that parameters of different magnitudes can be compared and analyzed on the same scale, which is conducive to improving the accuracy and efficiency of data processing. The normalized first-level conventional data information can be comprehensively evaluated in the desensitizing risk analysis module based on the first-level desensitizing risk factors and the second-level desensitizing risk factors to more accurately judge the network security situation.

[0067] When the data packet value in the first-level general data information is too large or too small, misjudgment may occur when the desensitization risk analysis module performs desensitization risk analysis. The probability of misjudgment can be reduced by normalizing and weighting the data packets in the first-level general data information. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 It is a flow chart of S101-S107 of a network security systematized defense supervision platform in a key infrastructure field of the present invention;

[0069] Figure 2 It is a flow chart of S201-S206 of a network security systematic defense and supervision platform in a key field of the present invention. DETAILED DESCRIPTION

[0070] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0071] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, "plurality" means two or more, unless otherwise clearly and specifically defined.

[0072] In the description of the present application, the term "for example" is used to mean "used as an example, illustration or description". Any embodiment described as "for example" in the present application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid unnecessary details to obscure the description of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present application.

[0073] Example 1

[0074] The present application provides a network security systematized defense and supervision platform in the key infrastructure field, including:

[0075] Data stream classification module, feature extraction and recognition module, desensitization risk analysis module;

[0076] The data stream classification module is used to classify different types of data streams transmitted from one node to another into web page data stream, text data stream, audio and video data stream, database access data stream and sensor data stream;

[0077] The feature extraction and recognition module is used to collect the first-level conventional data information and first-level conventional sub-label information of the web page data stream, text data stream and audio and video data stream transmitted in a unit time, and send it to the desensitization risk analysis module;

[0078] The feature extraction and recognition module is also used to collect the secondary conventional data information and secondary conventional sub-label information of the database access data stream and the sensor data stream transmitted in a unit time, and send it to the desensitization risk analysis module;

[0079] The unit time mentioned in this application can be set autonomously according to the data flow characteristics of the enterprise and organization network and the demand for data transmission, preferably half a day or a day;

[0080] The first-level conventional data information includes the data packet length YLi, the maximum data packet transmission unit YDi and the data block size YKi of the web page data stream, text data stream and audio and video data stream transmitted per unit time;

[0081] The data packet length, the maximum transmission unit of the data packet and the data block size in the first-level conventional data information are normalized, and the normalization formula is:

[0082] Normalization

[0083] Among them, YLi represents the current data packet length, YLi max Indicates the maximum length of the data packet, YLi min Indicates the minimum value of the data packet length;

[0084] Normalization

[0085] Among them, YDi represents the maximum transmission unit of the current data packet, YDi max Indicates the maximum value of the maximum transmission unit of the data packet, YDi min Indicates the minimum value of the maximum transmission unit of a data packet;

[0086] Normalization

[0087] Among them, YKi represents the current data block size, YKi max Indicates the maximum size of the data block, YKi min Indicates the minimum value of the data block size;

[0088] Different weight coefficients are assigned to the packet length, maximum transmission unit of the packet and data block size in the normalized first-level conventional data information. The weight of the packet length is preset to be w L , the weight of the maximum transmission unit of a data packet is w D , the weight of the data block size is w K , the normalized first-level conventional data information is YJi, and the calculation formula is:

[0089] YJi=w l ×Normalized YLi+w D×Normalized YDi+w K ×Normalized YKi;

[0090] In the formula, w L +w D +w K =1, each weight coefficient is adjusted according to actual needs;

[0091] The first-level conventional sub-label information includes the data packet source IP address, data packet source port number, data packet transmission rate YCi and data packet transmission delay YSi of the web page data stream, text data stream and audio and video data stream transmitted per unit time;

[0092] The desensitization risk analysis module is used to perform a first-level desensitization risk analysis on the normalized first-level conventional data information and first-level conventional sub-label information of the web page data stream, text data stream and audio and video data stream transmitted in a unit time to generate a first-level desensitization risk factor;

[0093] The desensitization risk analysis module is also used to perform a secondary desensitization risk analysis on the secondary conventional data information and the secondary conventional sub-label information of the database access data stream and the sensor data stream transmitted in a unit time to generate a secondary desensitization risk factor;

[0094] The secondary conventional data information includes user accounts accessed by the database access data stream and sensor data stream transmitted per unit time through the network, and devices transmitted by the database access data stream and sensor data stream transmitted per unit time;

[0095] The secondary conventional sub-label information includes the packet loss rate and throughput of the database access data stream and the sensor data stream transmitted per unit time.

[0096] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages:

[0097] Normalization processing can eliminate the dimensional differences between different parameters, so that parameters of different magnitudes can be compared and analyzed on the same scale, which is conducive to improving the accuracy and efficiency of data processing. The normalized first-level conventional data information can be comprehensively evaluated in the leakage prevention and supervision module based on the first-level desensitized risk factors and the second-level desensitized risk factors to more accurately judge the network security situation.

[0098] When the data packet value in the first-level general data information is too large or too small, misjudgment may occur when the desensitization risk analysis module performs desensitization risk analysis. The probability of misjudgment can be reduced by normalizing and weighting the data packets in the first-level general data information.

[0099] Example 2

[0100] In Example 1, the number of data packets per unit time is large, and the time for data collection, risk assessment and desensitization processing will be longer than the time required for data packet transmission, resulting in the calculated data packet being high risk, but the data packet has been transmitted and no desensitization processing has been performed.

[0101] Therefore, the embodiments of the present application are optimized based on the above embodiments.

[0102] In some embodiments, Figure 1 As shown, the feature extraction and recognition module collects in real time the data packet length YLi, the maximum data packet transmission unit YDi, the data block size YKi, the data packet transmission rate YCi and the data packet transmission delay YSi of the web page data stream, text data stream and audio and video data stream transmitted per unit time, and forms a pentagonal radar chart by simulating and calculating the five groups of data to obtain the desensitization risk coefficient and risk level;

[0103] Five sets of data should be extracted from one data package at the same time to ensure the synchronization of data collection.

[0104] The five sets of collected data are preprocessed, which specifically includes the following steps:

[0105] Remove outliers, missing values, and duplicate data from packet length, packet maximum transmission unit, data block size, packet transmission rate, and packet transmission delay.

[0106] Construct a pentagonal radar chart and calculate the area of ​​the pentagon, which includes the following steps:

[0107] S101: construct a pentagonal radar chart, which includes five data axes, and maps five sets of data, namely, preprocessed data packet length, maximum data packet transmission unit, data block size, data packet transmission rate, and data packet transmission delay, onto the five axes of the radar chart;

[0108] S102: Based on the specific values ​​in the five sets of data, draw points on the corresponding axes, and connect the five points on the five axes to form a pentagon;

[0109] S103: Calculate the area of ​​the connected pentagons using a geometric formula. For a convex pentagon, divide it into triangles, calculate the areas of the triangles respectively, and sum them up.

[0110] The relationship between the pentagon area and the desensitization risk value is calculated to judge the risk level, which specifically includes the following steps:

[0111] S104: establishing a mapping model between the pentagon area and the desensitization risk coefficient through historical data of data packet length, maximum data packet transmission unit, data block size, data packet transmission rate, and data packet transmission delay, and using a regression analysis method to train the model;

[0112] S105: According to the mapping model, different risk levels are set, and thresholds of the pentagonal area are set; for example, the risk levels are divided into three levels: high, medium, and low, and corresponding area thresholds are set;

[0113] S106: when collecting and processing five sets of data in real time, calculating the pentagonal area of ​​each data packet, and judging the risk level according to the set threshold value;

[0114] S107: Set a time window T, perform batch processing on the data packets in each time window, and at the end of the window, calculate the average area of ​​all data packets in the current window to judge the risk level of the current window.

[0115] In step S104, Figure 2 As shown, the mapping model between the pentagon area and the desensitization risk coefficient is established, and the regression analysis method is used to train the model. The specific method is:

[0116] S201: Take the pentagon area as the independent variable X and the desensitization risk coefficient as the dependent variable Y;

[0117] S202: setting the corresponding relationship between the area of ​​a part of the pentagon obtained through the pentagon radar chart and the desensitization risk coefficient, specifically, when the pentagon area is 100, the desensitization risk coefficient is 0.2; when the pentagon area is 150, the desensitization risk coefficient is 0.4; when the pentagon area is 200, the desensitization risk coefficient is 0.6; when the pentagon area is 250, the desensitization risk coefficient is 0.8; when the pentagon area is 300, the desensitization risk coefficient is 1.0;

[0118] S203: Construct a linear regression model, and the calculation formula is:

[0119] Y=β0+β1X+∈;

[0120] Among them, Y represents the desensitization risk coefficient, β0 represents the intercept, β1 represents the slope, X represents the area of ​​the pentagon, and ∈ represents the error term, which is the difference between the model prediction value and the actual observation value;

[0121] S204: Use the least squares formula to estimate the regression coefficients β0 and β1 by minimizing the residual sum of squares. The calculation formula is:

[0122]

[0123] Among them, RSS means minimizing the residual sum of squares, y i represents the actual observed desensitization risk coefficient value, x i represents the actual observed pentagon area value, n represents the number of observations, β0 represents the intercept, and β1 represents the slope;

[0124] S205: Solve the equation group to obtain the values ​​of β0 and β1. The calculation formula is:

[0125]

[0126] In the formula, represents the mean value of the desensitization risk coefficient actually observed, represents the actual observed mean area of ​​the pentagon, y i represents the actual observed desensitization risk coefficient value, x i represents the actual observed pentagon area value, n represents the number of observations, β0 represents the intercept, and β1 represents the slope;

[0127] S206: Calculate the above regression coefficient using statistical software to obtain the regression coefficient result. The regression equation is:

[0128] Y = -0.01 + 0.004 × X;

[0129] Among them, Y represents the desensitization risk coefficient and X represents the area of ​​the pentagon.

[0130] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages:

[0131] Five sets of data are extracted from one data packet at the same time, ensuring the synchronization of data collection, which helps to more accurately reflect the actual status of the data packet during transmission and provide a more reliable data basis for risk assessment.

[0132] By simulating and calculating five sets of data to form a pentagonal radar chart, and calculating the area of ​​the pentagon to judge the desensitization risk coefficient and risk level, it comprehensively considers the influencing factors of multiple dimensions, which is more comprehensive and accurate than the evaluation of a single indicator. By using historical data to establish a mapping model between the pentagon area and the desensitization risk coefficient, and training through regression analysis, the risk assessment model is more scientific and reasonable, and can more accurately predict and judge the risk level of the data packet.

[0133] By evaluating the risk level of data packets in real time and taking timely desensitizing measures when high-risk data packets are discovered, the risk of data leakage can be effectively reduced and user privacy and data security can be protected.

[0134] Example 3

[0135] In the first and second embodiments, when a large number of data packets need to be transmitted at the same time, the amount of calculation is huge, and the data packets need to be divided.

[0136] Therefore, the embodiments of the present application are optimized based on the above embodiments.

[0137] In some embodiments, the first-level conventional data information collected by the feature extraction and recognition module includes the data packet length, maximum data packet transmission unit, data block size, data packet transmission rate and data packet transmission delay of the web data stream, text data stream and audio and video data stream transmitted per unit time. By dividing the above data packet information according to time windows and calculating the pentagon area in a single time window, the interval risk value is obtained, which specifically includes the following steps:

[0138] 301: Set the time window T1, count the number of all data packets transmitted in a single time window, and record the size of each data packet;

[0139] The time window is set according to the specific situation and can be set to 10 minutes;

[0140] S302: according to the size of the data packet, a series of size intervals are set, such as 0-100KB, 100-500KB, 500-1MB, etc., and each data packet is allocated to a corresponding interval according to the size of the data packet;

[0141] If the number of packets in a certain interval exceeds the preset limit, the interval is further subdivided. If the number of packets in a certain interval is insufficient, they are extracted from the remaining time windows;

[0142] S303: Calculate the sum of the pentagonal areas of the data packets in each time window, and calculate the interval volume according to the sum of the pentagonal areas and the interval length, where the interval volume is the product of the sum of the pentagonal areas and the interval length, and set a risk threshold for each interval;

[0143] 304: For data packets in the interval exceeding the risk threshold, a separate desensitization risk assessment is performed using a linear regression model;

[0144] S305: Desensitizing the high-risk data packets according to the evaluation results.

[0145] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages:

[0146] By defining a reasonable time window to extract data packets, the amount of data processed at a single time can be effectively limited, avoiding the huge computing pressure caused by processing the entire network traffic at one time. This not only reduces the complexity and time consumption of a single processing, but also enables a more reasonable allocation and utilization of system resources. By partitioning data packets according to size, batch processing can be performed on data packets in each interval, rather than processing each data packet one by one, which can significantly reduce processing time and improve overall processing efficiency.

[0147] By partitioning the data packets, the system can dynamically adjust computing resources according to the data volume and risk level of different intervals. For intervals with higher risks, more resources can be allocated for in-depth analysis and processing; while for intervals with lower risks, resource input can be reduced to achieve optimal resource allocation; based on the data packet size partitioning, for intervals with lower risks or smaller data volumes, simpler processing methods can be adopted or processing frequency can be reduced to avoid unnecessary computing overhead.

[0148] Example 4

[0149] In the above embodiment, for the processing of a large number of data packets per unit time, the setting of the number of data packets within the interval can be completed in multiple times.

[0150] Therefore, the embodiments of the present application are optimized based on the above embodiments.

[0151] In some embodiments, dynamically adjusting the number of data packets collected by the feature extraction and identification module and issuing medium- and high-risk warnings specifically include the following steps:

[0152] S401: dividing the data packets according to step S302 so that each interval contains a maximum number of data packets;

[0153] S402: Perform risk assessment on each interval, calculate the interval risk value, set the risk warning threshold, and when the risk value of a certain interval exceeds the warning threshold, mark the interval as a medium-high risk interval;

[0154] S403: When a certain interval is marked as medium or high risk, the data packets in the interval are further grouped, and the data packets in the interval are divided into multiple sub-intervals according to specific attributes of the data packets, such as source IP address, destination port, etc., and the risk of each sub-interval is re-evaluated;

[0155] S404: If there are still medium- and high-risk areas in the sub-interval, continue to group the sub-interval into more fine-grained groups and repeat the risk assessment process;

[0156] Continue iterating the grouping until a sufficiently small risk area or single data packet is found;

[0157] S405: within the final risk sub-interval, perform detailed evaluation and calculation on each data packet to determine specific medium and high risk data packets;

[0158] Desensitize the identified medium and high risk data packets.

[0159] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages:

[0160] By dividing data packets into intervals and placing the maximum number of data packets in each interval, computing resources can be effectively utilized, processing delays caused by excessive number of data packets can be reduced, the number of data packets can be dynamically adjusted, and the interval size can be optimized based on real-time risk assessment results, so that data packet traffic can be managed more efficiently and system overload can be avoided.

[0161] Set risk warning thresholds, conduct risk assessments on intervals, and mark them as medium- and high-risk intervals when the risk value exceeds the threshold. This warning mechanism can detect potential security threats in a timely manner and provide strong support for subsequent processing measures. Further, by dividing sub-intervals and repeating risk assessments, the risk scope can be gradually narrowed, and specific medium- and high-risk data packets can be located. The precise positioning method can improve the accuracy and efficiency of risk processing.

[0162] Desensitizing identified medium- and high-risk data packets can effectively reduce the risk of data leakage and protect user privacy and data security. Desensitization strategies based on risk assessment are more efficient and accurate than traditional full-volume desensitization.

[0163] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0164] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0165] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0166] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0167] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0168] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0169] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A network security systematized defense and supervision platform in the key infrastructure field, characterized in that: include: Data stream classification module, feature extraction and recognition module, desensitization risk analysis module; The data stream classification module is used to classify different types of data streams transmitted from one node to another into web page data stream, text data stream, audio and video data stream, database access data stream and sensor data stream; The feature extraction and recognition module is used to collect the first-level conventional data information and first-level conventional sub-label information of the web page data stream, text data stream and audio and video data stream transmitted in a unit time, and send it to the desensitization risk analysis module; It is also used to collect the secondary conventional data information and secondary conventional sub-label information of the database access data stream and sensor data stream transmitted in a unit time, and send it to the desensitization risk analysis module; The first-level conventional data information includes the data packet length YLi, the maximum data packet transmission unit YDi and the data block size YKi of the web page data stream, text data stream and audio and video data stream transmitted per unit time; The data packet length, the maximum transmission unit of the data packet and the data block size in the first-level conventional data information are normalized, and the normalization formula is: Normalization Among them, YLi represents the current data packet length, YLi max Indicates the maximum length of the data packet, YLi min Indicates the minimum value of the data packet length; Normalization Among them, YDi represents the maximum transmission unit of the current data packet, YDi max Indicates the maximum value of the maximum transmission unit of the data packet, YDi min Indicates the minimum value of the maximum transmission unit of a data packet; Normalization Among them, YKi represents the current data block size, YKi max Indicates the maximum size of the data block, YKi min Indicates the minimum value of the data block size; Different weight coefficients are assigned to the packet length, maximum transmission unit of the packet and data block size in the normalized first-level conventional data information. The weight of the packet length is preset to be w L , the weight of the maximum transmission unit of a data packet is w D , the weight of the data block size is w K , the normalized first-level conventional data information is YJi, and the calculation formula is: YJi=w l ×Normalized YLi+w D ×Normalized YDi+w K ×Normalized YKi; In the formula, w L +w D +w K =1, and each weight coefficient is adjusted according to actual needs.

2. According to claim 1, a network security systematized defense and supervision platform in the key infrastructure field is characterized in that: The desensitization risk analysis module is used to perform a first-level desensitization risk analysis on the normalized first-level conventional data information and first-level conventional sub-label information of the web page data stream, text data stream and audio and video data stream transmitted within a unit time to generate a first-level desensitization risk factor; The desensitizing risk analysis module is also used to perform secondary desensitizing risk analysis on the secondary conventional data information and secondary conventional sub-label information of the database access data stream and sensor data stream transmitted within a unit time to generate secondary desensitizing risk factors.

3. According to claim 1, a network security systematized defense and supervision platform in the key infrastructure field is characterized in that: The feature extraction and identification module collects in real time the data packet length YLi, the maximum data packet transmission unit YDi, the data block size YKi, the data packet transmission rate YCi and the data packet transmission delay YSi of the data packets in the web page data stream, text data stream and audio and video data stream transmitted per unit time, and forms a pentagonal radar chart by simulating and calculating the five sets of data to obtain the desensitization risk coefficient and risk level.

4. According to claim 3, a network security systematized defense and supervision platform in the key infrastructure field is characterized in that: The five sets of collected data are preprocessed, which specifically includes the following steps: Remove outliers, missing values, and duplicate data from the five sets of data.

5. According to claim 4, a network security systematized defense and supervision platform in the key infrastructure field is characterized in that: Construct a pentagonal radar chart and calculate the area of ​​the pentagon, which includes the following steps: S101: construct a pentagonal radar chart, which contains five data axes, and map the five sets of preprocessed data onto the five axes of the radar chart; S102: Based on the specific values ​​in the five sets of data, draw points on the corresponding axes, and connect the five points on the five axes to form a pentagon; S103: Calculate the area of ​​the connected pentagons using a geometric formula. For a convex pentagon, divide it into triangles, calculate the areas of the triangles respectively, and sum them up.

6. According to claim 5, a network security systematized defense and supervision platform in the key infrastructure field is characterized in that: The relationship between the pentagon area and the desensitization risk value is calculated to judge the risk level, which specifically includes the following steps: S104: Establish a mapping model between the pentagon area and the desensitization risk coefficient through the historical data of the five groups of data, and use the regression analysis method to train the model; S105: setting different risk levels according to the mapping model, and setting a threshold value of the pentagon area; S106: when collecting and processing five sets of data in real time, calculating the pentagonal area of ​​each data packet, and judging the risk level according to the set threshold value; S107: Set a time window T, perform batch processing on the data packets in each time window, and at the end of the window, calculate the average area of ​​all data packets in the current window to judge the risk level of the current window.

7. According to claim 6, a network security systematized defense and supervision platform in the key infrastructure field is characterized in that: The mapping model between the pentagon area and the desensitization risk coefficient is established, and a regression analysis method is used to train the model. The specific method is: S201: Take the pentagon area as the independent variable X and the desensitization risk coefficient as the dependent variable Y; S202: Setting the corresponding relationship between the area of ​​a part of the pentagon obtained through the pentagon radar chart and the desensitization risk coefficient; S203: Construct a linear regression model, and the calculation formula is: Y=β0+β1X+∈; Among them, Y represents the desensitization risk coefficient, β0 represents the intercept, β1 represents the slope, X represents the area of ​​the pentagon, and ∈ represents the error term, which is the difference between the model prediction value and the actual observation value; S204: Use the least squares formula to estimate the regression coefficients β0 and β1 by minimizing the residual sum of squares. The calculation formula is: Among them, RSS means minimizing the residual sum of squares, y i represents the actual observed desensitization risk coefficient value, x i represents the actual observed pentagon area value, n represents the number of observations, β0 represents the intercept, and β1 represents the slope S205: Solve the equation group to obtain the values ​​of β0 and β1. The calculation formula is: In the formula, represents the mean value of the desensitization risk coefficient actually observed, represents the actual observed mean area of ​​the pentagon, y i represents the actual observed desensitization risk coefficient value, x i represents the actual observed pentagon area value, n represents the number of observations, β0 represents the intercept, and β1 represents the slope; S206: Calculate the above regression coefficient using statistical software to obtain the regression coefficient result. The regression equation is: Y = -0.01 + 0.004 × X; Among them, Y represents the desensitization risk coefficient and X represents the area of ​​the pentagon.

8. According to claim 7, a network security systematized defense and supervision platform in the key infrastructure field is characterized in that: The feature extraction and recognition module collects the first-level general data information, which includes five groups of data of the data packet. The interval risk value is obtained by dividing the above data packet according to the time window and calculating the pentagon area in a single time window.

9. A network security systematized defense and supervision platform in the key infrastructure field according to claim 8, characterized in that: The method of dividing the above data packets according to time windows and calculating the pentagonal area in a single time window to obtain the interval risk value specifically includes the following steps: 301: Set the time window T1, count the number of all data packets transmitted in a single time window, and record the size of each data packet; S302: according to the size of the data packet, set a series of size intervals, and allocate each data packet to a corresponding interval according to the size of the data packet; If the number of packets in a certain interval exceeds the preset limit, the interval is further subdivided. If the number of packets in a certain interval is insufficient, they are extracted from the remaining time windows; S303: Calculate the sum of the pentagonal areas of the data packets in each time window, and calculate the interval volume according to the sum of the pentagonal areas and the interval length, where the interval volume is the product of the sum of the pentagonal areas and the interval length, and set a risk threshold for each interval; 304: For data packets in the interval exceeding the risk threshold, a separate desensitization risk assessment is performed using a linear regression model; S305: Desensitizing the high-risk data packets according to the evaluation results.

10. A network security systematized defense and supervision platform in the key infrastructure field according to claim 9, characterized in that: Dynamically adjusting the number of data packets collected by the feature extraction and recognition module and issuing medium- and high-risk warnings specifically includes the following steps: S401: dividing the data packets according to step S302 so that each interval contains a maximum number of data packets; S402: Perform risk assessment on each interval, calculate the interval risk value, set the risk warning threshold, and when the risk value of a certain interval exceeds the warning threshold, mark the interval as a medium-high risk interval; S403: When a certain interval is marked as medium or high risk, further group the data packets in the interval, divide the data packets in the interval into multiple sub-intervals according to the specific attributes of the data packets, and re-evaluate the risk of each sub-interval; S404: If there are still medium- and high-risk areas in the sub-interval, continue to group the sub-interval in fine-grained manner and repeat the risk assessment process; Continue iterating the grouping until the smallest risk area or single data packet is found; S405: within the final risk sub-interval, each data packet is evaluated and calculated individually to determine medium and high risk data packets; Desensitize the identified medium and high risk data packets.

Citation Information

Patent Citations

  • Data center network equipment intelligent operation and maintenance system, method and terminal

    CN112187514A

  • Network security operation method based on security policy

    CN117081868A

  • Network security supervision system for preventing network data leakage

    CN117201206A