Beidou map information security monitoring system and monitoring method

By introducing isolation layers and information transit stations into cloud servers, combining identity authentication and behavioral feature analysis, the problem of difficult management of surveying and mapping data information in cloud servers is solved, and dynamic monitoring of information security and user experience is achieved.

CN118779848BActive Publication Date: 2024-12-03CHENGDU XIWU SECURITY SYST ALLIANCE
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411252546.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-09
Publication Date
2024-12-03
Estimated Expiration
2044-09-09

AI Technical Summary

Technical Problem

The information security of surveying and mapping data in existing cloud servers is difficult to dynamically manage, resulting in identity verification information leakage and illegal access, affecting information security, and frequent identity verification increases data access delay and affecting user experience.

Method used

Using the technology of combining isolation layer and information transit station, the user's access request is verified and recorded through the identity verification module, login information recording module and analysis node, the user's access behavior characteristics are generated, abnormal access is judged, and access is monitored through the neural network model. Act, timely discover illegal access and forged identity verification information.

Benefits of technology

It improves information security, reduces identity verification delay, improves system efficiency, promptly detects abnormal access behaviors, reduces the risk of information leakage, and enhances user experience.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118779848B_ABST
    Figure CN118779848B_ABST
Patent Text Reader

Abstract

The present application discloses a Beidou map information security monitoring system and monitoring method, which belongs to the field of information security technology. A Beidou map information security monitoring system and monitoring method, including: a cloud server, used to store surveying and mapping data of various regions; a terminal device, through which a user generates an access request, the access request includes user identity information and access requirements; an isolation layer, used to receive access requests, and verify the identity information in the access request, and send the verified access request to an information transfer station; the information transfer station obtains the user's access rights based on the user's identity information, and determines whether the access requirement exceeds the authority based on the access rights. In the technical solution provided by the present application, two nodes, an isolation layer and an information transfer station, are used to verify the access request respectively, which can prevent illegal access requests from bypassing one of the nodes and directly accessing the data in the server when passing a certain verification.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of information security monitoring technology, and more specifically, to a Beidou map information security monitoring system and monitoring method. Background Art

[0002] Beidou map refers to a map developed based on Beidou satellite positioning data. Because Beidou satellites have high-precision positioning capabilities, they can provide users with good positioning services. When developing Beidou satellite maps, the most critical data is the map's surveying and mapping data. When drawing maps, a large amount of surveying and mapping data is required. In order to facilitate the staff in various regions to obtain surveying and mapping data, various types of surveying and mapping data will be uploaded to the cloud server for use by staff in various regions.

[0003] At present, the surveying and mapping data in the cloud server are generally protected by identity authentication and permission management. Among them, identity authentication is to control the identity information of the people who access the cloud server, and only the visitors who pass the identity authentication can access the content in the cloud server. And permission management is to provide different access rights to different people, and determine the access scope of the visitor according to the access rights.

[0004] The above two solutions lack a dynamic management mechanism. When the identity authentication information of some staff members is leaked, it is difficult to detect in time when illegal persons use the leaked identity authentication information to access the cloud server, which in turn leads to the leakage of surveying and mapping data and affects information security. Summary of the invention

[0005] The content of this application is used to introduce concepts in a brief form, which will be described in detail in the detailed implementation section below. The content of this application is not intended to identify the key features or essential features of the technical solution claimed for protection, nor is it intended to limit the scope of the technical solution claimed for protection.

[0006] As a first aspect of the present application, in order to solve the technical problem of information security, the present application provides a Beidou map information security monitoring system, including:

[0007] Cloud servers are used to store surveying and mapping data of various regions;

[0008] Terminal device, through which the user generates an access request, the access request includes the user's identity information and access requirements;

[0009] The isolation layer is used to receive access requests, verify the identity information in the access requests, and send the verified access requests to the information transfer station;

[0010] The information transfer station obtains the user's access rights based on the user's identity information, and determines whether the access requirements exceed the authority based on the access rights;

[0011] When the access requirements do not exceed the authority, download the corresponding surveying and mapping data from the cloud server according to the access requirements, or upload new surveying and mapping data;

[0012] When the access requirement exceeds the authority, the access request is terminated;

[0013] The analysis node is connected to the isolation layer signal to obtain the access requests of each user on the previous day, and generates the access behavior characteristics of each user based on the access requests of each user on the previous day;

[0014] It is determined whether the current access request is an abnormal access request according to the access behavior characteristics of each user. When an abnormal access request exists, an instruction to terminate the access request is sent to the information transfer station.

[0015] In the technical solution provided by the present application, two nodes, an isolation layer and an information transfer station, are used to verify the access request respectively, which can prevent illegal access requests from bypassing one of the nodes and directly accessing the data in the cloud server when passing a certain verification. Then, in order to solve the problem of illegal users obtaining the identity authentication information of real users or forging identity authentication information, the present application also provides an analysis node, which will record the access habits of each user, compare the access behavior characteristics of each user with the characteristics of the current access request, and use this as a basis to determine whether it is the access behavior of the real user, so as to timely discover the illegal access behavior caused by illegal users obtaining the identity authentication information of real users or forging identity authentication information, thereby increasing the security of information.

[0016] In practice, the data categories accessed by users are different each time. It is difficult to find the user's access pattern directly based on the data categories accessed by users because the data categories are diverse, so it is difficult to reflect the actual characteristics of the data categories accessed by users, and thus it is impossible to accurately find the habits of each user when accessing data, and thus it is impossible to accurately determine whether there is an abnormal login. To address this problem, this application provides the following technical solutions:

[0017] Furthermore, the surveying and mapping data are divided into M data categories according to regions and data types, where M = a*b, a is the number of regions, and b is the number of data types.

[0018] In the technical solution provided in the present application, the surveying and mapping data are divided according to regions and types, so it is possible to well describe the user's habit of accessing data and accurately determine whether there is any abnormal login.

[0019] Furthermore, the access requirements include access time, access address, whether to upload data, whether to download data, and the category of data accessed.

[0020] Users need to access the cloud server multiple times during work, and each access to the cloud server requires identity verification, which will increase data access latency and affect user experience. To address this problem, this application provides the following technical solutions:

[0021] Isolation layer, including:

[0022] Authentication module, used to verify the user's identity information;

[0023] The login information recording module is used to record the identity information of each user who has completed the verification;

[0024] Login information update module, used to update each user's verification record;

[0025] The information forwarding module sends the access request of the user recorded in the login information recording module to the information transfer station;

[0026] The login information update module deletes from the login information recording module a user who has not sent an access request within a predetermined threshold, and does not delete from the login information recording module a user who has sent an access request at an interval less than the predetermined threshold;

[0027] The login information update module deletes users whose access addresses have changed from the login information recording module; users whose access addresses have not changed are not deleted from the login information recording module;

[0028] The login information updating module deletes the users whose login time exceeds the preset time from the login information recording module, and does not delete the users whose login time does not exceed the preset time from the login information recording module.

[0029] In the technical solution provided by the present application, when the user sends access requests continuously, it is only necessary to verify the identity information once, so the verification time of subsequent access requests is reduced, the data access delay is reduced, the system operation efficiency is improved, and the user experience is increased. In addition, the present application adopts this solution, and can also better determine the access habits of the current actual visitor. In practice, according to the time and address information, the actual visitor who sends the access request can be determined, and then the access behavior characteristics of the actual visitor can be obtained. If the access behavior characteristics of the actual visitor do not match the access behavior characteristics of the previous day, it can be indicated that the current actual visitor is most likely not the owner of the identity authentication information and is an illegal user. In this way, in this solution, the isolation layer designed not only reduces the user's verification process, but also allows a sample data to contain multiple access requests to accurately describe the user's access habits while ensuring that these access requests are necessarily generated by the same actual visitor, thereby accurately describing the access behavior characteristics of the actual visitor.

[0030] If the information stored in the cloud server can be directly accessed after verification, criminals will forge identity information and directly download the required data from the cloud server to bypass identity supervision, which will lead to information leakage.

[0031] Furthermore, the information transfer station includes:

[0032] User permission recording module, used to record the access rights of all users;

[0033] An information receiving module, used for receiving an access request sent from the isolation layer;

[0034] The information transfer module verifies the access request according to the access rights recorded in the user authority recording module. If the access request passes the verification, the corresponding mapping data is downloaded / uploaded from the cloud server according to the access request, and then forwarded to the isolation layer, which sends it to the corresponding terminal device.

[0035] In the technical solution provided by the present application, no terminal can directly access the cloud server. Only the information transfer station has the authority to download corresponding data from the cloud server or upload corresponding data to the cloud server. This avoids the problem of information leakage caused by directly obtaining data from the cloud server by forging identity information.

[0036] Furthermore, the analysis nodes include:

[0037] The information collection module is connected to the isolation layer signal to obtain all access requests that have passed the identity information verification, and the collected access requests are made into a training data set;

[0038] The monitoring module has a built-in neural network model. The monitoring module obtains the access behavior characteristics of each user based on the training data set;

[0039] The information acquisition module is connected to the isolation layer signal, obtains the access request that has passed the identity authentication from the isolation layer, and then sends the access request to the monitoring module;

[0040] When the difference between the access request and the user's access behavior characteristics exceeds a preset threshold, an alarm message is issued;

[0041] The intervention module obtains the user who generates the alarm information and then sends an instruction to terminate the access request to the information transfer station.

[0042] In the technical solution provided in the present application, a neural network model is used to extract the user's access behavior characteristics, which can promptly detect the user's abnormal access behavior, and then promptly issue an alarm when abnormal access behavior occurs, close the corresponding access window to reduce the amount of information loss, and promptly detect information leakage to increase information security.

[0043] As a second aspect of the present application, the present application provides a Beidou map information security monitoring method, which uses the aforementioned Beidou map information security monitoring system to monitor the calling of surveying and mapping data.

[0044] The beneficial effect of this application is that: on the basis of setting permission levels and identity authentication, the content accessed by users is recorded according to the behavioral habits of each user, and then timely warnings are issued when users abnormally access cloud server data to reduce the risk of information leakage. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The drawings constituting a part of this application are used to provide a further understanding of this application, so that other features, purposes and advantages of this application become more obvious. The illustrative embodiment drawings and their descriptions of this application are used to explain this application and do not constitute an improper limitation on this application.

[0046] In addition, throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that the components and elements are not necessarily drawn to scale.

[0047] In the attached picture:

[0048] Figure 1 Schematic diagram of Beidou map information security monitoring system. DETAILED DESCRIPTION

[0049] Embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms, and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are only for exemplary purposes and are not intended to limit the scope of protection of the present application.

[0050] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other.

[0051] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments.

[0052] Example 1: Reference Figure 1 The Beidou map information security monitoring system includes a cloud server, a terminal device, an isolation layer, an information transfer station, and an analysis node. The cloud server is connected to the information transfer station by signal, and the cloud server is used to store the surveying and mapping data of each area. The surveying and mapping data is divided into M data categories according to the area and the data type, where M=a*b, a is the number of areas, and b is the number of data types.

[0053] The area here mainly refers to the division of the physical area of ​​the map. For example, the area can be divided according to administrative divisions. Generally, the area is divided according to the scope of responsibility of each unit within the enterprise. Data type mainly refers to different types of data. In the same area, some surveying and mapping data are: "the distance from the corresponding position to the reference object", such as the width of the road surface. Some surveying and mapping data are: "the height difference from the corresponding position to the reference object", such as the undulation of the road surface. Based on the two dimensions of data type and area, M types of data categories of surveying and mapping data can be divided.

[0054] In this solution, the cloud server is only responsible for storing and updating information. When users need it, they can download the required surveying and mapping data from the cloud server.

[0055] Terminal devices are devices that users use to send requests to cloud servers to obtain relevant information from cloud servers. Terminal devices are mainly computers. In addition, they can also be electronic terminals that can connect to the Internet, such as mobile phones and tablets.

[0056] Before using a terminal device to generate an access request, the user needs to generate registration information. Only after successful registration can the data in the cloud server be obtained. The specific registration method and registration steps of the user will not be discussed here. This solution mainly manages users who have completed registration. The access rights of each user are related to the registration process, so the setting method of access rights will not be discussed further here. At the same time, this solution only discusses the situation where the access rights are fixed and do not change.

[0057] The isolation layer is used to receive access requests and verify the identity information in the access requests, and send the verified access requests to the information transfer station.

[0058] In this solution, the isolation layer is mainly used to confirm the user's identity information. The user's identity information generally includes the account number and password. The main content of the isolation layer verification is whether the entered password is correct.

[0059] In actual work, it is impossible for users to enter their identity information every time they access the surveying and mapping data in the cloud server. Therefore, the isolation layer also has a login function. After the terminal device is connected to the isolation layer, for the user, it is only necessary to verify the identity information once in a short period of time. The subsequent access requests can be directly verified by the isolation layer and then sent to the information transfer station.

[0060] Specifically: The isolation layer includes: an identity authentication module, a login information recording module, a login information update module, and an information forwarding module. Among them, the identity authentication module is connected to the terminal device signal, and the access request generated by the terminal device will be sent to the identity authentication module for identity information verification. If the identity information verification is passed, the identity authentication module will send the access request and the user corresponding to the access request to the login information recording module. The specific verification method of the identity authentication module is to check the account number and password in the identity information. The specific identity information verification method is not limited here.

[0061] The login information recording module is connected to the identity authentication module. The users recorded in the login information recording module are users who have passed the identity card verification. The access requests generated by such users will be directly sent to the information transfer station. Specifically, the login information recording module is used to record the identity information of each user who has completed the verification. For the same verified user, only the latest identity information is recorded.

[0062] For example, after user A sends an access request to the isolation layer, he immediately sends another access request to the isolation layer, regardless of whether the identity authentication module verifies the two access requests. When entering identity information into the login information recording module, only the most recently entered identity information of user A needs to be retained, and the identity information of the other user A will be automatically deleted. When saving the identity information of user A, the account number in the identity information is mainly used for identification, and the corresponding password will not be extracted from the identity information. In this way, subsequent access requests generated by users recorded in the login information recording module can be sent directly to the information transfer station.

[0063] In order to further increase information security, a login information update module is also set up, and the login information update module is connected to the login information recording module by signal. The login information update module is mainly used to delete the users recorded in the login information update module. The judgment logic during deletion is as follows: the login information update module deletes the users who have not sent access requests within a predetermined threshold from the login information recording module, and does not delete the users who have sent access requests with an interval less than the preset threshold from the login information recording module.

[0064] For example, if a user does not send a new access request within 5 minutes (predetermined threshold), the user's identity information will be deleted from the login information recording module. When the user needs to send an access request again, he needs to re-verify his identity information. If a user sends a new access request at least once within 5 minutes, the identity information will not be deleted from the login information recording module. Please note that this rule is only used to determine whether the user has sent access requests continuously. If the user has sent a new access request at least once within 5 minutes, but violates any of the following two regulations, the user's identity information will also be deleted from the login information recording module.

[0065] The login information update module will delete users whose access addresses have changed from the login information recording module. That is, the login information update module will conduct a preliminary verification of each access request. If the access address has changed, the user will be deleted from the login information recording module. For users whose access addresses have not changed, they will not be deleted from the login information recording module. It should be noted that this rule is only used to determine whether the user's access address has changed. It does not mean that if the access address has not changed, the user's identity information will not be deleted from the login information recording module. If the user violates any of the remaining two rules, the user's identity information will also be deleted from the login information recording module.

[0066] The login information updating module deletes the user who exceeds the preset time from the login information recording module, and does not delete the user who does not exceed the preset time from the login information recording module.

[0067] For example, regardless of whether the user continues to send access requests, after more than 10 minutes (predetermined time), the user's identity information will be deleted from the login information recording module. When the user needs to send access requests again, he or she needs to re-verify the identity information. Note that this rule is only used to determine whether the user's total access time exceeds the predetermined time. It does not mean that the user's identity information will not be deleted from the login information recording module as long as the total access time does not exceed. If the user violates any of the remaining two rules, the user's identity information will also be deleted from the login information recording module.

[0068] The information forwarding module is connected with the login information recording module by signal, and for the users recorded in the login information recording module, the access request thereof is directly sent to the information transfer station.

[0069] That is, after a user completes the login in the login information recording module, subsequent access requests do not need to undergo identity authentication and can be directly forwarded to the information transfer station.

[0070] The information transfer station includes: a user authority recording module, an information receiving module, and an information transfer module. Among them, the user authority recording module is used to record the access rights of all users. That is, in the user authority recording module, the scope of authority of each registered user will be recorded to determine the data categories that these users can access. The information receiving module is used to receive access requests sent from the isolation layer and then forward the access requests to the information transfer module.

[0071] The information transfer module verifies the access request according to the access rights recorded in the user authority recording module. If the access request passes the verification, the corresponding mapping data is downloaded / uploaded from the cloud server according to the access request, and then forwarded to the isolation layer, which sends it to the corresponding terminal device.

[0072] The process of verifying the access request here is to determine whether the access request is within the scope of authority. After the verification is passed, the information transfer module will perform the corresponding operation according to the access request. For example, if you need to download three types of surveying and mapping data from the cloud server, the information transfer module will send the download request of the corresponding surveying and mapping data to the cloud server, and after the download is completed, it will be forwarded to the isolation layer, and then forwarded to the terminal device by the isolation layer. Correspondingly, if the access request is to upload three data to the cloud server, the information transfer module will upload the corresponding surveying and mapping data to the cloud server.

[0073] It should be noted that the information transfer module here is not an intermediate medium for information transmission. Instead, it intercepts the access request and then sends the access request to the cloud server according to the time sequence. Therefore, the cloud server will only connect to the information transfer module signal, and illegal users cannot access the cloud server through other ports, thereby reducing the risk of cloud server information leakage.

[0074] The above two mechanisms can only meet the daily protection work. In practice, there will be leakage of user identity information, or obtaining data in the cloud server by forging identity information. To address this problem, this application provides the following technical solutions:

[0075] The analysis node includes an information collection module, a monitoring module, an information acquisition module, and an intervention module. The information collection module is connected to the isolation layer signal, obtains all access requests that have passed identity information verification, and makes the collected access requests into a training data set. That is, the information collection module is connected to the isolation layer signal, and all access requests that have passed identity information verification in the isolation layer will be collected by the information collection module and then used as training data.

[0076] The training data includes several training samples; each training sample includes all the access requirements of the user when the record in the login information recording module is not updated. For example, when user A sends an access request to the isolation layer, after the isolation layer successfully verifies, the identity information of user A will be recorded in the login information recording module, and the four subsequent access requests sent by user A do not need to be verified. Then when user A sends the sixth access request, because the access address changes, the sixth access request of user A needs to be verified. At this time, the five access requests sent by user A before are all the access requirements when the record in the login information recording module is not updated, and then a training sample is made with the five access requests sent by user A before.

[0077] Each user has several training samples, which contain the user's access habits. When the subsequent monitoring module monitors the access request, it obtains the characteristics of each user's current access request, and then compares it with the access behavior characteristics obtained by the training data to determine whether the user's current access request is reasonable, so as to further monitor the information in the cloud server. Therefore, each user needs to obtain the user's access behavior characteristics based on the user's previous access request.

[0078] In practice, this application uses the following scheme to divide the user's access requests. When the user is in normal use, all the access requests generated by the user on the previous day are used as training data, and the neural network model in the monitoring module is used to obtain the user's access behavior characteristics. Then, all the user's access requests on that day will be input into the monitoring module, and the access behavior will be feature extracted. The extracted features will be compared with the access behavior features, and an alarm will be issued if the threshold is exceeded.

[0079] In the above scheme, the content of each training sample and the specific warning method have been clearly mentioned. However, how to extract features and access behavior features is not further described. The following is a description of the feature extraction method:

[0080] If the training sample is directly used for feature extraction, it will be difficult to discover the user's access habits and highlight the user's access range characteristics. To address this problem, this application provides the following technical solutions:

[0081] The access requirements are converted into matrix data related to the data category division method. The horizontal columns in the matrix data represent the areas, and the vertical rows in the matrix data represent the data types. The value of each element in the matrix data is related to the order of access requirements. The data categories that have not been accessed are recorded as 0 in the matrix data. For the accessed data categories, the values ​​corresponding to the elements that are accessed earlier are smaller.

[0082] In the above records, it has been clarified that a training sample is when the user's record in the login information recording module is not updated. Therefore, a training sample will include all the access requirements of the user when the user is logged in to the login information recording module. The access requirements can be converted into data categories that need to be processed.

[0083] For example, a training sample includes 5 access requirements, and corresponding operations are performed for the corresponding 5 data categories. Therefore, it can be converted into the following matrix data.

[0084] First, establish a plane rectangular coordinate system and only keep the first quadrant of the plane rectangular coordinate system. Set the horizontal coordinate of the first quadrant of the plane rectangular coordinate system to the sequence number of the region. For example, if there are 10 regions in total, the origin of the horizontal coordinate represents the first region, the first point of the horizontal coordinate represents the second region, and so on. The range of the horizontal coordinate is 0~9;

[0085] For the first quadrant, the ordinate is set to the serial number of the data type. For example, if there are 10 data types, the origin of the ordinate represents the first data type, the first point of the ordinate represents the second data type, and so on. The range of the ordinate is also 0 to 9. Because the number of data types and the number of regions are both limited, a plane region of a*b can be obtained.

[0086] Then, the data categories accessed by the five access requirements are found in the first quadrant. For the data categories that are not accessed, the element values ​​in the first quadrant are recorded as 0, and for the data categories that are accessed, they are recorded as 1 to 5 according to the order of access requirements. In this way, a matrix data can be obtained, which can actually be understood as a grayscale image on the plane.

[0087] Therefore, in this plane area, if the access requirement does not access the corresponding data category, the element of this data category in the plane area is set to zero. For the data category that the access requirement needs to access, the earlier the element is accessed, the smaller the value corresponding to it. In fact, the position of the data category is set to the corresponding value according to the order of the access requirements. If an access requirement includes multiple data categories, the element values ​​corresponding to the data categories that need to be accessed are the same, and they are all set based on the order of the access requirements.

[0088] In this way, each training sample can be organized into an a*b matrix data. In this matrix data, the number of non-zero elements is the number of access requirements in the training sample. The range of non-zero elements is the range that needs to be accessed, and the change in the value of the element in the non-zero element is the order of access.

[0089] In the solution provided by this application, there is no need to consider that the same data category will be selected twice in a training sample. Because, in this solution, the content from the cloud server in the cloud is sent to the information transfer station, then sent to the isolation layer, and then sent to the terminal device by the isolation layer. Therefore, when the user has logged in (for a short time), the user rarely needs the same data category twice. If it exists, it will be sent directly to the user by the isolation layer based on historical information. Therefore, in this solution, there will be no overlap of two access requirements in the matrix data.

[0090] In this way, after the training samples of the present application are converted into matrix data, all training samples can be compared at the same scale, and the data that was originally difficult to present specific rules can be converted into a picture format, so that the user's range characteristics can be reflected according to the aggregation of elements in the picture.

[0091] In the above scheme, it has been made clear that users will be deleted from the login information recording module after a period of time, so the time length corresponding to each training sample is basically the same. When users are working normally, they must focus on data related to their work. When information is leaked, illegal users will download data frequently and irregularly in a short period of time, so this can be used to judge whether the user's access request to the cloud server is reasonable.

[0092] In this way, the information collection module can convert all users' access requests into training samples. For each user, since they need to access the data in the cloud server every day, each user has multiple training samples, and these training samples of users indicate the access habits of each user.

[0093] Therefore, the monitoring module only needs to input the training samples of the corresponding user into the neural network model to obtain the access behavior characteristics of the user.

[0094] Specifically, the monitoring module has a built-in neural network model, which is an LSTM neural network model.

[0095] In practice, it is difficult to directly extract matrix data from the LSTM neural network model. Therefore, in this solution, the following solution is also used to extract the matrix data corresponding to each training sample. Specifically:

[0096] S1: Collect all training samples of the user, convert all training samples into matrix data, and obtain Y1, Y2, ...Y i …、Y m ; Wherein, Y1 is the first matrix data of the user, i is the index of the matrix data of the user, and m represents the total number of matrix data of the user;

[0097] Preset the bias term k;

[0098] S2: For each matrix data Y i , calculate the average value A of all its pixels i and standard deviation B i ;

[0099] ;

[0100] Among them, p s represents the element value of the sth element in the matrix data, M represents the total number of elements in the matrix data, and s represents the index of the element in the matrix data;

[0101] ;

[0102] S3: According to each matrix data Y i The average value A i and standard deviation B i Calculate the texture features T of each matrix data i ;

[0103] T i =A i +kB i .

[0104] To this end, for each matrix data we convert it into sequential texture features T1, T2, ...Ti …、T m .

[0105] The method of extracting texture features in the above scheme is based on the discrete degree of pixel points in the matrix data, which can focus on the local features in each matrix data. In fact, in this scheme, each sample data described in sequence is converted into image data. Then, the texture features are extracted by the aforementioned method, so by calculating the average value of pixels in the local area, the scheme can smooth the image and effectively remove Gaussian noise. At the same time, by introducing the standard deviation and bias k, it can retain and highlight fine texture features while smoothing the image, achieving a balance between smoothing and retaining details. After obtaining accurate texture features, the rules of users accessing data can be found based on the texture features, thereby accurately describing the user's behavioral characteristics.

[0106] After using the above solution to extract the features of the matrix data, the LSTM neural network model can be used to obtain the user's access behavior characteristics.

[0107] T1, T2, ...T i …、T m Input to the LSTM neural network model for information extraction, the LSTM neural network model will extract information based on the input T1, T2, ...T i …、T m Related feature sequences generate user access behavior characteristics.

[0108] Specifically: The LSTM neural network model includes an input layer, an LSTM layer, a hidden layer, and an output layer.

[0109] Input layer, receiving feature sequence T1, T2, ...T i …、T m .

[0110] The LSTM layer is the core of the neural network model, which is responsible for processing the correlation between the input feature sequences.

[0111] The LSTM layer contains multiple LSTM units, each of which updates and transmits information through the following gating mechanism.

[0112] Forget gate: determines which old information (i.e. the cell state at the previous time step) should be forgotten. It outputs a value between 0 and 1 through the sigmoid function, indicating the proportion of information retained.

[0113] Input gate: determines which new information should be updated to the cell state. It also outputs a value through the sigmoid function to control the update degree of the candidate cell state generated by the tanh function.

[0114] Cell state: The core of LSTM, used to store long-term information. The update of cell state depends on the output of forget gate and input gate.

[0115] Output gate: determines which information in the cell state should be output to the hidden state of the current time step. It outputs a value through the sigmoid function and multiplies it with the cell state activated by the tanh function to obtain the final hidden state.

[0116] The LSTM neural network model in this solution also sets a hidden layer to further extract features or perform complex transformations. The hidden layer in this solution is a fully connected layer. Specifically, after the LSTM layer, a fully connected layer is added using the API provided by a deep learning framework (such as TensorFlow, PyTorch, etc.).

[0117] The output layer is used to output the feature sequences T1, T2, ...T in the LSTM layer. i …、T m The sequence of weight parameter items extracted from .

[0118] In a general LSTM neural network model, the output layer uses the activation function softmax to determine the difference between the input sample and the corresponding label, thereby using the LSTM neural network to complete the prediction task. In this solution, the output layer is modified to a sequence of weight parameter items, that is, the relevant information input to the output layer is not processed by the activation function, but converted into a standard sequence data. This sequence data is the access behavior feature extracted by the LSTM network.

[0119] The monitoring module obtains the access behavior characteristics of each user based on the training data set, that is, the aforementioned training data will be used to train the neural network model in the monitoring module, and then obtain the access behavior characteristics of each user.

[0120] Then, in the subsequent actual monitoring process, the monitoring module will convert all the access requests generated by the user in sequence on the day into sample data according to the aforementioned sample data conversion method, and input them into the monitoring module. Then, the access behavior characteristics of the day are obtained at the output layer, and the access behavior characteristics of the day are compared with the access behavior characteristics of the user on the previous day using the cosine distance. If it exceeds the preset threshold, it indicates that there is a risk.

[0121] For example, when a user records his identity information in the login information recording module for the fifth time on the day, the user actually has four feature sequences T1, T2, T3, and T4. At this time, for each first access demand generated by the user, a corresponding feature T5 will be generated. T5 will be updated as the access demand increases. Every time T5 is updated, T1, T2, T3, T4, and T5 need to be input into the prediction module to obtain the user's features for the day. Therefore, in practice, each time a user generates an access request, it needs to be verified once. In this solution, the LSTM neural network model is actually used to reduce the dimension of the data and organize the correlation, so the user's behavior habits can be better highlighted.

[0122] In this solution, the cosine distance is used to compare the access behavior features and the current features. In fact, the similarity distance comparison method can also be used, such as the Hamming distance. The specific feature similarity comparison method is an existing technology and will not be discussed here.

[0123] The information acquisition module is connected to the isolation layer signal, obtains the access request that has passed the identity authentication from the isolation layer, and then sends the access request to the monitoring module. When the difference between the access request and the access behavior characteristics of the user exceeds the preset threshold, an alarm message is issued; the intervention module obtains the user who generates the alarm message, and then sends an instruction to terminate the access request to the information transfer station.

[0124] Thus, the technical solution provided by the present application can timely discover some special access habits that cannot be directly discovered by using permission management means. Monitoring the access habits can reduce the risk of information leakage and increase the security of information.

[0125] Embodiment 2: A Beidou map information security monitoring method adopts the aforementioned Beidou map information security monitoring system to monitor the calling of surveying and mapping data.

[0126] The above description is only some preferred embodiments of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present application is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above invention concept. For example, the above features are replaced with (but not limited to) technical features with similar functions disclosed in the embodiments of the present application to form a technical solution.

Claims

1. A Beidou map information security monitoring system, characterized in that: include: Cloud servers are used to store surveying and mapping data of various regions; Terminal device, through which the user generates an access request, the access request includes the user's identity information and access requirements; The isolation layer is used to receive access requests, verify the identity information in the access requests, and send the verified access requests to the information transfer station; The information transfer station obtains the user's access rights based on the user's identity information, and determines whether the access requirements exceed the authority based on the access rights; When the access requirements do not exceed the authority, the corresponding surveying and mapping data is downloaded from the cloud server according to the access requirements; When the access requirement exceeds the authority, the access request is terminated; The analysis node is connected to the isolation layer signal to obtain the access requests of each user on the previous day, and generates the access behavior characteristics of each user based on the access requests of each user on the previous day; Determine whether the current access request is an abnormal access request based on the access behavior characteristics of each user, and when there is an abnormal access request, send an instruction to terminate the access request to the information transfer station; Isolation layer, including: Authentication module, used to verify the user's identity information; The login information recording module is used to record the identity information of each user who has completed the verification; Login information update module, used to update each user's verification record; The information forwarding module sends the access request of the user recorded in the login information recording module to the information transfer station; The login information update module deletes users who have not sent access requests within a predetermined threshold from the login information recording module, and does not delete users who have sent access requests at intervals less than a preset threshold from the login information recording module; A login information update module deletes users whose access addresses have changed from the login information recording module, and does not delete users whose access addresses have not changed from the login information recording module; A login information updating module, for users who exceed the preset time, deletes them from the login information recording module, and for users who do not exceed the preset time, does not delete them from the login information recording module; Information transfer stations include: User permission recording module, used to record the access rights of all users; An information receiving module, used for receiving an access request sent from the isolation layer; The information transfer module verifies the access request according to the access rights recorded in the user authority recording module. If the access request passes the verification, the corresponding mapping data is downloaded / uploaded from the cloud server according to the access request, and then forwarded to the isolation layer, which sends it to the corresponding terminal device; The analysis nodes include: The information collection module is connected to the isolation layer signal to obtain all access requests that have passed the identity information verification, and the collected access requests are made into a training data set; The monitoring module has a built-in neural network model. The monitoring module obtains the access behavior characteristics of each user based on the training data set; The information acquisition module is connected to the isolation layer signal, obtains the access request that has passed the identity authentication from the isolation layer, and then sends the access request to the monitoring module; When the difference between the access request and the user's access behavior characteristics exceeds a preset threshold, an alarm message is issued; The intervention module obtains the user who generates the alarm information and then sends an instruction to terminate the access request to the information transfer station.

2. The Beidou map information security monitoring system according to claim 1, characterized in that: The surveying and mapping data are divided into M data categories according to the region and data type, where M=a*b, a is the number of regions, and b is the number of data types.

3. The Beidou map information security monitoring system according to claim 1, characterized in that: The access requirements include access time, access address, whether to upload data, whether to download data, and the category of data accessed.

4. The Beidou map information security monitoring system according to claim 1, characterized in that: The neural network model is an LSTM neural network model.

5. The Beidou map information security monitoring system according to claim 1, characterized in that: The training data includes several training samples; Each training sample includes all the access requirements of the user when the record in the login information recording module is not updated.

6. The Beidou map information security monitoring system according to claim 5, characterized in that: Convert the access requirements into matrix data related to the data classification method, where the horizontal columns in the matrix data represent the regions and the vertical rows in the matrix data represent the data types; The value of each element in the matrix data is related to the order of access requirements; For data categories that have not been visited, they are recorded as 0 in the matrix data; For the accessed data category, the earlier the element is accessed, the smaller the value corresponding to it.

7. A Beidou map information security monitoring method, characterized in that: The Beidou map information security monitoring system according to any one of claims 1 to 6 is used to monitor the calling of surveying and mapping data.