Smart community dynamic combination access control verification method and system

Abstract

The invention discloses a smart community dynamic combination access control verification method and system, and the method comprises the following steps: obtaining corresponding user data through an access control system when a user carries out access control verification in any verification mode; according to the user data and the system data of the access control system, the historical recognition success rate of the user, the safety coefficient of the current time period, the safety level of the user and the recent suspicious activity index are calculated; calculating a user score according to the historical recognition success rate of the user, the safety coefficient of the current time period, the safety level of the user, the recent suspicious activity index and a preset weight; and determining a verification mode needing to be executed by the current user according to the calculated user score and a preset judgment rule. On the premise of not influencing the daily access control access experience of the user, the access control access safety coefficient is effectively improved, and the convenience and safety of different types of personnel are balanced.

Description

Technical Field

[0001] The present invention belongs to the technical field of smart community data processing, and in particular relates to a smart community dynamic combined access control verification method and system. Background Art

[0002] In modern smart communities, access control systems play a vital role as an important part of community security. Traditional access control systems mainly rely on a single verification method, such as card swiping, password, fingerprint, etc. However, these single verification methods have certain security risks in actual applications. For example, cards may be copied or stolen, passwords may be leaked, and fingerprint recognition may fail in certain special circumstances (such as finger injuries, blurred fingerprints, etc.). These problems make it difficult for community access control systems to strike a balance between security and convenience.

[0003] For example, the Chinese invention patent application with publication number CN118298536A discloses a door access authorization authentication method and system based on a smart community monitoring system. After the user obtains the door access authority, when the first camera of the passage that the user must pass through to the area to be authorized recognizes the user's target face, the navigation route from the area where the first camera is located to the area to be authorized and the time required for the user to navigate through the route are planned. After the required navigation time, and if no cancellation authorization instruction is received, a target authorization instruction is sent to the door access device in the area to be authorized, and the target authorization request corresponding to the target face is added to the door access authority whitelist of the door access device, so that the user can send a target authorization request to the door access device with the door access password, door access card or biometric information to unlock the door lock of the door access device.

[0004] In order to solve the above problems, some multi-factor authentication access control systems have emerged in recent years, combining multiple verification methods to improve security. However, most of these systems are static combinations, that is, a combination of certain verification methods is pre-set, and users must follow the established verification steps every time they enter or exit. Although this method improves security to a certain extent, it also brings inconvenience to user experience, especially when users frequently enter and exit, the cumbersome verification steps may cause user dissatisfaction and resistance. Therefore, it is necessary to design an access control verification solution that takes into account different personnel, providing as much convenience as possible for personnel with high credibility, and enhancing the security of personnel with relatively low credibility. Summary of the invention

[0005] The present invention aims to solve the problem that the existing technology cannot balance the convenience and security of different types of people, and provides a smart community dynamic combination access control verification method and system, which dynamically combines all access control entry and exit methods of the access control management system, which can not only improve the safety factor of access control entry and exit, but also will not affect the user's daily good experience of access control entry and exit.

[0006] In order to solve the above technical problems, the access control verification method proposed in the present invention comprises the following steps:

[0007] When a user uses any verification method to verify access control, the corresponding user data is obtained through the access control system;

[0008] Calculate the user confidence lower limit, the safety factor of the current time period, the user safety level and the recent suspicious activity index respectively according to the user data and the system data of the access control system;

[0009] The user score is calculated based on the user's lower confidence limit, the safety factor of the current time period, the user's safety level, the recent suspicious activity index and the preset weight;

[0010] The verification method to be performed by the current user is determined based on the calculated user score and the preset judgment rules.

[0011] Preferably, the user score is calculated as follows:

[0012] S=α·HSR+β·TSC+γ·USL+δ·RSA

[0013] Where S is the calculated user score, α, β, γ, and δ are the weights of the corresponding factors, and α+β+γ+δ=1, HSR is the user confidence lower limit, TSC is the safety factor of the current time period, USL is the user security level, and RSA is the recent suspicious activity index.

[0014] Preferably, the calculation method of the user confidence lower limit is:

[0015] HSR=μ success -σ success

[0016] In the formula, μ success is the average verification success rate of the user on all verification devices in the community, σ success The standard deviation of the user's verification success rate on all verification devices in the community.

[0017] Preferably, the calculation method of the safety factor of the current time period is:

[0018]

[0019] Where n is the number of time periods divided by the access control system, ωi is the weight of the i-th time period, TSC i is the safety demand coefficient of the i-th time period, n, ω i and TSC i It is preset manually when the access control system is initialized.

[0020] Preferably, the user's security level is calculated by using a hierarchical analysis method, and the calculation method is as follows:

[0021] Determine user security level factors, set a ratio scale for representing the importance of one factor relative to another factor, and construct a pairwise comparison matrix based on the determined factors and ratio scales;

[0022] Use the linear algebra library to calculate the eigenvalues ​​and eigenvectors of the comparison matrix;

[0023] The feature vector is normalized so that the sum of the elements in the feature vector is 1, and the elements in the normalized feature vector are assigned to the determined user security level factors.

[0024] Preferably, the normalization adopts the following calculation method:

[0025]

[0026] Where V′ is the normalized eigenvector, V is the original eigenvector before normalization, N is the number of elements in the original eigenvector, and a corresponds to the element in the original eigenvector.

[0027] Preferably, the recent suspicious activity index is calculated using an exponentially weighted average algorithm, and the calculation method is as follows:

[0028] RSA=ε·CA+(1-ε)·P_RSA

[0029] Where CA is the current suspicious activity index, P_RSA is the last RSA value of the current user, and ε is an artificially preset smoothing parameter used to balance the impact of current activities and historical activities.

[0030] Preferably, the preset determination rule is specifically:

[0031] If the calculated user score S satisfies S≤0.5, it is determined that the current user only needs one verification method and the verification is successful to pass the access control verification;

[0032] If the calculated user score S satisfies 0.5<S≤0.75, it is determined that the current user only needs two verification methods and both verifications are successful to pass the access control verification;

[0033] If the calculated user score S satisfies S>0.75, it is determined that the current user only needs three or more verification methods and all verifications are successful to pass the access control verification.

[0034] Preferably, the verification methods include password verification, card reading verification, face verification, fingerprint verification, palm vein verification, authorization code verification and mobile phone number verification.

[0035] Accordingly, the present invention also proposes a smart community dynamic combined access control verification system, the system executing the above-mentioned access control verification method, including:

[0036] The data acquisition module is used to obtain the corresponding user data through the access control system when the user uses any verification method for access control verification, and obtain the data required for calculating the user's lower confidence limit, the safety factor of the current time period, the user's safety level and the recent suspicious activity index from the access control system;

[0037] A human-computer interaction module, used to provide a user interface and obtain a preset value input by a user on the user interface;

[0038] A comprehensive calculation module, which calculates the user confidence lower limit, the safety factor of the current time period, the user security level and the recent suspicious activity index according to the user data and the system data of the access control system, and calculates the user score in combination with the preset weight;

[0039] The verification method generation module generates the verification method for the current user based on the calculated user score and preset judgment rules, and sends it to the access control system.

[0040] Compared with the prior art, the present invention has the following technical effects:

[0041] 1. The access control verification method proposed in the present invention calculates the final user score through the relevant user data in the access control system, determines the verification method that the user needs to use according to the user score, and dynamically combines all the access control access methods of the access control management system. Under the premise of not affecting the user's daily access control access experience, it effectively improves the safety factor of access control access and balances the convenience and safety of different types of personnel.

[0042] 2. The access control verification method proposed in the present invention calculates the user confidence lower limit, the safety factor of the current time period, the user security level and the recent suspicious activity index respectively, and obtains the user score based on the above calculations. Through the calculation of the above parameters, the types of different personnel can be accurately reflected, and the accuracy of personnel identification can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a flow chart of the verification method of the present invention. DETAILED DESCRIPTION

[0044] In order to make the objectives, technical solutions and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in combination with specific embodiments of the present application and with reference to the accompanying drawings.

[0045] Embodiment 1

[0046] like Figure 1 As shown, a smart community dynamic combined access control verification method includes the following steps:

[0047] When a user uses any verification method to verify access control, the corresponding user data is obtained through the access control system. When a user uses any verification method to verify access control, the access control verification system is triggered to obtain the corresponding user's data. The obtained data includes the number of successful door openings of the current user in the current device's history, the total number of door openings of the current user in a certain device's history, and the data directly provided by the system in subsequent calculations, such as preset weights, preset time periods and other parameters.

[0048] The user confidence lower limit, the safety factor of the current time period, the user safety level and the recent suspicious activity index are calculated respectively according to the user data and the system data of the access control system.

[0049] In this embodiment, the calculation of the user confidence lower limit introduces the standard deviation to take into account the fluctuation of the user verification success rate. The calculation method of the user confidence lower limit is:

[0050] HSR=μ success -σ success

[0051] In the formula, μ success is the average verification success rate of the user on all verification devices in the community, σ success The standard deviation of the user's verification success rate on all verification devices in the community.

[0052] Specifically, if there is a set of historical success rate data of the current user:

[0053] [0.9,0.85,0.87,0.9,0.88]

[0054] It comes from the ratio of the number of successful authentications of the user on a certain device to the number of successful authentications of the user on the same device. The average is calculated first:

[0055]

[0056] Then calculate the standard deviation:

[0057]

[0058] Then the final user confidence lower limit HSR = 0.88-0.01897 = 0.86103.

[0059] In this embodiment, the safety factor of the current time period is a weighted average based on the time period. The calculation method of the safety factor of the current time period is:

[0060]

[0061] Where n is the number of time periods divided by the access control system, ω i is the weight of the i-th time period, TSC i is the safety demand coefficient of the i-th time period, n, ω i and TSC i It is preset manually when the access control system is initialized.

[0062] As in one embodiment of the present invention, the time period is divided into daytime and nighttime, corresponding to the two time periods of 08:00-22:00 and 22:00-08:00 respectively. When the system is initialized, the weight of the daytime time period is preset to 0.3, and the weight of the night time period is preset to 0.7. At the same time, the safety demand coefficients of the daytime and night time periods are set to 0.3 and 0.7 respectively. According to the above data, TSC=0.3×0.3+0.7×0.7=0.58 can be calculated. In other embodiments of the present invention, in order to divide different time periods more finely, other time period division methods can be selected, such as dividing into morning, afternoon, and night, or dividing into morning, afternoon, the first half of the night, and the second half of the night, or making more detailed divisions according to local living habits such as nightlife. Correspondingly, the corresponding time period safety demand coefficient also corresponds to the meaning of the time period.

[0063] The user's security level is calculated by the hierarchical analysis method, and the calculation method is as follows:

[0064] Determine user security level factors, set a ratio scale for representing the importance of one factor relative to another factor, and construct a pairwise comparison matrix based on the determined factors and ratio scales;

[0065] Use the linear algebra library to calculate the eigenvalues ​​and eigenvectors of the comparison matrix;

[0066] The feature vector is normalized so that the sum of the elements in the feature vector is 1, and the elements in the normalized feature vector are assigned to the determined user security level factors.

[0067] Specifically, when constructing the comparison matrix, the user security level factor is the factor involved in the comparison. Taking three factors involved in the comparison as an example, they are set as A, B and C. Then, an importance ratio scale is set. This scale is used to indicate the importance of one factor relative to another factor, for example:

[0068] A ratio scale of 1 means that both factors are equally important;

[0069] A ratio scale of 3 means that one factor appears slightly more important than another factor;

[0070] A ratio scale of 5 indicates that one factor is significantly more important than another factor;

[0071] The ratio scale of 7 indicates that one factor appears very important compared to another factor;

[0072] A ratio of 9 indicates that one factor is extremely important compared to another factor;

[0073] The scales 2, 4, 6, and 8 represent the middle values ​​of the above judgments;

[0074] 1 / 3, 1 / 5, 1 / 7, 1 / 9 indicate opposite importance.

[0075] In the constructed comparison matrix, the element a of the matrix i j represents the importance of the i-th element relative to the j-th element. Taking the above three factors A, B and C as an example, the calculation of the user security level in the embodiment is explained. It is set that A is slightly more important than B, and the ratio is 3; A is more important than C, and the ratio is set to 5; B is slightly more important than C, and the ratio is set to 3. According to the above settings, the paired comparison matrix A is constructed:

[0076]

[0077] In this embodiment, A, B, and C are defined as community managers, community residents, and outsiders (such as couriers, riders, visitors, etc.), respectively.

[0078] Next, the eigenvectors and eigenvalues ​​are calculated based on the comparison matrix A. First, the characteristic polynomial is constructed:

[0079] det(A-λI)=0

[0080] Where I is the identity matrix, for the comparison matrix A:

[0081]

[0082] Need to calculate det(A-λI):

[0083]

[0084] At this point, the characteristic polynomial is obtained.

[0085] For the above characteristic polynomial, this embodiment uses the Laplace expansion method to calculate:

[0086]

[0087] Calculate each 2×2 determinant in the above formula separately:

[0088]

[0089] Substituting the above three calculation results into det(A-λI), we have:

[0090]

[0091] Solving the above polynomial yields the eigenvalues ​​λ=3, -0.533, 0.467.

[0092] For each eigenvalue λ, the eigenvector is obtained by the following equation:

[0093] (A-λI)V=0

[0094] Taking one of the eigenvalues ​​3 as an example, we can get the following equation:

[0095]

[0096] Solve the homogeneous system of equations above:

[0097]

[0098] Use Gaussian elimination to simplify the above system of equations:

[0099]

[0100] Converting the above system of equations into a simpler form:

[0101]

[0102] In the above formula, a specific value is selected for simplification, such as setting v2 and v3 to 1, and obtaining v1=0.066, and finally obtaining the eigenvector V.

[0103] The feature vector V is expressed as follows:

[0104]

[0105] In order to converge the discrete degree of the user security level, in this embodiment, a normalization operation is also performed on the feature vector V. The specific method is as follows:

[0106]

[0107] Where V′ is the normalized eigenvector, V is the original eigenvector before normalization, N is the number of elements in the original eigenvector, and a corresponds to the element in the original eigenvector.

[0108] Substitute the example's eigenvector V into the normalization formula:

[0109]

[0110] According to the above formula, the user security level USL = [0.066, 0.26, 0.673]. For outsiders, strict control is required, so the USL weight is the highest, taking 0.673; community residents are frequent visitors to the community, and the USL weight is in the middle, taking 0.26; community managers have the highest authority to manage public areas, and the USL weight is the lowest, taking 0.066. When a person performs access control identification, the system can determine the type of the current person based on the system database, and assign different USL weights (user security level USL) to the person based on the determined type.

[0111] The recent suspicious activity index is calculated using an exponentially weighted average algorithm, and the calculation method is as follows:

[0112] RSA=ε·CA+(1-ε)·P_RSA

[0113] Where CA is the current suspicious activity index, P_RSA is the last RSA value of the current user, and ε is an artificially preset smoothing parameter with a value of (0, 1) to balance the impact of current activities and historical activities.

[0114] In this embodiment, the current suspicious activity index comes from multiple sources, including the number of abnormal attempts in the access control system log, the door opening request in an abnormal time period, or an abnormal access pattern such as multiple consecutive failed authentication attempts. The selection of the smoothing parameter ε can be determined based on the analysis of historical data to balance the impact of current activities and historical activities. For example, a larger ε value means a higher sensitivity to current activities, while a smaller ε value pays more attention to historical activities.

[0115] Taking the above three sources that affect the suspicious activity index as an example, the process of obtaining the current suspicious activity in this embodiment is explained. Specifically, this embodiment defines the number of abnormal attempts (Attempt Count, AC), door opening requests in abnormal time periods (Off-hours Access, 0A), and abnormal access patterns (Continuous Failed Attempts, CFA). The number of abnormal attempts is the number of abnormal credential recognitions that appear in the device system log, which is usually related to the device access control system; door opening requests in abnormal time periods are usually non-working / abnormal community entry and exit times, such as between 22:00-6:00; multiple consecutive failed identification attempts, that is, the number of failed door opening recognition attempts with the same credential. Use the weighted average method to integrate the data from these sources:

[0116]

[0117] Among them, ω AC ,ω OA ,ω CFA are the weights of the corresponding sources, AC max OA max , CFA max are the historical maximum values ​​of the corresponding sources, respectively, and are used to normalize the data.

[0118] Take the following data as an example:

[0119] The number of abnormal attempts is 10 times, and the historical maximum is 50 times;

[0120] 5 door opening requests during non-normal time periods, with a historical maximum of 20;

[0121] The abnormal access pattern is 3 times, and the historical maximum is 10 times;

[0122] The weights are 0.4, 0.3, and 0.3 respectively.

[0123] Then there is

[0124] The user score is calculated based on the user's lower confidence limit, the safety factor of the current time period, the user's safety level, the recent suspicious activity index and the preset weight; the calculation method of the user score is:

[0125] S=α·HSR+β·TSC+γ·USL+δ·RSA

[0126] Wherein, S is the calculated user score, α, β, γ, δ are the weights of the corresponding factors, and α+β+γ+δ=1, HSR is the user confidence lower limit, TSC is the safety factor of the current time period, USL is the user security level, and RSA is the recent suspicious activity index. In one embodiment of the present invention, the weights of the corresponding factors are preset to 0.25. In some other embodiments of the present invention, different weights can be set according to different sensitivities, and it is only necessary to satisfy α+β+γ+δ=1. For example, if the sensitivity tends to be high towards the user confidence lower limit, the weight α of HSR can be preset to be greater than 0.25, and other weights are adjusted accordingly to satisfy α+β+γ+δ=1.

[0127] Finally, based on the user score calculated by the above method and the preset judgment rule, the verification method to be performed by the current user is determined. Specifically, the preset judgment rule is:

[0128] If the calculated user score S satisfies S≤0.5, it is determined that the current user only needs one verification method and the verification is successful to pass the access control verification;

[0129] If the calculated user score S satisfies 0.5<S≤0.75, it is determined that the current user only needs two verification methods and both verifications are successful to pass the access control verification;

[0130] If the calculated user score S satisfies S>0.75, it is determined that the current user only needs three or more verification methods and all verifications are successful to pass the access control verification.

[0131] In this embodiment, the verification methods include password verification, card reading verification, face verification, fingerprint verification, palm vein verification, authorization code verification and mobile phone number verification. When the system calculates that the user score is less than or equal to 0.5, since it is determined that the current user only needs one verification method and the verification is successful to pass the access control verification, the verification method triggered by the current user and the verification result (pass or fail) can be directly used as the current access control verification. If the calculated user score is greater than 0.5, since the system determines that the current user needs two or more verification methods, two or more of which include the currently triggered verification method, the remaining verification methods can be selected by the user, and the user only needs to meet the number of verification methods that pass the verification to meet the system requirements to pass. In some other embodiments of the present invention, the remaining verification methods can be generated by the access control verification system, specifying the verification methods that the current user needs to supplement.

[0132] Specifically, if a user is passing through the access control, he uses any of the verification methods such as password verification, card reading verification, face verification, fingerprint verification, palm vein verification, authorization code verification and mobile phone number verification. Taking card reading verification as an example, the user uses the access card for verification and identification, and the access control verification system receives the user verification request. If the final calculated user score does not exceed 0.5, the access card verification result is used as the basis for determining whether the user is allowed to pass; if the final calculated user score S satisfies 0.5<S≤0.75, in addition to the current access card verification method, the user needs to provide an additional verification method. One solution is to prompt the user to supplement the verification, and the user can choose any other verification method except the access card verification. The other solution is for the access control verification system to generate another verification method except the access card verification based on the current access control hardware equipment and user-related data. Both solutions require the user to pass all the verifications before the user is allowed to pass. For the user score S finally calculated to satisfy S>0.75, the supplementary verification method also includes the above two schemes, and the user is allowed to pass only when the current verification method and the supplementary verification method are both successfully verified.

[0133] Embodiment 2

[0134] A smart community dynamic combined access control verification system, the system executing the access control verification method as described in the first embodiment, comprising:

[0135] The data acquisition module is used to obtain the corresponding user data through the access control system when the user uses any verification method for access control verification, and obtain the data required for calculating the user's lower confidence limit, the safety factor of the current time period, the user's safety level and the recent suspicious activity index from the access control system;

[0136] A human-computer interaction module, used to provide a user interface and obtain a preset value input by a user on the user interface;

[0137] A comprehensive calculation module, which calculates the user confidence lower limit, the safety factor of the current time period, the user security level and the recent suspicious activity index according to the user data and the system data of the access control system, and calculates the user score in combination with the preset weight;

[0138] The verification method generation module generates the verification method for the current user based on the calculated user score and preset judgment rules, and sends it to the access control system.

[0139] The above is only a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several modifications and improvements without departing from the inventive concept of the present invention, which all belong to the protection scope of the present invention.

Claims

1. A smart community dynamic combined access control verification method, characterized in that: It includes the following steps: When a user conducts access control verification using any verification method, corresponding user data is obtained through the access control system; Calculate the lower confidence limit of the user, the safety factor of the current time period, the user safety level, and the recent suspicious activity index respectively according to the user data and the system data of the access control system; Calculate the user score according to the lower confidence limit of the user, the safety factor of the current time period, the user safety level, the recent suspicious activity index, and the preset weights; Determine the verification method that the current user needs to execute according to the calculated user score and the preset determination rules.

2. According to claim 1, a smart community dynamic combined access control verification method is characterized in that: The calculation method of the user score is as follows: S = α·HSR + β·TSC + γ·USL + δ·RSA In the formula, S is the calculated user score, α, β, γ, and δ are the weights of the corresponding factors respectively, and α + β + γ + δ = 1, HSR is the lower confidence limit of the user, TSC is the safety factor of the current time period, USL is the user safety level, and RSA is the recent suspicious activity index.

3. According to claim 2, a smart community dynamic combined access control verification method is characterized in that: The calculation method of the lower confidence limit of the user is as follows: HSR=μ success -s success In the formula, μ success is the average verification success rate of the user on all verification devices in the community, σ success The standard deviation of the user's verification success rate on all verification devices in the community.

4. According to claim 2, a smart community dynamic combined access control verification method is characterized in that: The calculation method of the safety factor of the current time period is as follows: Where n is the number of time periods divided by the access control system, ω i is the weight of the i-th time period, TSC i is the safety demand coefficient of the i-th time period, n, ω i and TSC i It is preset manually when the access control system is initialized.

5. According to claim 2, a smart community dynamic combined access control verification method is characterized in that: The safety level of the user is calculated by the analytic hierarchy process, and the calculation method is as follows: Determine the user safety level factors, set the ratio scale for representing the importance of one factor relative to another factor, and construct a pairwise comparison matrix according to the determined factors and the ratio scale; Use the linear algebra library to calculate the eigenvalues and eigenvectors of the comparison matrix; Normalize the eigenvector so that the sum of the elements in the eigenvector is 1, and allocate the elements in the normalized eigenvector to the determined user safety level factors.

6. A smart community dynamic combined access control verification method according to claim 5, characterized in that: The normalization adopts the following calculation method: In the formula, V′ is the normalized eigenvector, V is the original eigenvector before normalization, N is the number of elements in the original eigenvector, and a corresponds to the elements in the original eigenvector.

7. A smart community dynamic combined access control verification method according to claim 2, characterized in that: Use the exponential weighted average algorithm to calculate the recent suspicious activity index, and the calculation method is as follows: RSA = ε·CA + (1 - ε)·P_RSA In the formula, CA is the current suspicious activity index, P_RSA is the previous RSA value of the current user, and ε is the smoothing parameter preset manually, which is used to balance the influence of the current activity and the historical activity.

8. According to claim 1, a smart community dynamic combined access control verification method is characterized in that: The preset determination rules are specifically as follows: If the calculated user score S satisfies S ≤ 0.5, it is determined that the current user only needs one verification method and can pass the access control verification if the verification is successful; If the calculated user score S satisfies 0.5 < S ≤ 0.75, it is determined that the current user only needs two verification methods and can pass the access control verification if both verifications are successful; If the calculated user score S satisfies S > 0.75, it is determined that the current user needs three or more verification methods and can pass the access control verification only if all verifications are successful.

9. A smart community dynamic combined access control verification method according to claim 1, characterized in that: The verification methods include password verification, card reading verification, face verification, fingerprint verification, palm vein verification, authorization code verification, and mobile phone number verification.

10. A smart community dynamic combined access control verification system, characterized in that: The system executes the access control verification method described in any one of claims 1 - 9, including: The data acquisition module is used to obtain the corresponding user data through the access control system when the user uses any verification method for access control verification, and obtain the data required for calculating the user's lower confidence limit, the safety factor of the current time period, the user's safety level and the recent suspicious activity index from the access control system; A human-computer interaction module, used to provide a user interface and obtain a preset value input by a user on the user interface; A comprehensive calculation module, which calculates the user confidence lower limit, the safety factor of the current time period, the user security level and the recent suspicious activity index according to the user data and the system data of the access control system, and calculates the user score in combination with the preset weight; The verification method generation module generates the verification method for the current user based on the calculated user score and preset judgment rules, and sends it to the access control system.

Images