Intelligent security access method and system based on adaptive honeypot

Through the intelligent secure access method based on adaptive honeypot, the adversarial network optimization response actions and generate policy tables are used to solve the problem of insufficient flexibility and adaptability of traditional firewalls in the face of new attacks and advanced persistent threats, and improve the Cybersecurity response efficiency and internal threat defense capabilities.

CN118944948APending Publication Date: 2024-11-12ELECTRIC POWER RES INST STATE GRID SHANXI ELECTRIC POWER +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411152858.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-21
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

Traditional firewalls are difficult to effectively identify and respond when facing new, unknown attack methods and advanced persistent threats, and lack flexibility and adaptability, and cannot effectively monitor and defend against internal attacks.

Method used

Adopting intelligent secure access method based on adaptive honeypots, the policy table of response action set and honeypot state is initialized, attack session commands are obtained, response action is selected, and honeypot state is updated, and reward function is used to calculate the reward function to optimize the response action. , generate a policy table to deal with protection policies.

Benefits of technology

Improves the efficiency of responding to attack sessions, enhances flexible response capabilities to dynamic threats, and enables more efficient monitoring and defense against internal threats.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118944948A_ABST
    Figure CN118944948A_ABST
Patent Text Reader

Abstract

The embodiment of the invention provides an intelligent security access method and system based on a self-adaptive honeypot, and belongs to the technical field of network security. The access method comprises the following steps: initializing a strategy table based on a response action set and a honeypot state; acquiring an attack session; receiving commands of attack sessions according to the sequence; selecting a corresponding response action according to the command; updating the honey pot state according to the response action; adding the response action and the honeypot state into a strategy table; judging whether the command is an automatic command or not; under the condition that the command is judged to be the non-automatic instruction, judging whether an unread attack session exists at present or not; returning to execute the step of acquiring the attack session needing to be read under the condition of judging that the attack session which is not read exists at present; generating a strategy table under the condition of judging that the attack session which is not read does not exist at present; and taking the strategy table as a coping protection strategy of the current server.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of network security technology, and in particular to an intelligent security access method and system based on an adaptive honeypot. Background Art

[0002] In the modern network environment, network security issues are becoming more diverse and complex. With the continuous advancement of hacker attack technology, traditional firewall-based defense measures can no longer meet the growing security needs. As a typical border defense device, firewalls control the traffic in and out of the network through preset security policies to prevent potential threats and attacks. These policies usually consist of a series of rules, such as IP address filtering, port restrictions, and protocol management.

[0003] Although firewall-based defense systems are widely used due to their high efficiency and ease of deployment, they have obvious limitations. First, firewalls rely on pre-set security rules. Although these rules can handle known threat patterns, they are often unable to cope with new, unknown attack methods. Firewalls are usually unable to intelligently identify and respond to zero-day attacks or advanced persistent threats (APTs). Second, once the firewall policy is set, it lacks sufficient flexibility and adaptability to the dynamically changing network environment and ever-changing attack methods. In addition, firewalls mainly focus on protecting the network boundary, and have limited monitoring and defense capabilities for attacks launched from within or malicious behaviors that have already penetrated the network.

[0004] In view of this, the field of network security urgently needs a more flexible, intelligent and adaptable defense mechanism to dynamic network threats. Summary of the invention

[0005] The purpose of the embodiments of the present invention is to provide an intelligent security access method and system based on an adaptive honeypot, which can improve the effectiveness of the generated response strategy.

[0006] In order to achieve the above object, an embodiment of the present invention provides an intelligent secure access method based on an adaptive honeypot, comprising:

[0007] Initialize the policy table based on the response action set and the honeypot status;

[0008] Get the attack session that needs to be read;

[0009] receiving commands of the attack session in order;

[0010] Selecting a corresponding response action according to the command;

[0011] Update the honeypot state according to the response action;

[0012] Adding the response action and honeypot status to the policy table;

[0013] Determining whether the command is an automated instruction;

[0014] In the case where it is determined that the command is an automated instruction, returning to the step of sequentially receiving commands of the attack session;

[0015] In the case where it is determined that the command is not an automated instruction, determining whether there is currently an unread attack session;

[0016] In the case where it is determined that there is currently an attack session that has not been read, returning to the step of obtaining the attack session that needs to be read;

[0017] In the case of determining that there is no unread attack session currently, generating the policy table;

[0018] The strategy table is used as the response protection strategy of the current server.

[0019] Optionally, selecting a corresponding response action according to the command includes:

[0020] Calculate the reward function corresponding to each response action;

[0021] The response action is selected according to the reward function.

[0022] Optionally, calculate the reward function corresponding to each response action, including:

[0023] Encoding the command, the response action, and the honeypot state to obtain an encoding vector;

[0024] Inputting the encoding vector into the adversarial network to obtain the state of the honeypot at the next moment;

[0025] The honeypot state at the next moment is compared with the current honeypot state to obtain a reward function for the response action.

[0026] Optionally, inputting the encoding vector into an adversarial network to obtain the honeypot state at the next moment includes:

[0027] According to formula (1), the state of the honeypot at the next moment is determined.

[0028] Q t+1 (S t , a t )=Q t (S t , a t )+α[r t +γmax(Qt (S t , A)-Q t (S t , a t ))], (1)

[0029] Among them, Q t+1 (S t , a t ) is the state value of the honeypot state, S t is the honeypot state, a t is the response action, α is the learning rate, γ is the preset discount factor, r t is the immediate reward function, and A is the response action set.

[0030] Optionally, selecting the response action according to the reward function includes:

[0031] Select the first n response actions according to the order of the reward function from large to small;

[0032] Determining a transition probability of the selected response action;

[0033] The response action is selected according to the transition probability.

[0034] Optionally, determining the transition probability of the selected response action includes:

[0035] The transition probability is determined according to formula (2) to formula (4),

[0036] p i =p min +(ni)Δp, (2)

[0037]

[0038] p min ≠0, (4)

[0039] Among them, p i is the transition probability of the i-th response action selected in sequence, p min is the preset benchmark probability value, and Δp is the gradient probability.

[0040] Optionally, generating the policy table includes:

[0041] Determine the security level coefficient of the honeypot state of the last response action after each attack session in the strategy table ends;

[0042] Determining whether the security level coefficient is greater than a preset first threshold;

[0043] In the case where it is determined that the security level coefficient is greater than the first threshold, returning to the step of initializing a policy table based on a response action set and a honeypot state;

[0044] In the case where it is determined that the security level coefficient is less than or equal to the first threshold, determining an average difference of the security level coefficients;

[0045] Determining whether the average difference is greater than or equal to a preset second threshold;

[0046] In the case where it is determined that the average difference is greater than or equal to the second threshold, returning to the step of initializing the policy table based on the response action set and the honeypot state;

[0047] When it is determined that the average difference is smaller than the second threshold, the strategy table is output.

[0048] On the other hand, the present invention further provides an intelligent secure access system based on an adaptive honeypot, wherein the access system comprises a processor, and the processor is configured to execute any of the access methods described above.

[0049] In yet another aspect, the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, wherein the instructions are used to be read by a machine so that the machine executes any of the access methods described above.

[0050] Through the above technical solution, the embodiment of the present invention provides an intelligent security access method and system based on an adaptive honeypot, which realizes a step-by-step response to the attack session by configuring corresponding response actions for the commands in each attack session. Compared with the policy group solution of the prior art, the access method and system provided by the present invention have higher response efficiency.

[0051] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

[0053] Figure 1 is a flow chart of an intelligent secure access method based on an adaptive honeypot according to an embodiment of the present invention;

[0054] Figure 2 is an example diagram of an adversarial network according to one embodiment of the present invention;

[0055] Figure 3is a flow chart of a method for selecting a response action according to a reward function according to one embodiment of the present invention;

[0056] Figure 4 is a flowchart of a method for determining a generated policy table according to an embodiment of the present invention. DETAILED DESCRIPTION

[0057] The specific implementation of the embodiment of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the embodiment of the present invention, and is not used to limit the embodiment of the present invention.

[0058] like Figure 1 FIG. 1 is a flow chart of an intelligent secure access method based on an adaptive honeypot according to an embodiment of the present invention. Figure 1 The access method may include the following steps:

[0059] In step S10, a policy table based on a response action set and a honeypot state is initialized;

[0060] In step S11, the attack session to be read is obtained;

[0061] In step S12, commands for attacking the session are received in sequence;

[0062] In step S13, a corresponding response action is selected according to the command;

[0063] In step S14, the honeypot status is updated according to the response action;

[0064] In step S15, the response action and the honeypot status are added to the policy table;

[0065] In step S16, it is determined whether the command is an automation instruction;

[0066] If it is determined that the command is an automated instruction, returning to the step of sequentially receiving commands for attacking the session, that is, returning to step S12;

[0067] In step S17, when it is determined that the command is not an automated instruction, it is determined whether there is an unread attack session at present, that is, the execution returns to step S11;

[0068] If it is determined that there is an unread attack session, returning to the step of obtaining the attack session to be read;

[0069] In step S18, if it is determined that there is no unread attack session currently, a policy table is generated;

[0070] In step S19, the policy table is used as the protection strategy for the current server.

[0071] In this Figure 1 In the access method shown, step S10 can be used to initialize a policy table based on a response action set and a honeypot state. Specifically, the initialized response action set can be empty, and the honeypot state can be the initial state of the system.

[0072] Steps S11 to S17 can be used to obtain the attack sessions that need to be read one by one, and generate corresponding response actions for each command in the attack session. Among them, after reading the command, the specific method of selecting the corresponding response action according to the command in step S13 can be various forms known to those skilled in the art, such as presetting a database and associating the correlation between the command and the response action in the database. In one example of the present invention, in order to ensure the effectiveness of the selected response action, step S12 can be respectively calculating the reward function corresponding to each response action, and then selecting the response action according to the reward function. Further, for the specific calculation method of the reward function, in one example of the present invention, the calculation method can be first encoding the command, the response action and the honeypot state to obtain a coding vector. Then, for the obtained coding vector, the coding vector is input into a preset adversarial network to obtain the honeypot state at the next moment. Finally, the honeypot state at the next moment can be compared with the current honeypot state to obtain the reward function of the response action. As for the specific calculation method of the reward function, it can be various forms known to those skilled in the art. For example, multiple evaluation indicators of the honeypot state are preset, and the reward function is determined by weighted calculation.

[0073] The adversarial network may be in various forms known to those skilled in the art. In one example of the present invention, the adversarial network may be: Figure 2 The structure shown. Figure 2 In the example, the adversarial network may include a generator G, a database, and a discriminator D. The generator G generates data G(Z) by combining the input random noise. The discriminator D determines whether the generator G has been trained by judging the authenticity of the generated data and the real data in the database. The honeypot state at the next moment may also be expressed by the following formula (1):

[0074] Q t+1 (S t , a t )=Q t (S t , a t )+α[r t +γmax(Q t (S t, A)-Q t (S t , a t ))], (1)

[0075] Among them, Q t+1 (S t , a t ) is the state value of the honeypot state, S t For honeypot state, a t is the response action, α is the learning rate, γ is the preset discount factor, r t is the immediate reward function, and A is the response action set.

[0076] In addition, since the access method gradually selects the response action corresponding to each command, and the response actions themselves are interrelated, when selecting the response action according to the reward function, if the response action with the largest reward function is directly selected, the optimal strategy table may not be obtained. Therefore, in one example of the present invention, it may be possible to adopt the following method: Figure 3 The method shown in selects the response action. Specifically, in Figure 3 In the method, the method may include the following steps:

[0077] In step S20, the first n response actions are selected in descending order of the reward function;

[0078] In step S21, the transition probability of the selected response action is determined. The transition probability can be determined using the following formulas (2) to (4):

[0079] p i =p min +(ni)Δp, (2)

[0080]

[0081] p min ≠0, (4)

[0082] Among them, p i is the transition probability of the i-th response action selected in sequence, p min is the preset benchmark probability value, and Δp is the gradient probability.

[0083] In step S22, a response action is selected according to the transition probability.

[0084] After determining the response action corresponding to each command, in order to further determine the efficiency of the generated strategy table. In one example of the present invention, the method may also include: Figure 4 The steps shown in .

[0085] In this Figure 4In the method, the method may include the following steps:

[0086] In step S30, the security level coefficient of the honeypot state of the last response action after each attack session in the strategy table ends is determined;

[0087] In step S31, it is determined whether the security level coefficient is greater than a preset first threshold;

[0088] If it is determined that the security level coefficient is greater than the first threshold, returning to the step of initializing the policy table based on the response action set and the honeypot state, that is, returning to step S10;

[0089] In step S32, when it is determined that the security level coefficient is less than or equal to the first threshold, the average difference of the security level coefficient is determined;

[0090] In step S33, it is determined whether the average difference is greater than or equal to a preset second threshold;

[0091] If it is determined that the average difference is greater than or equal to the second threshold, returning to the step of initializing the policy table based on the response action set and the honeypot state, that is, returning to step S10;

[0092] In step S34, when it is determined that the average difference is smaller than the second threshold, the strategy table is output.

[0093] In this Figure 4 In the method shown, step S30 can be used to determine the security level coefficient of the honeypot state of the last response action after each attack session in the policy table ends. The security level system can be related to determining the security level of the honeypot state. The specific calculation method of the security level coefficient can be a variety of forms known to those skilled in the art. In one example of the present invention, the calculation method of the security level coefficient can be, for example, to construct a standard matrix of the authority levels of multiple ports, and then describe the honeypot state through the security matrix, thereby forming a security level matrix representing the honeypot state. On this basis, the security level coefficient is determined by respectively calculating the distance between the security level matrix and the preset multiple standard security level matrices. Steps S31 to S33 are used to determine whether the security level coefficient of the honeypot state meets the requirements, wherein step S31 is used to determine whether the minimum value of the security level coefficient meets the requirements, in order to ensure that the system will not temporarily enter a high-risk state when executing the policy table. Step S33 is used to determine whether the average difference of the security level coefficient meets the requirements, in order to avoid excessive fluctuations in the security authority of the system.

[0094] On the other hand, the present invention also provides an intelligent secure access system based on an adaptive honeypot, wherein the access system includes a processor, and the processor is configured to execute any of the access methods described above. Specifically, the access method may include the following steps:

[0095] In step S10, a policy table based on a response action set and a honeypot state is initialized;

[0096] In step S11, the attack session to be read is obtained;

[0097] In step S12, commands for attacking the session are received in sequence;

[0098] In step S13, a corresponding response action is selected according to the command;

[0099] In step S14, the honeypot status is updated according to the response action;

[0100] In step S15, the response action and the honeypot status are added to the policy table;

[0101] In step S16, it is determined whether the command is an automation instruction;

[0102] If it is determined that the command is an automated instruction, returning to the step of sequentially receiving commands for attacking the session, that is, returning to step S12;

[0103] In step S17, when it is determined that the command is not an automated instruction, it is determined whether there is an unread attack session at present, that is, the execution returns to step S11;

[0104] If it is determined that there is an unread attack session, returning to the step of obtaining the attack session to be read;

[0105] In step S18, if it is determined that there is no unread attack session currently, a policy table is generated;

[0106] In step S19, the policy table is used as the protection strategy for the current server.

[0107] In this Figure 1 In the access method shown, step S10 can be used to initialize a policy table based on a response action set and a honeypot state. Specifically, the initialized response action set can be empty, and the honeypot state can be the initial state of the system.

[0108] Steps S11 to S17 can be used to obtain the attack sessions that need to be read one by one, and generate corresponding response actions for each command in the attack session. Among them, after reading the command, the specific method of selecting the corresponding response action according to the command in step S13 can be various forms known to those skilled in the art, such as presetting a database and associating the correlation between the command and the response action in the database. In one example of the present invention, in order to ensure the effectiveness of the selected response action, step S12 can be respectively calculating the reward function corresponding to each response action, and then selecting the response action according to the reward function. Further, for the specific calculation method of the reward function, in one example of the present invention, the calculation method can be first encoding the command, the response action and the honeypot state to obtain a coding vector. Then, for the obtained coding vector, the coding vector is input into a preset adversarial network to obtain the honeypot state at the next moment. Finally, the honeypot state at the next moment can be compared with the current honeypot state to obtain the reward function of the response action. As for the specific calculation method of the reward function, it can be various forms known to those skilled in the art. For example, multiple evaluation indicators of the honeypot state are preset, and the reward function is determined by weighted calculation.

[0109] The adversarial network may be in various forms known to those skilled in the art. In one example of the present invention, the adversarial network may be: Figure 2 The structure shown. Figure 2 In the example, the adversarial network may include a generator G, a database, and a discriminator D. The generator G generates data G(Z) by combining the input random noise. The discriminator D determines whether the generator G has been trained by judging the authenticity of the generated data and the real data in the database. The honeypot state at the next moment may also be expressed by the following formula (1):

[0110] Q t+1 (S t , a t )=Q t (S t , a t )+α[r t +γmax(Q t (S t , A)-Q t (S t , a t ))], (1)

[0111] Among them, Q t+1 (S t , a t ) is the state value of the honeypot state, S t For honeypot state, at is the response action, α is the learning rate, γ is the preset discount factor, r t is the immediate reward function, and A is the response action set.

[0112] In addition, since the access method gradually selects the response action corresponding to each command, and the response actions themselves are interrelated, when selecting the response action according to the reward function, if the response action with the largest reward function is directly selected, the optimal strategy table may not be obtained. Therefore, in one example of the present invention, it may be possible to adopt the following method: Figure 3 The method shown in selects the response action. Specifically, in Figure 3 In the method, the method may include the following steps:

[0113] In step S20, the first n response actions are selected in descending order of the reward function;

[0114] In step S21, the transition probability of the selected response action is determined. The transition probability can be determined using the following formulas (2) to (4):

[0115] p i =p min +(ni)Δp, (2)

[0116]

[0117] p min ≠0, (4)

[0118] Among them, p i is the transition probability of the i-th response action selected in sequence, p min is the preset benchmark probability value, and Δp is the gradient probability.

[0119] In step S22, a response action is selected according to the transition probability.

[0120] After determining the response action corresponding to each command, in order to further determine the efficiency of the generated strategy table. In one example of the present invention, the method may also include: Figure 4 The steps shown in .

[0121] In this Figure 4 In the method, the method may include the following steps:

[0122] In step S30, the security level coefficient of the honeypot state of the last response action after each attack session in the strategy table ends is determined;

[0123] In step S31, it is determined whether the security level coefficient is greater than a preset first threshold;

[0124] If it is determined that the security level coefficient is greater than the first threshold, returning to the step of initializing the policy table based on the response action set and the honeypot state, that is, returning to step S10;

[0125] In step S32, when it is determined that the security level coefficient is less than or equal to the first threshold, the average difference of the security level coefficient is determined;

[0126] In step S33, it is determined whether the average difference is greater than or equal to a preset second threshold;

[0127] If it is determined that the average difference is greater than or equal to the second threshold, returning to the step of initializing the policy table based on the response action set and the honeypot state, that is, returning to step S10;

[0128] In step S34, when it is determined that the average difference is smaller than the second threshold, the strategy table is output.

[0129] In this Figure 4 In the method shown, step S30 can be used to determine the security level coefficient of the honeypot state of the last response action after each attack session in the policy table ends. The security level system can be related to determining the security level of the honeypot state. The specific calculation method of the security level coefficient can be a variety of forms known to those skilled in the art. In one example of the present invention, the calculation method of the security level coefficient can be, for example, to construct a standard matrix of the authority levels of multiple ports, and then describe the honeypot state through the security matrix, thereby forming a security level matrix representing the honeypot state. On this basis, the security level coefficient is determined by respectively calculating the distance between the security level matrix and the preset multiple standard security level matrices. Steps S31 to S33 are used to determine whether the security level coefficient of the honeypot state meets the requirements, wherein step S31 is used to determine whether the minimum value of the security level coefficient meets the requirements, in order to ensure that the system will not temporarily enter a high-risk state when executing the policy table. Step S33 is used to determine whether the average difference of the security level coefficient meets the requirements, in order to avoid excessive fluctuations in the security authority of the system.

[0130] In yet another aspect, the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, wherein the instructions are used to be read by a machine so that the machine executes any of the access methods described above.

[0131] Through the above technical solution, the embodiment of the present invention provides an intelligent security access method and system based on an adaptive honeypot, which realizes a step-by-step response to the attack session by configuring corresponding response actions for the commands in each attack session. Compared with the policy group solution of the prior art, the access method and system provided by the present invention have higher response efficiency.

[0132] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0133] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes 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 processor, 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.

[0134] 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.

[0135] 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 A step that specifies a function in one or more boxes.

[0136] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0137] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0138] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0139] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0140] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. An intelligent and secure access method based on adaptive honeypot, It is characterized in that The access method comprises: Initialize the policy table based on the response action set and the honeypot status; Get the attack session that needs to be read; receiving commands of the attack session in order; Selecting a corresponding response action according to the command; Update the honeypot state according to the response action; Adding the response action and honeypot status to the policy table; Determining whether the command is an automated instruction; In the case where it is determined that the command is an automated instruction, returning to the step of sequentially receiving commands of the attack session; In the case where it is determined that the command is not an automated instruction, determining whether there is currently an unread attack session; In the case where it is determined that there is currently an attack session that has not been read, returning to the step of obtaining the attack session that needs to be read; In the case of determining that there is no unread attack session currently, generating the policy table; The strategy table is used as the response protection strategy of the current server.

2. The access method according to claim 1, It is characterized in that Selecting a corresponding response action according to the command includes: Calculate the reward function corresponding to each response action; The response action is selected according to the reward function.

3. The access method according to claim 2, It is characterized in that Calculate the reward function corresponding to each response action, including: Encoding the command, the response action, and the honeypot state to obtain an encoding vector; Inputting the encoding vector into the adversarial network to obtain the state of the honeypot at the next moment; The honeypot state at the next moment is compared with the current honeypot state to obtain a reward function for the response action.

4. The access method according to claim 3, It is characterized in that Inputting the encoding vector into the adversarial network to obtain the honeypot state at the next moment includes: According to formula (1), the state of the honeypot at the next moment is determined. Q t+1 (S t ,a t )=Q t (S t ,a t )+α[r t +γmax(Q t (S t ,A)-Q t (S t ,a t ))], (1) Among them, Q t+1 (S t , a t ) is the state value of the honeypot state, S t is the honeypot state, a t is the response action, α is the learning rate, γ is the preset discount factor, r t is the immediate reward function, and A is the response action set.

5. The access method according to claim 2, It is characterized in that Selecting the response action according to the reward function includes: Select the first n response actions according to the order of the reward function from large to small; Determining a transition probability of the selected response action; The response action is selected according to the transition probability.

6. The access method according to claim 5, It is characterized in that Determining the transition probability of the selected response action includes: The transition probability is determined according to formula (2) to formula (4), p i =p min +(n-i)Δp, (2) p min ≠0, (4) Among them, p i is the transition probability of the i-th response action selected in sequence, p min is the preset benchmark probability value, and Δp is the gradient probability.

7. The access method according to claim 1, It is characterized in that Generating the strategy table includes: Determine the security level coefficient of the honeypot state of the last response action after each attack session in the strategy table ends; Determining whether the security level coefficient is greater than a preset first threshold; In the case where it is determined that the security level coefficient is greater than the first threshold, returning to the step of initializing a policy table based on a response action set and a honeypot state; In the case where it is determined that the security level coefficient is less than or equal to the first threshold, determining an average difference of the security level coefficients; Determining whether the average difference is greater than or equal to a preset second threshold; In the case where it is determined that the average difference is greater than or equal to the second threshold, returning to the step of initializing the policy table based on the response action set and the honeypot state; When it is determined that the average difference is smaller than the second threshold, the strategy table is output.

8. An intelligent security access system based on adaptive honeypot, It is characterized in that The access system comprises a processor, and the processor is configured to execute the access method according to any one of claims 1 to 7.

9. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores instructions, and the instructions are used to be read by a machine so that the machine executes the access method according to any one of claims 1 to 7.