Intelligent password automatic extraction and secure storage method

Abstract

The application discloses an intelligent password automatic extraction and safe storage method, comprising the following steps: dynamically evaluating environment risk by analyzing environment parameters and historical user behavior data, and intelligently adjusting password extraction and storage strategies according to the risk level; the application models user behavior through a hidden Markov model, identifies abnormal behavior in real time and stops the unsafe password extraction process, thereby effectively improving the intelligence and safety of password management, improving the adaptability and safety of the password extraction process, and guaranteeing the reliability of local and cloud storage through a password migration mechanism, and the application has high safety, flexibility and operability, and is suitable for digital platforms and intelligent devices with high safety requirements.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of password extraction and secure storage processing, and particularly relates to an intelligent password automatic extraction and secure storage method. BACKGROUND

[0002] With the widespread application of information technology, password protection has become one of the important research topics in the field of information security. As the core means of digital identity verification, passwords are widely used in network services, electronic payments, social platforms and other fields. However, with the continuous development of network attack technology, the security and reliability of traditional password management methods are facing severe challenges. Especially in the process of password storage and extraction, traditional methods often rely too much on user's active operation and memory, and there are significant shortcomings in the dynamic detection of environmental risks and abnormal behaviors. Most of the existing technologies focus on the encrypted storage and access control of passwords, and do not fully consider the influence of environmental changes and behavior patterns on password security during the password extraction process, resulting in certain security vulnerabilities in the password storage process.

[0003] Existing intelligent password management systems mostly manage and store passwords based on static rules, which have low security and adaptability. For example, although device-based storage encryption and user identity verification mechanisms can prevent external attacks, they ignore the dynamic changes of environmental risk factors such as user location, device network status, and surrounding light, which have a potential impact on password security. In addition, most of the existing intelligent password extraction methods only rely on user behavior patterns or device status, and fail to dynamically adjust according to the risk level of the environment during password extraction. Therefore, when facing complex usage scenarios and changing security requirements, existing technologies often fail to provide adequate security protection. SUMMARY

[0004] This section is intended to summarize some aspects of the embodiments of the present application and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the abstract and title of the specification to avoid obscuring the purpose of this section, the abstract and the title, and such simplifications or omissions cannot be used to limit the scope of the present application.

[0005] In view of the above existing problems, the present application is proposed. Therefore, the present application provides an intelligent password automatic extraction and secure storage method to solve the problems proposed in the background art.

[0006] To solve the above technical problems, the present application provides the following technical scheme: an intelligent password automatic extraction and secure storage method, comprising:

[0007] Obtaining a current environment parameter, quantifying the environment parameter, and assigning a corresponding weight value to the current environment parameter based on a historical user behavior model or a historical user behavior data set to obtain a total risk value of the current environment parameter;

[0008] According to the total risk value of the current environment parameter, an environment risk level is determined, and a user password is extracted by using a user memory benchmark and a current user behavior, and the extracted user password is stored according to the current environment risk level;

[0009] The extraction process of the user password is modeled by using a hidden Markov model, and whether the current user behavior is abnormal is identified.

[0010] As a preferred scheme of the intelligent password automatic extraction and safe storage method, the environment parameter includes:

[0011] The current user location, the time period of the current user access, the light intensity around the current user, and the network state when the current user accesses.

[0012] As a preferred scheme of the intelligent password automatic extraction and safe storage method, the quantification of the environment parameter and the assignment of the corresponding weight value to the current environment parameter based on the historical user behavior model or the historical user behavior data set to obtain the total risk value of the current environment parameter include:

[0013] The fluctuation of each environment parameter is considered by using a fuzzy comprehensive evaluation method, and the risk coefficient of each environment parameter is initialized, and the fluctuation is obtained by the standard deviation of each environment parameter;

[0014] The historical user behavior model or the historical user behavior data set is called to obtain the historical weight assigned by the user behavior, and the weight value of each environment parameter is adjusted according to the fluctuation, the risk coefficient, and the historical weight assigned by the user behavior;

[0015] According to the risk value of each environment parameter and the weight value of the environment parameter, a risk determination matrix is established;

[0016] The risk determination matrix is calculated by using a product sum method to obtain the total risk value of the current environment parameter.

[0017] As a preferred scheme of the intelligent password automatic extraction and safe storage method, according to the total risk value of the current environment parameter, the environment risk level is determined, including:

[0018] According to the total risk value of the current environment parameter, the highest risk value and the lowest risk value are defined;

[0019] when the total risk value of the current environment parameter exceeds a defined maximum risk value, determining that the current environment risk level is high risk;

[0020] when the total risk value of the current environment parameter is between a defined minimum risk value and a defined maximum risk value, determining that the current environment risk level is medium risk;

[0021] when the total risk value of the current environment parameter is below the defined minimum risk value, determining that the current environment risk level is low risk.

[0022] As a preferred scheme of the intelligent password automatic extraction and secure storage method, the user password is extracted by using the user memory benchmark and the current user behavior according to the environment risk level, including:

[0023] The user memory benchmark is used as an initial user password, and the initial user password is encrypted;

[0024] When the environment risk level is low risk, the user password is extracted according to the encrypted initial user password and the random number generated by the QRNG;

[0025] When the environment risk level is high risk, the user password is extracted according to the combination of the first 8 bits of the encrypted initial user password and the Fourier series of the sliding speed;

[0026] When the environment risk is medium risk, the user password is extracted by introducing environmental entropy into the encrypted initial user password according to the environment parameter.

[0027] As a preferred scheme of the intelligent password automatic extraction and secure storage method, the extracted user password is stored according to the current environment risk level, including:

[0028] The extracted user password is encrypted, and the encrypted user password is stored locally;

[0029] When the environment risk level changes from low to high, the user password migration mechanism is triggered, the local storage device is verified by signature, the user password in the local storage device that passes the signature verification is decrypted, the decrypted user password is stored in the cloud, and then the user password in the local storage device is destroyed.

[0030] As a preferred scheme of the intelligent password automatic extraction and secure storage method, the extraction process of the user password is modeled by using a hidden Markov model to identify whether the current user behavior is abnormal, including:

[0031] Define the current user behavior as a state variable of a hidden Markov model, and define the environment parameter as an observation value of the hidden Markov model;

[0032] By training the hidden Markov model, the observation probability of each state variable under the environment parameter and the transition probability between the state variables are generated, and the normal operation chain of the current user behavior is obtained;

[0033] The K-L divergence is used to identify the normal operation chain of the current user behavior, and when the mean value of the K-L divergence identification result plus the standard deviation of the identification result is exactly a multiple of the identification result, it indicates that the current user behavior is abnormal, and the user password extraction process is terminated.

[0034] Compared with the prior art, the application has the following beneficial effects:

[0035] 1. By introducing the comprehensive analysis of the environment parameter and the user historical behavior model, the risk level of the current environment can be dynamically evaluated, and the password extraction and storage strategy can be intelligently adjusted according to the risk level, which effectively avoids the security risks caused by ignoring the environmental changes in the traditional password management, and enhances the overall security of password extraction and storage.

[0036] 2. The hidden Markov model is used to accurately identify whether the current user behavior is abnormal, and the password extraction process is terminated immediately when the abnormality is identified, which effectively prevents the security risks caused by user behavior abnormalities or vulnerabilities in the password extraction process, and improves the intelligence and adaptability of the password management system. BRIEF DESCRIPTION OF DRAWINGS

[0037] In order to more clearly illustrate the technical solutions of the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor. Among them:

[0038] Figure 1 The overall flowchart of the intelligent password automatic extraction and secure storage method according to an embodiment of the application is described.

[0039] Figure 2 The dynamic environment risk assessment level distribution diagram of the intelligent password automatic extraction and secure storage method according to an embodiment of the application is described.

[0040] Figure 3 The HMM model abnormality detection ROC curve diagram of the intelligent password automatic extraction and secure storage method according to an embodiment of the application is described. DETAILED DESCRIPTION

[0041] In order to make the above objectives, characteristics and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application are described in detail below with reference to the drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work should fall within the protection scope of the present application.

[0042] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. The present application, however, can be practiced in a variety of ways beyond the specific details set forth herein without departing from the scope of the present application. It can be appreciated by those skilled in the art that the present application can be practiced without such specific details.

[0043] Secondly, "one embodiment" or "an embodiment" referred to herein means that a specific feature, structure or characteristic described can be included in at least one implementation of the present application. "In one embodiment" appearing in different places in the specification does not mean the same embodiment, nor does it mean that the embodiment is separate or alternative to other embodiments.

[0044] The present application is described in detail with reference to the drawings. In the detailed description of the embodiments of the present application, the cross-sectional view of the device structure is partially enlarged without the general proportion for the convenience of description, and the schematic view is only an example, which should not limit the scope of protection of the present application. In addition, three-dimensional spatial dimensions including length, width and depth should be included in actual manufacture.

[0045] Meanwhile, in the description of the present application, it should be noted that the terms "upper, lower, inner and outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the indicated device or element must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application. In addition, the terms "first, second or third" are only for the purpose of description, and cannot be understood as indicating or implying relative importance.

[0046] In the present application, unless otherwise explicitly specified and limited, the terms "mounting, connecting, connection" should be understood broadly, for example: it can be fixed connection, detachable connection or integral connection; it can also be mechanical connection, electrical connection or direct connection, it can also be indirectly connected through intermediate medium, or it can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0047] Example 1

[0048] Reference Figure 1For the first embodiment of the application, the embodiment provides an intelligent password automatic extraction and secure storage method, comprising:

[0049] S1, obtain the current environment parameter, quantize the environment parameter, and assign a corresponding weight value to the current environment parameter based on the historical user behavior model or the historical user behavior data set, to obtain the total risk value of the current environment parameter;

[0050] Specifically, the environment parameter is obtained in real time by various sensors and devices;

[0051] Specifically, the environment parameter includes the current user position (the user's current coordinates are obtained by GPS / base station positioning to obtain the user's current position), the time period of the current user's access (the time period of the user's current access is obtained by the current system time), the light intensity around the current user (light data is obtained by light sensor / environmental light sensor), and the network state when the current user accesses (obtained by monitoring the network bandwidth, delay and signal strength of the device, etc.);

[0052] It should be explained that the light intensity around the current user can be used as a potential security indicator, for example, in the case of weak light intensity, it means that the user is in a dim environment, and the use of the device in this environment is not common, combined with the light intensity to evaluate the user's behavior, the abnormal behavior can be effectively identified; For example, the user performs high-frequency password input operation in a very dark environment, which needs to pay attention to the potential risk of this behavior;

[0053] Further, the quantization method considers the volatility generated by each environment parameter by fuzzy comprehensive evaluation method, and initializes the risk coefficient of each environment parameter;

[0054] It should be noted that the volatility is obtained from the standard deviation of each environment parameter;

[0055] Further, the historical user behavior model or the historical user behavior data set is called to obtain the historical weight assigned to the user behavior, and the weight value of each environment parameter is adjusted according to the volatility, the risk coefficient and the historical weight assigned to the user behavior;

[0056] Specifically, the user behavior data set is collected by the log recording module, and is converted into a standardized data format (such as JSON, CSV or database record), the data collected by the log recording module is uploaded to the database for summarization, and the historical user behavior data set is obtained;

[0057] Specifically, load the trained historical user behavior model from local storage or cloud storage, where the loading process uses a machine learning framework (such as TensorFlow, PyTorch, Scikit-learn, etc.), reads the historical user behavior model from the database into memory, and initializes it to obtain a historical user behavior dataset;

[0058] Specifically, the weight value of each adjusted environmental parameter which can be represented by the following formula:

[0059]

[0060] wherein, is the standard deviation of each environmental parameter, is the risk coefficient of each environmental parameter (such as location sensitivity coefficient, network risk coefficient, etc.), represents the historical weight (i.e. the weight assigned according to the user's historical behavior);

[0061] Further, according to the risk value of each environmental parameter and the weight value of the environmental parameter, a risk judgment matrix is established;

[0062] Specifically, the risk judgment matrix is represented as:

[0063]

[0064] wherein, represents the risk value of each environmental parameter;

[0065] Further, the total risk value of the current environmental parameter is obtained by calculating the risk judgment matrix through the product sum method;

[0066] Specifically, for each environmental parameter i, the obtained weight value and the corresponding risk value are multiplied and summed to further obtain:

[0067]

[0068] It should be noted that combining the light intensity as an environmental parameter with the user's historical behavior can more comprehensively evaluate the overall environmental risk, and the purpose is to help the system provide more accurate judgment basis in the process of multi-dimensional comprehensive evaluation of user behavior and environmental changes, improve system security and reduce potential risks;

[0069] S2, determining the environment risk level according to the total risk value of the current environment parameter, extracting the user password by using the user memory benchmark and obtaining the current user behavior according to the environment risk level, and storing the extracted user password according to the current environment risk level;

[0070] It should be explained that since the fuzzy comprehensive evaluation method has been performed on the environment parameters with the greatest impact before the environment risk level is defined, the total risk value of the current environment parameter obtained here is equivalent to the risk range value, so it is not necessary to define rules or thresholds again, and the maximum and minimum values of the risk range value can be obtained;

[0071] Further, the highest risk value and the lowest risk value are defined according to the total risk value of the current environment parameter, and the rule definition of the environment risk level is given as follows:

[0072] Rule 1: When the total risk value of the current environment parameter exceeds the defined highest risk value, the current environment risk level is determined to be high risk;

[0073] Rule 2: When the total risk value of the current environment parameter is between the defined lowest risk value and the defined highest risk value, the current environment risk level is determined to be medium risk;

[0074] Rule 3: When the total risk value of the current environment parameter is lower than the defined lowest risk value, the current environment risk level is determined to be low risk;

[0075] For example, assuming that we have three environment parameters: location, light, and network status, their dynamic weights and risk values are as follows:

[0076] Location: , (for example, the user location deviates for the first time, i.e., appearing in an unusual place);

[0077] Light: , (for example, the light intensity is weak);

[0078] Network status: , (for example, the current network environment is poor);

[0079] Then R=(0.6x0.8)+(0.3x0.5)+(0.1x0.7)=0.7;

[0080] If the highest risk value is 0.8 and the lowest risk value is 0.5 at this time, according to the calculated 0.7, it can be determined that the current environment risk level is medium risk;

[0081] Further, the user memory reference is used as the initial user password, and the initial user password is encrypted, and the user password extraction method for each environment risk level is defined as follows:

[0082] It needs to be explained that the user memory reference refers to a string or phrase that is easy to remember and has personalized features, such as "user name in upper and lower case + current birth year", "user's favorite person", "user's personal experience", etc.; Its main role is to provide a reproducible input for the system initial user password generation, while avoiding the problem that the random password generated by the system is easy to forget;

[0083] When the environment risk level is low, the user password is extracted according to the encrypted initial user password and the random number generated by QRNG;

[0084] It needs to be explained that QRNG is a quantum random number generator, and the random number generated by it is unpredictable;

[0085] Specifically, the AES encryption algorithm is used, and the random number generated by QRNG is used as the salt value to encrypt the password for storage; When extracting the password, the encrypted user password and the salt value are decrypted;

[0086] When the environment risk level is high, the user password is extracted according to the first 8 bits of the encrypted initial user password and the Fourier coefficient combination of the sliding speed;

[0087] Specifically, the encryption method is similar to the above encryption method, and the AES encryption algorithm is used, and the Fourier coefficient of the sliding speed is used to increase the security of the encryption of the user password in the high-risk environment level; When extracting the password, the key generated by the AES encryption algorithm and the initial user password are decrypted;

[0088] When the environment risk is medium, the user password is extracted according to the encrypted initial user password by introducing environmental entropy through environmental parameters;

[0089] Specifically, the AES encryption algorithm is used, the key of the initial user password is retained, the environmental parameters are spliced into a string, the environmental parameters after splicing into a string are hashed to obtain environmental entropy, and the initial user password and the environmental entropy are combined to obtain the encrypted password; When the password needs to be extracted, the key of the initial user password and the initial user password are decrypted;

[0090] Further, if the extracted user password is encrypted, the encrypted user password needs to be stored locally;

[0091] Further, when the environmental risk level changes from low to high, a user password migration mechanism is triggered, the local storage device is verified by signature, the user password in the local storage device that passes the signature verification is decrypted (each decryption operation is as described above), the decrypted user password is stored to the cloud, and then the user password in the local storage device is destroyed;

[0092] It should be noted that the signature verification ensures that only the verified device can decrypt the password, not only protecting the security of the local storage, but also avoiding the security problem caused by the loss or damage of the device by migrating the password to the cloud; by storing the password in the cloud, the system can further improve the reliability and recoverability of the data, ensuring that the user data can be safely accessed in any device environment; in addition, the local storage password is destroyed during the password migration process, which greatly reduces the risk of password leakage when the local device is stolen;

[0093] S3, using a hidden Markov model to model the extraction process of the user password, to identify whether the current user behavior is abnormal;

[0094] It should be noted that the hidden Markov model (Hidden Markov Model, HMM) is a statistical model used to describe an observation sequence generated by a hidden state sequence;

[0095] Further, the current user behavior is defined as a state variable of the hidden Markov model, and the environmental parameter is defined as an observation value of the hidden Markov model;

[0096] Further, by training the hidden Markov model, the observation probability of each state variable under the environmental parameter and the transition probability between the state variables are generated, and the normal operation chain of the current user behavior is obtained;

[0097] Specifically, the observation probability of the state variable, for example, the user is currently at home, and the observed user location data is the probability of the frequently appearing location near the home; and the transition probability between the state variables, for example, the probability of the user transferring from home to office; the generated normal operation chain of the user behavior, for example, the user frequently transfers from home to office in the past few days, but suddenly jumps from home to a strange area (for example, a strange place, that is, a user location in an environmental parameter that has never appeared) today;

[0098] Further, the K-L divergence is used to identify the normal operation chain of the current user behavior, and when the mean value of the K-L divergence identification result plus the standard deviation of the identification result is exactly a multiple of the identification result, it means that the current user behavior is abnormal, and the user password extraction process is terminated;

[0099] It needs to be explained that the K-L divergence (Kullback-Leibler Divergence) is an index used to measure the difference between two probability distributions, and if the transition probability of the current state deviates too much from the transition probability of the normal behavior of the user, the value of the K-L divergence will be higher, indicating abnormal behavior;

[0100] Specifically, according to the 68-95-99.7 rule (a characteristic of normal distribution), about 68% of data points will fall within the range of mean ± 1 standard deviation, 95% of data points will fall within the range of mean ± 2 standard deviations, and 99.7% of data points will fall within the range of mean ± 3 standard deviations; Therefore, setting the threshold value as the mean plus the multiple of the standard deviation can effectively distinguish the abnormal behavior of the user when extracting the password, especially when the distribution of K-L divergence is close to the normal distribution;

[0101] It should be noted that in addition, the size of the multiple determines the sensitivity of detecting abnormal behavior, the smaller the multiple, the more stringent the system is in judging abnormal behavior, that is, when the K-L divergence deviates slightly from the mean, it will trigger an abnormal behavior warning, and the larger the multiple, the more relaxed the system is in judging abnormal behavior.

[0102] Embodiment 2

[0103] Reference Figure 2 and Figure 3 For the second embodiment of the present application, the embodiment provides an intelligent password automatic extraction and secure storage method, comprising: in order to verify the effectiveness of the intelligent password automatic extraction and secure storage method proposed by the present application, a set of experimental comparison schemes are designed to compare the performance differences of the present application and prior art under different environmental conditions; The experiment selects three typical environmental risk levels (low risk, medium risk, high risk), and simulates the user password extraction process under each risk level;

[0104] Experimental equipment and environment setting: the experiment uses a smart phone of "SecureTech" brand (model: STX2025) as the user equipment, which is equipped with environmental parameter sensors (including GPS, light intensity sensor, network status monitor, etc.) and encryption algorithm module; An encryption application based on Android operating system is used, which uses AES encryption algorithm and QRNG (quantum random number generator) to generate random numbers; The system has integrated a hidden Markov model (HMM) for abnormal behavior detection, and dynamically adjusts the password extraction strategy according to the environmental parameters (location, light, network status, etc.);

[0105] Experimental environment: low-risk environment: the user is in the office, the network state is stable, the light intensity is moderate, and the location is known; medium-risk environment: the user is in a public place, the network fluctuates greatly, the light changes rapidly, and the location is relatively unfamiliar; high-risk environment: the user is in an area with insufficient light, the network signal is unstable, and the location is unknown or abnormal;

[0106] Experimental steps: real-time acquisition of environmental parameters through sensors and recording of user behavior data; then the system uses fuzzy comprehensive evaluation method to quantify each environmental parameter and calculate the total risk value of the current environment to determine the risk level (low, medium and high); according to the change of the environmental risk level, the system uses QRNG random number encryption, Fourier coefficient encryption and environmental entropy encryption to extract the user password and store it in the local device and the cloud to ensure the security of the experimental data; analyze the user behavior through hidden Markov model and detect whether there is an anomaly, if there is an anomaly, the password extraction process is terminated; part of the test data results are shown in Table 1:

[0107] Experimental subjects Environmental risk level Password extraction method Password security Password storage security Abnormal behavior detection rate Prior art Low risk Based on fixed password and simple encryption 65% Local storage 70% Medium risk Based on user behavior model and encryption 70% Local storage 75% High risk Simple encryption and single authentication 50% Local storage 60% Invention scheme Low risk AES encryption + QRNG random number 95% Local + cloud storage 95% Medium risk AES encryption + environmental entropy 90% Local + cloud storage 90% High risk AES encryption + sliding speed Fourier coefficient 98% Local + cloud storage 98%

[0108] From the test data results in Table 1, it can be seen that the present application is better than the prior art in terms of password security comparison, password storage security comparison and user abnormal behavior detection, and by Figure 2 , the gray heat map and contour line show the joint influence of light intensity and network delay on risk level in the environmental parameters, wherein the weaker the light, the higher the risk, and the higher the network delay, the higher the risk, combined with dynamic weight distribution, the high-risk area (dark color) is more consistent with the actual scene, which is better than the traditional fixed threshold method; in addition, Figure 3 It can be seen that the ROC curve shows that the AUC of the HMM model of the present application is close to 1, which is significantly better than the traditional threshold method, when the threshold is set to mean+3σ, (K-L divergence=1.2), TPR=95% and FPR<5%, which verifies the high sensitivity and specificity of the HMM model in abnormal behavior detection;

[0109] In summary, through comparative experiments, the feasibility and beneficial effects of the present application are fully proved.

[0110] Those skilled in the art will appreciate that embodiments of the present application can be readily used as software, hardware, or a combination of software and hardware. In a software embodiment, various elements are implemented in hardware, software, or a combination of both hardware and software. In a software embodiment, the software implementation can be in a modular or a non- modular form. One skilled in the art will appreciate that the software implementation can be stored and transported on any computer readable medium, such as a compact diskette, a CD, a semiconductor memory, or a hard disk. A computer readable medium having such software or programs recorded on the medium is also within the scope of the present application. The software implementations of the present application are enabled to be accessed from the storage, to be loaded into and executed by a computer, or to be bypassed (in the case of a bypass mechanism).

[0111] The present application is described in relation to flow charts and / or block diagrams that are presented according to the methods, apparatus (systems) and computer program products of embodiments of the present application. It is to be understood that each block of the flow charts and / or block diagrams, and combinations of blocks in the flow charts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flow charts and / or block diagrams block or blocks. Figure 1 Figure 1

[0112] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the flow charts and / or block diagrams block or blocks. Figure 1 Figure 1

[0113] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the flow charts and / or block diagrams block or blocks. Figure 1 Figure 1

[0114] While the preferred embodiments of the application have been described, additional variations and modifications can be employed by those skilled in the art. Therefore, the appended claims are intended to cover all such variations and modifications as falling within the scope of the application. ​​​​​​

[0115] Obviously, numerous modifications and variations of the present application are possible in light of the above teachings. It is therefore to be understood that within the scope of the appended claims and their equivalents, the application can be practiced otherwise than as specifically described.

Claims

1. A method for intelligent automatic password extraction and secure storage, characterized in that, include: Obtain the current environmental parameters, quantify the environmental parameters, and assign corresponding weight values ​​to the current environmental parameters based on the historical user behavior model or historical user behavior dataset to obtain the total risk value of the current environmental parameters; The environmental parameters are quantified, and based on historical user behavior models or historical user behavior datasets, corresponding weight values ​​are assigned to the current environmental parameters to obtain the total risk value of the current environmental parameters, including: The fuzzy comprehensive evaluation method is used to consider the volatility of each environmental parameter and initialize the risk coefficient of each environmental parameter. The volatility is obtained by the standard deviation of each environmental parameter. By calling the historical user behavior model or inputting the historical user behavior dataset, the historical weights of user behavior are obtained. Based on the volatility, risk coefficient, and historical weights of user behavior, the weight values ​​of each environmental parameter are adjusted. A risk assessment matrix is ​​established based on the risk value of each environmental parameter and the weight value assigned to each environmental parameter; The total risk value of the current environmental parameters is obtained by calculating the risk judgment matrix through product summation. Based on the total risk value of the current environmental parameters, the environmental risk level is determined. Using the environmental risk level, the user's password is extracted by utilizing the user's memory benchmark and obtaining the current user behavior. At the same time, the extracted user password is stored according to the current environmental risk level. Based on the aforementioned environmental risk level, the user's password is extracted using the user's memory benchmark and current user behavior, including: The user's memory reference is used as the initial user password, and the initial user password is encrypted. When the environmental risk level is low, the user password is extracted based on the encrypted initial user password and the random number generated by QRNG. The password is encrypted using the AES encryption algorithm and stored using the random number generated by QRNG as a salt value. When the password is extracted, it is decrypted by combining the encrypted user password with the salt value. When the environmental risk level is high, the user password is extracted based on the first 8 digits of the encrypted initial user password and the Fourier coefficients of the sliding speed. The encryption method is the same as the AES encryption algorithm used in the low-risk scenario. When extracting the password, it can be decrypted using the key generated by the AES encryption algorithm and the initial user password. When the environmental risk is medium, the user password is extracted by introducing environmental entropy through environmental parameters based on the encrypted initial user password. Specifically, the key of the initial user password is retained through AES encryption algorithm, the environmental parameters are concatenated into a string, and the concatenated environmental parameters are hashed to obtain the environmental entropy. The initial user password and the environmental entropy are combined to obtain the encrypted password. When it is necessary to extract the password, it can be decrypted using the key of the initial user password and the initial user password. Hidden Markov Models are used to model the process of extracting user passwords and identify whether there are any anomalies in the current user behavior.

2. The intelligent password automatic extraction and secure storage method as described in claim 1, characterized in that, The environmental parameters include: The current user's location, the time period during which the current user accessed the site, the ambient light intensity around the current user, and the network status during the current user's access.

3. The intelligent password automatic extraction and secure storage method as described in claim 1, characterized in that, Based on the total risk value of the current environmental parameters, the environmental risk level is determined, including: Define the highest and lowest risk values ​​based on the total risk value of the current environmental parameters; If the total risk value of the current environmental parameters exceeds the defined maximum risk value, the current environmental risk level is determined to be high risk. If the total risk value of the current environmental parameter is between the defined minimum risk value and the defined maximum risk value, then the current environmental risk level is determined to be medium risk. If the total risk value of the current environmental parameters is lower than the defined minimum risk value, then the current environmental risk level is determined to be low risk.

4. The intelligent password automatic extraction and secure storage method as described in claim 1, characterized in that, Simultaneously, based on the current environmental risk level, the extracted user passwords are stored, including: The extracted user passwords are encrypted, and the encrypted user passwords are stored locally. When the environmental risk level changes from low to high, the user password migration mechanism is triggered. The signature verification is performed on the local storage device, the user password in the local storage device that has passed the signature verification is decrypted, the decrypted user password is stored in the cloud, and then the user password in the local storage device is destroyed.

5. The intelligent password automatic extraction and secure storage method as described in claim 1 or 2, characterized in that, Using a Hidden Markov Model (HMM), the process of extracting user passwords is modeled to identify whether there are any anomalies in the current user behavior, including: Define the current user behavior as a state variable of the Hidden Markov Model, and define the environmental parameters as an observation of the Hidden Markov Model; By training a Hidden Markov Model, the observation probability of each state variable under environmental parameters and the transition probability between state variables are generated to obtain the normal operation chain of the current user behavior. The normal operation chain of the current user behavior is identified by KL divergence. When the mean of the KL divergence identification result plus the standard deviation of the identification result is exactly a multiple of the identification result, it indicates that the current user behavior is abnormal, and the user password extraction process is terminated.

Images