User authentication methods, electronic devices and media

Abstract

This disclosure provides a user authentication method, electronic device, and medium. The method includes: acquiring authentication information input through a terminal and authentication action information indicating the input action when the authentication information is input; inputting the authentication information and authentication action information into a target authentication detection model to obtain a target authentication verification code output by the target authentication detection model; and sending authentication request information, which includes the authentication information and the target authentication verification code. This technical solution enables the receiving end to compare the authentication information with the target authentication verification code in the authentication request information to determine whether the authentication information in the authentication request information is the authentication information that meets the authentication requirements input by a specific user on a terminal carrying the target authentication detection model, thereby improving the reliability of the user authentication process and enhancing the user experience.

Description

Technical Field

[0001] This disclosure relates to the field of authentication technology, specifically to a user authentication method, electronic device, and medium. Background Technology

[0002] In recent years, with the development of communication technology, people's reliance on the internet has been increasing in all aspects of daily life. In daily use, before a terminal runs sensitive applications such as financial trading or social applications, it is necessary to authenticate the user to determine if they have the necessary permissions to use specific services through the applications running on the terminal.

[0003] In related technologies, servers can store usernames and passwords in plaintext in server configuration files or databases. When a user wants to use a service through a terminal, they can send an authentication request containing the user's identifier and password to the corresponding server. When the server receives the authentication request, it can retrieve the corresponding password from its local configuration file or database based on the user identifier carried in the authentication request and compare it with the password carried in the authentication request. If the password matches the password carried in the authentication request, the server can return an authentication success notification to the terminal and provide the corresponding service to the user through the terminal. If the password does not match the password carried in the authentication request, the server can send an authentication failure notification to the terminal.

[0004] While the above scheme can achieve user authentication, the user's ID and password can be easily seen by others when the user enters them through the terminal, which may lead to the leakage of the user ID and password and reduce the reliability of user authentication. Therefore, how to improve the reliability of user authentication is an urgent problem to be solved. Summary of the Invention

[0005] To address the problems in the related technologies, embodiments of this disclosure provide a user authentication method, an electronic device, and a medium.

[0006] In a first aspect, this disclosure provides a user authentication method, which is applied to a terminal and includes:

[0007] Acquire authentication information input through the terminal and authentication action information used to indicate the input action when inputting authentication information;

[0008] Input the authentication information and authentication action information into the target authentication detection model to obtain the target authentication verification code output by the target authentication detection model.

[0009] Send authentication request information, which includes authentication information and the target authentication verification code.

[0010] In one implementation of this disclosure, the input action includes a touch action, and the authentication action information includes at least one of the touch trajectory, touch force, and touch area of ​​the touch action when inputting authentication information.

[0011] In one implementation of this disclosure, authentication information and authentication action information are input into the target authentication detection model to obtain the target authentication verification code output by the target authentication detection model, including:

[0012] Input the authentication information, authentication action information, and terminal device identification information into the target authentication detection model to obtain the target authentication verification code output by the target authentication detection model.

[0013] The authentication request includes device identification information, authentication information, and the target authentication verification code.

[0014] In one implementation of this disclosure, before obtaining the authentication information input through the terminal and the authentication action information used to indicate the input action when inputting the authentication information, the method further includes:

[0015] Acquire registration information input through the terminal and registration action information used to indicate the input action when entering registration information;

[0016] Send a registration request message, which includes registration information and device identification information;

[0017] Receive the initial authentication detection model and the target registration verification code sent by the registration server in response to the registration request information;

[0018] The initial authentication detection model is trained by taking registration information, registration action information, and device identification information as input and the target registration verification code as output, in order to obtain the target authentication detection model.

[0019] In one implementation of this disclosure, before obtaining the authentication information input through the terminal and the authentication action information used to indicate the input action when inputting the authentication information, the method further includes:

[0020] Acquire registration information input through the terminal and registration action information used to indicate the input action when entering registration information;

[0021] Send a registration request, which includes registration information and device identification information;

[0022] Receive the initial authentication detection model and the target registration verification code sent by the registration server in response to the registration request;

[0023] Obtain a private authentication detection model based on the initial authentication detection model;

[0024] Receive the updated weight parameters sent by the edge server, and update the private authentication detection model according to the updated weight parameters;

[0025] The updated private authentication detection model is trained by taking registration information, registration action information, and device identification information as input and the target registration verification code as output.

[0026] When the trained private authentication detection model fails to converge, the gradient update vector is obtained based on the trained private authentication detection model and sent. The edge server aggregates the gradient update vector and updates the weight parameters of the shared authentication detection model on the edge server based on the aggregated gradient update vector to obtain the updated weight parameters.

[0027] When the trained private authentication detection model converges, the target authentication detection model is obtained based on the trained private authentication detection model.

[0028] In one implementation of this disclosure, before obtaining the private authentication detection model based on the initial authentication detection model, the method further includes:

[0029] Receive private data upload instructions;

[0030] In response to a private data upload command, it sends registration information, registration action information, and device identification information;

[0031] Receive the initial weight parameters sent by the edge server;

[0032] Obtain a private authentication detection model based on the initial authentication detection model, including:

[0033] The initial authentication detection model is updated based on the initial weight parameters to obtain a private authentication detection model.

[0034] In one implementation of this disclosure, before sending registration information, registration action information, and device identification information in response to a private data upload instruction, the method further includes:

[0035] Noise is added to registration information, registration action information, and device identification information to obtain noisy registration information, noisy registration action information, and noisy device identification information;

[0036] Send registration information, registration action information, and device identification information, including:

[0037] Send the registration information after adding noise, the registration action information after adding noise, and the device identification information after adding noise.

[0038] In one implementation of this disclosure, before receiving the private data upload instruction, the method further includes:

[0039] The system sends user characteristic information, which includes the user's gender, age, education level, left / right hand preference, authentication method preference, and application preference. The edge server uses this user characteristic information to identify multiple target terminals and sends private data upload instructions to the target terminals.

[0040] In a second aspect, embodiments of this disclosure provide an electronic device, which includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the method of any of the first aspects.

[0041] Thirdly, this disclosure provides a computer-readable storage medium having computer instructions stored thereon, wherein the computer instructions, when executed by a processor, implement the method of any one of the first aspects.

[0042] The technical solutions provided in this disclosure may have the following beneficial effects:

[0043] In the above technical solution, by acquiring the authentication information input through the terminal and the authentication action information used to indicate the input action when inputting the authentication information, and inputting the authentication information and authentication action information into the target authentication detection model, the target authentication verification code output by the target authentication detection model is obtained, and authentication request information is sent. The authentication request information includes the authentication information and the target authentication verification code. Since the target authentication detection model can be understood as a model that exists only on this terminal and has learned the pattern between the authentication information, authentication action information, and the target authentication verification code, meaning that the target authentication verification code can only be generated by the target authentication detection model on the corresponding terminal based on the input action information corresponding to the input action and the authentication information when a specific user inputs authentication information that meets the authentication requirements using their usual input action, the receiving end can compare the target authentication verification code in the authentication request information to determine whether the authentication information in the authentication request information is the authentication information that meets the authentication requirements input by a specific user on the terminal carrying the target authentication detection model. This helps to improve the reliability of the user authentication process and improve the user experience.

[0044] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0045] Other features, objects, and advantages of this disclosure will become more apparent from the following detailed description of non-limiting embodiments, taken in conjunction with the accompanying drawings. In the drawings:

[0046] Figure 1 A schematic structural block diagram of a user authentication system according to an embodiment of the present disclosure is shown;

[0047] Figure 2 A schematic flowchart illustrating a user authentication method according to an embodiment of the present disclosure is shown;

[0048] Figure 3 A schematic flowchart illustrating a user authentication method according to an embodiment of the present disclosure is shown;

[0049] Figure 4 A schematic flowchart illustrating a user authentication method according to an embodiment of the present disclosure is shown;

[0050] Figure 5 A schematic flowchart illustrating a user authentication method according to an embodiment of the present disclosure is shown;

[0051] Figure 6 A schematic structural block diagram of a user authentication device according to an embodiment of the present disclosure is shown;

[0052] Figure 7 A schematic structural block diagram of an electronic device according to an embodiment of the present disclosure is shown;

[0053] Figure 8 A schematic diagram of the structure of a computer system suitable for implementing the method according to embodiments of the present disclosure is shown. Detailed Implementation

[0054] In the following, exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings to enable those skilled in the art to readily implement them. Furthermore, for clarity, portions unrelated to the description of the exemplary embodiments have been omitted from the drawings.

[0055] In this disclosure, it should be understood that terms such as “comprising” or “having” are intended to indicate the presence of features, figures, steps, behaviors, components, parts or combinations thereof disclosed in this specification, and do not preclude the possibility of the presence or addition of one or more other features, figures, steps, behaviors, components, parts or combinations thereof.

[0056] It should also be noted that, unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other. This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0057] The details of the embodiments of this disclosure are described in detail below through specific examples.

[0058] In recent years, with the development of communication technology, people's reliance on the internet has been increasing in all aspects of daily life. Typically, the terminals people use can access the internet through network devices located nearby. These terminals can include smart mobile communication terminals, wearable devices, desktop computers, laptops, tablets, etc., and can also include wireless routers, wired routers, wireless gateways, etc.

[0059] In daily use, before running sensitive applications such as financial trading applications or social applications on the terminal, it is necessary to authenticate the user of the terminal to determine whether the user has the permission to use specific services through the specific application running on the terminal.

[0060] For example, before a user makes an order payment using the payment function of a payment application running on a terminal, the user using the terminal needs to be authenticated to confirm that the user using the terminal and the account corresponding to the order payment belong to that user.

[0061] In related technologies, servers can store usernames and passwords in plaintext in server configuration files or databases. When a user wants to use a service through a terminal, they can send an authentication request containing the user's identifier and password to the corresponding server. When the server receives the authentication request, it can retrieve the corresponding password from its local configuration file or database based on the user identifier carried in the authentication request and compare it with the password carried in the authentication request. If the password matches the password carried in the authentication request, the server can return an authentication success notification to the terminal and provide the corresponding service to the user through the terminal. If the password does not match the password carried in the authentication request, the server can send an authentication failure notification to the terminal.

[0062] While the above scheme can achieve user authentication, the user's ID and password can be easily seen by others when the user enters them through the terminal, which may lead to the leakage of the user ID and password and reduce the reliability of user authentication. Therefore, how to improve the reliability of user authentication is an urgent problem to be solved.

[0063] In view of the above-mentioned deficiencies, the technical solution provided in this disclosure obtains the authentication information input by the terminal and the authentication action information used to indicate the input action when inputting the authentication information. The authentication information and authentication action information are then input into a target authentication detection model to obtain the target authentication verification code output by the target authentication detection model. An authentication request is then sent, and the authentication request information includes the authentication information and the target authentication verification code. Since the target authentication detection model can be understood as a model that exists only on the terminal and has learned the relationship between the authentication information, authentication action information, and the target authentication verification code—meaning the target authentication verification code can only be generated by the target authentication detection model on the corresponding terminal based on the input action information corresponding to the input action and the authentication information when a specific user inputs authentication information that meets the authentication requirements using their usual input action—the receiving end can compare the target authentication verification code in the authentication request information to determine whether the authentication information in the authentication request information is the authentication information that meets the authentication requirements input by the specific user on the terminal carrying the target authentication detection model. This helps improve the reliability of the user authentication process and enhances the user experience.

[0064] Figure 1 A schematic structural block diagram of a user authentication system according to an embodiment of the present disclosure is shown. The user authentication system includes a terminal 101, a server 102, and a network 103. The network 103 is a medium for providing a communication link between the terminal 101 and the server 102. The network 103 may include various connection types, such as wired or wireless communication links or fiber optic cables, etc.

[0065] Terminal 101 refers to various electronic devices with network access capabilities, including but not limited to mobile communication terminals, desktop computers, tablet computers, laptop computers, wearable devices, etc.

[0066] Server 102 can be a cloud server or a server provided by a user authentication service provider. It should be noted that in this embodiment of the application, one server 102 can correspond to one or more terminals 101. For example, a user authentication service provider can divide the area it governs into multiple blocks, and multiple terminals 101 in each block correspond to one server 102.

[0067] Figure 2 A schematic flowchart illustrating a user authentication method according to an embodiment of the present disclosure is shown, the user authentication method being applied to a terminal. Figure 2 As shown, the user authentication method includes the following steps:

[0068] In step S101, the authentication information input through the terminal and the authentication action information used to indicate the input action when inputting the authentication information are obtained.

[0069] In one embodiment of this disclosure, the authentication information input through the terminal can be understood as information used to indicate the identity of the user using the terminal. For example, the authentication information may include username, password, device identification code, etc.

[0070] In one embodiment of this disclosure, inputting authentication information can be understood as instructing the user to input the authentication information through a human-computer interaction device on the terminal. The human-computer interaction device may include a keyboard, mouse, microphone, touch screen, handwriting tablet, etc.

[0071] In one embodiment of this disclosure, authentication action information can be understood as indicating at least one feature of the input action when the authentication information is input through a human-computer interaction device on a terminal. For example, when the human-computer interaction device is a keyboard, the input action of inputting authentication information through the terminal's human-computer interaction device may include keyboard typing, and the authentication action information can be understood as indicating the frequency, speed, etc., of keyboard typing; when the human-computer interaction device is a mouse, the authentication action information can be understood as indicating the mouse's movement trajectory, movement speed, movement acceleration, etc.

[0072] It should be noted that the authentication action information used to indicate the input action when inputting authentication information can be obtained based on the process of the user inputting authentication information once or multiple times.

[0073] In step S102, the authentication information and authentication action information are input into the target authentication detection model to obtain the target authentication verification code output by the target authentication detection model.

[0074] In one embodiment of this disclosure, the target authentication detection model may be pre-stored in the terminal or obtained from other devices or systems.

[0075] The target authentication detection model can be a neural network (NN) model, a convolutional neural network (CNN) model, or a long short-term memory (LSTM) model, etc.

[0076] In step S103, an authentication request is sent, which includes authentication information and a target authentication verification code.

[0077] In one embodiment of this disclosure, sending authentication request information can be understood as sending it to the end that processes the authentication request via wired or wireless means, such as sending the authentication request information to an authentication server.

[0078] In one embodiment of this disclosure, the end receiving the authentication request information can be a cloud server or an authentication server provided by an authentication request processing service operator. Upon receiving the authentication request information, the receiving end can search for the authentication information within the request information to determine a pre-acquired verification code corresponding to that authentication information. This verification code is then compared with the target authentication verification code to determine if they are identical. If they are identical, it can be determined that the authentication information in the request information is authentication information that meets the authentication requirements, input by a specific user on a terminal carrying the target authentication detection model.

[0079] The technical solution provided in this disclosure obtains authentication information input from the terminal and authentication action information indicating the input action when inputting authentication information. This authentication information and authentication action information are then input into a target authentication detection model to obtain a target authentication verification code output by the target authentication detection model. An authentication request is then sent, including the authentication information and the target authentication verification code. Since the target authentication detection model can be understood as a model that exists only on the terminal and has learned the relationship between authentication information, authentication action information, and the target authentication verification code—meaning the target authentication verification code can only be generated by the target authentication detection model on the corresponding terminal when a specific user inputs authentication information that meets the authentication requirements using their usual input action—the target authentication detection model on that terminal generates the code based on the input action information corresponding to that input action and the authentication information. Therefore, the receiving end can compare the target authentication verification code in the authentication request information to determine whether the authentication information in the authentication request information is the same authentication information that a specific user input on the terminal carrying the target authentication detection model, thus improving the reliability of the user authentication process and enhancing the user experience.

[0080] In one implementation of this disclosure, the input action includes a touch action, and the authentication action information includes at least one of the touch trajectory, touch force, and touch area of ​​the touch action when inputting authentication information.

[0081] In one embodiment of this disclosure, the touch trajectory can be understood as the trajectory of the user's limb movement in contact with the touch screen. For example, the speed, acceleration, and position of the user's limb in contact with the touch screen at different times can be determined based on the touch trajectory.

[0082] In one embodiment of this disclosure, touch force can be understood as the value of the force of a user's limb pressing the touch screen at different touch times and different touch positions. The touch force can be multiple point values ​​or one or more continuous values.

[0083] In one embodiment of this disclosure, the touch area can be understood as a value used to indicate the area of ​​the user's limb in contact with the touch screen at different touch times and different touch positions. The touch area distribution information can be multiple point values ​​or one or more continuous values.

[0084] In the technical solution provided in this disclosure, by limiting the input action to include touch action, and the authentication action information to include at least one of the touch trajectory, touch force, and touch area of ​​the touch action when inputting authentication information, it can be ensured that the authentication action information can accurately reflect the characteristics of the input action, which helps to improve the reliability of the authentication process.

[0085] In one implementation of this disclosure, Figure 3 A schematic flowchart illustrating a user authentication method according to an embodiment of the present disclosure is shown, such as... Figure 3 As shown, in step S102, the authentication information and authentication action information are input into the target authentication detection model to obtain the target authentication verification code output by the target authentication detection model, which can be achieved through step S1021:

[0086] In step S1021, the authentication information, authentication action information, and device identification information of the terminal are input into the target authentication detection model to obtain the target authentication verification code output by the target authentication detection model.

[0087] The authentication request includes device identification information, authentication information, and the target authentication verification code.

[0088] In one embodiment of this disclosure, device identification information can be understood as being used to indicate a terminal device, and a unique corresponding terminal device can be determined based on the device identification information.

[0089] In the technical solution provided in this disclosure, by inputting authentication information, authentication action information, and terminal device identification information into the target authentication detection model to obtain the target authentication verification code output by the target authentication detection model, and by limiting the authentication request to include device identification information, authentication information, and target authentication verification code, it can be ensured that the end receiving the authentication request information can compare it with the target authentication verification code in the authentication request information to determine whether the authentication information in the authentication request information is the authentication information that meets the authentication requirements entered by a specific user on a specified terminal. This helps to improve the reliability of the user authentication process and improve the user experience.

[0090] In one implementation of this disclosure, Figure 4 A schematic flowchart illustrating a user authentication method according to an embodiment of the present disclosure is shown, such as... Figure 4As shown, before obtaining the authentication information input through the terminal and the authentication action information used to indicate the input action when inputting the authentication information in step S101, the method further includes steps S104-S107:

[0091] In step S104, the registration information input through the terminal and the registration action information used to indicate the input action when inputting the registration information are obtained;

[0092] In step S105, a registration request is sent, which includes registration information and device identification information.

[0093] In step S106, the initial authentication detection model and the target registration verification code sent by the registration server in response to the registration request information are received.

[0094] In step S104, the registration information, registration action information, and device identification information are used as inputs, and the target registration verification code is used as the output to train the initial authentication detection model in order to obtain the target authentication detection model.

[0095] In one embodiment of this disclosure, the registration information input through the terminal can be understood as information used to indicate the identity of the user using the terminal. For example, the registration information may include a username, password, device identifier, etc. It should be noted that the format of the registration information should be the same as the format of the authentication information.

[0096] In one embodiment of this disclosure, inputting registration information can be understood as instructing the user to input the registration information through a human-computer interaction device on a terminal. The human-computer interaction device may include a keyboard, mouse, microphone, touch screen, handwriting tablet, etc.

[0097] In one embodiment of this disclosure, registration action information can be understood as indicating at least one feature of the input action when the registration information is entered through a human-computer interaction device on a terminal. For example, when the human-computer interaction device is a keyboard, the input action of entering registration information through the terminal's human-computer interaction device may include keyboard typing, and the registration action information can be understood as indicating the frequency, speed, etc., of keyboard typing; when the human-computer interaction device is a mouse, the registration action information can be understood as indicating the mouse's movement trajectory, movement speed, movement acceleration, etc.

[0098] In one embodiment of this disclosure, sending registration request information can be understood as sending the authentication request information to the end that processes the registration request, namely the registration server, via wired or wireless means.

[0099] In one embodiment of this disclosure, the initial authentication detection model may be pre-stored in the terminal or obtained from other devices or systems.

[0100] The initial authentication detection model can be a neural network (NN) model, a convolutional neural network (CNN) model, or a long short-term memory (LSTM) model, etc.

[0101] The technical solution provided in this disclosure involves acquiring registration information input through a terminal and registration action information indicating the input action when entering the registration information; sending registration request information, which includes registration information and device identification information; receiving an initial authentication detection model and a target registration verification code from the registration server in response to the registration request information; and training the initial authentication detection model using the registration information, registration action information, and device identification information as input and the target registration verification code as output to obtain a target authentication detection model. This ensures that the target authentication detection model can learn the relationship between the registration information, registration action information, device identification information, and the target authentication verification code through training, facilitating the subsequent acquisition of the target authentication verification code based on the target authentication detection model by inputting authentication information, authentication action information, and device identification information.

[0102] In one implementation of this disclosure, Figure 5 A schematic flowchart illustrating a user authentication method according to an embodiment of the present disclosure is shown, such as... Figure 5 As shown, before obtaining the authentication information input through the terminal and the authentication action information used to indicate the input action when inputting the authentication information in step S101, the method further includes steps S108-S115:

[0103] In step S108, the registration information input through the terminal and the registration action information used to indicate the input action when inputting the registration information are obtained;

[0104] In step S109, a registration request is sent, which includes registration information and device identification information;

[0105] In step S110, the initial authentication detection model and the target registration verification code sent by the registration server in response to the registration request are received.

[0106] In step S111, a private authentication detection model is obtained based on the initial authentication detection model;

[0107] In step S112, the update weight parameters sent by the edge server are received, and the private authentication detection model is updated according to the update weight parameters;

[0108] In step S113, the registration information, registration action information, and device identification information are taken as inputs, and the target registration verification code is taken as output to train the updated private authentication detection model.

[0109] In step S114, when the trained private authentication detection model has not converged, the gradient update vector is obtained based on the trained private authentication detection model and sent. The edge server is used to aggregate the gradient update vector and update the weight parameters of the shared authentication detection model of the edge server based on the aggregated gradient update vector to obtain the updated weight parameters.

[0110] In step S115, when the trained private authentication detection model converges, the target authentication detection model is obtained based on the trained private authentication detection model.

[0111] In one embodiment of this disclosure, both the private authentication detection model and the public authentication detection model can be neural network models, convolutional neural network models, or long short-term memory network models, etc.

[0112] In one embodiment of this disclosure, obtaining the target authentication detection model based on the trained private authentication detection model can be understood as storing the trained private authentication detection model as the target authentication detection model, or it can be understood as directly identifying the trained private authentication detection model as the target authentication detection model.

[0113] In this implementation, the updated weight parameters received by the terminal from the edge server are obtained by the edge server aggregating gradient update vectors sent by multiple terminals and updating the weight parameters of the shared authentication detection model based on the aggregated gradient update vectors. Therefore, the updated shared authentication detection model can reflect the common patterns learned by the edge server's shared authentication detection model in the previous training round, namely, the registration information, registration action information, device identification information, and target registration verification code obtained by multiple terminals. By using registration information, registration action information, and device identification information as input and the target registration verification code as output, the updated private authentication detection model is trained. Training the updated private authentication detection model allows it to not only learn the common user patterns but also to personalize the data collected during user authentication for the terminal itself. The training process enables the private authentication detection model to learn the private patterns between the registration information, registration action information, device identification information, and the target registration verification code obtained by the terminal itself. When the trained private authentication detection model fails to converge, it indicates that the model still needs further training. By obtaining and sending gradient update vectors from the trained model, the edge server can continue to obtain the corresponding update weight parameters based on the gradient update vectors uploaded by multiple terminals without disclosing the user's personal data. This allows for continued training of the private authentication detection models for each terminal. When the trained private authentication detection model converges, it can quickly output the target registration verification code based on the input authentication information, authentication action information, and device identification information. Therefore, this converged private authentication detection model is used as the target authentication detection model.

[0114] The technical solution provided in this disclosure involves: acquiring registration information input through a terminal and registration action information indicating the input action when entering the registration information; sending a registration request, which includes registration information and device identification information; receiving an initial authentication detection model and a target registration verification code sent by the registration server in response to the registration request; obtaining a private authentication detection model based on the initial authentication detection model; receiving updated weight parameters sent by the edge server and updating the private authentication detection model based on the updated weight parameters; and training the updated private authentication detection model using the registration information, registration action information, and device identification information as input and the target registration verification code as output. When the trained private authentication detection model has not converged, the gradient update vector is obtained from the trained private authentication detection model and sent to the edge server. The edge server aggregates the gradient update vector and updates the weight parameters of the shared authentication detection model on the edge server according to the aggregated gradient update vector to obtain the updated weight parameters. When the trained private authentication detection model converges, the target authentication detection model is obtained from the trained private authentication detection model. This ensures that the training process can be jointly executed by the edge server and the terminal. Furthermore, the target authentication detection model can learn not only the shared rules but also the private rules, improving training efficiency and enhancing user experience.

[0115] In one implementation of this disclosure, before obtaining the private authentication detection model based on the initial authentication detection model, the method further includes:

[0116] Receive private data upload instructions;

[0117] In response to a private data upload command, it sends registration information, registration action information, and device identification information;

[0118] Receive the initial weight parameters sent by the edge server;

[0119] Obtain a private authentication detection model based on the initial authentication detection model, including:

[0120] The initial authentication detection model is updated based on the initial weight parameters to obtain a private authentication detection model.

[0121] In the technical solution provided in this disclosure, by receiving a private data upload instruction and responding to it by sending registration information, registration action information, and device identification information, the edge server can perform preliminary training on its initial authentication detection model based on this registration information, registration action information, and device identification information to obtain a shared authentication detection model on the edge server. This shared authentication detection model can be understood as a model that has preliminarily learned the patterns between the registration information, registration action information, and device identification information obtained by multiple terminals and the corresponding target registration verification codes. Subsequently, by receiving initial weight parameters sent by the edge server, the initial authentication detection model is updated according to the initial weight parameters to obtain a private authentication detection model. This private authentication detection model can be understood as a model that has learned the patterns learned by the shared authentication detection model; that is, at this point, the private authentication detection model can also be understood as a model that has preliminarily learned the patterns of shared user authentication results from multiple terminals. This facilitates subsequent multi-round training of the private authentication detection model without needing to train based on the initial authentication detection model, thus reducing the training difficulty.

[0122] In one implementation of this disclosure, before sending registration information, registration action information, and device identification information in response to a private data upload instruction, the method further includes:

[0123] Noise is added to registration information, registration action information, and device identification information to obtain noisy registration information, noisy registration action information, and noisy device identification information;

[0124] Send registration information, registration action information, and device identification information, including:

[0125] Send the registration information after adding noise, the registration action information after adding noise, and the device identification information after adding noise.

[0126] In one embodiment of this disclosure, noise is added to the registration information, registration action information, and device identification information. This can be done by adding random noise or by adding Laplace noise.

[0127] The technical solution provided in this disclosure adds noise to registration information, registration action information, and device identification information to obtain noise-added registration information, noise-added registration action information, and noise-added device identification information. This process, along with sending the noise-added registration information, registration action information, and device identification information, can prevent the leakage of sensitive user personal information in the registration information, registration action information, and device identification information without compromising the accuracy of the relationship between these information and the target registration verification code. This improves the security of user personal information.

[0128] In one implementation of this disclosure, before receiving the private data upload instruction, the method further includes:

[0129] The system sends user characteristic information, which includes the user's gender, age, education level, left / right hand preference, authentication method preference, and application preference. The edge server uses this user characteristic information to identify multiple target terminals and sends private data upload instructions to the target terminals.

[0130] The user characteristic information can be stored in the terminal beforehand or obtained from other devices or systems.

[0131] In the technical solution provided in this disclosure, since the input habits of users with different user feature information may vary greatly, by sending user feature information, the edge server can determine multiple target terminals based on the user feature information and send private data upload instructions to the target terminals. This can ensure that the real-time user authentication logs obtained by the edge server are logs of users with relatively consistent input habits when performing user authentication operations, thereby improving the convergence speed of the private authentication detection model during training and reducing the training difficulty.

[0132] The following are embodiments of the apparatus disclosed herein, which can be used to execute embodiments of the method disclosed herein.

[0133] According to one embodiment of the present disclosure, the user authentication device is disposed on a terminal, and the user authentication device can be implemented as part or all of the electronic device through software, hardware or a combination of both. Figure 6 A schematic structural block diagram of a user authentication device according to an embodiment of the present disclosure is shown, such as Figure 6 As shown, the user authentication device 200 includes:

[0134] The authentication information acquisition module 201 is configured to acquire authentication information input through the terminal and authentication action information used to indicate the input action when inputting authentication information;

[0135] The verification code acquisition module 202 is configured to input authentication information and authentication action information into the target authentication detection model in order to obtain the target authentication verification code output by the target authentication detection model.

[0136] The authentication request sending module 203 is configured to send authentication request information, which includes authentication information and the target authentication verification code.

[0137] In this embodiment, authentication information input through the terminal and authentication action information indicating the input action when inputting authentication information are obtained. The authentication information and authentication action information are then input into the target authentication detection model to obtain the target authentication verification code output by the target authentication detection model. An authentication request is then sent, including the authentication information and the target authentication verification code. Since the target authentication detection model can be understood as a model that exists only on this terminal and has learned the relationship between the authentication information, authentication action information, and the target authentication verification code—meaning the target authentication verification code can only be generated by the target authentication detection model on the corresponding terminal based on the input action information corresponding to the input action and the authentication information when a specific user inputs authentication information that meets the authentication requirements using their usual input action—the receiving end can compare the target authentication verification code in the authentication request information to determine whether the authentication information in the authentication request information is the authentication information that meets the authentication requirements input by a specific user on the terminal carrying the target authentication detection model. This helps improve the reliability of the user authentication process and enhances the user experience.

[0138] This disclosure also discloses an electronic device, Figure 7 A schematic structural block diagram of an electronic device according to an embodiment of the present disclosure is shown, such as... Figure 7 As shown, the electronic device includes a memory and a processor; wherein the memory is used to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement any of the methods in the embodiments of this disclosure.

[0139] Figure 8 A schematic diagram of the structure of a computer system suitable for implementing the methods according to embodiments of the present disclosure is shown. Figure 8 As shown, the computer system includes a processing unit that can execute various methods described above based on a program stored in a read-only memory (ROM) or a program loaded from a storage portion into a random access memory (RAM). The RAM also stores various programs and data required for the operation of the computer system. The processing unit, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0140] The following components are connected to the I / O interface: input sections including keyboards, mice, etc.; output sections including cathode ray tubes (CRTs), liquid crystal displays (LCDs), and speakers; storage sections including hard disks, etc.; and communication sections including network interface cards such as LAN cards and modems. The communication section performs communication processes via a network such as the Internet. Drives are also connected to the I / O interface as needed. Removable media, such as disks, optical disks, magneto-optical disks, semiconductor memories, etc., are installed on the drive as needed so that computer programs read from them can be installed into the storage section as needed. The processing unit can be implemented as a CPU, GPU, TPU, FPGA, NPU, etc.

[0141] In particular, according to embodiments of this disclosure, the methods described above can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program tangibly embodied on a machine-readable medium, the computer program containing program code for performing the methods described above. In such embodiments, the computer program can be downloaded and installed from a network via a communication component, and / or installed from a removable medium.

[0142] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0143] The units or modules described in the embodiments of this disclosure can be implemented in software or programmable hardware. The described units or modules can also be located in a processor, and the names of these units or modules do not necessarily constitute a limitation on the unit or module itself.

[0144] In another aspect, this disclosure also provides a computer-readable storage medium, which may be a computer-readable storage medium included in the electronic device or computer system described above; or it may be a standalone computer-readable storage medium not assembled into a device. The computer-readable storage medium stores one or more programs, which are used by one or more processors to perform the methods described in this disclosure.

[0145] The above description is merely a preferred embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

Claims

1. A user authentication method, characterized in that, The method is applied to a terminal and includes: Acquire authentication information input through the terminal and authentication action information used to indicate the input action when inputting the authentication information; The authentication information and the authentication action information are input into the target authentication detection model to obtain the target authentication verification code output by the target authentication detection model; wherein, the target authentication verification code can only be generated by the target authentication detection model on the corresponding terminal based on the input action information corresponding to the input action and the authentication information when a specific user inputs authentication information that meets the authentication requirements using the user's usual input action; Send authentication request information, which includes the authentication information and the target authentication verification code.

2. The user authentication method according to claim 1, characterized in that, The input action includes a touch action, and the authentication action information includes at least one of the touch trajectory, touch force, and touch area of ​​the touch action when the authentication information is input.

3. The user authentication method according to claim 1, characterized in that, The step of inputting the authentication information and the authentication action information into the target authentication detection model to obtain the target authentication verification code output by the target authentication detection model includes: The authentication information, the authentication action information, and the device identification information of the terminal are input into the target authentication detection model to obtain the target authentication verification code output by the target authentication detection model; The authentication request includes the device identification information, the authentication information, and the target authentication verification code.

4. The user authentication method according to claim 3, characterized in that, Before acquiring the authentication information input through the terminal and the authentication action information used to indicate the input action when inputting the authentication information, the method further includes: Acquire registration information input through the terminal and registration action information used to indicate the input action when inputting the registration information; Send registration request information, the registration request information including the registration information and the device identification information; Receive the initial authentication detection model and the target registration verification code sent by the registration server in response to the registration request information; The initial authentication detection model is trained by taking the registration information, the registration action information, and the device identification information as inputs and the target registration verification code as output, so as to obtain the target authentication detection model.

5. The user authentication method according to claim 3, characterized in that, Before acquiring the authentication information input through the terminal and the authentication action information used to indicate the input action when inputting the authentication information, the method further includes: Acquire registration information input through the terminal and registration action information used to indicate the input action when inputting the registration information; Send a registration request, the registration request including the registration information and the device identification information; Receive the initial authentication detection model and the target registration verification code sent by the registration server in response to the registration request; Obtain a private authentication detection model based on the initial authentication detection model; Receive update weight parameters sent by the edge server, and update the private authentication detection model according to the update weight parameters; The updated private authentication detection model is trained by taking the registration information, the registration action information, and the device identification information as inputs and the target registration verification code as output. When the trained private authentication detection model fails to converge, a gradient update vector is obtained based on the trained private authentication detection model and sent. The edge server aggregates the gradient update vector and updates the weight parameters of the shared authentication detection model of the edge server based on the aggregated gradient update vector to obtain the updated weight parameters. When the trained private authentication detection model converges, the target authentication detection model is obtained based on the trained private authentication detection model.

6. The user authentication method according to claim 5, characterized in that, Before obtaining the private authentication detection model based on the initial authentication detection model, the method further includes: Receive private data upload instructions; In response to the private data upload instruction, the registration information, the registration action information, and the device identification information are sent. Receive the initial weight parameters sent by the edge server; The step of obtaining a private authentication detection model based on the initial authentication detection model includes: The initial authentication detection model is updated based on the initial weight parameters to obtain the private authentication detection model.

7. The user authentication method according to claim 6, characterized in that, Before sending the registration information, the registration action information, and the device identification information in response to the private data upload instruction, the method further includes: Noise is added to the registration information, the registration action information, and the device identification information to obtain the registration information, registration action information, and device identification information after noise addition. Sending the registration information, the registration action information, and the device identification information includes: Send the registration information after adding noise, the registration action information after adding noise, and the device identification information after adding noise.

8. The user authentication method according to claim 7, characterized in that, Before receiving the private data upload instruction, the method further includes: sending user feature information, which includes the user's gender, age, education level, left / right hand preference, user authentication method preference, and application preference information. The edge server is used to determine multiple target terminals based on the user feature information and send the private data upload instruction to the target terminals.

9. An electronic device, characterized in that, The electronic device includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the method according to any one of claims 1-8.

10. A readable storage medium, characterized in that, It stores computer instructions that, when executed by a processor, implement the steps of the method described in any one of claims 1-8.

Images