Identity authentication method, related device and system

Abstract

Embodiments of the present application provide an identity authentication method, related device and system. The method is applied to a distributed system composed of multiple electronic devices, the multiple electronic devices including a first device, a second device and a third device. The method includes: the first device authenticates a first user according to a first user feature of the first user to obtain a first authentication result of the first user; the first device shares the first authentication result with the electronic devices in the distributed system; if the first authentication result is passed, the second device shares a user model of the first user with the electronic devices in the distributed system, the user model of the first user including a user feature used to describe the first user; and if the third device determines that the first user is within a preset range according to the user model of the first user, the third device determines that the first user is legal. By using the embodiments of the present application, the number of authentication operations performed by the user when switching devices can be reduced while ensuring security, greatly facilitating the use of the user.

Description

Technical Field

[0001] This application relates to the field of information security technology, and in particular to an identity authentication method, related apparatus and system. Background Technology

[0002] When using electronic devices such as mobile phones and tablets, users often need to manually authenticate their identity first. Only after successful authentication can users access the functions and applications on the electronic device, preventing unauthorized users from stealing information and ensuring security. Authentication methods are diverse, including traditional methods such as password authentication and image authentication, and biometric authentication methods such as fingerprint authentication and facial recognition. Different electronic devices may use different authentication methods. When users switch between electronic devices, they often have to manually authenticate again for the new device, which is cumbersome and provides a poor user experience. Therefore, how to solve the problem of repeated authentication when switching devices while ensuring security is a problem that those skilled in the art are researching. Summary of the Invention

[0003] This application discloses an identity authentication method, related apparatus, and system, which can solve the problem of repeated identity authentication when users switch devices while ensuring security, reducing the number of authentication operations performed by users when switching devices and greatly facilitating user use.

[0004] In a first aspect, embodiments of this application disclose an identity authentication method applied to a distributed system composed of multiple electronic devices, the multiple electronic devices including a first device, a second device, and a third device. The method includes: the first device authenticating the first user based on a first user characteristic of the first user, obtaining a first authentication result of the first user; the first device sharing the first authentication result with the electronic devices in the distributed system; if the first authentication result indicates successful authentication, the second device sharing the user model of the first user with the electronic devices in the distributed system, the user model of the first user including user characteristics describing the first user; if the third device determines that the first user is within a preset range based on the user model of the first user, the third device determining that the first user is legitimate.

[0005] The second device can be multiple devices. The third device can be any device in the distributed system; that is, any device in the distributed system can have authentication requirements and can be used to determine whether a user is legitimate.

[0006] In this application, after the first user is authenticated by the first device, multiple devices in the distributed system can determine that the first user is a legitimate user. If the first user wants to use a third device, the third device can identify the user's identity based on the shared user model of the first user, and can directly exempt the legitimate first user from authentication, reducing the number of authentication operations performed when the legitimate user switches devices, achieving the effect of one-end authentication and multiple-end authentication exemption. Multiple devices in the distributed system can continuously and multi-dimensionally track the first user and obtain the user model of the first user. Compared with authentication methods based on single user features and one-time authentication methods, identifying user identity through the user model of the first user is more secure and reliable.

[0007] In one possible implementation, the method further includes: if the first authentication result is successful, the second device acquires the user characteristics of the first user at multiple time points, wherein the user characteristics of the first user at multiple time points are used to determine the user model of the first user.

[0008] In this application, multiple devices in a distributed system can continuously and multidimensionally track the first user and obtain the user model of the first user. Compared with authentication methods based on single user features and one-time authentication methods, the security and reliability of identifying user identity through the user model of the first user are higher.

[0009] In one possible implementation, the user characteristics of the first user at multiple time points include a first feature and a second feature, wherein the confidence level of the first feature is greater than the confidence level of the second feature; the method further includes: the second device determining a user model of the first user based on the user characteristics of the first user at multiple time points, wherein the user model of the first user includes a third feature and a fourth feature, the first feature and the third feature are of the same type, the second feature and the fourth feature are of the same type, the similarity between the first feature and the third feature is greater than or equal to a preset threshold, and the similarity between the second feature and the fourth feature is less than a preset threshold; the second device updating the user model of the first user using the second feature.

[0010] The update of the user model for the first user can be triggered when the first authentication result is successful, or it can be triggered within a preset period.

[0011] In this application, the second device can periodically update or trigger updates to the established user model, thereby increasing the consistency between the updated user model and the users described by that model, i.e., increasing the credibility of the user model. Identifying users through a more credible user model provides greater security and reliability.

[0012] In one possible implementation, before the third device determines that the first user is within a preset range based on the user model of the first user, and before the third device determines that the first user is legitimate, the method further includes: the second device determining that the first user is within the preset range based on the user characteristics of the first user at multiple time points, and sharing first indication information with the electronic devices in the distributed system, wherein the first indication information is used to indicate that the first user is within the preset range; the step of determining that the first user is legitimate based on the user model of the first user includes: the third device determining that the first user is within the preset range based on the user model of the first user and the first indication information, and the third device determining that the first user is legitimate.

[0013] In this application, the second device can predict the behavior of the first user to determine whether the first user wants to use any device in the distributed system. When the first device determines that the first user wants to use the third device, it notifies the third device. After receiving the notification, the third device can identify the user's identity based on the first user's user model and, if the first user is deemed legitimate, can use the third device without authentication. The entire process is seamless for the user; the user does not need to operate the third device and can use it without authentication, greatly facilitating the user experience.

[0014] In one possible implementation, the step of the third device determining that the first user is legitimate if the third device determines that the first user is within a preset range based on the user model of the first user includes: the third device receiving a first user operation from the first user; the third device determining that the first user is within the preset range based on the first user operation and the user model of the first user; and the third device determining that the first user is legitimate.

[0015] In one possible implementation, the method further includes: if the first authentication result is successful, the second device shares second indication information with the electronic devices in the distributed system, wherein the second indication information is used to indicate the association between the user model of the first user and the distributed system to which the first device belongs; the step of the third device determining that the first user is legitimate if the third device determines that the first user is within a preset range based on the user model of the first user includes: the third device determining that the first user is within the preset range based on the user model of the first user; the third device determining that the third device belongs to the distributed system associated with the user model of the first user based on the second indication information, and the third device determining that the first user is legitimate.

[0016] In one possible implementation, the method further includes: if the first authentication result is successful, the first device shares the first user characteristics of the first user with the electronic devices in the distributed system; if the third device determines that the first user is within a preset range based on the user model of the first user, the third device determines that the first user is legitimate, including: the third device determines that the first user is within a preset range based on the user model of the first user; the third device authenticates the first user based on the first user characteristics to obtain a second authentication result of the first user, the second authentication result being used to indicate whether the first user is legitimate.

[0017] In this application, after the first device authenticates the first user's first user characteristics, if the first user wants to use the third device, the third device can identify the user's identity based on the shared user model of the first user and authenticate the first user characteristics to obtain a second authentication result. The third device can then determine whether authentication is exempted based on the second authentication result. Users do not need to manually perform the authentication process, reducing the number of authentication operations required when switching devices and greatly facilitating user use.

[0018] In one possible implementation, the user characteristics of the first user at multiple points in time are collected at different times by one or more electronic devices in the distributed system.

[0019] In this application, the user features used to determine the user model are user features obtained from multiple electronic devices in the distributed system from different perspectives. Therefore, the consistency between the user model and the user described by the user model is higher, that is, the user model has higher credibility. Identifying user identities through a more credible user model has higher security and reliability.

[0020] In one possible implementation, the preset range is a range relative to the distributed system or a range relative to the third device.

[0021] In one possible implementation, any electronic device in the distributed system is a device that requires authentication of the user's identity before use.

[0022] In this application, any device in the distributed system can have authentication requirements and can be used to determine whether a user is legitimate; that is, any device can be a third-party device. Even if the third-party device does not have data collection capabilities, it can still identify the user's identity based on the shared user model of the first user, resulting in high availability.

[0023] Secondly, this application discloses another identity authentication method applied to a second device in a distributed system. The distributed system comprises multiple electronic devices, including a first device, a second device, and a third device. The method includes: the second device obtaining a first authentication result of a first user from the distributed system, wherein the first authentication result is obtained by the first device authenticating the first user based on the first user's first user characteristics; if the first authentication result indicates successful authentication, the second device shares the user model of the first user with the electronic devices in the distributed system, wherein the user model of the first user includes user characteristics describing the first user; the user model of the first user is used to track the first user, and the first user is considered legitimate when the first user is within a preset range.

[0024] In this application, after the first user is authenticated by the first device, multiple devices in the distributed system can determine that the first user is a legitimate user. If the first user wants to use a third device, the third device can identify the user's identity based on the shared user model of the first user, and can directly exempt the legitimate first user from authentication, reducing the number of authentication operations performed when the legitimate user switches devices, achieving the effect of one-end authentication and multiple-end authentication exemption. Multiple devices in the distributed system can continuously and multi-dimensionally track the first user and obtain the user model of the first user. Compared with authentication methods based on single user features and one-time authentication methods, identifying user identity through the user model of the first user is more secure and reliable.

[0025] In one possible implementation, the method further includes: if the first authentication result is successful, the second device acquires the user characteristics of the first user at multiple time points, wherein the user characteristics of the first user at multiple time points are used to determine the user model of the first user.

[0026] In this application, multiple devices in a distributed system can continuously and multidimensionally track the first user and obtain the user model of the first user. Compared with authentication methods based on single user features and one-time authentication methods, the security and reliability of identifying user identity through the user model of the first user are higher.

[0027] In one possible implementation, the user characteristics of the first user at multiple time points include a first feature and a second feature, wherein the confidence level of the first feature is greater than the confidence level of the second feature; the method further includes: the second device determining a user model of the first user based on the user characteristics of the first user at multiple time points, wherein the user model of the first user includes a third feature and a fourth feature, the first feature and the third feature are of the same type, the second feature and the fourth feature are of the same type, the similarity between the first feature and the third feature is greater than or equal to a preset threshold, and the similarity between the second feature and the fourth feature is less than a preset threshold; the second device updating the user model of the first user using the second feature.

[0028] The update of the user model for the first user can be triggered when the first authentication result is successful, or it can be triggered within a preset period.

[0029] In this application, the second device can periodically update or trigger updates to the established user model, thereby increasing the consistency between the updated user model and the users described by that model, i.e., increasing the credibility of the user model. Identifying users through a more credible user model provides greater security and reliability.

[0030] In one possible implementation, the method further includes: the second device determining the first user within the preset range based on the user characteristics of the first user at multiple time points, and sharing first indication information with the electronic device in the distributed system, wherein the first indication information is used to indicate that the first user is within the preset range, and the first indication information is used by the electronic device in the distributed system to determine that the first user is legitimate.

[0031] In this application, the second device can predict the behavior of the first user to determine whether the first user wants to use any device in the distributed system. When the first device determines that the first user wants to use the third device, it notifies the third device. After receiving the notification, the third device can identify the user's identity based on the first user's user model and, if the first user is deemed legitimate, can use the third device without authentication. The entire process is seamless for the user; the user does not need to operate the third device and can use it without authentication, greatly facilitating the user experience.

[0032] In one possible implementation, the method further includes: if the first authentication result is successful, the second device shares second indication information with the electronic device in the distributed system, wherein the second indication information is used to indicate the association between the user model of the first user and the distributed system to which the first device belongs; the second indication information is used by the electronic device in the distributed system to determine that the first user is legitimate when it determines that the first user is within a preset range.

[0033] In one possible implementation, the user characteristics of the first user at multiple points in time are collected at different times by one or more electronic devices in the distributed system.

[0034] In this application, the user features used to determine the user model are user features obtained from multiple electronic devices in the distributed system from different perspectives. Therefore, the consistency between the user model and the user described by the user model is higher, that is, the user model has higher credibility. Identifying user identities through a more credible user model has higher security and reliability.

[0035] In one possible implementation, the preset range is a range relative to the distributed system or a range relative to the third device.

[0036] In one possible implementation, any electronic device in the distributed system is a device that requires authentication of the user's identity before use.

[0037] In this application, any device in the distributed system can have authentication requirements and can be used to determine whether a user is legitimate; that is, any device can be a third-party device. Even if the third-party device does not have data collection capabilities, it can still identify the user's identity based on the shared user model of the first user, resulting in high availability.

[0038] Thirdly, this application discloses another identity authentication method applied to a third device in a distributed system. The distributed system consists of multiple electronic devices, including a first device, a second device, and the third device. The method includes: the third device obtaining a user model of a first user from the distributed system, wherein the user model of the first user is obtained by the second device when the first authentication result of the first user is successful, the user model of the first user includes user features describing the first user, and the first authentication result is obtained by the first device authenticating the first user based on the first user features; if the third device determines that the first user is within a preset range based on the user model of the first user, the third device determines that the first user is legitimate.

[0039] In this application, after the first user is authenticated by the first device, multiple devices in the distributed system can determine that the first user is a legitimate user. If the first user wants to use a third device, the third device can identify the user's identity based on the shared user model of the first user, and can directly exempt the legitimate first user from authentication, reducing the number of authentication operations performed when the legitimate user switches devices, achieving the effect of one-end authentication and multiple-end authentication exemption. Multiple devices in the distributed system can continuously and multi-dimensionally track the first user and obtain the user model of the first user. Compared with authentication methods based on single user features and one-time authentication methods, identifying user identity through the user model of the first user is more secure and reliable.

[0040] In one possible implementation, the step of determining that the first user is legitimate if the third device determines that the first user is within a preset range based on the user model of the first user includes:

[0041] The third device determines that the first user is within the preset range based on the user model of the first user and the first indication information. The third device determines that the first user is legitimate. The first indication information is used to indicate that the first user is within the preset range. The first indication information is obtained by the second device based on the user characteristics of the first user at multiple time points.

[0042] In this application, the second device can predict the behavior of the first user to determine whether the first user wants to use any device in the distributed system. When the first device determines that the first user wants to use the third device, it notifies the third device. After receiving the notification, the third device can identify the user's identity based on the first user's user model and, if the first user is deemed legitimate, can use the third device without authentication. The entire process is seamless for the user; the user does not need to operate the third device and can use it without authentication, greatly facilitating the user experience.

[0043] In one possible implementation, the step of the third device determining that the first user is legitimate if the third device determines that the first user is within a preset range based on the user model of the first user includes: the third device receiving a first user operation from the first user; the third device determining that the first user is within the preset range based on the first user operation and the user model of the first user; and the third device determining that the first user is legitimate.

[0044] In one possible implementation, the step of the third device determining the first user as legitimate if the third device determines that the first user is within a preset range based on the first user's user model includes: the third device determining that the first user is within the preset range based on the first user's user model; the third device determining that the third device belongs to the distributed system associated with the first user's user model based on second indication information; and the third device determining the first user as legitimate, wherein the second indication information is determined by the second device when the first authentication result is successful, and the second indication information is used to indicate that the first user's user model is associated with the distributed system to which the first device belongs.

[0045] In one possible implementation, the step of determining that the first user is legitimate if the third device determines that the first user is within a preset range based on the user model of the first user includes: the third device determining that the first user is within a preset range based on the user model of the first user; the third device authenticating the first user based on the characteristics of the first user to obtain a second authentication result of the first user, wherein the second authentication result is used to indicate whether the first user is legitimate.

[0046] In this application, after the first device authenticates the first user's first user characteristics, if the first user wants to use the third device, the third device can identify the user's identity based on the shared user model of the first user and authenticate the first user characteristics to obtain a second authentication result. The third device can then determine whether authentication is exempted based on the second authentication result. Users do not need to manually perform the authentication process, reducing the number of authentication operations required when switching devices and greatly facilitating user use.

[0047] In one possible implementation, the user characteristics of the first user at multiple points in time are collected at different times by one or more electronic devices in the distributed system.

[0048] In this application, the user features used to determine the user model are user features obtained from multiple electronic devices in the distributed system from different perspectives. Therefore, the consistency between the user model and the user described by the user model is higher, that is, the user model has higher credibility. Identifying user identities through a more credible user model has higher security and reliability.

[0049] In one possible implementation, the preset range is a range relative to the distributed system or a range relative to the third device.

[0050] In one possible implementation, any electronic device in the distributed system is a device that requires authentication of the user's identity before use.

[0051] In this application, any device in the distributed system can have authentication requirements and can be used to determine whether a user is legitimate; that is, any device can be a third-party device. Even if the third-party device does not have data collection capabilities, it can still identify the user's identity based on the shared user model of the first user, resulting in high availability.

[0052] Fourthly, embodiments of this application disclose a distributed system including multiple electronic devices, the multiple electronic devices including a first device, a second device, and a third device, wherein: the first device is used to authenticate the first user based on a first user characteristic of the first user, obtain a first authentication result of the first user, and share the first authentication result with the electronic devices in the distributed system; the second device is used to share the user model of the first user with the electronic devices in the distributed system when the first authentication result is successful, the user model of the first user including user characteristics describing the first user; the third device is used to determine that the first user is within a preset range based on the user model of the first user, and determine that the first user is legitimate.

[0053] In this application, after the first user is authenticated by the first device, multiple devices in the distributed system can determine that the first user is a legitimate user. If the first user wants to use a third device, the third device can identify the user's identity based on the shared user model of the first user, and can directly exempt the legitimate first user from authentication, reducing the number of authentication operations performed when the legitimate user switches devices, achieving the effect of one-end authentication and multiple-end authentication exemption. Multiple devices in the distributed system can continuously and multi-dimensionally track the first user and obtain the user model of the first user. Compared with authentication methods based on single user features and one-time authentication methods, identifying user identity through the user model of the first user is more secure and reliable.

[0054] Wherein, the first device is the first device described in the first aspect or any possible mode of the first aspect. The second device is the first and second aspects, or the second device described in any possible mode of the first and second aspects. The third device is the first and third aspects, or the third device described in any possible mode of the first and third aspects.

[0055] Fifthly, embodiments of this application disclose an electronic device, the electronic device including one or more memories and one or more processors, the one or more memories being coupled to the one or more processors, the one or more memories being used to store a computer program, and the one or more processors being used to invoke the computer program, the computer program including instructions, which, when executed by the one or more processors, cause the electronic device to perform the authentication method described in the first aspect, the second aspect, and the third aspect, or any possible manner of the first aspect, the second aspect, and the third aspect. The electronic device is a first device, a second device, or a third device.

[0056] Sixthly, embodiments of this application disclose a computer storage medium including a computer program, the computer program including instructions that, when executed on a processor, implement the authentication method described in the first aspect, the second aspect, and the third aspect, or any possible manner of the first aspect, the second aspect, and the third aspect.

[0057] In a seventh aspect, embodiments of this application disclose a chip system, the chip system including at least one processor, a memory and an interface circuit, the memory, the interface circuit and the at least one processor being interconnected by a circuit, the memory storing a computer program, the computer program being executed by the at least one processor to implement the authentication method described in the first aspect, the second aspect and the third aspect, or any possible manner of the first aspect, the second aspect and the third aspect. Attached Figure Description

[0058] The accompanying drawings used in the embodiments of this application are described below.

[0059] Figure 1 This is a schematic diagram of an identity authentication scenario provided in an embodiment of this application;

[0060] Figure 2 This is a schematic diagram of the architecture of an identity authentication system provided in an embodiment of this application;

[0061] Figures 3-4 These are schematic diagrams of the structures of some electronic devices provided in the embodiments of this application;

[0062] Figure 5 This is a schematic diagram of the architecture of another identity authentication system provided in the embodiments of this application;

[0063] Figures 6-10 This is a flowchart illustrating some identity authentication methods provided in the embodiments of this application. Detailed Implementation

[0064] The technical solutions in the embodiments of this application will be clearly and thoroughly described below with reference to the accompanying drawings. The terminology used in the implementation section of the embodiments of this application is only used to explain the specific embodiments of this application and is not intended to limit this application.

[0065] Please see Figure 1 , Figure 1 This is a schematic diagram of an identity authentication scenario 10 provided in an embodiment of this application. Scenario 10 can be a home scenario, an office scenario, etc., and multiple electronic devices (hereinafter referred to as devices) can exist in scenario 10. User 100 can exist in scenario 10, and user 100 can carry a first device 101 and a second device 102. There can be three rooms in scenario 10: a first room 110, a second room 120, and a third room 130, and any one of these rooms can contain devices. Specifically, the first room 110 can contain the third device 111, the fourth device 112, and the fifth device 113; the second room 120 can contain the sixth device 121 and the seventh device 122; and the third room 130 can contain the eighth device 131 and the ninth device 132. User 100 can be located at any position in scenario 10.

[0066] In scenario 10, multiple devices can form at least one system. This application uses the example of multiple devices forming a system in scenario 10 for illustration. For specific examples, please refer to [link to example]. Figure 2 The first system 20 is shown.

[0067] When user 100 uses any device in scenario 10, that device can perform a user authentication process. For example, when user 100 uses the first device 101, the first device 101 can perform continuous authentication based on touch screen behavior. The first device 101 can continuously acquire user 100's touch screen behavior information (such as the location and area of ​​the touch area, timestamp, number of touches, pressure, etc.), and then compare this touch screen behavior information with pre-established touch screen behavior samples to determine if they match. If they match, the first device 101 confirms user 100 as a legitimate user; if they do not match, the first device 101 confirms user 100 as an illegitimate user. The touch screen behavior samples can be learned by the first device 101 through multiple acquisitions of touch screen behavior information and can be used to represent the touch screen behavior information of legitimate users. The reliability of this authentication method is closely related to the quality of the aforementioned user touch screen behavior samples. If the samples are few or of poor quality, the reliability of the authentication results will be low, and the false positive rate will be high. For another example, when user 100 uses the first device 101, the first device 101 can perform continuous authentication based on facial information. Specifically, the first device 101 can continuously acquire the facial information of user 100 and determine whether the user within the image acquisition area has changed based on this facial information. This authentication method is relatively simple, only used to determine whether the user within the image acquisition area has changed, resulting in low security. Furthermore, the two authentication methods in the above example have high power consumption requirements, easily causing the device's battery to drain too quickly, thus reducing its practicality. Moreover, different devices can use different authentication methods. When user 100 switches to other devices in scenario 10, those other devices still need to perform the user authentication process, which is cumbersome and results in a poor user experience.

[0068] This application provides an identity authentication method applicable to an identity authentication system. This system can be a distributed system comprising multiple devices. The system continuously performs multi-dimensional human body tracking on users in the near-field three-dimensional space of the system, i.e., continuously collecting various types of user features and identifying the user's identity based on these features. The results of human body tracking (hereinafter referred to as tracking results) are used to characterize the user's identity. Tracking results can be shared within the identity authentication system and used by any device within the system to identify the user's identity and determine whether authentication is exempted. If the user is a legitimate user, authentication exemption can be determined. A legitimate user can be a user who has been authenticated by any device in the identity authentication system. Therefore, legitimate users can use any device in the identity authentication system without authentication, thus achieving the effect of one-end authentication and multi-end authentication exemption. Furthermore, the tracking results are obtained based on continuously collected multi-dimensional user features, which is more accurate, reliable, and secure than authentication methods based on single user features. Multiple devices in the distributed system collaboratively acquire tracking results, resulting in lower power consumption requirements and better availability.

[0069] The distributed system shown in the following embodiments is an identity authentication system used to implement the identity authentication method in this application.

[0070] It's important to clarify that "authentication-free" refers to eliminating the need for users to manually authenticate their identity (hereinafter referred to as manual authentication). Users can directly view the user interface after successful authentication through the device. For users, manual authentication includes, for example, placing their finger on the device's fingerprint sensor area, or entering a password. For devices, manual authentication includes, for example, the device automatically collecting and authenticating fingerprint features, or the device acquiring and authenticating the user's entered password. Therefore, the process of identifying users based on tracking results shared by the distributed system and determining authentication-free is seamless for the user, enhancing the user experience.

[0071] For example, when user 100 uses a device in scenario 10 for the first time, such as the first device 101, they can manually perform identity authentication (e.g., fingerprint authentication, facial recognition, etc.). If the authentication is successful, the first device 101 can determine that user 100 is a legitimate user and share the authentication result with other devices in scenario 10. Multiple devices in scenario 10 can collect user features of user 100 from various angles in near-field three-dimensional space and obtain tracking results based on these features. The tracking results indicate that the currently tracked user is user 100 and that user 100 is a legitimate user. The tracking results can be shared to provide a basis for cross-device authentication. For example, when user 100 is in the first room 110, the third device 111 and the fourth device 112 can obtain the location of user 100 through near-field positioning technologies (e.g., Bluetooth, Wireless Fidelity (Wi-Fi), Ultra Wideband (UWB), etc.), and the fifth device 113 can collect an image of user 100. When user 100 leaves room 110 and enters room 130, device 131 can collect user 100's voiceprint, and device 132 can collect user 100's image. The user's image can be used to obtain facial, body, behavioral, and location information. When user 100 triggers authentication on device 131 (e.g., user 100 presses a button on device 131), device 131 can determine, based on shared tracking results, that the user to be authenticated is user 100, and that user 100 is a legitimate user. At this point, device 131 can directly display the user interface displayed upon successful authentication, such as its desktop. In other words, multiple devices in a distributed system can achieve continuous, user-unnoticed device access control based on tracking results, ensuring security while improving user experience.

[0072] In some embodiments, the tracking result is a user model. That is, the distributed system can share user models, one user model can represent one user, and different users have different user models. Optionally, the user model can be a dataset, which can include various user features used to describe the user. These user features can be feature data used to characterize the user's identity, obtained by analyzing and processing the collected user features. Examples include biometrics, physical features, and behavioral features. Biometrics include, but are not limited to, fingerprints, voiceprints, faces, heart rate, and pulse. Physical features include, but are not limited to, height and limb length. Behavioral features include, but are not limited to, signatures, gait, and touch screen behavior (such as key press rhythm). In other words, this application can quantify a user through a dataset of user features. This application uses the tracking result as a user model as an example for illustration.

[0073] Understandably, Figure 1 The scenario 10 shown is just an example. In a real implementation, the number of devices, rooms, users, and devices carried by users in scenario 10 can be more or less.

[0074] The devices involved in the embodiments of this application may include, but are not limited to, smart TVs, smart cameras, smart speakers, smart projectors, smart routers, smart gateways and other home devices, smart bracelets, smart glasses and other wearable devices, or other mobile phones, tablets, handheld computers, personal digital assistants (PDAs), desktops, laptops, notebook computers, ultra-mobile personal computers (UMPCs), netbooks, smart screens and other devices.

[0075] For example, Figure 1 In the scenario 10 shown, the first device 101 is a mobile phone, the second device 102 is a smart bracelet, the third device 111 is a smart router, the fourth device 112 is a smart projector, the sixth device 121 is a smart speaker, the eighth device 131 is a smart TV, and the fifth device 113, the seventh device 122, and the ninth device 132 are all smart cameras.

[0076] Please see Figure 2 , Figure 2 This is a schematic diagram of the architecture of a first system 20 provided in an embodiment of this application. The first system 20 is a distributed system.

[0077] like Figure 2As shown, the first system 20 may include multiple devices that can connect and communicate with each other via wired and / or wireless means. Wired means may include at least one of the following: Universal Serial Bus (USB), twisted pair, coaxial cable, fiber optic cable, and gateway devices (e.g., routers, access points, APs). Wireless means may include at least one of the following: Wi-Fi, Bluetooth, and cellular communication technologies.

[0078] In some embodiments, multiple devices and connection media under the first system 20 can form a local area network and communicate through the local area network.

[0079] In some embodiments, multiple devices within the first system 20 can be trusted devices to each other. For example, terminal devices such as mobile phones, tablets, and smart bracelets can have applications installed for communication (e.g., but not limited to Huawei Smart Home). These applications can log in to accounts and are hereinafter referred to as account applications. The network mentioned above can include an application server corresponding to the account application (hereinafter referred to as an account application server). Devices with the account application installed can log in to the same or associated account through the account application, thereby communicating through the account application server. Other devices besides terminal devices can connect to the account application server wirelessly via Bluetooth or wired via USB. For example, a user can manually add smart home devices via Bluetooth on Huawei Smart Home. The account application server can identify whether any electronic device is trustworthy; for example, devices that have connected to or are currently connected to the account application server within a preset time period are considered trustworthy.

[0080] Not limited to this, in specific implementations, devices can also undergo verification first, and only those that pass verification are considered trusted devices, i.e., belong to the first system 20. For example, a device can only access a Wi-Fi network (i.e., belong to the first system 20) after successful password verification. Alternatively, multiple devices in the first system 20 can achieve single sign-on (SSO) through cookies, JSONP, etc. This application does not limit the communication method of multiple devices in the first system 20.

[0081] For example, the first system 20 can be the Internet of Things (IoT) in a home setting, or a group of smart devices.

[0082] The first system 20 in this application can be a distributed system, meaning that multiple devices in the system work collaboratively. For example, multiple devices in the first system 20 collaboratively collect user characteristics, and the collected user characteristics can be shared. Distributed systems achieve resource integration among multiple devices, reducing the processing pressure on individual devices and resulting in higher availability.

[0083] In some embodiments, at least one central control device may exist in the distributed system for comprehensively scheduling the implementation of the identity authentication process. For ease of description, this application uses the example of a central control device in the first system 20. Exemplarily, the central control device distributes the process of determining the tracking result to multiple devices in a distributed manner, that is, multiple devices in the first system 20 collaboratively determine the tracking result based on shared user characteristics. The tracking result can also be shared. The shared data can be distributed and stored on multiple devices in the first system 20.

[0084] In some embodiments, the distributed storage of the first system 20 may be stored by multiple devices in the first system 20 through local memory. For example, some devices store user models, some devices store collected user features, and some devices store authenticated user features (which may be called authentication credentials).

[0085] Not limited to this, in specific implementations, the distributed storage of the first system 20 is also implemented through at least one server. For example, multiple devices in the first system 20 can connect to at least one server, which can be a hardware server or a cloud server. Optionally, this at least one server can be a server with a database installed, and any device in the first system 20 can download data from the server and upload data to the server. This at least one server can store collected user characteristics, established user models, relevant information of the user models, tracking results, etc.

[0086] The following describes an exemplary electronic device provided in the embodiments of this application.

[0087] Please see Figure 3 , Figure 3 An exemplary schematic diagram of an electronic device 30 is shown. The electronic device 30 may be... Figure 1 Any device in scenario 10 shown can also be Figure 2 Any one of the devices in the first system 20 shown. Electronic device 30 may include processor 310, memory 320 and transceiver 330, which can be interconnected via a bus.

[0088] Processor 310 may be one or more central processing units (CPUs). If processor 310 is a CPU, the CPU may be a single-core CPU or a multi-core CPU. Memory 320 includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), or compact disc read-only memory (CD-ROM). Memory 320 is used to store related computer programs and data.

[0089] Transceiver 330 is used to receive and transmit data. In some embodiments, transceiver 330 can provide wireless communication solutions, including 2G / 3G / 4G / 5G, for use on electronic device 30. In some embodiments, transceiver 330 can provide wireless communication solutions, including wireless local area networks (WLAN) (such as Wi-Fi networks), Bluetooth (BT), global navigation satellite system (GNSS), frequency modulation (FM), near field communication (NFC), and infrared (IR) technologies, for use on electronic device 30. Electronic device 30 can communicate with networks and other devices via transceiver 330 using wireless communication technologies. Wireless communication technologies may include Global System for Mobile Communications (GSM), General Packet Radio Service (GPRS), Code Division Multiple Access (CDMA), Wideband Code Division Multiple Access (WCDMA), Time-Division Code Division Multiple Access (TD-SCDMA), Long Term Evolution (LTE), BitTorrent, GNSS, WLAN, NFC, FM, and / or IR technologies. The GNSS may include Global Positioning System (GPS), Global Navigation Satellite System (GLONASS), BeiDou Navigation Satellite System (BDS), Quasi-Zenith Satellite System (QZSS), and / or Satellite Based Augmentation Systems (SBAS).

[0090] In some embodiments, the electronic device 30 may further include a display screen. The display screen is used to display images, videos, text, etc. The display screen includes a display panel. The display panel may be a liquid crystal display (LCD), an organic light-emitting diode (OLED), an active-matrix organic light-emitting diode (AMOLED), a flexible light-emitting diode (FLED), a MiniLED, a MicroLED, a Micro-OLED, a quantum dot light-emitting diode (QLED), etc. Optionally, the electronic device 30 may include one or N display screens, where N is a positive integer greater than 1.

[0091] In some embodiments, the electronic device 30 may further include at least one acquisition module, any one of which can be used to acquire at least one user feature. This application uses an acquisition module to acquire one user feature as an example for illustration.

[0092] For example, electronic device 30 may include one or N cameras, where N is a positive integer greater than 1. The cameras are used to capture still images or videos. Electronic device 30 can also capture a user's facial features through the cameras.

[0093] For example, electronic device 30 may include a touch sensor, also known as a "touch device." Optionally, the touch sensor may be disposed on the display screen, and the touch sensor and the display screen constitute a touch screen, also known as a "touchscreen." When a touch operation is applied to the display screen, electronic device 30 can detect the intensity, location, etc., of the touch operation through the touch sensor, and transmit the detected touch operation to processor 310 to determine the type of touch event. Optionally, electronic device 30 may also provide visual output related to the touch operation through the display screen. Not limited to this, the touch sensor may also be disposed on the surface of electronic device 30, in a different location than the display screen. Electronic device 30 can collect user touch screen behavior characteristics (e.g., the location and area of ​​the touch area, timestamp, number of touches, pressure magnitude, etc.) through the touch sensor.

[0094] For example, electronic device 30 may include a pulse sensor. In some embodiments, the pulse sensor can detect pressure changes generated during arterial pulsation and convert them into electrical signals. There are many types of pulse sensors, such as piezoelectric pulse sensors, piezoresistive pulse sensors, and photoelectric pulse sensors. Piezoelectric and piezoresistive pulse sensors can convert the pressure process of a pulse beat into a signal output using micro-pressure materials (such as piezoelectric elements, bridges, etc.). Photoelectric pulse sensors can convert changes in the transmittance of blood vessels during a pulse beat into a signal output through reflection or transmission, i.e., acquiring pulse signals through photoplethysmography (PPG). Electronic device 30 can collect the user's pulse characteristics through the pulse sensor.

[0095] For example, electronic device 30 may include a heart rate sensor. In some embodiments, the heart rate sensor may acquire heart rate signals via PPG. The heart rate sensor may convert changes in vascular dynamics, such as changes in pulse rate (heart rate) or blood volume (cardiac output), into signal outputs via reflection or transmission. In some embodiments, the heart rate sensor may measure signals of electrical activity induced in cardiac tissue via electrodes attached to the skin, i.e., acquire heart rate signals via electrocardiography (ECG). Electronic device 30 may acquire the user's heart rate characteristics via the heart rate sensor.

[0096] For example, electronic device 30 may include at least one microphone. The microphone is used to convert sound signals into electrical signals. When making a phone call or sending a voice message, the user can speak by bringing their mouth close to the microphone, inputting sound signals into the microphone. Electronic device 30 can collect the user's voiceprint characteristics through the microphone.

[0097] In this application, the electronic device 30 can belong to a distributed system. The electronic device 30 can connect and communicate with other devices in the distributed system via transceiver 330, for example, by sharing authentication results and tracking results. When a first user triggers authentication of the electronic device 30, for example, when the processor 310 receives a detection signal from a touch sensor, the processor 310 can identify whether the first user is a legitimate user based on the shared tracking results, thereby determining whether authentication is exempted. A legitimate user is a user who has been authenticated by any device in the distributed system and whose authentication result is successful. The processor 310 can determine whether the user is a legitimate user based on the shared authentication results. When the first user is a legitimate user, the processor 310 can determine that authentication is exempted and instruct the display screen to show the user interface when authentication is successful, such as the desktop of the electronic device 30 or a specific interface of an application. Even if the electronic device 30 does not include a data acquisition module, it can still identify the user's identity and determine whether authentication is exempted based on the shared tracking results, resulting in high usability.

[0098] The processor 310 in the electronic device 30 can be used to read computer programs and data stored in the memory 320 and execute them. Figures 6-10 The identity authentication method shown can be any device in the distributed system, where electronic device 30 can be any device.

[0099] Please see Figure 4 , Figure 4 An exemplary schematic diagram of another electronic device 30 is shown. Electronic device 30 may be... Figure 1 Any device in scenario 10 shown can also be Figure 2 Any device in the first system 20 shown. Electronic device 30 belongs to a distributed system.

[0100] like Figure 4 As shown, the electronic device 30 may include an authentication unit 401. The authentication unit 401 is used to identify the user's identity based on the tracking results shared by the distributed system and to determine whether authentication is exempted. The tracking results are used to characterize the identities of users currently being tracked by multiple devices in the distributed system; the tracking results can be user models. When the user to be authenticated is determined to be a legitimate user, the authentication unit 401 determines that authentication is exempted; when the user to be authenticated is determined to be an illegitimate user, the authentication unit 401 triggers a manual authentication process to obtain an authentication result. Here, a legitimate user is a user who has been authenticated by any device in the distributed system and whose authentication result is successful.

[0101] In some embodiments, the electronic device 30 may further include a data acquisition unit for acquiring at least one user feature. Optionally, the data acquisition unit may include at least one data acquisition subunit, each subunit being used to acquire one user feature; for example, a data acquisition subunit may include... Figure 3The illustrated electronic device 30 includes a data acquisition module. Exemplarily, a face acquisition unit is used to acquire facial features. A gait acquisition unit is used to acquire gait features. A pulse acquisition unit is used to acquire pulse features. A heart rate acquisition unit is used to acquire heart rate features. A touchscreen behavior acquisition unit is used to acquire touchscreen behavior features. A location acquisition unit is used to acquire the user's location features.

[0102] When authentication unit 401 triggers the manual authentication process for the first user, the collection unit can collect at least one user characteristic of the first user and provide it to authentication unit 401 for authentication to obtain the authentication result of the first user. This authentication result can be shared with multiple devices in the distributed system. When the authentication result is successful, the multiple devices in the distributed system can determine that the first user is a legitimate user.

[0103] In some embodiments, the electronic device 30 may further include a model unit 402. The model unit 402 can determine the user model (i.e., tracking result) of the first user based on the user characteristics of the first user shared by the distributed system, wherein the user characteristics of the first user shared by the distributed system are collected by multiple devices in the distributed system at different times. The multiple devices in the distributed system may also share at least one established user model. Optionally, when there is no user model of the first user among these at least one user models, the model unit 402 can perform modeling training based on the aforementioned shared user characteristics of the first user (implemented, for example, using a support vector machine (SVM)) to obtain the user model of the first user. Optionally, when there is a user model of the first user among these at least one user models, the model unit 402 can select the user model of the first user from these at least one user models based on the aforementioned shared user characteristics of the first user.

[0104] In some embodiments, the electronic device 30 may further include a model management unit 403. When the authentication unit 401 determines that the authentication result of the first user is successful, the authentication unit 401 may send the authentication result to the model management unit 403. The model unit 402 may also send the user model (i.e., tracking result) of the first user to the model management unit 403. The model management unit 403 may determine that the first user is a legitimate user based on the user model of the first user and the authentication result of the first user.

[0105] In some embodiments, when the authentication unit 401 determines that the authentication result of the first user is successful, the authentication unit 401 may also send the user characteristics (referred to as authentication credentials) of the first user authenticated through the manual authentication process to the model management unit 403. The model management unit 403 can bind the first user's authentication credentials and the first user's user model together. The authentication unit 401 can determine the user to be authenticated as the first user based on the first user's user model (i.e., tracking result) and authenticate the first user's authentication credentials. When authentication is successful, the first user is a legitimate user. When authentication fails, the authentication unit 401 can trigger a manual authentication process.

[0106] In some embodiments, electronic device 30 may belong to multiple distributed systems. Electronic device 30 may share the authentication result of the first user with devices in the multiple distributed systems to which it belongs. If the authentication result of the first user is successful, the devices in these multiple distributed systems may all identify the first user as a legitimate user. At this time, the user model of the first user may be associated with these multiple distributed systems. Model management unit 403 may be used to manage the association relationship between user models and systems. Authentication unit 401 may cooperate with model management unit 403 to identify the identity of the user to be authenticated: that is, firstly, the user to be authenticated is determined to be the first user based on the user model of the first user (i.e., the tracking result), and then the association between the user model of the first user and the distributed system to which electronic device 30 belongs is determined. When the user model of the first user and the distributed system to which electronic device 30 belongs are associated, authentication unit 401 may determine that the first user is a legitimate user and determine that authentication is not required.

[0107] Understandably, a system can also be associated with multiple user models. Optionally, a distributed system can implement device access control for different users based on their user models. For example, the first system 20 has an access policy: users can only use entertainment devices (such as televisions, computers, mobile phones, etc.) in the first system 20 if their height is higher than a preset height threshold. Assume the first system 20 is associated with the user models of the first user and the second user, both of which include the user's height feature. The first user's height is below the preset height threshold, while the second user's height is above the preset height threshold. Therefore, when the first user triggers authentication for an entertainment device in the first system 20, the entertainment device will not display the user interface required for successful authentication, even if the first user is deemed legitimate. However, when the second user triggers authentication for an entertainment device in the first system 20, the entertainment device can be used without authentication.

[0108] In some embodiments, each user model may have a validity period, and the model management unit 403 may also be used to manage the validity period of the user models. The start time of the validity period may be the moment when the user represented by the user model most recently triggered authentication and was identified as a legitimate user. Alternatively, the start time of the validity period may also be the moment when the user represented by the user model most recently triggered a manual authentication process and determined that the authentication result was successful. The duration of the validity period may be agreed upon by multiple devices in the distributed system, for example, the validity period of each user model is a preset duration. Optionally, after obtaining the user model (i.e., the tracking result) of the first user, the authentication unit 401 may determine whether the user model is within the validity period. If the user model is within the validity period, the authentication unit 401 may determine that the first user is a legitimate user; if the user model is not within the validity period, the authentication unit 401 may trigger a manual authentication process. If the authentication result of the first user obtained by the manual authentication process is that the authentication is successful, the model management unit 403 can reset the validity period of the first user's user model, that is, set the start time of the validity period of the user model to the time when the authentication result of the first user is determined to be successful (for example, the time when the model management unit 403 receives the authentication result of the first user sent by the authentication unit 401).

[0109] Beyond the scenarios listed above, in specific implementations, the validity period can also be determined by the distributed system based on the frequency of user authentication triggers as represented by the user model. For example, if the frequency of user authentication triggers is greater than a first threshold or less than a second threshold (i.e., too fast or too slow), the distributed system can shorten the validity period of that user's user model. Alternatively, the validity period can also be determined by the distributed system based on the duration the user has been tracked, such as... Figure 1 The example shows the duration of user 100's presence in scenario 10. For instance, if a user is tracked for less than the third threshold, the distributed system can shorten the validity period of that user's user model. This application does not specify the exact method for determining the validity period.

[0110] In some embodiments, authentication credentials may also have an expiration date. The model management unit 403 can also be used to manage the expiration date of authentication credentials. The expiration date of authentication credentials is similar to that of user models and will not be described again.

[0111] In some embodiments, the electronic device 30 may further include a system management unit 404. The system management unit 404 manages at least one distributed system to which the electronic device 30 belongs and multiple devices included in each distributed system. The electronic device 30 can obtain information about the at least one distributed system it belongs to through the system management unit 404, such as system identifier, number of included devices, device identifiers, device addresses, etc. For example, the electronic device 30 can communicate with other devices in the distributed system through the system management unit 404. The system management unit 404 may include... Figure 3 The electronic device 30 shown includes a transceiver 330.

[0112] In some embodiments, the electronic device 30 may further include a storage unit, which can be used to store data shared by the distributed system, such as tracking results, at least one established user model, the association between the user model and the distributed system, and authentication credentials bound to the user model. Any unit in the electronic device 30 can retrieve the stored data from the storage unit to perform an authentication process. Optionally, the storage unit includes... Figure 3 The electronic device 30 shown includes a memory 320.

[0113] In this application, even if the electronic device 30 has low processing performance (e.g., only includes an authentication unit and a storage unit), it can still identify the user's identity and determine whether authentication is exempted based on the tracking results and authentication credentials shared by the distributed system, thereby reducing the impact on power consumption and achieving high availability.

[0114] Understandably, Figure 4 The electronic device 30 shown includes an authentication unit 401, a model unit 402, a model management unit 403, and a system management unit 404, which can belong to... Figure 3 The electronic device 30 shown includes a processor 310.

[0115] Please see Figure 5 , Figure 5 An exemplary schematic diagram of the architecture of another first system 20 is shown. The first system 20 is a distributed system.

[0116] like Figure 5 As shown, the first system 20 may include a device acquisition layer 501, a model tracking layer 502, an authentication layer 503, and a database 504. The descriptions of each layer are as follows:

[0117] The device acquisition layer 501 is used to collect user characteristics. The device acquisition layer 501 may include acquisition units of multiple devices with acquisition capabilities in the first system 20, such as the acquisition unit of device 1, the acquisition unit of device 2, ..., the acquisition unit of device n-1. Figure 5Taking device n in the first system 20 as an example, which does not have data acquisition capabilities (i.e., does not include a data acquisition unit), the data acquisition units of multiple devices in the device acquisition layer 501 can work collaboratively to collect multi-dimensional user features of the first user in real time. This means collecting various types of user features of the first user at multiple time points, which are then used by the model tracking layer 502 to determine the user model of the first user. Optionally, the multi-dimensional user features collected in real time by the device acquisition layer 501 can be sent to the database 504. Any device in the distributed system can retrieve data from the database 504 to achieve the identity authentication process.

[0118] For example, the device acquisition layer 501 includes devices such as mobile phones, tablets, televisions, large screens, and cameras for acquiring image data; devices such as routers, smartwatches, smart bracelets, mobile phones, and tablets for acquiring location data; and devices such as headphones, smart bracelets, smartwatches, mobile phones, and tablets for acquiring motion-sensing data. The image data can be used to obtain features such as facial information, body shape information, behavioral information, and location information. Motion-sensing data includes features such as body shape information, behavioral information, heart rate information, and voiceprint information.

[0119] like Figure 5 As shown, the model tracking layer 502 may include processing units such as feature extraction, user model building and updating, and situation tracking. Wherein:

[0120] Feature extraction is used to process user features acquired in real time by the device acquisition layer 501. The processed user features are used for user model building and updating, and situational tracking. For ease of explanation, this application refers to the user features acquired in real time by the device acquisition layer 501 as the first user features, and the user features obtained by the feature extraction of the model tracking layer 502 after processing the first user features as the second feature data. For example, the facial features acquired in real time by the device acquisition layer 501 are provided as the first feature data to the feature extraction of the model tracking layer 502. Feature extraction can extract the relative position and relative size of the most representative parts of the face (such as eyebrows, eyes, nose, mouth, etc.) as the second feature data, and then supplement it with the shape information of the facial contour as the second feature data.

[0121] For example, feature extraction may include: obtaining location information based on image data, obtaining location information through near-field communication technologies such as Bluetooth, Wi-Fi, and UWB, extracting biometric features (such as obtaining facial information based on image data), and extracting features from somatosensory data.

[0122] The establishment and updating of the user model are used to train the model on the second feature data obtained from feature extraction (implementation method such as SVM) to obtain the corresponding user model. Optionally, the establishment and updating of the user model can also update the existing user model based on the second feature data obtained from feature extraction, so that the user model can more accurately quantify users. Optionally, the user model obtained above and the updated user model can be sent to the database 504.

[0123] Situation tracking is used to determine the corresponding user model from existing user models based on the second feature data obtained from feature extraction. The existing user model can be obtained from a database 504 by situation tracking. For example, situation tracking can process the second feature data obtained from feature extraction to obtain a corresponding dataset. Then, situation tracking can obtain the similarity between this dataset and the existing user model (for example, if both the dataset and the existing user model include facial features, the cosine distance or Euclidean distance of the facial feature vectors can be used as the similarity). When the similarity between the dataset and the first user model is greater than a preset similarity threshold, situation tracking can determine that the first user model is used to represent the user to whom the first feature data and the second feature data belong; that is, the first user model is the tracking result.

[0124] In some embodiments, the model tracking layer 502 may further include behavior prediction. Behavior prediction is used to determine whether a user has triggered device authentication based on second feature data obtained from feature extraction. For example, the second feature data belongs to a first user and includes location features. When the distance between the location features and device n in the first system 20 gradually decreases and becomes less than a preset distance, behavior prediction can determine that the first user has triggered device n authentication. When behavior prediction determines that the first user has triggered device n authentication, it can send a notification to the authentication layer 503, so that the authentication unit of device n in the authentication layer 503 can identify the identity of the first user and determine whether authentication is exempted.

[0125] Understandably, Figure 5 The model tracking layer 502 shown may include model units of multiple devices in the first system 20.

[0126] like Figure 5 As shown, the authentication layer 503 is used to identify the user's identity based on the tracking results obtained from the model tracking layer 502 and to determine whether authentication is exempt. The authentication layer 503 may include authentication units of multiple devices with authentication capabilities in the first system 20, such as the authentication unit of device 1, the authentication unit of device 2, ..., the authentication unit of device n. These multiple authentication units can collaboratively identify the user's identity based on the tracking results, or any single device's authentication unit can identify the user's identity based on the tracking results.

[0127] In some embodiments, the authentication layer 503 may further include model management. Model management is used to manage the association between user models and the system. When any device in the first system 20 determines that the authentication result of the first user's manual authentication process is successful, model management can determine that the first user's user model is associated with the first system 20. At this time, if the authentication unit of any device in the first system 20 determines that the user to be authenticated is the first user based on the tracking result, then collaborative model management can determine that the first user is associated with the first system 20, thus determining that the first user is a legitimate user and providing the user with a seamless, authentication-free function. Optionally, the above-mentioned association between the user model and the system can be sent to the database 504.

[0128] In some embodiments, the authentication layer 503 may further include system management. System management is used to manage the first system 20, such as managing the identifier of the first system 20, the number of devices included in the first system 20, device identifiers, device addresses, etc. For example, the central control device in the first system 20 can determine whether a device to be authenticated belongs to the first system 20 based on the identifier of the device to be authenticated through system management. Alternatively, the central control device can obtain the identifier of the first system 20 through system management and determine the user model associated with the first system 20 through model management.

[0129] For example, system management can perform device discovery in the first system 20, such as actively searching for devices that meet certain criteria and identifying those devices as included in the first system 20. System management can also perform device authentication in the first system 20, such as authenticating devices accessing the first system 20 (e.g., verifying the correctness of a Wi-Fi password), and confirming that the device is included in the first system 20 upon successful authentication. System management can also enable trusted transmission between devices in the first system 20, such as managing a first key used to encrypt and decrypt data transmitted between multiple devices in the first system 20.

[0130] Understandably, Figure 5 The model management in the authentication layer 503 shown may include model management units of multiple devices in the first system 20, and the system management may include system management units of multiple devices in the first system 20.

[0131] In some embodiments, model management can also be used to manage authentication credentials bound to user models. When any device in the first system 20 determines that the authentication result of the first user's manual authentication process is successful, the authentication unit of that device can send the user characteristics authenticated in the manual authentication process as authentication credentials to model management. Model management can then bind these authentication credentials to the user model (i.e., the tracking result) obtained by the model tracking layer 502. At this time, if the authentication unit of any device in the first system 20 determines that the user to be authenticated is the first user based on the tracking result, the model management will then authenticate the authentication credentials bound to the first user's user model. When authentication is successful, the authentication unit can determine that the first user is a legitimate user and provide the user with a seamless, authentication-free function. Optionally, the authentication credentials bound to the user model can be sent to the database 504.

[0132] In some embodiments, model management can also be used to manage the validity period of user models. When performing authentication, the authentication layer 503 can first determine, in conjunction with model management, whether the user model obtained from situation tracking is within its validity period. If the user model is not within its validity period, the authentication layer 503 can determine that the user corresponding to the user model is illegitimate and can trigger a manual authentication process. If the user model is within its validity period, the authentication layer 503 then determines whether the user corresponding to the user model is legitimate and whether authentication is exempt.

[0133] In some embodiments, model management can also be used to manage the validity period of authentication credentials. When performing authentication, the authentication layer 503 can first determine, in conjunction with model management, whether the authentication credentials corresponding to the user model obtained from situation tracking are within their validity period. If the authentication credentials are not within their validity period, the authentication layer 503 can determine that the user corresponding to the user model is illegitimate and can trigger a manual authentication process. If the authentication credentials are within their validity period, the authentication layer 503 then determines whether the user corresponding to the user model is legitimate and whether authentication is exempt.

[0134] In authentication layer 503, any device's authentication unit can also perform a manual authentication process, i.e., authenticate the collected user features of the first user and obtain the authentication result of the first user. In some embodiments, when the authentication result of the first user is successful, a model update can be triggered. That is, when the authentication result of the synchronized first user is successful, the establishment and updating of the user model in model tracking layer 502 can update the user model of the first user, for example, by adding the collected user features of the first user to the user model of the first user. In some embodiments, when the authentication result of the first user is successful, model management can also be triggered to update the relevant information of the user model of the first user, such as updating the validity period of the user model of the first user and the authentication credentials bound to the user model of the first user.

[0135] like Figure 5 As shown, database 504 is used to store at least one of the following: the user model established by model tracking layer 502, the updated user model, the authentication credentials used by the authentication unit of authentication layer 503 during the manual authentication process, the association between the user model and the system determined by model management, and the authentication credentials bound to the user model. Database 504 can be a storage unit including multiple devices in the first system 20, but is not limited to this, and can also be a database connected to multiple devices (such as...). Figure 1 (at least one server mentioned above). Database 504 can be used to synchronize the data stored above, and any device in the first system 20 can retrieve the data stored above from database 504.

[0136] Not limited to the cases listed above, in specific implementations, situational tracking can also be a neural network model. The input of this model is either first feature data or second feature data, and the output is a user model, which is used to represent the user to whom the first feature data or second feature data belongs. This application does not limit the specific method for determining the tracking result.

[0137] Figure 5 For descriptions of the acquisition unit, authentication unit, model unit, model management unit, and system management unit shown, please refer to [link to relevant documentation]. Figure 4 The electronic device 30 shown includes a data acquisition unit, an authentication unit 401, a model unit 402, a model management unit 403, and a system management unit 404.

[0138] Understandably, the authentication in this application can be system-level authentication or application-level authentication. System-level authentication, for example, is screen unlocking; before successful authentication, the device displays the lock screen interface, and after successful authentication, the device displays the desktop or the user interface that was displayed before the lock screen. Application-level authentication, for example, is authentication of system applications (such as settings, file vault, SMS, gallery, etc.); only after successful authentication can the device display the specific interface of the system application. Application-level authentication can also be authentication of applications hosted on application servers (such as payment applications, banking applications, social applications, etc.); only after successful authentication can the device display the specific interface of such applications.

[0139] For applications with an application server, the device can obtain application data from the application server and use the obtained data to display the corresponding user interface. In some embodiments, the application server requires user authentication before sending application data. When the device's authentication unit determines the user model and system association of the user to be authenticated, it can send a notification message to the application server. The application server can then determine that the user to be authenticated is a legitimate user based on this notification message and send the corresponding data to the device.

[0140] The authentication unit of any device in authentication layer 503 may include a local authentication unit and an application authentication unit. For example, the authentication unit of device 1 includes a local authentication unit and an application authentication unit. After obtaining the authentication result, the local authentication unit can directly use it for system-level authentication of device 1. After obtaining the authentication result, the application authentication unit can directly use it for application-level authentication of device 1. Optionally, after obtaining the authentication result, the local authentication unit can send it to the application authentication unit for application-level authentication of device 1.

[0141] In some embodiments, the model management of authentication layer 503 is also used to manage authentication credentials bound to user models. After the authentication unit of the device in authentication layer 503 obtains the user model (i.e., the tracking result) of the user to be authenticated, it can authenticate the authentication credentials bound to the user model. If the authentication is successful, the user to be authenticated is determined to be legitimate, and authentication exemption is determined. Optionally, for applications with an application server, the device can send the authentication credentials to the application server for authentication. When the application server determines that the authentication is successful, it will send the corresponding data to the device. At this time, the device determines that authentication exemption is granted, that is, it uses the data sent by the application server to display the corresponding user interface.

[0142] The following describes the identity authentication method provided in the embodiments of this application.

[0143] Please see Figure 6 , Figure 6 This application provides an identity authentication method. This method can be applied to... Figure 1 Scenario 10 is shown. This method can be applied to distributed systems that include multiple devices, such as... Figure 2 The first system 20 shown Figure 5 The first system 20 is shown. These multiple devices may include a first device, a second device, and a third device.

[0144] It should be noted that the terms "first device," "second device," and "third device" are used only to distinguish the roles of the devices performing the authentication method. For example, the first device is the device that the user manually authenticates before using the third device. The second device is used to share tracking results (i.e., the user model determined based on real-time collected user characteristics). The third device is the device the user wants to authenticate. In specific implementations, a device can have multiple roles; for example, a device can be both the first and second device. Multiple devices can also share the same role; for example, three devices can all be the second device, and these three devices collaboratively acquire tracking results and share them with other devices in the distributed system. The role of any device in the distributed system can be determined according to the actual situation, and this application does not impose any limitations on this.

[0145] For example, the first user is located in a scenario where multiple devices exist within a distributed system, such as a distributed system... Figure 2 The first system 20 shown is a distributed system in which multiple devices are located. Figure 1 Scenario 10 is shown. Multiple devices in the distributed system can continuously track the first user, i.e., collect the user characteristics of the first user at multiple time points, and determine the user model of the first user (i.e., tracking results) based on the user characteristics. The first device can be the device that triggers authentication for the first user in Scenario 10 for the first time. The first device can perform a manual authentication process and obtain the authentication result of the first user. Both the tracking result and the authentication result can be shared within the distributed system. When the first user triggers other devices in the distributed system, such as a third device, the third device can determine that the first user is a legitimate user based on the tracking result and the authentication result, and determine that authentication is not required (i.e., display the user interface when authentication is successful). This achieves the effect of authentication at one end and authentication exemption at multiple ends, reducing the number of authentication operations performed when the user switches devices.

[0146] This method may include, but is not limited to, the following steps:

[0147] S101: The first device authenticates the first user based on the first user's first user characteristics and obtains the first user's authentication result.

[0148] Specifically, a first user can trigger a manual authentication process on a first device. This involves the first device collecting the first user's first user features and authenticating those features to obtain the authentication result. For example, if the first user feature is a facial feature, the first device can measure the similarity between the collected facial features and pre-recorded facial feature samples (e.g., using cosine or Euclidean distance between facial feature vectors). When the similarity is greater than or equal to a preset similarity threshold, the first device determines the authentication result as successful; when the similarity is less than the preset threshold, the first device determines the authentication result as unsuccessful. If the authentication result is successful, the first device can display the user interface, such as a system device or application-specific interface. The first user features can include multiple types of user features, making authentication more accurate and reliable with a lower false recognition rate compared to single-feature authentication methods.

[0149] S102: The first device shares the authentication result of the first user with the devices in the distributed system.

[0150] Specifically, the authentication result obtained by any device in a distributed system through a manual authentication process can be shared with multiple devices in the distributed system; this process can be called synchronous authentication. Synchronous authentication can be performed once every preset time interval, or it can be performed every time an authentication result is obtained.

[0151] S103: If the authentication result of the first user is successful, the second device shares the user model of the first user with the devices in the distributed system.

[0152] Specifically, multiple devices in a distributed system can collect user characteristics of the first user at multiple points in time. For example, device 1 (such as a mobile phone or tablet) collects motion sensor data and touch screen sensor data; device 2 (such as a smartwatch, smart bracelet, or other wearable device) collects motion sensor data, biometric data, and signal distance data; and device 3 (which may include a camera, such as a camera or a large screen) collects situational data and location data. See specific examples for further details. Figure 1 This describes how multiple devices in scenario 10 collect user characteristics of user 100 from various angles in the near-field three-dimensional space. The user characteristics collected can be synchronously transmitted to multiple devices in the distributed system.

[0153] Then, multiple devices in the distributed system, such as a second or third device, can determine the user model (i.e., tracking results) of the first user based on the user characteristics of the first user collected at multiple time points. The user characteristics (e.g., location characteristics) of the first user at multiple time points and the tracking results can be shared with any device in the distributed system. Therefore, any device in the distributed system can determine the current status of the tracked user, such as the location of the first user.

[0154] In this model, a user model can represent a single user, with different user models corresponding to different users. Optionally, a user model can be a dataset that includes various user features used to describe the user. These user features can be obtained by analyzing and processing collected user characteristics, serving as feature data to characterize the user's identity. Examples include biometric features, physical features, and behavioral features. Biometric features include, but are not limited to, fingerprints, voiceprints, facial features, heart rate, and pulse. Physical features include, but are not limited to, height and limb length. Behavioral features include, but are not limited to, signatures, gait, and touchscreen behavior (such as key press rhythm). The user characteristics of the first user at multiple time points can also include multiple user features. Determining the user model through multi-dimensional user features is more accurate and reliable, with a lower false positive rate.

[0155] Specifically, the distributed system may share at least one user model, which can be obtained by devices in the distributed system through modeling and training based on collected user features. In some embodiments, if the user model of the first user is not among these at least one user models, then multiple devices in the distributed system can build a user model of the first user based on the collected user features of the first user at multiple time points. In other embodiments, multiple devices in the distributed system can select a user model of the first user from these at least one user models based on the collected user features of the first user at multiple time points.

[0156] For example, the second device can process the user characteristics of the first user collected at multiple time points to obtain a corresponding dataset. Then, the second device can obtain the similarity between at least one user model shared by the distributed system and the dataset. When the similarity between the first user model and the dataset is greater than a preset threshold, the second device can determine that the dataset belongs to the user represented by the first user model, that is, the currently tracked user is the user represented by the first user model.

[0157] Understandably, when multiple devices in a distributed system are tracking a first user, they can combine near-field communication technologies (such as Bluetooth, Wi-Fi, and UWB) to obtain the first user's location features. These location features can be used to more accurately identify the user. For example, if the currently tracked users include a first user and a second user, and the locations of the two users are usually not the same, the multiple devices in the distributed system can distinguish between the first user and the second user based on the acquired location features. Optionally, user identification can also be based on whether the location features are continuous. The distance difference between the first user's location within two consecutive seconds should be less than a second distance threshold, and the distance difference between the second user's location within two consecutive seconds should also be less than a second distance threshold. The multiple devices in the distributed system can distinguish between the first user and the second user based on location features acquired at multiple time points.

[0158] S104: If the third device determines that the first user is within the preset range based on the user model of the first user, the third device determines that the first user is legitimate.

[0159] Specifically, the third device can determine whether the first user is within a preset range based on the first user's user characteristics (e.g., location characteristics) at multiple points in time and the first user's user model. When the third device determines that the first user is within the preset range, it can determine that the first user has triggered authentication by the third device, meaning that the first user has already initiated or is about to initiate an access request. Since the first device in the distributed system has determined that the first user's authentication result is successful, the third device can then determine that the first user is a legitimate user and determine that authentication is not required (e.g., directly displaying the user interface when authentication is successful).

[0160] The preset range can be a range relative to the distributed system. Optionally, the distance between the location of the first user and the location of any device in the distributed system is less than a first distance threshold. For example, multiple devices in the distributed system exist... Figure 1 In scenario 10, any position of the first user within scenario 10 indicates that the first user is within a preset range. This preset range can also be a range relative to the third device. Optionally, the distance between the location of the first user and the location of the third device is less than a second distance threshold. For example, multiple devices in a distributed system exist... Figure 1 In scenario 10, the third device is device 131 or device 132 in scenario 10, and the first user being in the third room 130 indicates that the first user is within the preset range.

[0161] In some embodiments, the method may further include: if the first authentication result is successful, multiple devices in the distributed system, such as a second device or a third device, may determine the user model of the first user and the association between the distributed system to which the first device belongs.

[0162] Specifically, a device can belong to at least one distributed system; for example, the first device not only belongs to... Figure 2 The first system 20 shown also belongs to other systems. Therefore, when the first device determines that the authentication result of the first user is successful, multiple distributed systems to which the first device belongs can all determine that the first user is a legitimate user. This can be characterized as multiple distributed systems being associated with the user model of the first user. Then, in S104, when the third device determines that the first user is within a preset range, it can determine whether the user model of the first user is associated with the distributed system to which the third device belongs. Only when the user model of the first user is associated with the distributed system to which the third device belongs will the third device determine that the first user is a legitimate user. Optionally, the tracking results shared by the second device can also carry indication information as to whether the user model of the first user is associated with the distributed system to which the third device belongs.

[0163] In some embodiments, the method may further include: if the first authentication result is successful, the first device may share the first user characteristics of the first user with the devices in the distributed system.

[0164] Specifically, the first user characteristic can be used as an authentication credential and bound to the first user's user model. In S104, when the third device determines that the first user is within a preset range, it can use the first user characteristic to authenticate the first user to obtain a second authentication result for the first user. When the second authentication result is successful, the third device determines that the first user is a legitimate user.

[0165] Understandably, multiple devices in a distributed system can track the first user after the first user has successfully completed manual authentication. However, this is not limited to tracking the first user; multiple devices in a distributed system can also track the first user when the first user first enters the scenario where the distributed system is located, or when the first user initiates an access request to a device in the distributed system (i.e., when the first user is determined to be within a preset range). For example, the tracking process may include: multiple devices in the distributed system performing behavioral similarity analysis (e.g., similarity analysis of collected motion sensor data, touch screen sensor data, and situational data), biometric data similarity analysis, and location analysis on the first user based on collected user characteristics to obtain tracking results. Then, when a device in the distributed system determines that the first user is within a preset range, it can combine the tracking results to determine the user's identity and whether authentication is exempted. If authentication is exempted, the user directly accesses the device; if authentication is not exempted, the device can perform a manual authentication process.

[0166] In some embodiments, prior to S104, the method may further include: multiple devices in the distributed system, such as a second device or a third device, determining that the first user is within a preset range based on the collected user characteristics of the first user at multiple time points, i.e., determining that the first user triggers authentication by the third device. In this case, the multiple devices in the distributed system may send indication information to the third device to enable the third device to determine that the first user is within the preset range, thereby triggering authentication. For example, the second device determines that the first user is within the preset range when the distance between the location of the first user and the location of the third device gradually decreases and is less than a third distance threshold.

[0167] In some embodiments, prior to S104, the method may further include: a third device receiving a first user operation from a first user. The third device may determine that the first user is within a preset range based on the first user operation and the tracking result. The first user operation may be, for example, but not limited to, a click operation, a swipe operation, a hover operation, a lift operation, etc., performed on the third device's display screen or a button.

[0168] In some embodiments, when multiple devices in a distributed system determine the tracking result, they can update the user model. For example, the user features of the first user collected at multiple time points include a first feature and a second feature, and the user model of the first user includes a third feature and a fourth feature. The first and third features are of the same type, and the second and fourth features are of the same type. For example, the first and third features are both fingerprint features, and the second and fourth features are both facial features. Since the confidence level of the first feature is greater than that of the second feature, when the similarity between the first and third features is greater than or equal to a preset similarity threshold, even if the similarity between the second and fourth features is less than the preset similarity threshold, the currently tracked user can still be identified as the first user. At this time, multiple devices in the distributed system can use the second feature to update the user model of the first user, for example, by adding the second feature to the user model of the first user, or by replacing the fourth feature with the second feature.

[0169] In some embodiments, multiple devices in the distributed system can also collect environmental parameters in the near-field space where the distributed system is located, such as the environmental parameters in scenario 10. These parameters include, but are not limited to, whether lights are on, brightness, object placement, and current time. Multiple devices in the distributed system can combine the real-time collected environmental parameters with the user characteristics of the first user to determine the tracking result. For example, the first user's user model includes two types of facial features: facial features at 6 AM without lights on, and facial features at 6 PM with lights on. The real-time collected environmental parameters are: lights on, current time 8 PM. The real-time collected user characteristics of the first user include facial features. Combining the environmental parameters, it can be determined that the facial features at 6 PM with lights on have a high confidence level. Therefore, the similarity between the facial features at 6 PM with lights on and the real-time collected facial features of the first user can be obtained. When this similarity is greater than a preset similarity threshold, the tracking result is determined to be the user model of the first user.

[0170] Beyond the scenarios listed above, in practical implementations, multiple devices in a distributed system can periodically acquire environmental parameters in the near-field space where the distributed system resides, and update the user model based on these parameters. For example, multiple devices in a distributed system can acquire the placement of objects to determine if there are obstructions around the devices. If multiple devices in the distributed system determine that there are obstructions around the device used to acquire facial features (such as a camera), the confidence level of the facial features in the user model can be lowered.

[0171] In some embodiments, the distributed system can also update authentication credentials. For example, each time the authentication result for the first user is successful after a manual authentication process, the distributed system can update the authentication credentials using the user characteristics used in this authentication process. That is, it can bind the user characteristics as authentication credentials to the first user's user model, or replace the authentication credentials already bound to the first user's user model with the user characteristics.

[0172] For a detailed implementation of the above authentication method, please refer to [link / reference]. Figures 7-8 Example shown.

[0173] exist Figure 6 In the method shown, after the first user is authenticated by the first device, multiple devices in the distributed system can determine that the first user is a legitimate user and perform a continuous, multi-dimensional tracking process to obtain tracking results. Any device in the distributed system can have authentication requirements; that is, any device can be a third-party device. Even if the third-party device lacks data collection capabilities, it can still identify the user's identity based on the shared tracking results, resulting in high availability. The third-party device can directly exempt the legitimate first user from authentication, reducing the number of authentication operations performed when the legitimate user switches devices, achieving the effect of one-end authentication and multi-end authentication exemption. Compared to authentication methods based on single user characteristics and one-time authentication methods, identifying user identity through tracking results obtained through continuous tracking offers higher security and reliability.

[0174] The following describes the application scenarios involved in the embodiments of this application and the human-computer interaction diagrams in these scenarios. The following embodiments use the architecture of the first system 20 as an example. Figure 5 The architecture shown is used as an example for description. The following embodiments only show a portion of the processing units in each layer.

[0175] First, let's introduce the various levels in the first system 20. Figure 6 The cooperative relationships in S101-S103 are shown in Figure 7.

[0176] 1. In authentication layer 503, the authentication unit of device 1 (i.e., the first device) determines that the first user triggers the manual authentication process.

[0177] 2. The acquisition unit of device 1 (i.e., the first device) in the device acquisition layer 501 acquires the first user characteristics of the first user.

[0178] 3. The acquisition unit of device 1 sends the first user feature to the authentication unit of device 1.

[0179] 4. The authentication unit of device 1 authenticates the characteristics of the first user and obtains the authentication result of the first user.

[0180] 5. The authentication unit of device 1 sends the authentication result of the first user to the database 504, that is, it shares the authentication result of the first user with the devices in the first system 20.

[0181] 6. The device acquisition layer 501 collects the user characteristics of the first user at multiple time points, such as continuously collecting the multi-dimensional second user characteristics of the first user.

[0182] 7. The device acquisition layer 501 sends the second user feature to the feature extraction of the model tracking layer 502.

[0183] 8. Feature extraction in model tracking layer 502 processes the second user features to obtain the third user features. An example of the processing method can be found in [link to example]. Figure 5 The illustration shows the feature extraction of the tracking layer 502 in the model.

[0184] 9. The feature extraction of the model tracking layer 502 sends the third user features to the situation tracking.

[0185] 10. The situation tracking of the model tracking layer 502 determines the user model (i.e., tracking result) of the first user based on the third user characteristics. Optionally, the situation tracking establishes the user model of the first user based on the third user characteristics. Optionally, the situation tracking obtains at least one user model shared in the database 504, and determines the user model of the first user from these at least one user model based on the third user characteristics.

[0186] 11. The situational tracking layer 502 obtains the authentication result of the first user shared in the database 504.

[0187] 12. If the authentication result of the first user is successful, the situation tracking of the model tracking layer 502 sends the user model of the first user to the database 504, that is, the user model (i.e., tracking result) of the first user is shared with the devices in the first system 20.

[0188] 13. The model management of the authentication layer 503 obtains the authentication result of the first user and the user model of the first user shared in the database 504.

[0189] 14. If the authentication result of the first user is successful, the model management of the authentication layer 503 determines the association between the user model of the first user and the first system 20 to which device 1 belongs.

[0190] 15. The model management of the authentication layer 503 sends the second instruction information to the database 504, that is, the second instruction information is shared with the devices in the first system 20, wherein the second instruction information is used to indicate the association between the user model of the first user and the first system 20 to which the device 1 belongs.

[0191] Among them, 1-5 correspond to Figure 6The order of S101-S102.1-5 and 6-10 is not limited and can be executed simultaneously.

[0192] In some embodiments, after the situational tracking of the model tracking layer 502 determines the user model (i.e., 10) of the first user, the establishment and updating of the user model in the model tracking layer 502 can update the user model of the first user based on the characteristics of the third user.

[0193] The following section introduces the various levels within System 20. Figure 6 The cooperative relationship shown in S104 is as follows: Figure 8 As shown.

[0194] 16. Device acquisition layer 501 collects user characteristics of the first user at multiple time points, such as continuously collecting multi-dimensional fourth user characteristics of the first user. The fourth user characteristics may be obtained when the first user approaches device n (i.e., the third device) or before.

[0195] 17. The device acquisition layer 501 sends the fourth user feature to the feature extraction of the model tracking layer 502.

[0196] 18. The feature extraction of the model tracking layer 502 processes the fourth user feature to obtain the fifth user feature.

[0197] 19. The feature extraction of the model tracking layer 502 sends the fifth user feature to the behavior prediction.

[0198] 20. The behavior prediction of the model tracking layer 502 can determine whether the first user is within a preset range based on the characteristics of the fifth user.

[0199] 21. The behavior prediction of the model tracking layer 502 sends a first notification to the authentication unit of device n in the authentication layer 503. The first notification is used to indicate that the first user is within a preset range.

[0200] 22. In authentication layer 503, the authentication unit of device n determines that the first user is within a preset range based on the first notification.

[0201] 23. In authentication layer 503, the authentication unit of device n obtains the user model of the first user shared in database 504.

[0202] 24. In authentication layer 503, the authentication unit of device n obtains the association between the user model of the first user and the system from the model management.

[0203] 25. In authentication layer 503, the authentication unit of device n obtains the system to which device n belongs from the system management.

[0204] 26. The authentication unit of device n in authentication layer 503 determines, based on the data obtained above, that device n belongs to the first system 20 associated with the user model of the first user, that is, the first user is determined to be legitimate. Therefore, the first user is exempt from authentication and the user interface when authentication is successful is directly displayed.

[0205] It should be noted that, Figure 7 The second user characteristics shown and Figure 8 The fourth user characteristic shown can be user characteristics at different points in time.

[0206] It should be noted that the first system 20 is implemented. Figure 8 During the process shown, the first user can be continuously tracked, i.e., during execution. Figure 7 6-10 shown. Figure 8 The user model of the first user obtained in step 24 is actually the tracking result obtained by the first system 20 continuously tracking the first user. If the first user is not in the scenario where the first system 20 is located when device n executes step 24, the obtained tracking result may not be the user model of the first user, but rather the user model of the user in the scenario where the first system 20 is located. The first system 20 will perform corresponding operations based on the current tracking result, such as determining whether the currently tracked user is a legitimate user.

[0207] In some embodiments, Figure 6 Before S101, Figure 6 The authentication method described may further include: multiple devices in the distributed system collecting user characteristics of the first user at multiple time points, and determining a user model where the first user does not exist based on the collected user characteristics. When the first device determines that the first user has triggered authentication, the first device can determine that the first user is illegitimate and trigger a manual authentication process (e.g., executing S101). For example, this is the scenario where the first user enters the distributed system for the first time and multiple devices are present. An example of the above process for determining and triggering the manual authentication process is as follows. Figure 9 As shown.

[0208] 1. When the first user enters scenario 10 where multiple devices are located in the first system 20, the device acquisition layer 501 continuously collects the first user's multi-dimensional sixth user characteristics.

[0209] 2. The device acquisition layer 501 sends the sixth user feature to the feature extraction of the model tracking layer 502.

[0210] 3. The feature extraction of the model tracking layer 502 processes the sixth user feature to obtain the seventh user feature.

[0211] 4. The feature extraction of the model tracking layer 502 sends the seventh user feature to the situation tracking.

[0212] 5. The situational tracking of the model tracking layer 502 retrieves at least one established user model from the database 504.

[0213] 6. The situational tracking of the model tracking layer 502 determines, based on the seventh user feature, the user model in which the first user does not exist among the above-mentioned user models.

[0214] 7. The situational tracking of the model tracking layer 502 sends a second notification to the database 504, that is, the second notification is shared with the devices in the first system 20. The second notification is used to indicate that there is no user model for the first user. The devices in the first system 20 can determine that the currently tracked first user is illegitimate based on the second notification.

[0215] 8. When the authentication unit of device 1 in authentication layer 503 determines that the first user is within a preset range, it can obtain the second notification shared in database 504. The preset range can be relative to the distributed system or relative to device 1. The authentication unit of device 1 can determine that the first user is within the preset range based on the location characteristics of the first user collected in real time by device acquisition layer 501.

[0216] 9. If the authentication unit of device 1 in authentication layer 503 determines that the currently tracked first user is illegitimate based on the second notification, then it determines to trigger the manual authentication process.

[0217] It should be noted that, Figure 9 The sixth user feature shown and Figure 7 The second user characteristic shown Figure 8 The fourth user characteristic shown can be user characteristics at different points in time.

[0218] Not limited to Figure 9 In the specific implementation, the situational tracking of the model tracking layer 502 can determine the user model of the first user based on the seventh user characteristic. In the authentication layer 503, the authentication unit of device 1 can obtain the association between the first user's user model and the system from the model management, and obtain the system to which device 1 belongs from the system management. Then, the authentication unit of device 1 can determine, based on the data obtained above, that the first user's user model and the first system 20 to which device 1 belongs do not have an association relationship, thus determining that the first user is illegitimate. At this time, the authentication unit of device 1 can trigger a manual authentication process. That is to say, although the first user's user model exists in the database 504 of the first system 20, if the first user has not been authenticated by any device in the first system 20, then when the first user triggers authentication, it will not be exempt from authentication, but will trigger a manual authentication process.

[0219] Alternatively, each user model can have a validity period. The situational tracking of the model tracking layer 502 can determine the first user's user model based on the seventh user characteristic. In the authentication layer 503, the authentication unit of device 1 can determine whether the first user's user model is within its validity period. If the first user's user model is within its validity period, the third device performs other judgment processes (such as whether the first user's user model is associated with the system to which device 1 belongs). If the first user's user model is not within its validity period, the third device can determine to trigger a manual authentication process. In other words, although the first user's user model exists in the database 504 of the first system 20, if the first user's user model is not within its validity period, the first user will not be exempt from authentication when authentication is triggered, but rather a manual authentication process will be triggered.

[0220] Alternatively, authentication credentials can also have an expiration date. The situational tracking of the model tracking layer 502 can determine the user model of the first user based on the seventh user characteristic. In the authentication layer 503, the authentication unit of device 1 can determine whether the authentication credentials bound to the user model of the first user are valid. If the authentication credentials are valid, the third device will then authenticate the credentials; if the authentication credentials are not valid, the third device can determine to trigger a manual authentication process. In other words, although the user model of the first user exists in the database 504 of the first system 20, if the authentication credentials bound to the user model are not valid, the first user will not be exempt from authentication when authentication is triggered, but rather a manual authentication process will be initiated.

[0221] Understandably, the user model of the first user exists in the database 504 of the first system 20, which can be understood as the scenario where the first user has entered multiple devices in the distributed system.

[0222] In some embodiments, the first user performs a manual authentication process in the distributed system, and the authentication result is successful (e.g., authentication passed). Figure 6 After steps S101-S103, the first user can leave the scenario where multiple devices in the distributed system reside. When the first user subsequently returns to this scenario, the multiple devices in the distributed system can re-track the first user to obtain the tracking results. When the first user triggers authentication on any device in the distributed system, that device can determine the legitimacy of the first user based on the tracking results, thereby determining whether authentication exemption is granted. See [link to specific example] for details. Figure 10 .

[0223] 1. When the first user enters scenario 10 where multiple devices are located in the first system 20, the device acquisition layer 501 continuously collects the first user's multi-dimensional eighth user characteristics.

[0224] 2. The device acquisition layer 501 sends the eighth user feature to the feature extraction of the model tracking layer 502.

[0225] 3. The feature extraction of the model tracking layer 502 processes the features of the eighth user to obtain the features of the ninth user.

[0226] 4. The feature extraction of the model tracking layer 502 sends the ninth user feature to the situation tracking.

[0227] 5. The situational tracking of the model tracking layer 502 retrieves at least one established user model from the database 504.

[0228] 6. The situational tracking of the model tracking layer 502 determines the user model of the first user from at least one of the above user models based on the ninth user feature.

[0229] 7. The situational tracking of the model tracking layer 502 sends the user model of the first user to the database 504, that is, the user model of the first user is shared with the devices in the first system 20.

[0230] 8. When device n receives the first user operation from the first user, the authentication unit of device n in the authentication layer 503 obtains the user model of the first user shared in the database 504.

[0231] 9. In the authentication layer 503, the authentication unit of device n determines that the first user is within a preset range based on the first user's operation, the first user's user model (optionally, and the first user's location characteristics collected in real time by the device acquisition layer 501).

[0232] 10. In authentication layer 503, the authentication unit of device n obtains the association between the user model of the first user and the system from the model management.

[0233] 11. In authentication layer 503, the authentication unit of device n obtains the system to which device n belongs from the system management.

[0234] 12. When the authentication unit of device n in authentication layer 503 determines that device n belongs to the first system 20 associated with the user model of the first user based on the data obtained above, it determines that the first user is legitimate. Therefore, the first user is exempt from authentication and the user interface when authentication is successful is directly displayed.

[0235] For example, user A's home represents the scenario of multiple devices in a distributed system. This distributed system can include a mobile phone, smartwatch, camera, and large screen. When user A uses their mobile phone at home, the phone performs facial recognition on user A and confirms successful authentication. This successful authentication result is shared with multiple devices in the distributed system. Before, during, or after user A's facial recognition, the multiple devices in the distributed system can continuously perform multi-dimensional human body tracking on user A. That is, the mobile phone can collect user A's location information, the smartwatch can collect user A's heart rate and pulse information, the camera can collect user A's posture information, and the large screen may not collect any information. The multiple devices in the distributed system can build a user model of user A based on the collected user features and distribute and store it across multiple devices. Subsequently, user A can use any device in the distributed system directly without manual authentication. For example, when user A returns home, the multiple devices in the distributed system can track user A and obtain user A's user model. When user A walks in front of the large screen, the large screen can determine that user A wants to use the large screen. Even if the large screen does not have data collection capabilities (e.g., it does not include a camera), it is still possible to determine that user A is a legitimate user and to determine that authentication is not required based on the user model of user A obtained from the above tracking and the above authentication results.

[0236] However, manual authentication is required when other users use devices in the distributed system. For example, when user B enters user A's home for the first time, multiple devices in the distributed system can track user B and determine that user B's user model does not exist. When user B walks in front of the large screen, the screen can determine that user B wants to use it. Since the tracking shows that user B's user model does not exist, the screen can determine that user B is illegitimate and trigger manual authentication. When user C enters user A's home for the second time, multiple devices in the distributed system can track user C and obtain user C's user model. However, user C did not pass authentication by any device in the distributed system the first time. Therefore, when user C walks in front of the large screen, the screen can determine that user C is illegitimate based on the user model obtained from the tracking and trigger manual authentication.

[0237] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When these computer program instructions are loaded and executed on a computer, the processes or functions described in this application are generated, in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive).

[0238] In summary, the above description is merely an embodiment of the technical solution of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made according to the disclosure of the present invention should be included within the scope of protection of the present invention.

[0239] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. An identity authentication method, characterized in that, The method, applicable to a distributed system consisting of multiple electronic devices, including a first device, a second device, and a third device, comprises: The first device authenticates the first user based on the first user's first user characteristics and obtains the first authentication result of the first user. The first device shares the first authentication result with the electronic devices in the distributed system; If the first authentication result is successful, the second device shares the user model of the first user with the electronic devices in the distributed system. The user model of the first user includes user characteristics for describing the first user. If the third device determines that the first user is within a preset range based on the first user's user model, the third device determines that the first user is legitimate. The first user's user model is used by the third device to determine whether a user within the preset range is the first user. The user within the preset range is the user determined by the third device to trigger the third device's authentication.

2. The method as described in claim 1, characterized in that, The method further includes: If the first authentication result is successful, the second device acquires the user characteristics of the first user at multiple time points, wherein the user characteristics of the first user at multiple time points are used to determine the user model of the first user.

3. The method as described in claim 2, characterized in that, The user characteristics of the first user at multiple time points include a first feature and a second feature, wherein the confidence level of the first feature is greater than the confidence level of the second feature; the method further includes: The second device determines the user model of the first user based on the user characteristics of the first user at multiple time points. The user model of the first user includes a third feature and a fourth feature. The first feature and the third feature are of the same type. The second feature and the fourth feature are of the same type. The similarity between the first feature and the third feature is greater than or equal to a preset threshold. The similarity between the second feature and the fourth feature is less than a preset threshold. The second device uses the second feature to update the user model of the first user.

4. The method as described in claim 2, characterized in that, Before the third device determines that the first user is within a preset range based on the user model of the first user, and before the third device determines that the first user is legitimate, the method further includes: The second device determines the first user within the preset range based on the user characteristics of the first user at multiple time points, and shares first indication information with the electronic devices in the distributed system, wherein the first indication information is used to indicate that the first user is within the preset range; If the third device determines that the first user is within a preset range based on the first user's user model, the third device determines that the first user is legitimate, including: The third device determines that the first user is within the preset range based on the first user's user model and the first indication information, and the third device determines that the first user is legitimate.

5. The method according to any one of claims 1-4, characterized in that, If the third device determines that the first user is within a preset range based on the first user's user model, the third device determines that the first user is legitimate, including: The third device receives the first user operation from the first user. The third device determines that the first user is within the preset range based on the first user's operation and the first user's user model, and the third device determines that the first user is legitimate.

6. The method according to any one of claims 1-4, characterized in that, The method further includes: If the first authentication result is successful, the second device shares second indication information with the electronic devices in the distributed system, wherein the second indication information is used to indicate the association between the user model of the first user and the distributed system to which the first device belongs; If the third device determines that the first user is within a preset range based on the first user's user model, the third device determines that the first user is legitimate, including: The third device determines the first user within the preset range based on the first user's user model; The third device determines, based on the second indication information, that it belongs to the distributed system associated with the user model of the first user, and the third device determines that the first user is legitimate.

7. The method according to any one of claims 1-4, characterized in that, The method further includes: If the first authentication result is successful, the first device shares the first user characteristics of the first user with the electronic devices in the distributed system; If the third device determines that the first user is within a preset range based on the first user's user model, the third device determines that the first user is legitimate, including: The third device determines that the first user is within a preset range based on the first user's user model; The third device authenticates the first user based on the first user's characteristics and obtains a second authentication result for the first user. The second authentication result is used to indicate whether the first user is legitimate.

8. The method according to any one of claims 2-4, characterized in that, The user characteristics of the first user at multiple time points are collected by one or more electronic devices in the distributed system at different times.

9. The method according to any one of claims 1-4, characterized in that, The preset range is a range relative to the distributed system, or a range relative to the third device.

10. The method according to any one of claims 1-4, characterized in that, In the distributed system, any electronic device is required to authenticate the user's identity before use.

11. An identity authentication method, characterized in that, A second device is applied in a distributed system, the distributed system comprising multiple electronic devices, the multiple electronic devices including a first device, a second device, and a third device, the method comprising: The second device obtains the first authentication result of the first user from the distributed system. The first authentication result is obtained by the first device from authenticating the first user based on the first user's first user characteristics. If the first authentication result is successful, the second device shares the user model of the first user with the electronic device in the distributed system. The user model of the first user includes user characteristics for describing the first user. The user model of the first user is used by the third device to determine whether a user within a preset range is the first user. The user within the preset range is the user determined by the third device to trigger the authentication of the third device. The user model of the first user is used by the third device to determine that the first user is legitimate when the first user is within the preset range.

12. The method as described in claim 11, characterized in that, The method further includes: If the first authentication result is successful, the second device acquires the user characteristics of the first user at multiple time points, wherein the user characteristics of the first user at multiple time points are used to determine the user model of the first user.

13. The method as described in claim 12, characterized in that, The user characteristics of the first user at multiple time points include a first feature and a second feature, wherein the confidence level of the first feature is greater than the confidence level of the second feature; the method further includes: The second device determines the user model of the first user based on the user characteristics of the first user at multiple time points. The user model of the first user includes a third feature and a fourth feature. The first feature and the third feature are of the same type. The second feature and the fourth feature are of the same type. The similarity between the first feature and the third feature is greater than or equal to a preset threshold. The similarity between the second feature and the fourth feature is less than a preset threshold. The second device uses the second feature to update the user model of the first user.

14. The method as described in claim 12, characterized in that, The method further includes: The second device determines that the first user is within the preset range based on the user characteristics of the first user at multiple time points, and shares first indication information with the electronic devices in the distributed system. The first indication information is used to indicate that the first user is within the preset range, and the first indication information is used by the electronic devices in the distributed system to determine that the first user is legitimate.

15. The method according to any one of claims 11-14, characterized in that, The method further includes: If the first authentication result is successful, the second device shares second indication information with the electronic devices in the distributed system, wherein the second indication information is used to indicate the association between the user model of the first user and the distributed system to which the first device belongs; The second indication information is used by electronic devices in the distributed system to determine that the first user is legitimate when the first user is within a preset range.

16. The method according to any one of claims 12-14, characterized in that, The user characteristics of the first user at multiple time points are collected by one or more electronic devices in the distributed system at different times.

17. The method according to any one of claims 11-14, characterized in that, The preset range is a range relative to the distributed system, or a range relative to the third device.

18. The method according to any one of claims 11-14, characterized in that, In the distributed system, any electronic device is required to authenticate the user's identity before use.

19. An identity authentication method, characterized in that, A third device is applied in a distributed system, the distributed system comprising multiple electronic devices, the multiple electronic devices including a first device, a second device, and the third device, the method comprising: The third device obtains the user model of the first user from the distributed system. The user model of the first user is obtained by the second device when the first authentication result of the first user is successful. The user model of the first user includes user features to describe the first user. The first authentication result is obtained by the first device to authenticate the first user based on the first user features. If the third device determines that the first user is within a preset range based on the first user's user model, the third device determines that the first user is legitimate. The first user's user model is used by the third device to determine whether a user within the preset range is the first user. The user within the preset range is the user determined by the third device to trigger the third device's authentication.

20. The method as described in claim 19, characterized in that, If the third device determines that the first user is within a preset range based on the first user's user model, the third device determines that the first user is legitimate, including: The third device determines that the first user is within the preset range based on the first user's user model and the first indication information. The third device determines that the first user is legitimate. The first indication information is used to indicate that the first user is within the preset range. The first indication information is obtained by the second device based on the user characteristics of the first user at multiple time points.

21. The method as described in claim 19 or 20, characterized in that, If the third device determines that the first user is within a preset range based on the first user's user model, the third device determines that the first user is legitimate, including: The third device receives the first user operation from the first user. The third device determines that the first user is within the preset range based on the first user's operation and the first user's user model, and the third device determines that the first user is legitimate.

22. The method as described in claim 19 or 20, characterized in that, If the third device determines that the first user is within a preset range based on the first user's user model, the third device determines that the first user is legitimate, including: The third device determines the first user within the preset range based on the first user's user model; The third device determines, based on the second indication information, that it belongs to the distributed system associated with the user model of the first user, and determines that the first user is legitimate. The second indication information is determined by the second device when the first authentication result is successful, and the second indication information is used to indicate that the user model of the first user is associated with the distributed system to which the first device belongs.

23. The method as described in claim 19 or 20, characterized in that, If the third device determines that the first user is within a preset range based on the first user's user model, the third device determines that the first user is legitimate, including: The third device determines that the first user is within a preset range based on the first user's user model; The third device authenticates the first user based on the first user's characteristics and obtains a second authentication result for the first user. The second authentication result is used to indicate whether the first user is legitimate.

24. The method as described in claim 20, characterized in that, The user characteristics of the first user at multiple time points are collected by one or more electronic devices in the distributed system at different times.

25. The method as described in claim 19 or 20, characterized in that, The preset range is a range relative to the distributed system, or a range relative to the third device.

26. The method as described in claim 19 or 20, characterized in that, In the distributed system, any electronic device is required to authenticate the user's identity before use.

27. A distributed system, characterized in that, It includes multiple electronic devices, wherein the multiple electronic devices include a first device, a second device, and a third device, wherein: The first device is configured to authenticate the first user based on the first user's first user characteristics, obtain the first authentication result of the first user, and share the first authentication result with the electronic devices in the distributed system; The second device is configured to share the user model of the first user with the electronic device in the distributed system when the first authentication result is successful. The user model of the first user includes user characteristics describing the first user. The third device is used to determine that the first user is within a preset range based on the user model of the first user, and to determine that the first user is legitimate. The user model of the first user is used by the third device to determine whether a user within the preset range is the first user. The user within the preset range is the user determined by the third device to trigger the authentication of the third device.

28. An electronic device, characterized in that, The electronic device includes one or more memories and one or more processors, the one or more memories being coupled to the one or more processors, the one or more memories being used to store a computer program, the one or more processors being used to invoke the computer program, the computer program including instructions that, when executed by the one or more processors, cause the electronic device to perform the method of any one of claims 11-18, or the method of any one of claims 19-26.

29. A computer storage medium, characterized in that, The method includes a computer program comprising instructions that, when executed on a processor, implement the method as described in any one of claims 11-18, or the method as described in any one of claims 19-26.

Images