Method and apparatus for authenticating a user of a mobile device

Abstract

The present disclosure generally relates to methods and devices for authenticating a user of a mobile device, and authenticating the user of the mobile device based on motion data of the mobile device. Embodiments provide a method, device, and computer program for authenticating a user, a mobile device including such a device, and a system. The method includes detecting the user's equipment. The method includes analyzing the user's gait using a machine learning model that uses motion data of the mobile device as an input to the machine learning model. The analysis is based on the identified user's equipment. The method includes authenticating the user based on the analysis of the user's gait.

Description

Technical Field

[0001] The present disclosure generally relates to authenticating a user of a mobile device based on motion data of the mobile device. Background Art

[0002] To access restricted resources, a user typically has to authenticate himself or herself to the resource, e.g., enable a gatekeeper of the resource to determine whether the user is authorized to access the resource. There are many ways to authenticate a user, e.g., physical keys, passwords, biometric scanners, etc. For example, a user may use a physical key to open a door or a cabinet, or a user may enter a password to log in to a computer. One way to authenticate a user is based on an analysis of the user's gait. Summary of the Invention

[0003] An object of the present disclosure is to provide a method, a device, and a computer program for authenticating a user, including a mobile device including such a device or executing such a method, and a corresponding system, which can improve the authentication of the user based on the user's gait.

[0004] Embodiments of the present disclosure provide a method for authenticating a user. The method includes detecting the user's equipment. The method includes analyzing the user's gait using a machine learning model, where the machine learning model uses motion data of the mobile device as an input to the machine learning model. The analysis is based on the identified user's equipment. The method includes authenticating the user based on the analysis of the user's gait.

[0005] Embodiments of the present disclosure provide a computer program having program code that, when executed on a computer, a processor, or a programmable hardware component, is adapted to execute a method for authenticating a user. The method includes detecting the user's equipment. The method includes analyzing the user's gait using a machine learning model that uses motion data of the mobile device as an input to the machine learning model. The analysis is based on the identified user's equipment. The method includes authenticating the user based on the analysis of the user's gait.

[0006] Embodiments of the present disclosure provide a device for authenticating a user. The device includes a circuit configured to obtain motion data from sensors of the mobile device. The circuit is configured to detect the user's equipment. The circuit is configured to analyze the user's gait using a machine learning model. The motion data is used as an input to the machine learning model. The analysis is based on the identified user's equipment. The circuit is configured to authenticate the user based on the analysis of the user's gait.

[0007] Embodiments of the present disclosure provide a mobile device including a device for authenticating a user. The device includes circuitry configured to obtain motion data from sensors of the mobile device. The circuitry is configured to detect the user's equipment. The circuitry is configured to analyze the user's gait using a machine learning model. The motion data is used as an input to the machine learning model. The analysis is based on the identified user's equipment. The circuitry is configured to authenticate the user based on the analysis of the user's gait.

[0008] Embodiments of the present disclosure provide a system including a transmitter device and a mobile device, the mobile device including a device for authenticating a user. The transmitter device is adapted to be attached to an equipment assembly. The transmitter device is configured to transmit a radio frequency signal to the mobile device. The mobile device is configured to authenticate the user based on the transmitted radio frequency signal. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Some examples of the device and / or method will be described hereinafter only by way of example and with reference to the drawings, wherein,

[0010] Figure 1a and Figure 1b a flowchart showing an embodiment of a method for authenticating a user;

[0011] Figure 1c a block diagram showing a device for authenticating a user, a mobile device including a device for authenticating a user, and a system including a transmitter device and a mobile device, the mobile device including a device for authenticating a user;

[0012] Figure 2 a schematic diagram showing a concept for authenticating a user; and

[0013] Figure 3 a block diagram showing a concept for selecting a gait authentication model for different configurations of equipment. DETAILED DESCRIPTION

[0014] Various examples will now be described more fully with reference to the drawings, in which some examples are shown. In the drawings, for clarity, the thickness of lines, layers, and / or regions may be exaggerated.

[0015] Accordingly, although further examples may have various modifications and alternative forms, some of its specific examples are shown in the drawings and will subsequently be described in detail. However, this detailed description does not limit further examples to the specific forms described. Further examples may cover all modifications, equivalents, and alternatives falling within the scope of the present disclosure. Throughout the description of the drawings, the same or similar numerals refer to the same or similar elements, which may be implemented in the same or modified form while providing the same or similar functions when compared to each other.

[0016] It should be understood that when an element is referred to as being "connected" or "coupled" to another element, these elements can be connected or coupled directly or via one or more intermediate elements. If two elements A and B are combined using "or", and not otherwise defined explicitly or implicitly, this should be understood to disclose all possible combinations, i.e., only A, only B, and A and B. Another way of phrasing the same combination is "at least one of A and B" or "A and / or B". With necessary modifications, this also applies to combinations of more than two elements.

[0017] The terms used herein to describe specific examples are not intended to limit further examples. Whenever the singular forms such as "a", "an", and "the" are used and the use of only a single element is neither explicitly nor implicitly defined as mandatory, further examples can also be implemented using multiple elements to achieve the same function. Similarly, when a function is subsequently described as being implemented using multiple elements, further examples can be implemented using a single element or processing entity to achieve the same function. It will be further understood that the terms "comprises", "comprising", "includes", and / or "including", when used, specify the presence of the stated features, integers, steps, operations, processes, acts, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, processes, acts, elements, components, and / or any combination thereof.

[0018] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the ordinary meaning in the field to which these examples belong.

[0019] Figure 1a and Figure 1b A flowchart showing an embodiment of a method for authenticating a user is presented. The method includes detecting 110 the user's equipment. The method includes analyzing 120 the user's gait using a machine learning model that uses the motion data of a mobile device as input to the machine learning model. The analysis is based on the identified user's equipment. The method includes authenticating 130 the user based on the analysis of the user's gait. The method can be executed by a mobile device 100.

[0020] Figure 1c A block diagram showing a corresponding device 10 for authenticating a user is presented. Device 10 includes a circuit 12. For example, circuit 12 can be configured to perform in conjunction with Figure 1a and / or Figure 1bThe methods described. For example, circuit 12 is configured to obtain motion data from sensor 20 of mobile device 100. Circuit 12 is configured to detect the user's equipment. Circuit 12 is configured to use a machine learning model to analyze the user's gait. The motion data is used as input to the machine learning model. The analysis is based on the identified user's equipment. The circuit is configured to authenticate the user based on the analysis of the user's gait. Figure 1c Mobile device 100 including device 10 is further illustrated. Mobile device 100 also includes sensor 20. Figure 1c System 1000 including mobile device 100 (having device 10) and transmitter device 30 is further illustrated.

[0021] The following description relates to Figure 1a the methods of 1a and / or 1b and Figure 1c device 10, mobile device 100, and system 1000 of

[0022] At least some embodiments of the present disclosure relate to a method, device, and computer program for authenticating a user based on an analysis of the user's gait (i.e., the way of walking). The user's gait can change according to the environment, for example, according to road conditions or according to other people who impede the user's walking. Another major factor affecting the user's gait is the equipment worn or carried by the user. For example, the user can walk in different ways wearing different types of shoes, for example, walk in a different way wearing formal shoes compared to sandals, or walk in a different way wearing high heels compared to sports shoes. Another major factor can be the bag worn or carried by the user. For example, if the user is carrying a (heavy) backpack, his or her gait may be less elastic; if the user wears a handbag on the shoulder or holds a travel bag in the hand, the walking may be slightly tilted, resulting in a change in the user's gait. In addition, the user walking in thick jeans can walk differently from walking in a skirt or shorts. Therefore, embodiments can detect the user's equipment and use the user's equipment in gait analysis, for example, by selecting a suitable machine learning model, or by using the detected equipment as input to a single machine learning model.

[0023] At least some embodiments are performed by mobile device 100. For example, the mobile device can be a mobile personal computing device, such as a smartphone, a tablet computer, or a laptop computer. Alternatively, the mobile device can be a wearable device, such as a smartwatch, a fitness tracker, a heart tracking device, smart glasses, or smart jewelry.

[0024] The method includes detecting the equipment of 110 users. In an embodiment, the equipment may include, for example, a bag worn and / or carried by the user, and / or shoes worn by the user. More generally, the equipment may include one or more components (i.e., equipment components). For example, one or more components of the equipment may include shoes worn by the user, a bag worn or carried by the user, and at least one of a pair of pants, a skirt, and socks worn by the user. For example, detecting the equipment of 110 users may include identifying the shoes worn by the user, detecting the bag worn or carried by the user, and identifying at least one of the pants, skirt, and socks worn by the user. The shoes worn, the pants worn, and the bag worn or carried may affect the user's gait. Therefore, the accuracy of authentication can be improved by identifying the shoes and / or pants and detecting the bag.

[0025] In at least some embodiments, the method includes identifying 111 the shoes worn by the user. For example, identifying 111 whether the user is wearing shoes may include identifying the type of shoes worn by the user (i.e., whether the shoes are sports shoes, sandals, dress shoes, high heels, etc.). Additionally or alternatively, identifying 111 whether the user is wearing shoes may include identifying a specific pair of shoes worn by the user (e.g., after registering a specific pair of shoes with the device or entity (e.g., mobile device) performing the method).

[0026] In various embodiments, detecting 110 the equipment may include detecting 112 the bag worn or carried by the user. For example, detecting 112 the bag may include identifying the type or location of the bag, i.e., whether the bag is a backpack worn on the user's back, or whether the bag is a handbag, sling bag, or suitcase carried on the user's shoulder or in the user's hand. In these cases, the user's back, shoulder, and hand may be the locations of the bag, and the types "backpack", "handbag", "sling bag", or "suitcase" may be the types of the bag. In some embodiments, detecting 112 the bag may include identifying 113 a specific bag (e.g., after registering a specific bag with the device or entity (e.g., mobile device) performing the method).

[0027] A variety of devices can be used to detect equipment. For example, at least a portion of the equipment can be detected based on radio frequency signals emitted by equipment components. Radio frequency (RF) is the rate of oscillation of an alternating current or voltage, or an electric or electromagnetic field, in the frequency range of approximately 20 kHz to approximately 300 GHz. For example, the radio frequency signal can be a wireless radio frequency signal. For example, circuit 12 can include processing circuitry (e.g., a processor) and (optionally) interface circuitry. For example, the interface circuitry can be coupled to a radio frequency receiver 14 of the mobile device. In other words, the mobile device 100 can include a radio frequency receiver 14 coupled to circuit 12. The radio frequency signal can be received via the radio frequency receiver 14 of the mobile device 100. A user can be authenticated based on the emitted radio frequency signal. In other words, the mobile device can be configured to authenticate the user based on the emitted radio frequency signal. Circuit 12 can be configured to obtain the radio frequency signal from a transmitter device 30 of the equipment component. The radio frequency signal can be used for equipment detection.

[0028] In at least some embodiments, the radio frequency signal can be a near field communication (NFC)-based radio frequency signal. Accordingly, the radio frequency receiver can be or include an NFC receiver. Alternatively, the radio frequency signal can be a Bluetooth-based radio frequency signal. Accordingly, the radio frequency receiver can be or include a Bluetooth receiver. In some cases, the radio frequency signal can be a radio frequency identification (RFID)-based radio frequency signal. Accordingly, the radio frequency receiver can be or include an RFID receiver or reader. In some embodiments, more than one radio frequency signal using more than one radio frequency technology can be used.

[0029] For example, the transmitter device can be adapted to be attached to the equipment component. For example, the transmitter device can be permanently attached to the equipment component, such as sewn to, glued to, or stapled to the equipment component. Alternatively, the transmitter device can be removably attached to the equipment component (e.g., via a clip or snap). The transmitter device can be configured to transmit a radio frequency signal to the mobile device. For example, the transmitter device can be one of an NFC beacon, a Bluetooth beacon, and an RFID tag.

[0030] The method can include monitoring 114 a radio frequency band to detect a user's equipment. For example, the radio frequency band can be monitored 114 to determine whether the user is wearing or carrying a bag to identify the bag, or to identify the shoes the user is wearing. By monitoring the radio frequency band, a radio frequency signal emitted by a transmitter device of the equipment component can be detected and received. Using the received radio frequency signal, the equipment component can be detected or identified.

[0031] In at least some embodiments, the method may include detecting the presence of the 116 equipment components. For example, the equipment may include one or more components. The equipment can be detected by detecting the presence of each of the one or more components of the equipment. For example, the method may include receiving 115 radio frequency signals from the transmitter device of the equipment component. The method may include detecting the presence of the equipment component based on the radio frequency signal. In other words, the radio frequency signal received from the equipment component may indicate the presence of the equipment component. The radio frequency signal enables the detection of the presence of the component and thus enables the detection of the equipment. For example, the method may include determining whether the user (currently) is wearing or carrying a bag based on the received radio frequency signal. For example, if a radio frequency signal is received from the bag, the determination of whether the user (currently) is wearing or carrying the bag may be affirmative, otherwise negative.

[0032] In addition, the method may include determining the type or specific item of the equipment component based on the received radio frequency signal. For example, identifying the bag carried by the user or the shoes worn. For example, the radio frequency signal may include identification information (e.g., one or more identifiers) of one or more components of the equipment. The type or specific item of the equipment component can be determined based on the identification information of one or more components of the equipment. For example, the identification information may be registered 140 with the device 10 or the entity executing the method (e.g., the mobile device 100). The method may include determining the type or specific item of the component based on the registered identification information. In other words, the bag or shoes can be identified based on the registered identification information and the identification information received in the radio frequency signal. The same identification information can be used to detect the presence of the equipment component (e.g., by comparing the identifier received together with the radio frequency signal with the registered identification information).

[0033] Alternatively or additionally, the detection 110 of the equipment may be based on another machine learning model. The motion data of the mobile device can be used as the input of the machine learning model. This can achieve the detection of the equipment if the equipment component is not equipped with a transmitter device.

[0034] The method includes analyzing 120 the user's gait using a machine learning model that uses the motion data of the mobile device as the input of the machine learning model.

[0035] Machine learning refers to algorithms and statistical models that computer systems can use to perform specific tasks, without using explicit instructions, but relying on models and inferences. For example, in machine learning, data transformations inferred from the analysis of historical and / or training data can be used instead of rule-based data transformations. For example, machine learning models or machine learning algorithms can be used to analyze the content of images. To enable a machine learning model to analyze the content of an image, training images can be used as input and training content information can be used as output to train the machine learning model. By training the machine learning model with a large number of training images and associated training content information, the machine learning model "learns" to recognize the content of the images, and thus the machine learning model can be used to recognize image content not included in the training images. The same principle can also be applied to other types of sensor data, for example, motion data in the embodiments: by using training sensor data and the desired output to train the machine learning model, the machine learning model "learns" the transformation between the sensor data and the output, and this transformation can be used to provide an output based on non-training sensor data provided to the machine learning model.

[0036] Training input data is used to train the machine learning model. The examples specified above use a training method called "supervised learning". In supervised learning, multiple training samples are used to train the machine learning model, where each sample can include multiple input data values and multiple desired output values, that is, each training sample is associated with a desired output value. By specifying the training samples and the desired output values, the machine learning model "learns" to provide which output value based on input samples similar to the t samples provided during training. In an embodiment, the motion data of the mobile device 100 can be used as the training input, and a (binary) training output indicating whether the authentication is successful or a training output indicating the confidence value of the authentication can be used as the training output to train the machine learning model. For example, the motion data of the mobile device 100 can be used as the training input, and an equipment (e.g., equipment components) can be used as the training output to train another machine learning model.

[0037] Machine learning algorithms are generally based on machine learning models. In other words, the term "machine learning algorithm" can represent a set of instructions that can be used to create, train, or use a machine learning model. The term "machine learning model" can, for example, indicate a data structure and / or set of rules representing the learned knowledge based on the training performed by the machine learning algorithm. In an embodiment, using a machine learning algorithm can mean using the underlying machine learning model (or multiple underlying machine learning models). Using a machine learning model can mean that the machine learning model and / or the data structure / set of rules as the machine learning model are trained by the machine learning algorithm.

[0038] For example, the machine learning model can be an artificial neural network (ANN). An ANN is a system inspired by biological neural networks (e.g., found in the brain). An ANN includes multiple interconnected nodes and multiple connections between the nodes, i.e., the so-called edges. There are generally three types of nodes: input nodes that receive input values, hidden nodes that are (only) connected to other nodes, and output nodes that provide output values. Each node can represent an artificial neuron. Each edge can transmit information from one node to another node. The output of a node can be defined as a (non-linear) function of the sum of its inputs. The input of a node can be used in the function based on the "weights" of the edges or nodes providing the input. The weights of the nodes and / or edges can be adjusted during the learning process. In other words, the training of an artificial neural network can include adjusting the weights of the nodes and / or edges of the artificial neural network, i.e., achieving a desired output for a given input.

[0039] Alternatively, the machine learning model can be a support vector machine. A support vector machine (i.e., a support vector network) is a supervised learning model with an associated learning algorithm that can be used to analyze data (e.g., in classification or regression analysis). A support vector machine can be trained by providing multiple training input values belonging to one of two classes to the input. A support vector machine can be trained to assign new input values to one of the two classes. Alternatively, the machine learning model can be a Bayesian network, which is a probabilistic directed acyclic graphical model. A Bayesian network can use a directed acyclic graph to represent a set of random variables and their conditional dependencies. Alternatively, the machine learning model can be based on a genetic algorithm, which is a search algorithm and heuristic technique that mimics the process of natural selection.

[0040] In an embodiment, the machine learning model can be used to determine whether a person carrying or wearing a mobile device is the user for which the machine learning model is trained using motion data as the input to the machine learning model. In other words, the machine learning model can be used to determine the match or correspondence between the gait of a person carrying or wearing a mobile device (resulting in motion data based on the gait of the user carrying or wearing the mobile device) and the gait of the user for which the machine learning model is trained.

[0041] Motion data is used as the input to the machine learning model. For example, the motion data can include or be based on at least one of accelerometer data, barometric pressure data, gyroscope data, and compass data. The motion data can indicate the movement of the mobile device caused by the user's gait (e.g., caused by the user walking). For example, the motion data can be provided by the sensor 20 of the mobile device 100. The sensor 20 can correspond to or include at least one of a motion sensor, an accelerometer, a barometric pressure sensor, a gyroscope, and a compass. The sensor 20 is coupled to the circuit 12 via the interface circuit of the circuit 12, for example.

[0042] This analysis is based on the equipment of the identified user. For example, a machine learning model can be selected from multiple machine learning models based on the detected equipment, or the detected equipment can be used as the input of the machine learning model.

[0043] In at least some embodiments, the method includes selecting a machine learning model from multiple machine learning models based on the detected equipment. The analysis of the gait can be based on the selected machine learning model. Each of the multiple machine learning models can be provided for different equipment of the user. In other words, for each of the multiple different potential equipment of the user, a specific machine learning model can be created and / or trained. The method can include creating and / or training a specific machine learning model for each of the multiple different potential equipment of the user. By using dedicated machine learning models based on different equipment, the models can be trained individually, so that more accurate authentication results can be provided for specific equipment.

[0044] In an embodiment, different granularities can be used to provide multiple machine learning models. In some embodiments, a granularity limited to the type of equipment components can be used. For example, in an embodiment where the equipment includes shoes and an optional bag, multiple machine learning models can be provided for the types of shoes without a bag (e.g., dress shoes, beach shoes, sports shoes, high heels) or the combination of the type of shoes and the type of bag (backpack, handbag, sling bag, suitcase).

[0045] In other words, the multiple machine learning models can be based on the combination of different types of shoes and different types of bags (or without a bag). Alternatively, a higher granularity can be used. In some embodiments, a granularity based on the combination of specific components of the equipment can be used. For example, in an embodiment where the equipment includes shoes and an optional bag, multiple machine learning models can be provided for a specific shoe without a bag (e.g., registered by the user 140) or the combination of a specific shoe and a specific bag (e.g., registered by the user 140). In both of these granularities, the presence of the bag can be used as a selection criterion in the selection 121 of the machine learning model. The method can include selecting a machine learning model based on the detected bag. For example, if no bag is detected, the machine learning model is selected by selecting from the first subset of the multiple machine learning models, and if a bag is detected, the machine learning model is selected by selecting from the second subset of the multiple machine learning models.

[0046] For example, detecting a user's equipment 110 may include identifying the shoes 111 worn by the user, detecting the bag (or not carrying) 112 worn or carried by the user, and / or identifying the bag 113 worn or carried by the user. The method may include selecting 121 a machine learning model based on the type of shoes or based on specific shoes owned by the user. For example, multiple machine learning models may be based on multiple different types of shoes or based on multiple specific shoes owned by the user. The selection of the machine learning model may include selecting one machine learning model among the multiple machine learning models that matches the type or specific shoes of the shoes 111 identified in the detection 110 of the user's equipment. By selecting the machine learning model based on the type of shoes, the number of machine learning models required can be reduced. By selecting the machine learning model based on specific shoes owned by the user, higher accuracy can be achieved. The method may include selecting 121 a machine learning model based on the type of bag or based on specific bags owned by the user. For example, multiple machine learning models may be based on multiple different types of bags (or the location of the bag) or based on multiple specific bags owned by the user. The selection of the machine learning model may include selecting one machine learning model among the multiple machine learning models that matches the type (or the location of the bag) or specific bag of the bag 113 identified in the detection 110 of the user's equipment. By selecting the machine learning model based on the detected and / or identified bag, a more specialized machine learning model can be selected.

[0047] In at least some embodiments, the detected user equipment may be used as an input to a machine learning model. For example, the detected user equipment may be used as an input to a machine learning model without selecting a dedicated machine learning model. This may allow for the use of a single model. Alternatively, a machine learning model may be selected based on whether the user is carrying a bag, and the type or identification of the detected bag may be used as an input to the machine learning model. In some embodiments, a machine learning model may be selected based on the type of shoes worn by the detected user, and the identification of the detected shoes may be used as an input to the machine learning model. In other words, in some embodiments, a machine learning model may be selected based on a first granularity of the detected user equipment, and a second granularity of the detected user equipment may be used as an input to the machine learning model, the second granularity being higher than the first granularity. For example, the method may include encoding the user's equipment to obtain one or more numerical values representing the user's equipment. The one or more numerical values may be provided as input values to the machine learning model. In at least some embodiments, the type of shoes worn by the detected user and / or the specific shoes of the detected user may be provided as an input to the machine learning model.

[0048] The method includes authenticating a user 130 based on an analysis of the user's gait. In an embodiment, authentication of the user 130 can determine whether (or not) the person using the mobile device is the user (i.e., the user who trained the machine learning network). For example, the user can be authenticated such that if the person using the mobile device matches the user who trained the machine learning model, the authentication is successful. The user can be authenticated to the mobile device, i.e., the authentication can be performed by and be valid for the mobile device. For example, if the user is successfully authenticated, the user can be authorized (e.g., by unlocking the mobile device) to access the mobile device, or to access another device or service. In other words, the method can include unlocking the mobile device based on the authentication of the user. The method can include providing access to another device or service based on the authentication of the user. This can enable the user to use another device or service, such as an automated teller machine (ATM), a door, or a vehicle. For example, another device can be an ATM, a door, a computer, another mobile device, a vehicle, etc. Another service can be a service accessed via the mobile device, such as a mobile banking service, an application of a car sharing company, or a service for accessing confidential and / or personal messages.

[0049] In some embodiments, the authentication can be performed continuously (i.e., periodically at a predetermined time interval or whenever new motion data is generated) or upon request of the user. For example, a machine learning model can be used to continuously analyze 120 the user's gait. In other words, the motion data can be continuously (i.e., periodically or whenever new motion data is generated) provided to the machine learning model upon generation. Based on the continuous analysis of the user's gait, the user can be continuously (i.e., periodically or whenever the output of the machine learning model changes) authenticated 130. This can avoid delays in user authentication and can avoid storing motion data.

[0050] Alternatively, the analysis 120 of the user's gait can be triggered by an authentication request. For example, the authentication request can be triggered by an unlocking process of the mobile device, or by an application request of the mobile device that indicates a desire or need for (additional) authentication of the user. The analysis of the user's gait can be based on the cumulative motion data of the mobile device. The cumulative motion data can include motion data for a predetermined time interval (or predetermined amount) prior to the authentication request. For example, the motion data can be stored or accumulated by a motion processor or a motion co-processor of the mobile device. For example, circuit 12 can include a motion processor or a motion co-processor. The predetermined time interval or predetermined amount of time can be based on the storage capacity of the motion (co-)processor, or the predetermined time interval or predetermined amount of time can be based on parameters of the analysis of the user's gait. This can avoid continuous analysis of the user's gait, thereby reducing the power consumption of the mobile device.

[0051] In some embodiments, the authentication of a user can be further based on the presence of an equipment component. For example, the presence of the equipment component being detected can provide a further indication that the user is actually being authenticated. In other words, the presence of the equipment component can indicate that the user owns the equipment component. The method can include using the detected presence of the equipment component as a second factor in user authentication. For example, the first factor in authentication can be the analysis of the user's gait, and the second factor can be the detected presence of the equipment component. The detected presence of the equipment component as the second factor in two-factor authentication can enable the detection of the equipment and the presence of the user.

[0052] In at least some embodiments, the method can include registering one or more equipment components with a mobile device. For example, the one or more equipment components include shoes (e.g., one or more pairs of shoes) and / or bags (e.g., one or more bags). For example, one or more equipment components can be registered by receiving radio frequency signals including identification information from a transmitter device of the one or more components and registering the one or more equipment components based on the one or more radio frequency signals. For example, the identification information can include an identification of the component (e.g., the serial number and / or model number of the corresponding component) and / or information about the type of component (e.g., whether the component is a shoe, a bag), etc. In some embodiments, the identification information can be universal identification information. For example, the universal identification information can indicate that the transmitter device from which the universal identification information is received is a transmitter device that can be attached to any component of the equipment. The user can then, for example, use an application of the mobile device to associate specific identification information with the universal identification information. The association of the specific identification information with the universal identification information can link the universal identification information with the specific identification information, for example, to indicate that whenever the universal identification information is received, the specific shoe or specific bag attached to the transmitter device providing the universal identification information is part of the user's equipment. In embodiments where additional machine learning networks are used to detect the user's equipment, one or more equipment components can be registered manually, for example, using an application of the mobile device. The method can include using the registered one or more equipment components to detect the equipment. For example, the identification information received during the registration of one or more components of the equipment can be used to detect the equipment, for example, by comparing the identification information received during the registration with the identification information received during the equipment detection by the component. By registering one or more equipment components with a mobile device, the method and / or device can detect the equipment.

[0053] In some embodiments, the method may include creating 150 a new machine learning model after registering an equipment component. For example, the method may include creating 150 one or more new machine learning models after registering an equipment component. For example, each new equipment combination in a plurality of different potential equipments has one model. The method may include training the new machine learning model (or one or more new machine learning models) based on the motion data of the mobile device, and obtaining the motion data while the user wears or carries the newly registered equipment component. This may enable subsequent selection of the trained machine learning model for gait analysis. When the user wears the corresponding equipment, the machine learning model and / or the plurality of machine learning models may be continuously improved. For example, the method may include continuously training the machine learning model for analyzing the user's gait based on the motion data of the mobile device. This may enable continuous improvement of the corresponding machine learning model.

[0054] The circuit 12 (e.g., a processing circuit) may be implemented using one or more processing units, one or more processing devices, any device for processing, e.g., a processor, a computer, or a programmable hardware component operable with appropriately adapted software. In other words, the described functions of the processing circuit may also be implemented in software and then executed on one or more programmable hardware components. Such hardware components may include a general-purpose processor, a digital signal processor (DSP), a microcontroller, etc. The interface circuit may correspond to one or more inputs and / or outputs for receiving and / or sending information within a module, between modules, or between modules of different entities, and the information may be a digital (bit) value according to a specified code.

[0055] More details and aspects of the method and / or device 10 are mentioned in connection with the proposed concept or one or more examples described above or below (e.g., Figures 2 to 3 ). The method and / or device 10 may include one or more additional optional features corresponding to one or more aspects of the proposed concept or one or more examples described above or below.

[0056] At least some embodiments are based on adjusting a gait authentication model (e.g., a plurality of machine learning models) according to the shoes worn and the bags carried.

[0057] Some use cases are based on continuously authenticated users. For this purpose, gait authentication may be used. Several factors may affect the accuracy of the gait authenticator. For example, the gait varies in different days, different walking speeds, carrying a handbag, a backpack, etc. and / or wearing different shoes. In an embodiment, the gait authenticator model may be adjusted for different shoe configurations and carrying a backpack or a handbag. By using such adjustment, the accuracy of the gait authenticator can be improved.

[0058] Figure 2A schematic diagram showing the concept for authenticating a user. In Figure 2 's concept, the user's shoes and / or bag are used as inputs to the gait authenticator 240. The gait authenticator 240 can be used to unlock 250 something, such as, for example, the mobile device 100 or another device or service. For example, as shown in Figure 2 210, the user's sports shoes can be detected and used as an input to the gait authenticator 240, high heels can be detected in Figure 2 220 and used as an input to the gait authenticator 240, and low heels and the carried bag can be detected in Figure 2 230 and used as an input to the gait authenticator 240.

[0059] At least some embodiments include identifying the shoes worn by the user and the bag carried by the user. During registration (e.g., registration 140), each worn configuration (e.g., sports shoes and a backpack, as shown in Figure 3 316) can be identified. Additionally, a (machine learning) model can be created and / or adjusted (i.e., trained) for each configuration. During evaluation, the configuration (i.e., the equipment) can be identified (e.g., by detecting the equipment 110), and a (machine learning) model can be selected for the configuration. The model can be used to authenticate the user, resulting in the use of a more specific model with expected higher accuracy. Additionally, owning the identified shoes or bag can be considered a second factor for authentication.

[0060] Figure 3 A block diagram showing the concept of selecting a gait authentication model for different configurations of equipment. Figure 3 Shows gait authentication models for different configurations 310 (e.g., multiple machine learning models), which include a first configuration 312 based on a backpack, a second configuration 314 based on high heels and a handbag, and a third configuration 316 based on sports shoes and a lightweight backpack. In Figure 3 , the configuration of the user 330 is detected by the configuration detector 320 (e.g., by detecting the equipment 110), and the gait authenticator 340 selects a machine learning model (in this case, the machine learning model based on the second configuration 314) based on the movement of the user 330 (e.g., movement data) and uses the machine learning model. The gait authenticator 340 can be used to unlock 250 something, such as, for example, the mobile device 100 of the user 330 or another device or service.

[0061] Embodiments can be based on hardware-based methods, such as, for example, methods based on tags / beacons (e.g., transmitter devices). The tags / beacons can be attached to the bag or shoes. For example, the tag or beacon can be based on Bluetooth or NFC, and can be wirelessly detected using a smartphone or smartwatch (e.g., the mobile device 100). During registration of the shoes / bag, the user can register the shoes / bag by bringing the shoes / bag close to the smartphone (e.g., the mobile device).

[0062] When walking, the authenticator (e.g., in combination with Figure 1a / Figure 1b the method or device 10 introduced) can train a new model for this configuration. During evaluation, the shoe / bag can be automatically recognized by a smartphone (e.g., by detecting the equipment). This can be used as two-factor authentication. The gait authenticator can load the specific model for this configuration. The prediction (e.g., user authentication) can be based on this model.

[0063] Alternatively or additionally, embodiments can be based on a software-based method, which can be based on signal analysis (e.g., motion data). Specific hardware may not be required. A mobile phone (e.g., mobile device 100) can analyze the motion data. During registration, the user can manually mark the shoe / bag to indicate the time when the user wears the shoe / carries the bag. When walking, the authenticator can train a new model for this configuration. During evaluation, using machine learning techniques (e.g., additional machine learning models), the algorithm can automatically detect the type of shoe the user is wearing and whether the user is carrying any bag. The corresponding configuration can be detected based on the detected type of shoe and based on whether the user is carrying a bag. The gait authenticator can load the specific model for this configuration. The prediction (e.g., user authentication) can be based on this model.

[0064] In a hardware-based method, the shoe / bag can include specific hardware, such as a Bluetooth or NFC tag (e.g., a transmitter device). The gait authenticator may need to pair the smartphone (e.g., mobile device) with the shoe / bag. The manual of the gait authenticator can include instructions for increasing accuracy when using a specific configuration based on a hardware-based NFC / Bluetooth tag.

[0065] The following examples relate to further embodiments:

[0066] (1) A method for authenticating a user, the method comprising:

[0067] Detecting the equipment of the user;

[0068] Analyzing the gait of the user using a machine learning model, the machine learning model using the motion data of the mobile device as the input of the machine learning model, the analysis being based on the identified equipment of the user;

[0069] Authenticating the user based on the analysis of the gait of the user.

[0070] (2) The method according to (1), wherein detecting the equipment of the user includes identifying at least one of the shoes worn by the user and detecting the bag worn or carried by the user.

[0071] (3) The method according to any one of (1) or (2), wherein the method includes selecting a machine learning model from a plurality of machine learning models based on the detected equipment, each of the plurality of machine learning models being provided for a different piece of equipment of the user, and the analysis of the gait is based on the selected machine learning model.

[0072] (4) The method according to (3), wherein for each of a plurality of different potential pieces of equipment of the user, a specific machine learning model is created and / or trained.

[0073] (5) The method according to any one of (3) or (4), wherein detecting the user's equipment includes identifying the shoes worn by the user, and the method includes selecting a machine learning model based on the type of shoes or based on a specific pair of shoes owned by the user.

[0074] (6) The method according to any one of (3) to (5), wherein detecting the user's equipment includes detecting a bag worn or carried by the user, and the method includes selecting a machine learning model based on the detected bag.

[0075] (7) The method according to any one of (5) or (6), further including identifying the bag, and the method includes selecting a machine learning model based on the type of bag or based on a specific bag owned by the user.

[0076] (8) The method according to any one of (1) to (7), wherein the detected user's equipment is used as an input to the machine learning model.

[0077] (9) The method according to any one of (1) to (8), wherein the method includes monitoring a radio frequency band to detect the user's equipment.

[0078] (10) The method according to (9), wherein the radio frequency band is monitored to determine whether the user is wearing or carrying a bag so as to identify the bag, or to identify the shoes worn by the user.

[0079] (11) The method according to any one of (1) to (10), wherein the method includes detecting the presence of an equipment component, the presence of the equipment component indicating that the user owns the equipment component, and the method includes using the detected presence of the equipment component as a second factor in user authentication.

[0080] (12) The method according to (11), wherein the method includes receiving a radio frequency signal from a transmitter device of the equipment component, and the method includes detecting the presence of the equipment component based on the radio frequency signal.

[0081] (13) The method according to any one of (1) to (12), wherein the detection of the equipment is based on another machine learning model, and the motion data of the mobile device is used as an input to the machine learning model.

[0082] (14)The method according to any one of (1) to (13) further includes registering one or more components of the equipment with the mobile device, and the method includes detecting the equipment using the one or more registered components of the equipment.

[0083] (15)The method according to (14), wherein the one or more equipment components include one or more pairs of shoes and / or one or more bags.

[0084] (16)The method according to any one of (14) or (15), the method includes creating a new machine learning model after registering the equipment components, and the method includes training the new machine learning model based on the motion data of the mobile device obtained while the user wears or carries the newly registered equipment components.

[0085] (17)The method according to any one of (1) to (16), wherein the method includes continuously training a machine learning model for analyzing the user's gait based on the motion data of the mobile device.

[0086] (18)The method according to any one of (1) to (17), wherein the user's gait is continuously analyzed using the machine learning model, and the user is continuously authenticated based on the continuous analysis of the user's gait.

[0087] (19)The method according to any one of (1) to (17), wherein the analysis (120) of the user's gait is triggered by an authentication request, and the analysis of the user's gait is based on the cumulative motion data of the mobile device, and the cumulative motion data includes the motion data of a predetermined time interval before the authentication request.

[0088] (20)The method according to any one of (1) to (19), the method includes unlocking the mobile device based on the user's authentication.

[0089] (21)The method according to any one of (1) to (20), the method includes providing access to another device or service based on the user's authentication.

[0090] (22)A computer program having program code that, when the computer program is executed on a computer, a processor, or a programmable hardware component, is used to execute the method according to any one of the foregoing examples.

[0091] (23)A device for authenticating a user, the device includes:

[0092] A circuit configured to:

[0093] Obtain motion data from a sensor of the mobile device,

[0094] Detect the user's equipment,

[0095] Analyze the gait of a user using a machine learning model, with motion data serving as the input to the machine learning model, the analysis being based on the identified equipment of the user, and

[0096] Authenticate the user based on the analysis of the user's gait.

[0097] (24) The device according to (23), wherein the circuit is configured to obtain a radio frequency signal from a transmitter device of the equipment component, the radio frequency signal being used for the detection of the equipment.

[0098] (25) A mobile device comprising the device according to any one of (23) or (24).

[0099] (26) A system comprising the mobile device according to (25) and a transmitter device, the transmitter device being adapted to be attached to the equipment component, the transmitter device being configured to transmit a radio frequency signal to the mobile device, the mobile device being configured to authenticate the user based on the transmitted radio frequency signal.

[0100] Aspects and features mentioned and described in conjunction with one or more of the previously detailed examples and figures may also be combined with one or more other examples so as to replace similar features of other examples or to introduce that feature additionally into other examples.

[0101] The examples may also be or relate to a computer program having program code which, when the computer program is executed on a computer or processor, is for performing one or more of the above methods. The steps, operations or processes of the various above methods may be formed by a programmed computer or processor. The examples may also cover a program storage device (e.g., a digital data storage medium) which is machine, processor or computer readable and encodes an instruction program that is machine executable, processor executable or computer executable. These instructions perform or cause to perform some or all of the actions of the above methods. The program storage device may comprise or may be, for example, a digital memory, a magnetic storage medium such as magnetic disks and tapes, a hard disk drive or an optically readable digital data storage medium. Further examples may also cover a computer, processor or control unit programmed to perform the actions of the above methods, or a (field) programmable logic array (F(PLA)) or a (field) programmable gate array (F(PGA)) programmed to perform the actions of the above methods.

[0102] The description and the figures merely illustrate the principles of the present disclosure. In addition, all examples enumerated herein are mainly explicitly for illustrative purposes only to help the reader understand the principles of the present disclosure and the concepts contributed by the inventors to advance the art. All statements of the principles, aspects and examples of the present disclosure and their specific examples enumerated herein are intended to cover their equivalents.

[0103] A functional block represented as a "means for" performing a particular function can refer to circuitry configured to perform that particular function. Thus, a "means for" something can be implemented as a "means configured or adapted for" something, e.g., an apparatus or circuitry configured or adapted for the corresponding task.

[0104] The functions of the various elements shown in the figures, including any functional blocks labeled "means", "means for providing a signal", "means for generating a signal", etc., can be implemented in the form of dedicated hardware (e.g., a "signal provider", "signal processing unit", "processor", "controller", etc.) and hardware capable of executing software associated with appropriate software. When provided by a processor, the functions can be provided by a single dedicated processor, a single shared processor, or multiple individual processors, some or all of which can be shared. However, the term "processor" or "controller" is far from being limited to hardware specifically capable of executing software, but can include digital signal processor (DSP) hardware, network processors, application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), read only memories (ROMs) for storing software, random access memories (RAMs), and non-volatile memories. It can also include other conventional and / or custom hardware.

[0105] For example, a block diagram can show a high-level circuit diagram implementing the principles of the present disclosure. Similarly, a flowchart, process diagram, state transition diagram, pseudocode, etc. can represent various processes, operations, or steps, which can, for example, be substantially represented on a computer-readable medium and executed by a computer or processor, whether or not such a computer or processor is explicitly shown. The methods disclosed in the specification or claims can be implemented by an apparatus having means for performing each corresponding action of these methods.

[0106] It should be understood that the disclosure of multiple acts, processes, operations, steps, or functions in the specification or claims should not be construed as in a particular order unless explicitly or implicitly stated otherwise, e.g., for technical reasons. Thus, the disclosure of multiple acts or functions does not limit these to a particular order unless these acts or functions are not interchangeable for technical reasons. Further, in some examples, a single act, function, process, operation, or step can respectively include or can be divided into multiple sub-acts, functions, processes, operations, or steps. Such sub-acts can be included and are part of the disclosure of that single act unless explicitly excluded.

[0107] In addition, the appended claims are hereby incorporated into the detailed description, where each claim can stand on its own as a separate example. Although each claim can stand on its own as a separate example, it should be noted that although a dependent claim can refer to a particular combination with one or more other claims in the claims, other examples can also include combinations of the dependent claim with the subject matter of each other dependent or independent claim. Such combinations are expressly set forth herein unless stated not to be intended. In addition, it is also intended to include the features of any other independent claim, even if the claim is not directly dependent on the independent claim.

Claims

1. A method for authenticating a user, the method comprising: Detect the user's equipment; Analyze the user's gait using a machine learning model, where the machine learning model uses the motion data of the mobile device as the input to the machine learning model, and the analysis is based on the identified equipment of the user; Authenticate the user based on the analysis of the user's gait.

2. The method according to claim 1, wherein, Detecting the user's equipment includes identifying at least one of the shoes worn by the user and detecting the bag worn or carried by the user.

3. The method according to claim 1, wherein, The method includes selecting a machine learning model from a plurality of machine learning models based on the detected equipment, where each of the plurality of machine learning models is provided for different equipment of the user, and the analysis of the gait is based on the selected machine learning model.

4. The method according to claim 3, wherein, For each of a plurality of different potential equipment of the user, create and / or train a specific machine learning model.

5. The method according to claim 3, wherein, Detecting the user's equipment includes identifying the shoes worn by the user, and the method includes selecting the machine learning model based on the type of the shoes or based on specific shoes owned by the user, and / or where detecting the user's equipment includes detecting the bag worn or carried by the user, and the method includes selecting the machine learning model based on the detected bag.

6. The method according to claim 5, further comprising identifying the package, the method comprising selecting the machine learning model based on the type of the package or based on a specific package owned by the user.

7. The method according to claim 1, wherein, The detected equipment of the user is used as the input to the machine learning model.

8. The method according to claim 1, wherein, The method includes monitoring a radio frequency band to detect the user's equipment.

9. The method according to claim 8, wherein, Monitor the radio frequency band to determine whether the user is wearing or carrying a bag to identify the bag, or to identify the shoes worn by the user.

10. The method according to claim 1, the method comprising detecting the presence of an equipment component, the presence of the equipment component indicating that the user owns the equipment component, the method comprising using the detected presence of the equipment component as a second factor in user authentication.

11. The method according to claim 10, the method comprising receiving a radio frequency signal from a transmitter device of the equipment component, the method comprising detecting the presence of the equipment component based on the radio frequency signal.

12. The method according to claim 1, wherein, The detection of the equipment is based on another machine learning model, and the motion data of the mobile device is used as the input to the machine learning model.

13. The method according to claim 1, further comprising registering one or more equipment components with the mobile device, the method comprising detecting the equipment using the registered one or more equipment components.

14. The method according to claim 13, wherein, One or more of the equipment components include one or more pairs of shoes and / or one or more bags.

15. The method according to claim 1, the method comprising creating a new machine learning model after registering the equipment component, the method comprising training the new machine learning model based on the motion data of the mobile device obtained while the user wears or carries the newly registered equipment component.

16. The method according to claim 1, wherein, Continuously analyze the user's gait using the machine learning model, and continuously authenticate the user based on the continuous analysis of the user's gait, or where the analysis of the user's gait is triggered by an authentication request, and the analysis of the user's gait is based on the cumulative motion data of the mobile device, and the cumulative motion data includes the motion data of a predetermined time interval before the authentication request.

17. A computer program product having program code for performing the method according to claim 1 when the computer program is executed on a computer, a processor or a programmable hardware component.

18. A device for authenticating a user, the device comprising: A circuit, configured to: Obtain motion data from the sensors of the mobile device, Detect the user's equipment, Analyze the user's gait using a machine learning model, where the motion data is used as the input to the machine learning model, and the analysis is based on the identified equipment of the user, and Authenticate the user based on the analysis of the user's gait.

19. A mobile device comprising the device according to claim 18.

20. A system comprising the mobile device according to claim 19 and a transmitter device adapted to be attached to an equipment component, the transmitter device being configured to transmit a radio frequency signal to the mobile device, the mobile device being configured to authenticate a user based on the transmitted radio frequency signal.

Images