Method for biometric authentication through vascular monitoring
Patent Information
- Application Number
- TW110143175
- Authority / Receiving Office
- TW · TW
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2021-08-27
- Filing Date
- 2021-11-19
- Publication Date
- 2026-11-11
- Estimated Expiration
- 2041-11-18
AI Technical Summary
Biometric authentication methods using vein patterns are vulnerable to spoofing and require direct access to the back of the hand, which can be inconvenient or impossible in certain situations, such as wearing gloves or being far from the scanning device.
A method for authenticating individuals based on vascular dynamics, including changes in blood vessel properties over time, using electromagnetic radiation to detect and analyze blood vessel patterns without direct contact, allowing for hands-free authentication using devices like mobile phones or wearable devices.
Provides a secure and convenient authentication method that is resistant to spoofing and does not require direct access to the body, enabling seamless transactions and access control.
Smart Images

Figure TWG2TB001919902_001 
Figure TWG2TB001919902_002 
Figure TWG2TB001919902_003
Abstract
Description
Technical Field
[0001] This disclosure relates to biometric authentication in computer security, and more specifically, to techniques for examining physiological characteristics to allow for minimally destructive authentication. [Cross-reference to related applications]
[0002] This application claims priority to U.S. Provisional Patent Application No. 63 / 115,941, filed November 19, 2020, entitled “Arterial Pocket Authentication,” the entire contents of which are incorporated herein by reference. Prior Technology
[0003] Biometric authentication procedures verify an individual's identity through biometrics. The term "biometric" refers to physical or behavioral characteristics that can be used as a means of verifying identity. Biometrics are difficult to fake and convenient because the corresponding individual does not need to remember passwords or management tokens. Instead, the authentication mechanism becomes an integral part of the individual.
[0004] Fingerprints have historically been the most common biometric modality. However, with technological advancements, other biometric modalities have emerged. As an example, vascular pattern recognition (also known as "vein pattern authentication") uses near-infrared light to create images of subcutaneous blood vessels. These subcutaneous vessels collectively define the "vascular pattern" that can be used for authentication. Vein pattern authentication has shown promise because vascular patterns are not only unique to the corresponding individual, but also undergo minimal change as the individual ages.
[0005] Vein pattern authentication typically involves identifying and then analyzing the pattern of blood vessels along the back of the hand. For example, near-infrared light generated by a light-emitting diode (LED) can be emitted toward the back of the hand, penetrating the skin. Because blood vessels and other tissues have different absorbance, the near-infrared light reflects toward the skin at different depths. The vascular pattern can be inferred based on the analysis of the reflected near-infrared light, and features such as branch locations and angles can be determined from the vascular pattern (and then used for authentication).
[0006] Vein pattern authentication has been touted as a contactless option for biometric authentication, as vein patterns are relatively difficult to reconstruct and therefore less susceptible to forgery. However, vein pattern authentication has several drawbacks. While individuals may not need to touch the sensing surface as in the case of fingerprint authentication, they must still present the back of their hand for analysis. This can be difficult in some situations (e.g., when the individual is not near the payment system) and cumbersome in others (e.g., when the individual is wearing gloves). Summary of the Invention
[0007] and Simple Explanation of the Diagram
[0008] [Figure 1] includes advanced instructions for traditional authentication procedures, which prompt an unknown person to present their hand to the blood vessel scanner.
[0009] [Figure 2A] includes a high-order representation of the system, which can be used to authenticate the identity of an unknown person whose vascular distribution is available for scanning.
[0010] [Figure 2B] includes a high-order representation of the system, which can be used to authenticate the identity of an unknown person based on gesture recognition.
[0011] [Figure 2C] includes a high-order representation of the system, which can be used to identify unknown persons based on location.
[0012] [Figure 3] shows an example of an electronic device capable of implementing an authentication platform designed to authenticate the identity of an unknown person based on data generated by a signal sensor.
[0013] [Figures 4A-B] include flowcharts of the process for authenticating users of an authentication platform based on radar detection of vascular dynamics monitored using electronic devices close to the body.
[0014] [Figures 5A-B] include flowcharts of the registration and usage phases in an embodiment of an authentication platform determining whether to authenticate a user based on gesture recognition.
[0015] [Figures 6A-B] include flowcharts of the registration and usage phases in an embodiment of an authentication platform determining whether to authenticate a user based on location identification.
[0016] [Figures 7A-B] include flowcharts of the registration and usage phases of an example of a certification process that depends on vascular dynamics.
[0017] [Figure 8] includes a scenario in which the unknown person is still authenticated despite leaving the electronic device in their front trouser pocket.
[0018] [Figure 9] shows a schematic diagram of an example of an authentication platform that can authenticate unknown persons based on vascular dynamics.
[0019] [Figure 10] depicts a flowchart of the process for biometric authentication of the identity of a person who claims to be a given individual.
[0020] [Figure 11] depicts a flowchart of the process for generating a biometric signature for an individual that can then be used for authentication.
[0021] [Figure 12] depicts a flowchart of the process for determining whether a person should be identified as a given individual based on the vascular dynamics of blood vessels over time.
[0022] [Figure 13] is a block diagram illustrating an example of a processing system in which at least some of the operations described herein can be implemented.
[0023] By studying the embodiments in conjunction with the drawings, the various features of the technology described herein will become more apparent to those skilled in the art. Embodiments are shown in the drawings by way of example rather than limitation. Although various embodiments are depicted in the drawings for illustrative purposes, those skilled in the art will recognize that alternative embodiments may be employed without departing from the principles of the technology. Therefore, while specific embodiments are shown in the drawings, various modifications can be made to the technology. Implementation
[0024] To register for an authentication program that relies on vein matching, an individual (also known as a “user”) may initially be prompted to present their hand to a vascular scanner. The term “vascular scanner” can be used to refer to an imaging instrument that includes (i) a transmitter operable to emit electromagnetic radiation (e.g., in the near-infrared range) into the body, and (ii) a sensor operable to sense electromagnetic radiation reflected by physiological structures within the body. Typically, the image is created based on the reflected electromagnetic radiation used as a reference sample. In general, the reference sample represents a “live” pattern of blood vessels that can be used for authentication.
[0025] Figure 1 provides a high-level illustration of a traditional authentication process, where an unknown person is prompted to present their hand to a vascular scanner. As shown in Figure 1, the vascular scanner emits electromagnetic radiation onto the hand and then creates an image (also known as a “scan”) based on the electromagnetic radiation reflected from the blood vessels in the hand. This image indicates the vascular pattern of the hand and can therefore be verified against a reference sample created for a given individual during the enrollment phase (also known as the “registration phase”). If the image matches the reference sample, the unknown person is authenticated as that given individual. However, if the image does not match the reference sample, the unknown person will not be authenticated as that given individual.
[0026] Because vein scanners do not require contact with the body part being scanned, vein matching has become an attractive option for biometric authentication. However, vein matching has proven vulnerable to spoofing. For example, Jan Krissler and Julian Albrecht demonstrated at the 2018 Chaos Communication Congress how to bypass a vein scanner using a wax prosthetic hand. While spoofing is unlikely to succeed in most real-world conditions, any concerns associated with vulnerabilities could hinder the adoption of reliable biometric authentication technologies.
[0027] Therefore, this paper introduces methods for identifying unknown individuals based on changes in vascular properties over time. In general, these methods rely on monitoring vascular dynamics to identify unknown individuals. The term "vascular dynamics" refers to changes in vascular properties over time. Examples of properties include the location, size, volume, and pressure of blood vessels, as well as the velocity and acceleration of blood flowing through them.
[0028] To prevent deception, an authentication platform (also known as an "authentication system") can determine the similarity between the vascular dynamics of an unknown person whose identity is to be authenticated and those of a given individual. For example, suppose an unknown person wishes to authenticate themselves as a given individual. In this scenario, the unknown person might be prompted to place a body part near a vascular scanner. The vascular scanner then transmits a series of signals to the body part, generating data representing signals reflected by physiological structures (including blood vessels) located within the body part. By analyzing this data, the authentication platform can determine, derive, or otherwise obtain a measure indicative of the unknown person's vascular dynamics. This measure may be referred to as the unknown person's "biometric signature" or "vascular signature." The authentication platform can then compare this biometric signature with at least one registered biometric signature associated with a given individual to determine whether the unknown person should be authenticated as that individual. Therefore, the authentication platform can determine the probability that the unknown person is the given individual based on a comparison of their vascular dynamics.
[0029] Authentication based on vascular dynamics offers many of the same benefits as vein matching, namely, high accuracy, reliability, and consistency, because the information to be "read" is inside the body. However, vascular dynamics are less susceptible to deception because both vascular patterns and changes in vascular properties can be used as unique biological characteristics.
[0030] For illustrative purposes, embodiments may be described in the context of monitoring the vascular dynamics of the femoral artery. However, the methods described herein can be similarly applied to any artery, arteriole, microvessel, venule, or vein. Similarly, the methods described herein can be similarly applied to any set of vessels. Therefore, in cases where authentication is sought for sensitive tasks, the authentication platform can monitor the vascular dynamics of more than one vessel.
[0031] Although not strictly required, implementation is described below in the context of instructions that can be executed by an electronic device. The term "electronic device" is often used interchangeably with the term "computing device" and can therefore be used to refer to computer servers, POS systems, tablet computers, wearable devices (e.g., fitness trackers and watches), mobile phones, etc.
[0032] While some modules may be described as being exclusively or primarily executed by a single electronic device, some implementations operate in a distributed environment where modules are shared among multiple electronic devices linked together via a network. For example, a POS system might require an unknown person to initiate an authentication process, but the measurement of vascular dynamics could be performed by a mobile phone located near the unknown person. Similarly, while a mobile phone could be responsible for measuring the unknown person's vascular dynamics, the decision to authenticate the unknown person could be made by an authentication platform residing on a computer server connected to the mobile phone's communications. the term
[0033] References to "an embodiment" or "some embodiments" in this disclosure mean that the described features, functions, structures, or characteristics are included in at least one embodiment. The appearance of such phrases does not necessarily refer to the same embodiment, nor does it necessarily refer to mutually exclusive alternative embodiments.
[0034] Unless the context explicitly requires otherwise, the terms “comprise,” “comprising,” and “comprised of” should be interpreted as inclusive rather than exclusive (i.e., “including but not limited to”). The term “based on” should also be interpreted as inclusive rather than exclusive. Therefore, unless otherwise stated, the term “based on” is intended to mean “at least partially based on.”
[0035] The terms “connected,” “coupled,” and variations thereof are intended to include any connection or coupling between two or more elements, whether direct or indirect. This connection / coupling can be physical, logical, or a combination thereof. For example, elements may be electrically or communicatively coupled to each other even without sharing a physical connection.
[0036] The term "module" can be used broadly to refer to software, firmware, and / or hardware. A module is typically a functional element that produces one or more outputs based on one or more inputs. A computer program may include or utilize one or more modules. Therefore, a computer program may utilize multiple modules responsible for performing different tasks or a single module responsible for performing all tasks.
[0037] When used to refer to a list of multiple items, the term "or" is intended to cover all of the following interpretations: any item in the list, all items in the list, and any combination of items in the list. The term "and / or" should be interpreted in a similar manner. For example, suppose an embodiment is described as including "a first item, a second item, and / or a third item." In this case, the phrase is intended to cover a single first item, a single second item, a single third item, a combination of the first and second items, a combination of the first and third items, and a combination of the second and third items.
[0038] The order of steps performed in any process described herein is exemplary. However, unless contrary to physical possibility, these steps can be performed in various orders and combinations. For example, steps can be added to or removed from the steps described herein. Similarly, steps can be substituted or reordered. Therefore, any description of a process is intended to be open-ended. Authentication through analysis of vascular information
[0039] This section introduces an authentication platform that uses vascular dynamics as a biometric to verify that an unknown person is a given individual. As discussed further below, vascular dynamics of an unknown person can be measured using probe signals emitted into the body. The reflections of these probe signals can be detected by sensors located close to the body. These reflections can represent “return signals,” from which changes in vascular properties can be determined.
[0040] Authentication platforms can be used to secure biometric-driven transactions, such as payments authenticated via hands-free interfaces. For example, suppose an unknown person wants to authenticate themselves to complete a transaction. Instead of prompting the unknown person to place a body part (e.g., their hand) near a vascular scanner, biometric authentication can be performed using an electronic device already near their body. For example, this electronic device could be located in a pocket of clothing worn on the body. As an example, a mobile phone located in the pocket of an unknown person's trousers could include a transmitter operable to emit a detection signal and a sensor operable to sense the reflection of the detection signal from physiological structures within the body. While this authentication method relies on the analysis of blood vessels beneath the skin, the mobile phone does not need to touch the skin, nor does it need to "see" the skin. Therefore, authentication platforms can allow people to authenticate themselves in a minimally invasive manner by relying on information about vascular dynamics.
[0041] Several different authentication methods are discussed in more detail below. These methods include: • Vessel pattern authentication: Authentication is based on the vessel pattern determined from the returned signal; • Dynamic vascular authentication: Authentication is based on measured vascular characteristics; and • Vascular Deformity Authentication: Authentication is based on vascular characteristics measured when an unknown person interacts with the environment, for example, by performing gestures that cause vascular deformation.
[0042] In some embodiments, the authentication platform operates independently to authenticate the identity of an unknown person, while in other embodiments, it operates in conjunction with another system. For example, a payment system may interface with the authentication platform to ensure that transactions are completed securely and smoothly. As an example, the authentication platform may facilitate contactless payment procedures, where an unknown person can initiate or complete a transaction by making a part of their body available for scanning. As described above, an unknown person can make a part of their body available for scanning by simply placing a vascular scanner near their body (e.g., in the pocket of the unknown person's trousers).
[0043] It is important to note that while the embodiments may discuss authentication within the context of initiating or completing a transaction, authentication can be useful in a variety of contexts. For example, suppose individuals are invited one after another to a network-accessible meeting where sensitive information will be shared. Each person attempting to enter that network-accessible meeting may need to be authenticated by an authentication platform to be allowed access. Overview of the authentication platform
[0044] Figure 2A shows a high-order representation of system 200A, which can be used to authenticate the identity of an unknown person whose vascular distribution is available for scanning. As shown in Figure 2A, system 200A includes an authentication platform 202A that can access a user interface (UI) 204, a signal transmitter 206, a signal sensor 208, and a processor 210. As discussed further below, these components of system 200A can be embedded in the same electronic device or distributed among multiple electronic devices. For example, the authentication platform 202A can reside partially or entirely on a network-accessible server system, while the UI 204, signal transmitter 206, signal sensor 208, and processor 210 can reside on another electronic device located near the unknown person.
[0045] UI 204 represents the interface through which an unknown person can interact with system 200. UI 204 may be a voice-driven graphical user interface (GUI) displayed on the screen of an electronic device. Alternatively, UI 204 may be a non-voice-driven GUI displayed on the screen of an electronic device. In such an embodiment, UI 204 may visually indicate the body parts to be presented for authentication purposes. For example, UI 204 may visually prompt the unknown person to present their hand to signal transmitter 206 and signal sensor 208 for scanning by presenting an icon identical to the unknown person's hand.
[0046] Signal transmitter 206 (also referred to as a “signal generator”) can be configured to emit electromagnetic radiation in the form of pulses toward a body part presented by an unknown person. For example, suppose UI 204 instructs the unknown person to present her hand for authentication. In this scenario, signal transmitter 206 can emit a signal toward the hand over a time interval. Typically, this time interval is between 0.1 and 1.0 seconds, however, it can be longer or shorter in some embodiments. Meanwhile, signal sensor 208 can be any sensing device capable of generating data representing signals reflected by physiological structures located within the body. In general, signal transmitter 206 can be responsible for emitting a signal into the body over a time interval, while signal sensor 208 can be responsible for collecting the reflected signals. For convenience, the signal emitted into the body can be referred to as a “detection signal” or a “measurement signal,” and the reflected signal can be referred to as a “return signal.”
[0047] Those skilled in the art will recognize that reflection occurs on the surface to which the signal is projected. The reflective surface is typically the boundary between two structures (e.g., blood vessels and tissues) with different electromagnetic properties. Ideally, the sensing device should be able to detect the returned signal with sufficient resolution to distinguish minute differences (e.g., on the micrometer scale) in order to detect changes in vascular properties, as discussed further below. As an example, for large arteries with wall diameters of 0.5–1.0 mm, a resolution on the order of tens of micrometers is often sufficient to detect changes caused by variations in blood flow. Smaller vessels tend to require higher resolutions, e.g., on the micrometer or smaller scale.
[0048] In some embodiments, the signal transmitter 206 and the signal sensor 208 are part of an active sensing system. The active sensing system can be, for example, an integrated circuit (IC) designed to transmit and sense electromagnetic radiation within a predetermined frequency range (e.g., 6 to 8 GHz, 6 to 8.5 GHz, 6 to 9 GHz). An IC capable of transmitting and sensing electromagnetic radiation over a wide frequency range (e.g., greater than 2 GHz) may be referred to as an "ultra-wideband (UWB) IC".
[0049] It is worth noting that the signal transmitter 206 and the signal sensor 208 can operate together to capture information about the distribution of blood vessels (and more specifically, the nature of certain blood vessels) at a distance through obstacles. Therefore, an unknown person to be authenticated may not need to actually contact the signal transmitter 206 or the signal sensor 208, as discussed further below.
[0050] As described above, the signal transmitter 206 and signal sensor 208 can be embedded in a single electronic device. In some embodiments, the electronic device is associated with an unknown person. For example, the signal transmitter 206 and signal sensor 208 can be embedded in a mobile phone associated with an unknown person. In other embodiments, the electronic device is not associated with an unknown person. For example, the signal transmitter 206 and signal sensor 208 can be embedded in a POS system through which an unknown person attempts to complete a transaction.
[0051] As shown in Figure 2A, the authentication platform 202A may include a vascular dynamics (VD) signal feature algorithm 212, a return signal evaluation algorithm 214, a VD signature processing algorithm 216, an authentication algorithm 218, and a biometric database 220. The biometric database 220 may store biometric data representing collected information related to the vascular properties that can be used to identify unknown individuals. As discussed further below, the biometric data in the biometric database 220 may vary depending on the authentication method employed by system 200A. The biometric data in the biometric database 220 may be encrypted, hashed, or otherwise obfuscated to prevent unauthorized access.
[0052] For example, in an embodiment of authentication platform 200A that determines the suitability of authentication based on vascular dynamics, biometric database 220 may include reference values for different vascular properties. Therefore, biometric database 220 may include data indicating the temporal changes in vascular properties for a given vascular vessel. As discussed further below, this data can be used for authentication because the temporal changes can not only be uniquely associated with an individual but also vary across different vascular vessels within that individual.
[0053] As another example, biometric database 220 may include models indicating the deformation of a given blood vessel when a gesture is performed by a given individual or different individuals. Each model may represent a series of discrete locations indicating how the shape of a given blood vessel changes over time as the gesture is performed. These models may be stored in profiles associated with different individuals. Profiles may include a single model associated with a single gesture, multiple models associated with a single gesture, or multiple models associated with different gestures.
[0054] Therefore, the biometric database 220 may include one or more biometric signatures. Each biometric signature may represent one or more values indicating temporal changes in the vascular properties of a given blood vessel in a given individual. As an example, a biometric signature may contain a vector of length N, where each element is an intensity value based on the magnitude of the frequency shift between the probe signal and the corresponding return signal, thereby indicating the amplitude of the blood vessel wall. N may represent the number of samples acquired over time. In other words, N may represent the number of probe signals emitted into the body during the authentication process.
[0055] Biometric signatures in biometric database 220 can be associated with a single individual, in which case authentication platform 202A can be limited to authenticating an unknown person as that individual. Alternatively, these biometric signatures can be associated with multiple individuals, in which case authentication platform 202A may be able to authenticate an unknown person as any of those individuals. Therefore, a single individual may have multiple biometric signatures in biometric database 220. These biometric signatures may correspond to different blood vessels and / or different electronic devices used for collection. For example, a single electronic device (e.g., a mobile phone) may be located near multiple anatomical areas (e.g., in different pockets), and each anatomical area has a different biometric signature. As another example, an individual may be associated with multiple electronic devices (e.g., mobile phones and wearable devices), and each electronic device may have a different biometric signature.
[0056] In some embodiments, each biometric signature is assigned a quality level that indicates its reliability for authentication. The quality level can be based on factors such as the signal-to-noise ratio (SNR) of the original returned signal and the repeatability of the original measurement. In embodiments where multiple biometric signatures are used for authentication, appropriate quality levels can be used to assign appropriate weights to those biometric signatures. Generally, more weight can be assigned to biometric signatures with higher quality levels.
[0057] When executed by processor 210, the algorithm implemented in authentication platform 202A allows individuals to generate biometric signatures during the registration phase. Subsequently, the algorithm implemented in authentication platform 202A allows verification during the usage phase. The registration and usage phases are further described below with reference to Figures 4A-7B.
[0058] VD signal characterization algorithm 212 can be responsible for defining the properties of the probe signal to be transmitted into the body of an unknown person to be authenticated. For example, VD signal characterization algorithm 212 can define characteristics such as the frequency of the probe signal (e.g., to control penetration depth), amplitude (e.g., to return a stronger signal at a higher risk of noise), samples per second, transmission timing, transmission duration, or any combination thereof. VD signal characterization algorithm 212 can take any of the following as input: • The nature of the electronic device, such as its known location or power availability; • Properties of the signal transmitter 206, such as power requirements, maximum transmission amplitude, or maximum transmission frequency; • The properties of the signal sensor 208, such as power requirements, sensitivity, or frequency range; or • The output generated by the return signal evaluation algorithm 214 indicates, for example, the problem of data acquisition.
[0059] The return signal evaluation algorithm 214 can be responsible for evaluating the return signal by examining the data generated by the signal sensor 208. For example, the return signal evaluation algorithm 214 can evaluate each return signal to generate and then assign a quality level. If one or more return signals are deemed unacceptable (e.g., based on the corresponding quality level), the return signal evaluation algorithm 214 can prompt the VD signal characteristic algorithm 212 to change one or more characteristics of the probe signal. For example, the return signal evaluation algorithm 214 can trigger the VD signal characteristic algorithm 212 to change the probe signal in response to finding that a certain number or percentage of return signals are unacceptable. The data acquisition process can then restart. Therefore, the return signal evaluation algorithm 214 can initiate a feedback loop to optimize the characteristics of the probe signal.
[0060] VD signature processing algorithm 216 can be responsible for examining the data generated by signal sensor 208 to produce a biometric signature that can be used for authentication. As discussed further below, the biometric signature can be created during (i) the registration phase of an authentication program supported by individual registration authentication platform 202A and (ii) the usage phase during which an unknown person is authenticated by authentication platform 202A. In general, a biometric signature can indicate a metric related to vascular dynamics. One example of a biometric signature is a series of values indicating the location, size, volume, or pressure of a blood vessel as it changes over time. Another example of a biometric signature is a series of values indicating the velocity or acceleration of blood flowing through a blood vessel as it changes over time. While the term "biometric signature" can be used to refer to multiple values corresponding to different points in time, a single value can also be used as a biometric signature. For example, the depth of a given blood vessel relative to the skin surface can be used as a biometric signature.
[0061] During the usage phase, authentication algorithm 218 can be invoked to perform authentication by comparing a new biometric signature generated for the unknown person with one or more reference biometric signatures associated with a given individual. As mentioned above, the reference biometric signatures can be stored in biometric database 220. The output produced by authentication algorithm 218 can represent a final decision on whether the unknown person should be authenticated as a given individual. Authentication algorithm 218 can output discontinuous or continuous outputs, such as probability measures (e.g., the likelihood of specifying the unknown person as a given individual), binary outputs (e.g., authenticated or unauthenticated), classifications (e.g., specifying whether the unknown person is being authenticated with low, medium, or high confidence), etc.
[0062] Figure 2B includes a high-order representation of system 200B, which can be used to authenticate the identity of an unknown person based on gesture recognition. In general, system 200B of Figure 2B operates in a similar manner to system 200A of Figure 2A. However, here, the authentication platform 200B determines whether to authenticate the unknown person based on the analysis of data generated by signal sensor 208 when the unknown person performs a physical action (referred to as a "gesture"). Alternatively, the data may be based on or represent the output generated by signal sensor 208 when the unknown person performs a gesture.
[0063] Therefore, biometric database 220 may include data related to the execution of a given gesture, as well as other data related to vascular properties. For example, biometric database 220 may include a “gesture signature,” which represents data regarding the execution of a gesture by a given individual. The data may include one or more values of vascular properties during the execution. The gesture signature may be associated with anatomical location (and therefore one or more blood vessels) and / or electronic devices. As an example, when a given individual bends their wrist down and then up, the data may specify changes in the size of the radial artery, as determined by a wrist-worn electronic device. In this context, the gesture signature is associated with a specific gesture (i.e., bending the wrist down and then up), a specific blood vessel (i.e., the radial artery), and a specific electronic device (i.e., the wrist-worn electronic device).
[0064] A gesture can be any action that requires clenching or activating one or more muscles, resulting in changes in the blood supply to those muscles. These changes in blood supply affect the vascular dynamics of blood vessels within the same anatomical region. Each gesture signature can contain a vector of length N, where each element is an intensity value based on the magnitude of the frequency shift between the probe signal and the corresponding return signal. The frequency shift may be related to the dilation or constriction of the corresponding blood vessel, which in turn is related to the activation of certain muscles during the execution of the gesture.
[0065] As shown in Figure 2B, the authentication platform 202B may include a gesture recognition module 222 operable to facilitate vascular dynamics-based authentication when a gesture is performed. The gesture recognition module 222 may include a gesture processing algorithm 224 and a gesture recognition algorithm 226. In general, these algorithms are capable of determining whether the data generated by the signal sensor 208 when an unknown person performs a gesture is sufficiently similar to data generated by the signal sensor 208 (or another signal sensor) during one or more previous executions. This approach allows the gesture recognition module 222 to determine whether the unknown person has performed the gesture as requested, thereby determining whether the unknown person should be authenticated as a given individual.
[0066] The gesture processing algorithm 224 can be responsible for examining the data generated by the signal sensor 208 to generate a gesture signature that can be used for authentication. For example, the gesture processing algorithm 224 can analyze the data to identify a series of values corresponding to the gesture execution, and then define a gesture signature based on that series of values. As discussed further below, the gesture processing algorithm 224 can be assigned the task of generating gesture signatures during the registration and use phases.
[0067] Conversely, gesture recognition algorithm 226 can only be executed during the usage phase. During the usage phase, gesture recognition algorithm 226 can be responsible for identifying the gesture being performed based on the analysis of data generated by signal sensor 208. For example, gesture recognition algorithm 226 can identify a gesture by comparing a new gesture signature generated for an unknown person with one or more reference gesture signatures stored in biometric database 220. Gesture recognition algorithm 226 can then generate an output indicating whether the specified unknown person performed the gesture, for example, as requested by UI 202. In some embodiments, authentication platform 202B determines whether to authenticate an unknown person as a given individual solely based on whether the output indicates that the gesture was performed by an unknown person. In other embodiments, authentication platform 202B employs a more granular approach. For example, authentication platform 202B may attempt to determine the similarity between the gesture performance of an unknown person and past gesture performances of a given individual.
[0068] Figure 2C includes a high-order representation of system 200C, which can be used for location-based authentication of an unknown person. In general, system 200C of Figure 2C operates in a similar manner to system 200A of Figure 2A. However, here, the authentication platform 200C determines whether to authenticate the unknown person based on whether the signal sensor 208 is in a desired position relative to the unknown person's body.
[0069] In this embodiment, at least some biometric signatures in the biometric database 220 can be associated with values indicating locations. These values may be referred to as “location tags” or “location labels.” Location labels may include alphanumeric identifiers or descriptions of locations. Examples of suitable descriptions include “right front trouser pocket,” “left back trouser pocket,” “left chest,” and “right wrist.” Thus, each location label can indicate that the signal sensor 208 was located in a nearby anatomical region when the corresponding biometric signature was generated.
[0070] As shown in Figure 2C, the authentication platform 202C may include a location recognition module 228, which is operable to facilitate authentication based on whether the vascular dynamics indicator signal sensor 208 is in a desired location. The location recognition module 228 may include a registration system 230, a location algorithm 232, and a location rule database 234.
[0071] The location rule database 234 may contain rules associated with each available location tag. These rules can be defined by users of the authentication platform 200C, administrators responsible for managing the authentication platform 200C, or the authentication platform 200C itself. As an example, rules can specify purchase licenses, purchase conditions (e.g., maximum purchase value), or authentication conditions. Therefore, the location recognition module 228 (and thus the authentication platform 200C) can prevent, prohibit, or otherwise restrict authentication based on the rules in the location rule database 234. Rules can also be general instructions associated with a specific computer program. As an example, rules can define the standards that must be met when playing songs through a music-focused computer program, or rules can be defined as the standards that must be met when starting a step counter through a fitness-focused computer program. Rules can be configured (e.g., created or editable) through the UI 202.
[0072] Registration system 230 allows individuals (e.g., users or administrators) to assign rules to specific location tags and update these rules once they are stored in location rule database 234. As an example, a user might want to specify that transactions requiring authentication should be limited to a certain amount (e.g., no more than $50) unless a given electronic device (e.g., a mobile phone) is located in a given location (e.g., the left front trouser pocket). As discussed above, authentication platform 200C may be able to infer whether a given electronic device is located in a given location by examining data generated by signal sensor 208 during the usage phase. More specifically, authentication platform 200C can analyze the data to determine whether its values indicate a blood vessel (e.g., the femoral artery) known to be near the given location.
[0073] Simultaneously, the location algorithm 232 can be responsible for retrieving location tags associated with biometric signatures from the biometric database 214 as needed. Furthermore, the location algorithm 232 can use those location tags to retrieve appropriate rules from the location rule database 234.
[0074] Figure 3 illustrates an example of an electronic device 300 capable of implementing an authentication platform 314, designed to authenticate the identity of an unknown person based on data generated by a signal sensor 310. As discussed above, a signal transmitter 308 can transmit a probe signal into the body of the unknown person, and the signal sensor 310 can generate data based on the reflection of the probe signal by physiological structures within the body. In some embodiments, the probe signal represents a radio wave, and therefore the return signal detected by the signal sensor 310 can also be a radio wave. Note that the signal transmitter 308 can also be configured to transmit a discontinuous series of probe signals, referred to as "pulses," at time intervals.
[0075] In some embodiments, the authentication platform 314 is embodied as a computer program executed by the electronic device 300. For example, the authentication platform 314 may reside on a mobile phone capable of obtaining information that can determine whether the authentication is appropriate. As another example, the authentication platform 314 may reside on a wearable device capable of obtaining information that can make a determination. Examples of wearable devices include watches, fitness trackers, and head-mounted displays. In other embodiments, the authentication platform 314 is embodied as a computer program executed by another electronic device communicatively connected to the electronic device 300. In this embodiment, the electronic device 314 may transmit data to another electronic device for processing. For example, while authentication of an unknown person may be requested by a POS system used to initiate a transaction, the data may be generated by a mobile phone located near the unknown person. The data may be provided to the POS system or another electronic device (e.g., a computer server) for processing, or it may be processed by the mobile phone before being transmitted to the POS system or other electronic device. Those skilled in the art will recognize that the form of the authentication platform 314 may also be distributed across multiple electronic devices.
[0076] Electronic device 300 may include processor 302, memory 304, user interface (UI) output mechanism 306, signal transmitter 308, signal sensor 310, and communication module 312. Communication module 312 may be, for example, a wireless communication circuit designed to establish a communication channel with other electronic devices. Examples of wireless communication circuits include integrated circuits (also called “chips”) configured for Bluetooth, Wi-Fi, NFC, etc. Processor 302 may have general-purpose characteristics similar to a general-purpose processor, or processor 302 may be an application-specific integrated circuit (ASIC) that provides control functions for electronic device 300. As shown in Figure 3, processor 302 may be directly or indirectly coupled to all components of electronic device 300 for communication purposes.
[0077] Memory 304 can be composed of any suitable type of storage medium, such as static random access memory (SRAM), dynamic random access memory (DRAM), electrically erasable programmable read-only memory (EEPROM), flash memory, or temporary registers. In addition to storing instructions that can be executed by processor 302, memory 304 can also store data generated by signal sensor 310 and data generated by processor 302 (e.g., when executing the module of authentication platform 314). Note that memory 304 is merely an abstract representation of the storage environment. Memory 304 can be composed of actual memory chips or modules.
[0078] As discussed above, signal transmitter 308 can be configured to emit electromagnetic radiation into the body of an unknown person to be authenticated, for example, in the form of radio waves. Typically, signal transmitter 308 will only emit electromagnetic radiation when instructed to do so. For example, authentication platform 314 can be configured to determine whether authentication is required. In this scenario, authentication platform 314 can generate an output from prompting processor 302 instructing signal transmitter 308 to emit a detection signal with specific characteristics. As discussed above, these characteristics can be determined by authentication platform 314.
[0079] The signal sensor 310 can be any sensing device capable of collecting reflections of detection signals, such as those reflected by physiological structures within the body. Data on vascular dynamics that can be determined from it can be based on or represent these reflections.
[0080] The communication module 312 can manage communication between components of the electronic device 300. The communication module 312 can also manage communication with other electronic devices. Examples of electronic devices include mobile phones, tablets, personal computers, wearable devices, POS systems, and network-accessible server systems consisting of one or more computer servers. For example, in an embodiment where the electronic device 300 is a mobile phone, the communication module 312 can be communicatively connected to a network-accessible server system responsible for checking data generated by the signal sensor 310.
[0081] For convenience, authentication platform 314 may be referred to as a computer program residing in memory 304. However, authentication platform 314 may consist of software, firmware, or hardware components implemented in or accessible to electronic device 300. According to the embodiments described herein, authentication platform 314 may include various algorithms and modules (collectively, “elements”) as discussed above with reference to Figures 2A-C. These elements may be integral parts of authentication platform 314. Alternatively, these elements may be logically separate from authentication platform 314 but operate “together” with it. Together, these elements may enable authentication platform 314 to authenticate the identity of an unknown person based on an analysis of vascular dynamics determined from data generated from signal sensor 310. As an example, after obtaining data generated by signal sensor 310, authentication platform 314 may generate a biometric signature for an unknown individual to whom signal transmitter 308 emits pulse signals. Authentication platform 314 may then compare the biometric signature with at least one reference biometric signature associated with a given individual stored in biometric database 316. The authentication platform 314 can determine whether to authenticate an unknown person as a given individual based on the similarity between a biometric signature and a reference biometric signature. In Figure 3, the biometric database 318 is located in the memory 304 of the electronic device 300. However, the biometric database 318 may alternatively or additionally be located in remote memory accessible via a network by the electronic device 300.
[0082] Other elements may also be included as part of the authentication platform 314. For example, a UI module may be responsible for generating content to be output by the UI output mechanism 306 for presentation to an unknown person. The form of the content may depend on the nature of the UI output mechanism 306. For example, if the UI output mechanism 306 is a speaker, the content may include audio instructions to bring the electronic device 300 closer to the anatomical area or to indicate that the authentication phase should continue. As another example, if the UI output mechanism 306 is a display, the content may include visual instructions to bring the electronic device 300 closer to the anatomical area or to indicate that the authentication phase should continue. Authentication methods
[0083] Figures 4A-B include flowcharts of the process for authenticating a user of an authentication platform based on radar detection of vascular dynamics monitored using electronic devices close to the body. Figures 5A-B include flowcharts of these processes in a gesture recognition context, while Figures 6A-B include flowcharts of these processes in a location recognition context. Note that unless otherwise stated, the steps of these processes can be combined with steps of other processes. Therefore, when an authentication platform is assigned the task of authenticating an unknown person, it can utilize vascular dynamics recognition, gesture recognition, or location recognition methods.
[0084] As shown in Figures 4A-B, the authentication process has two phases: registration phase 400 and usage phase 450. These phases can be designed to allow for hands-free authentication without requiring the user to interact with any electronic device.
[0085] To begin registration phase 400, the user can position the electronic device, including (i) a signal transmitter and (ii) a signal sensor, in an initial position. Note that in some cases, the user may need to place the electronic device in the initial position. For example, the user may need to keep the electronic device in the initial position. In other cases, the user may only need to keep the electronic device in its current position. For example, the user may be instructed to place the electronic device in a trouser pocket where it is already in place.
[0086] The user can be prompted to position the electronic device in an initial location via an electronic device or another electronic device. For example, a user can indicate their interest in registering with the authentication platform through an interface displayed on the electronic device's screen. Typically, the initial location is close to the anatomical area of interest (and therefore, blood vessels). For instance, the authentication platform could instruct the user to position the electronic device in their front trouser pocket to be near the femoral artery, or it could instruct the user to position the electronic device near their wrist to be near the radial artery.
[0087] In some embodiments, users can authenticate themselves before the registration phase 400. For example, users may be prompted to provide sensitive information that can serve as proof of identity (e.g., name, address, or payment card number).
[0088] The authentication platform's VD signal feature algorithm can then generate an initial set of signal features. In some embodiments, this initial set of signal features represents preset signal features, while in other embodiments, it represents preset signal features modified based on conditions of registration phase 400. For example, the VD signal feature algorithm can modify a preset frequency range for scanning based on the initial position (and therefore, possibly the required penetration depth). As another example, the VD signal feature algorithm can modify a preset spike amplitude based on the minimum power requirements of the signal sensors included in the electronic device.
[0089] This initial set of signal characteristics can be provided to a signal transmitter included in the electronic device, enabling the transmission of a probe signal based on these initial characteristics. Therefore, the signal transmitter can transmit the probe signal within a range of intensity, duration, timing conditions, and frequency defined by the VD signal characteristic algorithm. As shown in Figure 4A, the signal sensor then detects the return signal generated by the reflection of the probe signal within the device. The return signal can be provided as input to either a return signal evaluation algorithm or a VD signature processing algorithm.
[0090] The return signal evaluation algorithm can be responsible for evaluating the quality of the returned signal. For example, the return signal evaluation algorithm can compare one or more properties of the returned signal to a predefined benchmark to evaluate the quality value. Such properties may include amplitude, SNR, and repeatability. For each benchmark, the return signal evaluation algorithm can assign a value based on the difference between the corresponding property and the benchmark (e.g., between 0 and 1, between 1 and 100, etc.). These values can then be summed and divided by the total number of properties to establish a quality grade for the overall returned signal. In some embodiments, weights can be assigned to the properties. In such embodiments, some properties may have a greater impact on the quality grade than others. If any value is below a threshold that defines a minimum acceptable standard, the return signal evaluation algorithm can output a report indicating that each property does not meet the corresponding benchmark. The VD signal characterization algorithm can use this report to update the signal characteristics of the probe signal and then initiate another scan.
[0091] After obtaining an acceptable return signal, the VD signature processing algorithm can process the return signal to generate a biometric signature. For example, the VD signature processing algorithm can, for each of N timestamped samples acquired within a time interval (also known as a "measurement interval" or "scan interval"), obtain the frequency difference (called a "frequency shift") between the probe signal and the return signal. The frequency shift can then be normalized and stored in a vector of length N. In some embodiments, the vector also includes information such as the time "step" between each frequency shift. The vector can represent the biometric signature generated for the return signal.
[0092] As shown in Figure 4A, the quality level generated by the return signal evaluation algorithm and the biometric signature generated by the VD signature processing algorithm can be stored in a biometric database accessible to the authentication platform. For example, the quality level and biometric signature can be populated into entries representing a data structure of the biometric database. In some embodiments, the biometric database is maintained on the same electronic device as the authentication platform, while in other embodiments, the biometric database is maintained on a different electronic device. For example, the authentication platform may reside on an electronic device close to the user's body, while the biometric database may reside on a network-accessible server system to which the electronic device is communicatively connected.
[0093] This process may be completed multiple times during Phase 400 of the registration process. For example, a user may locate an electronic device in different locations to generate multiple biometric signatures, or a user may locate different electronic devices in the same or different locations to generate multiple biometric signatures. Regardless of the number of biometric signatures generated during Phase 400, these biometric signatures can be stored in a biometric database for future use during authentication.
[0094] In phase 450, the user may be prompted to position the electronic device at one of the predetermined locations defined in phase 400 for the biometric signature. For example, suppose the user is attempting to complete a transaction using a POS system. In this scenario, the POS system can send a request to authenticate the user directly or indirectly to the authentication platform. For instance, the POS system can send the authentication request directly to the electronic device via a wireless channel established using short-range communication protocols such as Bluetooth, Near Field Communication (NFC), or Wi-Fi. As another example, the POS system can send an instruction requiring authentication to a server system via a network, and the server system can then send an authentication request to the electronic device.
[0095] Upon receiving the request, the authentication platform can perform the steps discussed above regarding the registration phase 400 to generate a biometric signature. As mentioned above, these steps may need to be performed one or more times to generate a biometric signature of acceptable quality.
[0096] One or more reference biometric signatures can be retrieved from a biometric database using information about the electronic device or user to be authenticated. For example, criteria such as the device's geographic location, its identifier, and the user's claimed identity can be used to identify the reference biometric signatures. The VD authentication algorithm can then compare the biometric signature with the reference biometric signatures. For example, the VD authentication algorithm can identify equivalent features between the biometric signature and the reference biometric signature. An example of such a feature is a sequence of elements in two vectors where the frequency shift is increasing, representing the start of a pulse. Starting with equivalent features, the value in each element of the biometric signature can be compared with the corresponding element in the reference biometric signature. This can be done to identify differences in values and / or the rate of change of values. This process can be repeated for one or more features determined by the VD authentication algorithm to make the biometric signature and the reference biometric signature approximately or substantially equal. The VD authentication algorithm can then check these differences against an acceptable tolerance. For example, if these differences fall within 0.25, 0.5, or 0.75 standard deviations, the VD authentication algorithm can consider the biometric signature to be sufficiently similar to the reference biometric signature.
[0097] The VD authentication algorithm can make authentication decisions based on, for example, the number of differences within an acceptable tolerance and / or the average magnitude of the deviations. While authentication platforms may have varying degrees of confidence in their decisions, as mentioned above, the authentication decision is typically binary. In other words, the authentication decision usually indicates whether authentication has passed or failed. The authentication decision can be transmitted from the authentication platform to its destination. For example, if the authentication request was submitted by a POS system as described above, the authentication decision can be sent to the POS system.
[0098] Figures 5A-B include flowcharts of the registration phase 500 and usage phase 550 in an embodiment where an authentication platform determines whether to authenticate a user based on gesture recognition. These flowcharts illustrate how the authentication platform can detect gesture performance based on how those gestures change the user's vascular dynamics.
[0099] In general, the registration phase 500 and usage phase 550 of Figures 5A-B are equivalent to the registration phase 400 and usage phase 450 of Figures 4A-B. However, here, the user can be prompted to perform a gesture when the electronic device is close to an anatomical area of the body. The gesture can be selected (e.g., from several potential gestures) because the performance of the gesture may result in observable changes in the vascular dynamics of blood vessels located in the anatomical area. For example, the gesture might require the user to contract their thigh muscles once per second over a period of several seconds, or the gesture might require the user to bend their wrist downwards and then upwards over a period of several seconds.
[0100] As discussed above, the signal transmitter can transmit a probe signal with characteristics defined by the VD signal feature algorithm into the anatomical region, and the signal sensor can sense a return signal representing the probe signal reflected by physiological structures within the anatomical region. After the return signal evaluation algorithm determines that an acceptable return signal has been obtained, the gesture signature processing algorithm can process the return signal to generate a gesture signature. For example, the gesture signature processing algorithm can determine the frequency shift between the probe signal and the return signal for each of N timestamped samples acquired within a measurement interval. The frequency shift can then be normalized and stored in a vector of length N. As mentioned above, the vector can also include information such as the time “step” between each frequency shift. If the user is asked to perform an unknown gesture, the frequency shift can also be associated with or “marked” by an individual action. The vector can represent the gesture signature generated for the return signal.
[0101] In the registration phase 500, the gesture signature can be simply stored in a biometric database. However, in the usage phase 550, a gesture recognition algorithm can compare the gesture signature with at least one reference gesture signature. Reference gesture signatures can be identified using criteria such as the geographic location of the electronic device, the device's identifier, the user's claimed identity, and the gesture itself. For example, the gesture recognition algorithm can identify equivalent features between the gesture signature and the reference gesture signature. One example of such a feature is a sequence of elements in two vectors where the frequency shift is increasing, indicating the start of the gesture.
[0102] Starting with elements of equal characteristics, the value in each element of the gesture signature can be compared with the corresponding element of the reference gesture signature. This can be done to identify similarities and differences in properties such as time, rate of change, or frequency shift magnitude. In the vector, differences in such properties can be examined to determine if the difference falls within a certain tolerance (e.g., within 0.5 standard deviations). Based on these differences, the gesture recognition algorithm can generate a gesture confirmation indicating whether the gesture signature matches the reference gesture signature. In the event of a mismatch in gesture confirmation indications, the authentication platform may not authenticate the user, or the authentication platform may request the user to perform the gesture again (and then reuse stage 550).
[0103] Figures 6A-B include flowcharts of the registration phase 600 and usage phase 650 in an embodiment where an authentication platform determines whether to authenticate a user based on location identification. These flowcharts illustrate how the authentication platform can assign rules, settings, or functions to locate an electronic device in a given anatomical region by first identifying the location from vascular dynamics and then retrieving relevant rules from a biometric database.
[0104] As shown in Figure 6A, registration phase 600 can be very similar to registration phase 400 in Figure 4A. However, after a biometric signature has been generated, the user may be able to assign location tags to the biometric signature. The user, authentication platform, or a separate computer program executed by the electronic device can then use the registration system to output one or more location rules for the newly assigned location tags. For example, a location algorithm can generate a function table for each location tag in a biometric database. These functions can be called via an application programming interface (API) to apply the settings to a computer program executing on the electronic device or to call functions within those programs.
[0105] In phase 650, the authentication platform can generate a biometric signature, which can be compared with a reference biometric signature, as discussed above with reference to Figure 4B. After generating the biometric signature, the location algorithm can retrieve location tags associated with the reference biometric signature from a biometric database. The location algorithm can then use the location tags to retrieve corresponding location rules from a location rule database, if available. These location rules can then be used by the authentication platform (e.g., to determine if authentication is appropriate) or by another computer program executing on the electronic device. For example, in the case of a transaction completed through a computer program, location rules can be used to determine whether the transaction meets certain criteria before authentication. Additionally or alternatively, location rules can serve or facilitate the UI. For example, step counting monitored by the electronic device can be initiated by calling relevant functions through an API corresponding to a fitness-related computer program.
[0106] Figure 7A-B includes flowcharts of registration and use phases 700 and 750 of an example of a certification process that depends on vascular dynamics.
[0107] In registration phase 700, the UI may initially prompt the individual to position the electronic device in an initial location (step 701). Typically, the UI is presented by the electronic device. For example, the electronic device may provide audible or visual instructions for the individual to position the electronic device in the initial location. However, the UI may also be presented by another electronic device. For example, the UI may be presented by a mobile phone on which the individual has expressed interest in completing registration phase 700; however, the UI may also instruct the individual to position a wearable device (e.g., a watch or fitness tracker) in a given anatomical area (e.g., the wrist).
[0108] The authentication platform can then determine the initial characteristics of the probe signal to be emitted into the body (step 702). As described above, the initial characteristics can be based on the individual, the electronic device, or the initial location. The authentication platform can then cause (i) a signal transmitter to emit the probe signal into the body and (ii) a signal sensor to sense the return signal generated by the reflection of the probe signal (step 703). For example, the authentication platform can generate an instruction specifying the initial characteristics of the probe signal and then send the instruction to the operating system of the electronic device. Upon receiving the instruction, the operating system can operate the signal transmitter contained in the electronic device so that the probe signal with the initial characteristics is emitted. Data representing the return signal generated by the signal sensor can be obtained by the authentication platform (e.g., through the operating system).
[0109] The authentication platform can evaluate the data (and therefore the return signal) to generate a quality level (step 704). If the quality level is below a threshold, the authentication platform can adjust the detection signal settings as described above. However, if the quality level exceeds the threshold, the authentication platform can use the return signal to generate a biometric signature (step 705). Furthermore, the authentication platform can store the quality level and biometric signature in a biometric database (step 706).
[0110] In phase 750, the authentication platform first receives input indicating an authentication request from the source (step 751). In some embodiments, the source is a computer program running on the same electronic device as the authentication platform. For example, if the authentication platform resides on a mobile phone, the authentication request may originate from a mobile application through which the user attempts to perform the authentication request activity. In other embodiments, the source originates from another electronic device. For example, suppose a user attempts to complete a transaction using a POS system associated with a merchant. In this scenario, the POS system may require authentication, but the scanning can be performed by the electronic device associated with the user.
[0111] Subsequently, the authentication platform can perform steps equivalent to steps 701-705 of the registration phase (step 752). Therefore, the authentication platform can generate a biometric signature related to the vascular dynamics of blood vessels in the anatomical region based on data generated by signal sensors contained in an electronic device near the user's anatomical region. Next, the authentication platform can retrieve at least one reference biometric signature from a biometric database (step 753). The reference biometric signature can be associated with the individual the user will be authenticated as. Therefore, to determine whether to authenticate the user as that individual, the authentication platform can assess the similarity between the biometric signature and the reference biometric signature (step 754). The authentication platform can then transmit the authentication decision to the source (step 755). As discussed above, the authentication decision can be based on the degree of similarity between the biometric signature and the reference biometric signature.
[0112] Figure 8 illustrates a scenario where an unknown person is authenticated despite leaving an electronic device in their front trouser pocket. In this embodiment, the vascular dynamics of the femoral artery are used as the sole means of identifying the unknown person. Those skilled in the art will recognize that in other embodiments, authentication can be based on the vascular dynamics of another readily accessible vessel. For example, if the electronic device to be scanned is near the wrist or forearm, authentication can be based on the vascular dynamics of the radial or ulnar artery. Authentication can also be based on the vascular dynamics of multiple vessels as described above. For example, if the electronic device to be scanned is near the hip area (e.g., in a back trouser pocket), authentication can be based on the common iliac artery, internal iliac artery, and external iliac artery.
[0113] In general, this method allows for handless authentication because the authentication platform can define the characteristics of the probe signals to be emitted by the electronic device into an unknown human body. These characteristics can be generated through algorithms to allow for the analysis of returned signals, representing probe signals reflected by physiological structures within the body, to obtain information about vascular dynamics. As discussed above, the authentication platform can reside on the electronic device or another electronic device. For example, the authentication platform can reside on a network-accessible server system or POS system that is communicatively connected to the electronic device.
[0114] Typically, the detection signal represents electromagnetic radiation emitted within a 3.1–10.6 GHz bandwidth, although in some embodiments the electromagnetic radiation may be outside this range. Therefore, as shown in Figure 8, the detection signal can not only pass through clothing before entering the body, but also penetrate skin, fat, and muscle to a certain depth. Similarly, the return signal generated by reflecting the detection signal may be able to pass through muscle, fat, skin, and clothing to reach the electronic device. For these reasons, during authentication, an unknown person may be able to simply leave the electronic device in its normal position (e.g., a front pocket of pants).
[0115] This authentication method has several advantages. First, it can use electronic devices including signal transmitters and sensors without requiring conscious action to initiate or complete authentication. In fact, the method described herein may not require the user to wear or place the electronic device in a location that would disrupt normal daily use. Second, this method represents a novel means of positioning electronic devices along the body surface, namely, by examining the vascular dynamics of scanned blood vessels. Third, while this method allows gestures to be performed as a means of authentication, the gestures do not have to be attention-grabbing or flamboyant. Instead, the gestures can be subtle (e.g., clenching muscles or moving limbs), so that others are barely aware of them.
[0116] Figure 9 illustrates an example of a schematic implementation of an authentication platform 900 capable of authenticating an unknown person 902 based on vascular dynamics. Initially, the authentication platform 900 receives input indicating a request to authenticate the unknown person (also referred to as the "user" of the authentication platform 900). In Figure 9, this request originates from a POS system 904 that the unknown person 902 is using to complete a transaction. In other embodiments, this request may originate from other locations as discussed above.
[0117] The authentication platform 900 can then instruct the unknown person 902 to position the electronic device 906 at a given location corresponding to the anatomical region of interest. This instruction can be presented by the POS system 904, or it can be presented by the electronic device 906. The manner in which the instruction is presented can depend on how the authentication platform 900 is deployed. In other words, whether the instruction is presented by the POS system 904 or by the electronic device 906 (or some other electronic device) can depend on the location of the authentication platform 900.
[0118] After the electronic device 906 is positioned at a given location, it can transmit a probe signal to the anatomical region of interest. Typically, the probe signal is generated according to instructions from the authentication platform 900. These instructions can specify characteristics of the probe signal, such as frequency, amplitude, samples per second, transmission timing, transmission duration, or any combination thereof. Furthermore, the electronic device 906 can generate a return signal, representing the probe signal reflected by physiological structures within the anatomical region of interest.
[0119] Subsequently, the authentication platform 900 can examine the returned signal to determine whether to authenticate the unknown person 902 as a given individual. Specifically, the authentication platform 900 can generate a biometric signature indicating the vascular dynamics of blood vessels in an anatomical region of interest, and then compare this biometric signature with at least one reference biometric signature known to be associated with the given individual. The authentication decision can be based on the degree of similarity between the biometric signature and the reference biometric signature.
[0120] Additionally or alternatively, the authentication platform 900 may determine, based on the returned signal, whether the unknown person 902 has performed the expected gesture or whether the electronic device 906 is in the expected location. Therefore, the authentication platform 900 may determine whether to authenticate the unknown person 902 based on (i) vascular dynamics determined from the returned signal, (ii) the gesture determined from the vascular dynamics, or (iii) the location determined from the vascular dynamics.
[0121] After determining whether the unknown person 902 should be certified as the given individual she claims to be, the certification platform 900 can generate a notification instructing the certification decision. This notification can be transmitted to the POS system 904, which can then determine whether to allow the transaction based on the notification.
[0122] Figure 10 illustrates a flowchart of a process 1000 for biometric authentication of a person claiming to be a given individual. Initially, the authentication platform may receive input indicating a request for authentication of a person possessing an electronic device comprising (i) a signal transmitter and (ii) a signal sensor (step 1001). The signal transmitter and signal sensor may be, for example, part of a UWB IC capable of generating and detecting signals extending over a frequency range exceeding several GHz (e.g., 6 to 8 GHz, 6 to 8.5 GHz, 6 to 9 GHz). In some embodiments, the input is received from a computer program executing on the electronic device. In other embodiments, the input is received from another electronic device. For example, the input may be received from a POS system or a network-accessible server system to which the electronic device is communicatively connected.
[0123] Then, the authentication platform can cause the electronic device to transmit a detection signal into the human body within a time interval (step 1002). For example, the authentication platform can determine appropriate characteristics of the detection signal based on the electronic device or the person, and then the authentication platform can cause the signal transmitter to transmit a detection signal with appropriate characteristics. This can be achieved by sending an instruction to the operating system of the electronic device to operate the signal transmitter specifying the appropriate characteristics. Afterwards, the authentication platform can acquire data representing the detection signal reflected by blood vessels located in the body (step 1003). This data can be generated by a signal sensor contained in the electronic device that transmits the detection signal, or the data can be based on the output generated by the signal sensor contained in the electronic device that transmits the detection signal.
[0124] The authentication platform can determine the changes in vascular properties over time based on this data (step 1004). Therefore, the authentication platform can determine how blood vessels change over time rather than simply observing them in a "snapshot" corresponding to a single point in time. This property can be (i) the size of the blood vessel, (ii) the location of the blood vessel, (iii) the velocity of blood flowing through the blood vessel, or (iv) the pressure exerted by the blood on the blood vessel.
[0125] Then, the authentication platform can determine the probability that the person is a given individual based on the analysis of the property changes (step 1005). Therefore, the authentication platform can determine whether to authenticate the person as a given individual based on the user's vascular dynamics determined from the data. For example, the authentication platform can compare the property changes with reference values contained in a profile associated with the given individual, and then generate a score based on the similarity between the property changes and the reference values, indicating the probability that the person is the given individual.
[0126] In some embodiments, the authentication platform is programmed to generate biometric signatures representing variations of the properties described above. In this embodiment, the authentication platform may examine a biometric database containing biometric signatures associated with different individuals and then identify a reference biometric signature associated with a given individual from among those biometric signatures. Each biometric signature in the biometric database may represent a reference value for a property already established for the corresponding individual. The biometric database may be stored in data storage on an electronic device used to scan the body, or it may be stored in data storage accessible via a network to the electronic device. To determine the likelihood that the person is a given individual (and therefore whether to authenticate the person), the authentication platform may compare the biometric signature with the reference biometric signature.
[0127] Other steps may also be included. As an example, the authentication platform can infer from the data that the electronic device is located near an anatomical region of the body. More specifically, the authentication platform can analyze the data to identify patterns of values indicating a given blood vessel known to be located in that anatomical region. For example, if the authentication platform finds a value representing the vascular dynamics of the femoral artery, it can infer that the electronic device is located near the hip. As another example, the authentication platform can receive a second input indicating that the person has performed a gesture. In this embodiment, the authentication platform can determine whether the person has performed a gesture based on the analysis of the data. The authentication platform can determine that the person has performed the gesture in response to finding a pattern of values indicating gestures in the data.
[0128] Figure 11 illustrates a flowchart of process 1100 for generating a biometric signature for an individual that can subsequently be used for authentication. Initially, the authentication platform may receive a first input that instructs a request to create a biometric signature for the individual for authentication (step 1101). The first input itself may not request the creation of a biometric signature, but may simply represent a request for the individual to register for authentication services provided, facilitated, or supported by the authentication platform.
[0129] The authentication platform can then present instructions to position the electronic device near an anatomical region of the body (step 1102). In some embodiments, the individual may need to actively place the electronic device near the anatomical region, while in other embodiments, the individual may be able to leave the electronic device in its current position. For example, if (i) the anatomical region is the front or outer thigh and (ii) the electronic device is already in a front or back pocket of the pants, the electronic device may not need to be repositioned. As another example, if (i) the anatomical region is the front or back of the wrist and (ii) the electronic device is already secured to the wrist, the electronic device may not need to be repositioned. As another example, if (i) the anatomical region is the chest region and (ii) the electronic device is already in a breast pocket, the electronic device may not need to be repositioned. Other examples of anatomical regions include the neck region (also known as the “neck region”) and the temporal region, which can be monitored using electronic devices such as network-connected necklaces (e.g., smart necklaces), headphones, and hearing aids. Subsequently, the authentication platform can cause the electronic device to emit a detection signal into the individual's body (step 1103) and acquire data representing the detection signal reflected by blood vessels located in the body (step 1104). Steps 1103-1104 in Figure 11 can be similar to steps 1002-1003 in Figure 10.
[0130] Subsequently, the authentication platform can provide the data as input to an algorithm, which generates a biometric signature for the individual as output (step 1105). The biometric signature can represent one or more values indicating temporal changes in the vascular properties of a blood vessel. As an example, the biometric signature can contain a vector of length N, where each element is an intensity value based on the magnitude of the frequency shift between a probe signal and a corresponding return signal. In some embodiments, the biometric signature conveys information about the size of the blood vessel. In other embodiments, the biometric signature conveys information about the depth of the blood vessel under the skin, the blood flow velocity through the blood vessel, the pressure exerted by the blood on the blood vessel, etc. In other embodiments, the biometric signature represents the vibrational profile of the blood vessel over the time interval during which a probe signal is emitted into the body.
[0131] The authentication platform can then store the biometric signature in a profile associated with the individual (step 1106). This profile can represent a single entry or multiple entries corresponding to the individual in a biometric database. The biometric database can be stored on an electronic device used to scan the body and / or on another electronic device communicatively connected to that electronic device.
[0132] These steps can be performed in various sequences and combinations. For example, if the authentication platform requires the electronic device to be repositioned to another location or to another electronic device located in that location, these steps can be performed a second time. This ensures that authentication is robust to changes in the electronic device's location. Therefore, the authentication platform can generate biometric signatures for different anatomical regions (and therefore different blood vessels), and these biometric signatures can be used individually or collectively for future authentication.
[0133] Figure 12 illustrates a flowchart of process 1200 for determining whether to authenticate a person as a given individual based on vascular dynamics over time. Initially, the authentication platform may present an instruction to position an electronic device near the anatomical region of the person to be authenticated (step 1201). This instruction may be audible or visual in nature. For example, the instruction may audibly request the electronic device to be placed near the anatomical region, or the instruction may visually request (e.g., via text instruction or visual representation) the electronic device to be placed near the anatomical region. In some embodiments, the instruction is presented by the electronic device, while in other embodiments, the instruction is presented by another electronic device (e.g., a POS system). Subsequently, the authentication platform may acquire data generated by a signal sensor installed in the electronic device (step 1202). This data may represent a return signal based on (e.g., generated from) the reflection of a probe signal emitted by the electronic device to the person over a time interval.
[0134] The authentication platform can estimate changes in the properties of blood vessels located in the anatomical region over the time interval based on this data (step 1203). The estimated changes can represent changes in the shape, location, flow rate, or pressure of the blood vessels over the time interval. The authentication platform can then determine whether to authenticate the person as the given individual based on a comparison of the estimated changes with a profile associated with that individual (step 1204). This profile can represent one or more entries in a biometric database. Each entry can include a reference value indicating the degree of change in properties when the given individual completes the registration phase. Therefore, the authentication platform can compare the estimated changes with at least one reference value to determine the probability that the person is the given individual. Processing system
[0135] Figure 13 is a block diagram illustrating an example of a processing system 1300 in which at least some of the operations described herein can be implemented. For example, the components of the processing system 1300 may be mounted on an electronic device including a signal transmitter and a signal sensor. As another example, the components of the processing system 1300 may be mounted on an electronic device including an authentication platform.
[0136] The processing system 1300 may include a processor 1302, main memory 1306, non-volatile memory 1310, a network adapter 1312 (e.g., a network interface), a video display 1318, input / output devices 1320, control devices 1322 (e.g., a keyboard, pointing device, or mechanical input, such as buttons), a drive unit 1324 including storage media 1326, or a signal generation device 1330 communicatively connected to bus 1316. Bus 1316 is described as an abstraction representing one or more physical buses and / or point-to-point connections via appropriate bridges, adapters, or controllers. Therefore, bus 1316 may include system bus, peripheral component interconnect (PCI) bus, PCI-Express bus, HyperTransport bus, industry standard architecture (ISA) bus, small computer system interface (SCSI) bus, universal serial bus (USB), inter-integrated circuit (I2C) bus, or bus conforming to IEEE standard 1394.
[0137] Processing system 1300 may share a computer processor architecture similar to that of a computer server, router, desktop computer, tablet computer, mobile phone, video game console, wearable electronic device (e.g., watch or fitness tracker), network-connected (“smart”) device (e.g., television or home assistant device), augmented or virtual reality system (e.g., head-mounted display), or another electronic device capable of executing a set of instructions (sequential or otherwise) specifying actions to be taken by processing system 1300.
[0138] Although main memory 1306, non-volatile memory 1310, and storage medium 1326 are shown as a single medium, the terms "storage medium" and "machine-readable medium" should be understood to include either a single medium or multiple media storing one or more sets of instructions 1328. The terms "storage medium" and "machine-readable medium" should also be understood to include any medium capable of storing, encoding, or carrying a set of instructions for execution by processing system 1300.
[0139] Generally, routines executed to implement embodiments of the present disclosure can be implemented as part of an operating system or a particular application, component, program, object, module, or sequence of instructions (collectively, "computer programs"). Computer programs typically contain one or more instructions (e.g., instructions 1304, 1308, 1328) set at various times in various memories and storage devices within a computing device. When read and executed by processor 1302, the instructions cause processing system 1300 to perform operations to execute various forms of the present disclosure.
[0140] Although embodiments have been described in the context of a full-featured computing device, those skilled in the art will understand that various embodiments can be distributed as computer products in various forms. This disclosure applies regardless of the specific type of machine or computer-readable media used to actually cause the distribution. Further examples of machine and computer-readable media include recordable media, such as volatile and non-volatile memory devices 1310, removable disks, hard disk drives, optical discs (e.g., read-only optical disc memory (CD-ROMs) and digital optical discs (DVDs)), cloud-based storage, and transmission media, such as digital and analog communication links.
[0141] Network adapter 1312 enables processing system 1300 to coordinate data in network 1314 with entities outside processing system 1300 via any communication protocol supported by processing system 1300 and external entities. Network adapter 1312 may include network adapter cards, wireless network interface cards, switches, protocol converters, gateways, bridges, hubs, receivers, repeaters, or transceivers including integrated circuits (e.g., enabling communication via Bluetooth or Wi-Fi). Remark
[0142] For illustrative and descriptive purposes, the foregoing description of various embodiments of the claimed subject matter is provided. It is not intended to be exhaustive or to limit the claimed subject matter to the precise form disclosed. Many modifications and variations will be apparent to those skilled in the art. The embodiments were chosen and described in order to best describe the principles of the invention and its practical application, thereby enabling those skilled in the art to understand the claimed subject matter, the various embodiments, and the various modifications suitable for the particular purpose considered.
[0143] While the embodiments described present certain examples and considered best practices, the technology can be practiced in a variety of ways, however detailed the embodiments may be. Examples may differ considerably in their implementation details, but are still covered by the specification. Specific terms used in describing certain features or manner of the various embodiments should not be construed as implying that a term is redefined herein as limited to any particular characteristic, feature, or manner of the technology associated with that term. In general, the terms used in the following claims should not be construed as limiting the technology to the specific embodiments disclosed in the specification, unless those terms are expressly defined herein. Therefore, the actual scope of the technology includes not only the disclosed embodiments but also all equivalent ways of implementing or practicing the embodiments.
[0144] The language used in this specification has been chosen primarily for readability and pedagogical purposes. It may not have been chosen to describe or limit the subject matter of the application. Therefore, the scope of the technology is intended not to be limited by the implementation methods, but rather by the scope of any patent application published based on this invention. Thus, the disclosure of various embodiments is intended to illustrate, and not limit, the scope of the technology as set forth in the claims below.
[0145] 200A: System 200B: System 200C: System 202A: Authentication Platform 202B: Authentication Platform 202C: Authentication Platform 204: User Interface (UI) 206: Signal transmitter 208: Signal Sensor 210: Processor 212: Algorithm for Vascular Dynamics (VD) Signal Feature Analysis 214: Return Signal Evaluation Algorithm 216: VD Signature Processing Algorithm 218: Authentication Algorithm 220: Biometric Database 222: Gesture Recognition Module 224: Gesture Processing Algorithm 226: Gesture Recognition Algorithm 228: Location Recognition Module 230: Registration System 232: Position Algorithm 234: Location Rule Database 300: Electronic devices 302: Processor 304: Memory 306: User Interface (UI) Output Mechanism 308: Signal transmitter 310: Signal sensor 312: Communication Module 314: Authentication Platform 316: Biometric Database 400: Registration Phase 450: Usage Phase 500: Registration Phase 550: Usage Phase 600: Registration Phase 650: Usage Phase 700: Registration Phase 701~706: Steps 750: Usage Phase 751~755: Steps 900: Authentication Platform 902: Unknown Person 904: POS System 906: Electronic Devices 1000: Process 1001~1005: Steps 1100: Process 1101~1106: Steps 1200: Process 1201~1204: Steps 1300: Processing System 1302: Processor 1304: Instruction 1306: Main Memory 1308: Instruction 1310: Non-volatile memory 1312: Network Adapter 1314: Internet 1316: Busbar 1318: Video Monitor 1320: Input / Output Device 1322: Control device 1324: Drive Unit 1326: Storage Media 1328: Instruction 1330: Signal generating device
Claims
1. A method for authenticating a person as a given individual, the method comprising: Receive input via an electronic device instructing the person possessing the electronic device to authenticate the device. The electronic device transmits signals to the person's body over a time interval; acquires data representing the signals reflected by blood vessels located within the body; infers, based on the data, that the electronic device is located in a pocket of clothing near an anatomical region of the body, wherein the inference is based on a pattern of values in the data indicating a known blood vessel located in that anatomical region; determines, based on the data, a change in the properties of that blood vessel over the time interval; and determines, based on the analysis of the change in these properties, the probability that the person is the given individual.
2. The method of request item 1 also includes: The electronic device receives and confirms that a second input has been prompted the person to perform a gesture. And through the analysis of the data by the electronic device, it can be determined whether the person performed the gesture.
3. The method of request item 2, wherein in response to finding a second pattern in the data indicating the value of the gesture, it is determined that the person has performed the gesture.
4. The method of request item 1, wherein the determination includes: The change in this property is compared with a reference value contained in a profile associated with the given individual, and a score is generated based on the similarity between the change in this property and the reference value, the score indicating the probability that the person is the given individual.
5. The method of claim 4, wherein the property includes (i) the size of the blood vessel, (ii) the location of the blood vessel, (iii) the flow rate of blood through the blood vessel and / or (iv) the pressure exerted by the blood on the blood vessel.
6. The method of claim 4, wherein the configuration file is stored on a network-accessible server system that is communicatively connected to the electronic device.
7. The method of claim 1, wherein the data is generated by an ultra-wideband (UWB) integrated circuit in the electronic device, the UWB integrated circuit being capable of detecting signals in a frequency range from 6 GHz to 9 GHz.
8. The method of request item 1, wherein the determination includes: Examine a biometric database containing biometric signatures associated with different individuals, where each biometric signature represents a reference value for that property established for the corresponding individual, identify a reference biometric signature associated with the given individual from the biometric signatures contained in the biometric database, and compare the change in that property with the reference biometric signature to determine the probability.
9. The method of claim 8, wherein the biometric database is stored in a data storage accessible via a network by the electronic device.
10. The method of request item 1, wherein the input is received from a payment system used by the person to initiate a transaction that requires authentication.
11. The method of claim 1, wherein the anatomical region is the anterior or lateral region of the thigh, and wherein the vessel is the femoral artery.
12. The method of claim 1, wherein the anatomical region is the buttock region, and wherein the vessel is the common iliac artery, internal iliac artery, or external iliac artery.
13. The method of request 1, wherein the blood vessel and the given blood vessel are the same blood vessel.
Citation Information
Patent Citations
Bluetooth matching method and device
CN105992129A
Terminal and fingerprint unlocking method and device thereof
CN109409054A
Image taking device, image taking method, and image taking program
US20100074476A1
Method and apparatus for authenticating user using vein pattern
US20160117563A1
Biometric authentication device
US20190370527A1
