阅读说明：本技术 用于识别用户的方法和设备 (Method and apparatus for identifying user ) 是由 A.德拉弗雷斯特迪沃恩 P.莱维昂奈斯 于 2019-12-19 设计创作，主要内容包括：本发明涉及一种用于识别用户的方法,该用户的身体能够重新发射呈电磁波(R)形式的电磁信号(A)。该方法在收发器设备(M)上实施,并且包括该设备上的以下步骤：-发射(E1)电磁脉冲信号(A)；-当该用户在该设备附近时获得(E2)重新发射的信号(R),该信号取决于所发射的脉冲信号；-比较(E7)该重新发射的信号(R)与该用户的至少一个参考信号；-如果该重新发射的信号接近该参考信号,则识别(E9)该用户。(The invention relates to a method for identifying a user whose body is capable of re-emitting an electromagnetic signal (A) in the form of an electromagnetic wave (R). The method is implemented on a transceiver device (M) and comprises the following steps on the device: -emitting (E1) an electromagnetic pulse signal (a); -obtaining (E2) a re-transmitted signal (R) when the user is in the vicinity of the device, the signal depending on the transmitted pulse signal; -comparing (E7) the retransmitted signal (R) with at least one reference signal of the user; -identifying (E9) the user if the retransmitted signal is close to the reference signal.)

1. A method for identifying a user, the method being implemented on a transceiver device (M) and comprising the following steps on the device:

-emitting (E1) a pulsed electromagnetic signal (a);

-obtaining (E2) a signal (R) re-emitted by his or her body when the user is close to the device, said signal depending on the emitted pulse signal;

-comparing (E7) the retransmitted signal (R) with at least one reference signal of the user;

-identifying (E9) the user if the retransmitted signal is close to the reference signal.

2. An identification method as claimed in claim 1, wherein the identification step is followed by a step of updating the reference Signal (SIG) of the user by taking into account the retransmitted signal (R).

3. An identification method as claimed in any preceding claim, wherein the identification step is followed by the step of unlocking the transceiver device.

4. An identification method as claimed in any preceding claim, wherein the identification step is followed by a step of authorising the transaction.

5. An identification method as claimed in any preceding claim, wherein the at least one reference signal is associated with an identifier of a user, and wherein the identification step is followed by the step of selecting a user identified by his or her identifier.

6. An identification method as claimed in any preceding claim, wherein the step of comparing the retransmitted signal with at least one reference signal of the user comprises the sub-steps of:

-digitizing the received signal (R);

-normalizing the digitized signal (R);

-measuring the distance between the two signals.

7. A method for learning a Signature (SIG) of a user to make him or her recognized, the method being implemented on a transceiver device (M) and comprising the following steps on the device:

-emitting (E1) at least one pulsed electromagnetic signal (a);

-obtaining (E2) at least one signal (R) re-emitted by his or her body when the user is close to the device, said signal depending on the emitted pulse signal;

-generating (E6) a reference signal, called signature, from said at least one re-transmitted signal (R) received;

-saving (E6) the reference signal as the signature of the user.

8. The method for learning a Signature (SIG) of a user of claim 7, wherein the reference signal is saved in association with an identifier of the user.

9. A method for learning a user's Signature (SIG) according to claim 7 or 8, wherein the reference signal also depends on a verification action performed by the user when he or she is close to the device.

10. An apparatus for identifying a user, the apparatus comprising:

-means (CLF, ANT) for emitting a pulsed electrical signal (a);

-means (ANT, CLF) for obtaining a signal (R) re-emitted by his or her body when the user is close to the device, said signal depending on the emitted pulse signal;

-means (PG, UT) for comparing the retransmitted signal (R) with at least one reference signal of the user;

-means (PG, UT) for identifying the user when the retransmitted signal is close to the reference signal.

11. A mobile terminal or access control device comprising the device of claim 10.

12. An apparatus for learning a Signature (SIG) of a user, the apparatus comprising:

-means (CLF, ANT) for emitting a pulsed electrical signal (a);

-means (ANT, CLF) for obtaining a signal (R) re-emitted by his or her body when the user is close to the device, said signal depending on the pulse signal;

-means (PG, UT) for generating (E6) a reference signal, called signature, from said at least one re-transmitted signal (R) received;

-means (PG, UT) for saving (E6) said reference signal as signature of the user.

13. A mobile terminal or access control device comprising the apparatus of claim 12.

14. A computer program comprising instructions for implementing a method for identifying a user or a method for learning a signature of a user as claimed in any one of claims 1 to 9 when the program is executed by a processor.

Technical Field

The present invention relates to the identification of a human or animal user carrying a near field communication enabled device. More particularly, the present invention relates to authentication or authentication of a user carrying a mobile terminal.

Background

In order to authenticate or authenticate a user carrying a mobile terminal (such as for example a smartphone), a number of solutions are known: entering a secret code, facial recognition, fingerprint, iris authentication, etc. These solutions make it possible, among other things, to unlock a connected and powered terminal (such as, for example, a smartphone) and to carry out a secure transaction reserved for a single authenticated or authenticated user.

However, these methods are premised on a specific action of a user part: a face placed next to the screen, a finger placed on a fingerprint reader, etc. In addition, the user may be reluctant to provide information that he or she deems sensitive, such as, for example, his or her fingerprint, which may be reused for dishonest purposes.

Disclosure of Invention

The present invention improves upon the prior art. To this end, the invention proposes a method for identifying a user whose body can emit electromagnetic signals in the form of electromagnetic waves, the method being implemented on a transceiver device and comprising the following steps on the device:

-transmitting a pulsed electromagnetic signal;

-obtaining a retransmitted signal when the user is close to the device, said signal depending on the transmitted pulse signal;

-comparing the retransmitted signal with at least one reference signal of the user;

-identifying the user if the retransmitted signal is close to the reference signal.

According to the invention, a method for identifying a user is proposed. Advantageously, such a device allows the user to be identified, and thus authenticated or authenticated, in order to verify the transaction very simply, since the user can be identified without performing any specific action. The signal retransmitted and received by the device is dependent on the user and knowledge of the user allows him or her to be identified. The form of the signal generated and emitted via the body of the user actually depends on a certain number of characteristics (body shape, age, sex, moisture of the tissue, etc.) specific to the carrier. Analysis of such signals (form, power, etc.) makes it possible to extract therefrom characteristics specific to the user and thus identify the user by comparison with a known similar reference signal (signature). If another user engages the terminal, he or she will not be recognized because he or she does not have the same biometric characteristic.

"identification" is understood herein to mean the identification of a signature of a user in a broad sense. Identification may involve authentication, i.e. verifying the legitimacy of the user of the terminal (or identifying the fact that the user is indeed the owner of the phone), or authentication of the user, i.e. determining the identity of the user (indeed Jacques instead of Paul); naturally, authentication may be followed by authentication (Jacques may legally use the terminal) and vice versa.

A "transceiver device" is understood to mean any device capable of emitting a pulsed signal in the form of an electromagnetic wave, for example a device of the NFC type. Such a device may be any terminal associated with a user.

By means of the invention, the identification is automatic and secure. In fact, according to the prior art, the user who wants to be identified has to perform a specific action for this purpose, such as for example entering a code, or presenting his or her fingerprint, etc.

A "reference signal" is understood to be a signal itself or a set of data that allows the signal to be represented.

"close proximity" is understood to mean a distance (e.g. less than a few centimetres) small enough for establishing communication through human channels. It is noted that the skin of the user does not need to be in contact with the terminal to establish communication; the user's hand also does not necessarily need to be in physical contact with the antenna of the device.

According to a particular embodiment of the invention, the identification step of the method is followed by a step of updating the reference signal of the user by taking into account the retransmitted signal.

Advantageously, according to the present embodiment, the signal for identification is used to update the signature of the user. In fact, the user's physical imprint is inherent to him or her and may change over time (with his or her age, etc.). This signature can therefore be updated regularly, ensuring a higher level of security than static biometric parameters (fingerprints, iris prints, etc.) which do not change over time and which might otherwise be stolen.

According to a particular embodiment of the invention, the identifying step of the method is followed by the step of unlocking the transceiver device.

Advantageously, according to this embodiment, the device can be unlocked automatically as soon as the user grips the device (or approaches it) and automatically re-emits the electromagnetic waves already emitted by the device via his or her body.

According to another particular embodiment of the invention, the step of identifying of the method is followed by a step of authorizing the transaction.

Advantageously, according to the present embodiment, transactions that require authentication of the user (such as, for example, banking transactions) can be performed automatically once the user is close to the device, and there is no need to implement complex authentication procedures as proposed in the prior art.

According to another particular embodiment of the invention, the at least one reference signal is associated with an identifier of the user, and the step of identifying is followed by the step of selecting the user identified by his or her identifier.

Advantageously, according to the present embodiment, the user is authenticated (except when necessary for authentication), that is, he or she is identified from among several users. This makes it possible to carry out personalized transactions: selecting a profile for the user, displaying personalization information, personalization for the device or another device to which the device is linked, and the like.

According to another particular embodiment of the invention, the step of comparing the retransmitted signal with at least one reference signal of the user comprises the sub-steps of:

-digitizing the received signal;

-normalizing the digitized signal;

-measuring the distance between the two signals.

This embodiment of the invention makes it possible to simply carry out the step of comparing between the signal obtained in return and the reference signal. Any type of distance calculation available to those skilled in the art will be able to be used. This distance can typically be compared to a predetermined threshold or an acceptable maximum distance between the two signals. Since the two signals are not necessarily aligned, e.g. because the user is wearing different clothing, or because the terminal is at a more or less distance from the body, it is desirable to first use a processing algorithm that is capable of normalizing the second signal in order to take into account any amplitude differences between the signals.

The invention also relates to a method for learning a signature of a user whose body can re-emit an electromagnetic signal in the form of an electromagnetic wave so that he or she is identified, the method being implemented on a transceiver device and comprising the following steps on the device:

-transmitting at least one pulsed electromagnetic signal;

-obtaining at least one retransmitted signal when the user is close to the device, said signal depending on the transmitted pulse signal;

-generating a reference signal, called signature, from said at least one re-transmitted signal received;

-saving said reference signal as the signature of the user.

According to the present invention, a method for learning a signature of a user so that he or she is recognized is proposed. The method makes it possible to save the signature of the user in order to subsequently compare it with the response of the user's body to the impulse signal emitted by the device.

According to another particular embodiment of the invention, the learning method is further characterized in that the reference signal is saved in association with an identifier of the user.

According to another particular embodiment of the invention, the learning method is further characterized in that the reference signal depends on a verification action performed by the user when he or she is close to the device.

Advantageously, according to this embodiment, when the user is close to the terminal, his or her biometric signature can be combined with a signal relating to a specific action of the user's part. This carefully considered approach to IBC terminals has been described in particular in the application published under number WO 2016/001506 a 1. The user's action in fact changes the nature of the received and/or retransmitted wave, so that the signature carries an indication of the action: the signal will not be re-transmitted in the same manner depending on whether the user is approaching quickly or slowly, speeding up or slowing down, etc.

According to a hardware aspect, the invention also relates to a device for identifying a user whose body can re-emit an electromagnetic signal in the form of an electromagnetic wave, wherein the device comprises:

-means for transmitting a pulsed electrical signal;

-means for obtaining a retransmitted signal when the user is close to the device, said signal depending on the transmitted pulse signal;

-means for comparing the retransmitted signal with at least one reference signal of the user;

-means for identifying the user when the retransmitted signal is close to the reference signal.

According to another hardware aspect, the invention also relates to a mobile terminal comprising such an identification device.

According to another hardware aspect, the invention also relates to an access control device comprising such an identification device.

According to a hardware aspect, the invention also relates to a device for learning a signature of a user whose body can re-emit an electromagnetic signal in the form of an electromagnetic wave, wherein the device comprises:

-means for transmitting a pulsed electrical signal;

-means for obtaining a retransmitted signal when the user is close to the device, said signal depending on the pulsed signal;

-means for generating a reference signal, called signature, from said at least one re-transmitted signal received;

-means for saving said reference signal as the signature of the user.

According to another hardware aspect, the invention also relates to a mobile terminal comprising such a learning device.

According to another hardware aspect, the invention also relates to an access control device comprising such a learning device.

The invention also relates to a computer program comprising instructions for implementing one of the above-mentioned methods according to any one of the preceding specific embodiments when said program is executed by a processor. The method may be implemented in various ways, in particular in a hard-wired form or in a software form. This program may use any programming language and may be in the form of source code, object code, or intermediate code between source and object code, such as in partially compiled form, or in any other desired form.

The invention is also directed to a storage medium or information medium readable by a computer and comprising instructions of a computer program as described above. The storage medium mentioned above may be any entity or device capable of storing the program. For example, the medium may include a storage device such as a ROM (e.g., a CD ROM or a microelectronic circuit ROM), or even a magnetic storage device (e.g., a hard disk). Also, the storage medium may correspond to a transmissible medium (such as an electrical or optical signal) which may be conveyed wirelessly or otherwise via an electrical or optical cable. The program according to the invention may in particular be downloaded via an internet-type network.

Alternatively, the storage medium may correspond to an integrated circuit incorporating the program, the circuit being adapted to perform, or for use in the performance of, the relevant method.

Drawings

Other characteristics and advantages of the invention will become clearer from reading the following description of a specific embodiment, given as a simple schematic and non-limiting example, and the accompanying drawings, in which:

figure 1 illustrates an example of an environment in which the present invention may be practiced according to a particular embodiment,

figure 2 shows the architecture of a system implementing the invention according to an embodiment,

figure 3 illustrates the steps of a method for learning a signature according to a particular embodiment of the invention,

figure 4 illustrates the steps of an identification method according to a particular embodiment of the invention,

fig. 5 shows an example of a received impulse response of a user.

Detailed Description

5.1 general principles of the invention

The identification method described herein allows a user to be identified (i.e., authenticated or authenticated) on a transceiver device (hereinafter referred to as a terminal). This terminal is powered on and is preferably connected and mobile, such as for example a smartphone, a connected bracelet, a collar, etc.

The identification method allows a user to be authenticated by a smartphone, unlock the smartphone, and/or use value-added services that require strong authentication, such as payment, transportation, or other applications. Advantageously, the user does not need to perform a specific action to be identified.

The invention is based on the use of near field technology of the NFC type. It should be remembered that near field communication, commonly referred to as NFC, which is mainly based on the ISO (international organization for standardization) 14443 standard, uses wireless technology to allow the exchange of information between two remote peripheral devices over short distances, typically less than a few centimeters. This type of communication offers many applications, for example in the fields of payment and transportation.

Recently, new wireless communications based on the same concept have appeared, but these new wireless communications additionally use the human body as a communication channel. For these technologies combined together under the general term IBC (intra body communication) or even BCC (for body channel communication), the human body acts as an element for transmitting information from one point to another. These methods use electromagnetic coupling between the system and the human body and are applicable to proximity communication, which does not necessarily require physical contact with the device.

When the user approaches the terminal, the pulsed signal emitted by the terminal is emitted into his or her body at a certain frequency, preferably corresponding to that of NFC. Then, the human body acts as an antenna and re-emits electromagnetic waves having characteristics (resistance/capacitance/etc.) specific to each person. Thus, the response to the NFC pulse is unique and allows the user to be authenticated. It is still possible to combine this new recognition pattern with other already existing recognition patterns. It is also contemplated that the NFC device is carried by an animal rather than a human (e.g., a cat, which is used to lock its electronic cat hole).

5.2 specific embodiments of the invention

Fig. 1 illustrates an example of an environment in which the invention is implemented that allows a user (U) (designated herein as a "carrier") of a mobile terminal (M) to authenticate him or her via his or her cell phone to conduct a transaction.

According to this example, the user (U) or the bearer of the terminal (M) is a human being.

The terminal (M) according to the invention is a handheld device which is naturally capable of transmitting and receiving radio carriers through the body of the user (U) via an antenna. For this purpose, the terminal (M) is in close proximity to the user (U), without necessarily being in direct contact with the user. For example, the terminal (M) is placed in the user's hand, or if he or she is approaching the terminal, the terminal is placed a few centimeters from his or her hand. In these configurations, it is estimated that the antenna of the terminal (M) is not more than a few centimeters from the body of the user. This distance is for example less than 5 cm. The terminal (M) is equipped with a battery or battery unit for autonomous operation. According to this example, the terminal is a mobile terminal equipped with an NFC antenna (not shown) adapted to receive modulated electrical signals in the form of electromagnetic waves through the body of the user in IBC mode. Suitably is understood to mean that its antenna can be amplified or that the antenna has a sufficiently high gain. It is within the purview of one skilled in the art to acquire or amplify such an antenna.

Transaction is understood to mean any type of transaction, such as monetary transaction, purchase, ticket validation, environment customization, unlocking, etc.

According to a first example, it is assumed that a user wants to verify a purchase via a network (I), e.g. a mobile network or the internet. The terminal can establish a secure communication with a server (S) of the network in order to verify the monetary transaction; the user must authenticate to the server, that is, determine at the end of the method that he or she is indeed the owner of the terminal.

According to another example, the user (U) wants to customize the object (connected object, workstation, etc.) via his or her mobile device, so that the object performs the appropriate action (locking, personalized display, etc.) according to the person controlling it via the mobile terminal; in this case, the user must be authenticated and, if necessary, authenticated, that is, at the end of the method, he or she can be distinguished from several persons.

According to yet another example, the user wants to unlock his or her mobile terminal. For this purpose, it is sufficient that he or she is in the vicinity (e.g. while holding) in order to trigger the identification process and, if authentication is successful, to unlock the mobile device.

In all cases, the process according to the invention is carried out in two distinct phases or phases:

-a first phase: and learning the signature.

In a first phase, which corresponds to a so-called learning phase, the user performs several times (hereinafter referred to as N times, where N is a natural integer) on his or her terminal to trigger the operation of the learning module. The purpose of this step is to recover, at the terminal (or, alternatively, at another device with which the terminal can exchange data), a plurality (N) of signals corresponding to the signals generated by the person (U) in response to the pulses triggered by the learning method of the mobile device. These signals correspond to the characteristics of the user, but the variations are small, since the same mechanical/dynamic and physiological parameters of the user may vary over time, resulting in variations in the signals that are propagated by the body and then re-emitted. However, for a given person receiving the pulses from a given terminal, all signals have a very similar form overall and represent a kind of biometric print of the user, which will be referred to as "characteristic print" or "signature" of the user in the following. Thus, the characteristic signature represents an intrinsic characteristic of the user's body; it is in fact well known that certain biological factors of the user, such as for example age, humidity of the body tissue, etc., may influence its transmission characteristics. Reference may be made, for example, to the article "Intra-Body Communication Model Based on Variable Biological Parameters" in vivo Communication Model Based on Variable Biological Parameters (Khorshid et al, 2015, 49 th aid signals, systems and computer conference). Additionally, the stamp may be characteristic of his or her motion control if the user performs a particular gesture toward the mobile device while establishing proximity.

The characteristic Signature (SIG) can be obtained by N slightly different measurements delegated to a learning module responsible for calculating the "average" of the different signals or typical signals corresponding to the characteristic signature. For example, the module is an auto-learning or Machine Learning (ML) module. It should be recalled that machine learning or statistical learning involves the design, analysis, development, and implementation of methods that allow machines to evolve (broadly) through systematic processes, and thus accomplish difficult or problematic tasks through more traditional algorithmic means. One possible example of machine learning is classification, which aims to label data by associating each data with a class. The use of neural networks, wavelet analysis and decomposition methods, etc. are also contemplated.

According to this embodiment, the learning module calculates a characteristic footprint from the different signals returned by the user (e.g., taking the average of all valid tests, establishing a characteristic parameter set for the footprint, etc.). It then saves the reference signal of the user, possibly identified by the identifier, in a database. Once learning is performed, the resulting signature may advantageously be saved on the user's terminal. If the terminal is used by several users, several characteristic signatures may be saved, for example, together with the identifiers of these users if there is an interest in distinguishing each user.

-a second stage: the signature is used to identify the user.

In the second phase (of the implementation of the service), the IBC mobile terminal user, who wants to unlock his or her terminal or to verify the transaction, approaches the terminal and activates the application (unlock, payment, verification, personalization, etc.). The terminal transmits an impulse signal that propagates within the body of the user. The return signal is received by the terminal. The checking module of the terminal or a module linked to the terminal (e.g. a module on an external server) checks the signature of the user. Typically, a typical signal profile corresponding to the re-transmitted signal may be compared to a signal profile corresponding to the user signature, which was previously stored on the terminal or in a database accessible to the terminal.

If his or her signature is identified, the user is authenticated or authenticated and the transaction may be performed.

The present invention presents a fundamental advantage in terms of functional differentiation and security, since it allows a person who wants to access a security service to be authenticated and/or authenticated via his or her signature without having to perform specific gestures (facial or iris or fingerprint recognition, code entry, etc.).

This exemplary embodiment has been given by way of illustration and not limitation. Many variations may be applied to this exemplary embodiment. It is noted that another device (e.g., an external server) may perform learning and/or identification upon receiving data from the terminal. Furthermore, the recognition phase and the learning phase can be combined to produce an update of the signature: after the user has been identified, the characteristic imprint can advantageously be updated with the signal re-emitted by the body of the user during the identification phase. This allows, among other things, the signature of the user to change over time to take account of variations in his or her biological characteristics (e.g. due to age) and mechanical characteristics (e.g. due to his or her speed of approach to the terminal).

Fig. 2 illustrates a terminal M configured to implement a method for identifying a user according to a specific embodiment of the present invention. Such a terminal is configured to implement the identification method and the learning method according to any one of the particular embodiments of the invention described above.

According to a particular embodiment of the invention, the device M has the conventional architecture of a smartphone-type handset and comprises, inter alia, a memory MEM, a processing unit UT equipped with, for example, a processor and driven by a computer program PG stored in the memory MEM. The computer program PG comprises instructions for implementing the steps of the learning method or the recognition method as described before, when the program is executed by a processor.

At initialization, the code instructions of the computer program PG are loaded into a memory, for example, before being executed by a processor.

The device M comprises an NFC communication module configured to establish contactless communication and in particular to transmit a pulsed signal intended to be received by the body of the user and in return to receive an electromagnetic signal that has passed through and been re-transmitted by the body of the user. This module generally comprises an NFC Antenna (ANT) adapted to receive signals through a radio channel and via the human body, so that modulated electrical signals transmitted by the body of the user can be received by an antenna in a terminal located in the vicinity of the human body and by a demodulator, not shown, intended to receive the modulated electrical signals via the antenna and to convert them into digital signals intended to be transmitted to a processing unit, a Controller (CLF) and software components (firmware, etc.) necessary to implement IBC communication.

The memory MEM is configured to store a list or basis of signatures (USERS). According to a particular embodiment of the invention, such a basis associates an identifier with a signature to identify a user from a number of users.

According to a particular embodiment of the invention, the terminal M comprises a communication module COM configured to establish communication with the IP and/or the mobile network to carry out the transaction.

Fig. 3 illustrates the steps of a signature learning method in accordance with certain embodiments of the present invention.

The pulse signal (D) is repeatedly transmitted by the terminal and learning is completed by recovering the response signal that has passed through the body of the user. For example, a user may be located in a shop of a telecommunications carrier and is preparing to save his or her characteristic print, which will be his or her reference signature, and thereafter he or she will be able to use the characteristic print in his or her IBC service.

It is assumed that all prerequisites necessary for the learning method have been executed during the initialization step E0, such as for example loading and starting an application responsible for implementing the learning method on the terminal, and that the terminal is equipped with an NFC/IBC antenna with a suitable amplification.

In step E1, the terminal transmits a pulsed signal (a (f)) suitable for transmission via the human body, and this pulsed signal may advantageously be of the NFC type (13.56MHz), but may even have any suitable frequency. For example, frequencies around 10MHz are also suitable for the human body. The user's body receives the waves and transforms them into an antenna, that is, it re-emits electromagnetic waves.

In step E2, the terminal receives an impulse response r (i) from the user's body (where i represents a signal index in the plurality of received signals); to this end, the mobile device is positioned (e.g., by an application) in an electromagnetic wave reception mode. It should be remembered that the antenna must be able to receive signals with sufficient gain. The terminal receives and demodulates the received signal.

Then, in step E3, the terminal performs sampling and digitization of the signal and formation of an impulse response; at the end of this processing, the processed signal R' (i) (represented in time or frequency, or both, etc.) carries the characteristic value of the signal received as an echo. The index i corresponds to the iteration counter and may take a value between 1 and N and be incremented in each new iteration of the learning algorithm.

The terminal then stores the signal in a memory (represented here, for example, in the form of a database (6)) in step E4. Alternatively, the terminal may also transmit the signal R' (i) to an external learning server.

Step E5 corresponds to a test of the number of iterations N; as long as it has not reached the desired number of iterations, the mobile device retransmits the impulse response (step E1) and receives a new signal (step E2), which it processes and then stores with the other signal R' (i) (step E3). For example, the counter N is set to 3 and the three valid signals R ' (1), R ' (2) and R ' (3) must be received and saved (at least temporarily in the working memory). When the desired number of iterations is reached, step E5 will be followed by a step E6 of calculating a signature (reference signal). It should be noted that the number of iterations N may be predefined (e.g., N-3) or defined by the algorithm itself: for example, if the signals R' (i) differ too much from each other, the number N may be increased according to statistical criteria (standard deviation, variance, etc.); the number of times can be reduced if the signals R' (i) are very close.

According to one example, the following algorithm may be used:

-acquiring two signals R (1) and R (2) in response to pulses emitted by the mobile device and processing these signals;

-calculating the distance (possibly corrected) between the two signals. Such calculation examples are conventional to those skilled in the art of signal processing: for example, the euclidean distance between two first signals may be calculated. It is also known practice to compare two signals with each other (whether they are analog or digital signals) to determine a correlation function between the two signals, and to check whether there is identity between the signals according to the value of the function. For this purpose, a digital correlation function calculation method is generally used.

-if the distance is below a certain threshold, calculating the signature, otherwise acquiring the third signal R (3) and calculating the distance between the three curves, or calculating a second distance between the third signal and the mean of the first two signals, or calculating the distance between each curve and the statistical mean of the three curves, etc.

-and so on.

According to another example, a neural network may be used, as described in the article "authentication et identity de visage bands des Sur des Ondeletes et les Seaux de Neurones" (Facial authentication and authentication based on wavelets and neural networks) by M.BELAHCENE-BENATIA Mbarka (Revue science de mati rieuraux, Laboratoire LARHS N ° 02, 9 months 2014, pages 01 to 08). The method described on the basis of the conversion of a two-dimensional image of a face into a vector of size N obtained by linking together the rows (or columns) of the corresponding images, and then establishing a covariance matrix between the different images, can be easily adapted to the samples of the digital signal obtained from the signal r (i).

In step E6, the learning program calculates the signature SIG from all (N) signals r (i) received. Any method for obtaining one signal representative of N signals within the scope of the skilled person may be used, for example: averaging, or learning to identify the user using a (convolutional) neural network, or classifying the received signals by placing them in subsets corresponding to the user's signals using an SVM (support vector machine) system, etc.

The signal may typically take the form of an analog or digital signal, that is, a function representing the variation of the signal over a time interval (e.g., a few seconds) corresponding to an average response from the user. Such signals are represented by way of example in fig. 5. Alternatively, the signature may take the form of any set of data characteristics of the signal generated by the user's movements, as explained above, depending on both the biometric characteristics of the user and the characteristics of the terminal: a digital data set; the indices in the existing signature dictionary, for example, correspond to the classification of the system user (according to their age, gender, body shape, etc.).

The signature thus calculated is preferably stored in a memory, or in a database (5), or in the mobile device, or in an external database, together with the user's identifier, such as his or her name, date of birth, telephone number, MAC address of his or her terminal, his or her bank account number, etc.

Fig. 4 illustrates the steps of an identification method according to a specific embodiment of the present invention.

According to this embodiment, the user wants to authenticate with his or her mobile terminal, for example, in order to unlock the mobile terminal or to perform a transaction, for example, of a currency type. Without loss of generality, it is assumed here that the goal is to unlock the terminal or smartphone. The user grasps the smartphone, or at least approaches the smartphone, and then proceeds with the automatic authentication method without he or she having to enter his or her fingerprint or enter a PIN code, or present his or her face to the camera of the smartphone, or the like.

It is assumed that the learning phase described previously on the basis of fig. 3 has been performed and that the signature of the user is located on the mobile terminal (it should be remembered that the signature can be located elsewhere, in a database external to the terminal).

Step E' 0 is identical to step E0 described previously for the learning method, except for the application programs that have to be executed: here an identification application. This application may be initiated by the user, or run in the background, or when a grip on or proximity to the mobile device is detected (e.g., using a gyroscope, accelerometer, sensor, etc.), or may even be initiated or cycled/periodically by the user's action (pressing an unlock button or screen).

The steps E1, E2, E3 are identical to the corresponding steps described previously on the basis of fig. 3.

In step E7, the terminal accesses a memory (or database) to read the user's "signature" (characteristic impulse response). If the accessed memory is not located on the terminal, but on, for example, a server, a communication channel may be opened using the network for accessing the memory.

In a comparison step E8, the received and processed signal (R/R') is compared with the Signature (SIG) of the user. This makes it possible to check that the user who is indeed the terminal is gripping the terminal, in other words this step performs the identification (authentication or authentication) of the carrier. Several types of comparisons may be performed, such as, for example:

-calculating the "distance" between the received signal (R') and the signatures stored in the database. If the distance between the two signals is below a threshold, the user is authenticated. For example, a point-to-point correlation can be performed on the two signals (the received candidate signal and the signal corresponding to the signature) by calculating the difference of the respective values of the two curves, possibly by moving the received signal onto a reference signal. If the two curves are very similar, the minimum obtained should be close to zero.

Wavelet method with decomposition and analysis of time-frequency matching;

using (standard or convolutional) neural networks, where the multi-layer neuron system learns to identify humans from the obtained signals by adjusting internal weights. Such methods are known and can be accessed via standard signal processing software such as MATLAB.

At the end of this comparison step, if the received signal corresponds to a signature, the user is authenticated and step E8 may be followed by step E9 of conducting an unlocking and/or transaction (e.g., verifying payment). Otherwise, that is, if the impulse response R/R' does not correspond to the signature, it is possible to return to step E1 and retransmit the impulse signal, for example. According to a variant, a predefined number of retransmissions (e.g. 2) may be authorized before canceling the transaction.

Step E9 may also include the substeps of updating the signature: the signal R' obtained after processing can be used to modify the signature of the user, which may vary, inter alia, for biological reasons (he or she gets older, gains weight, etc.) or for mechanical reasons (he or she has a different way of dealing with the mobile device).

Fig. 5 shows an example of a received impulse response for a given user. The top curve of fig. 5 represents the time signal corresponding to a pulse. Such a signal is naturally given as an example, the pulses in particular being able to have a shorter duration without loss of generality. The bottom curve represents the time response to a given human user when the body has re-emitted the pulse signal. The two waves can add up cleanly. The difference (frequency, amplitude, envelope, etc.) between the two curves is the transmission characteristic of the signal within the human body.

It goes without saying that the embodiments that have been described above have been given by way of indication and not limitation only, and that numerous modifications can be easily made by those skilled in the art without in any way departing from the framework of the invention.