RU2817264C9 - Method of protecting electronic multifunctional mobile device from unauthorized access - Google Patents

Abstract

FIELD: physics.

SUBSTANCE: invention relates to mobile device user authentication. Technical result is achieved due to the continuity and multifactorial nature of the user authentication process. After activation of the mobile device by means of the primary identifier, binding to a specific authorized user is performed, teaching the device to recognize various user-related features, for example biometric, by means of the RFID tag located with the user, the distance between the user and a mobile device is cyclically recorded based on which a value of the probability of finding the device in physical possession of an authorized user is determined. This value is compared with a priori specified threshold values of probability, and depending on the result of comparison, the user is allowed to all functions of the device, or to some functions, or access is prohibited.

EFFECT: high reliability of authenticating a user of a mobile device.

16 cl, 1 dwg

Description

FIELD OF TECHNOLOGY TO WHICH THE INVENTION RELATES

The invention relates to the field of user authentication of an electronic multifunctional mobile device (EMMU), for example, a mobile phone (smartphone), tablet, laptop, and is intended to protect the EMMU from unauthorized access (including due to loss or theft) to the contents of the EMMU (stored in the EMMU information), to the functions implemented on the EMMU, connected applications and services through the implementation of continuous multi-factor user authentication during the use of the EMMU for its intended purpose, which does not interfere with this use.

BACKGROUND OF THE ART

There are known methods for authenticating users of electronic devices using a password, in particular, in the form of a PIN code or a graphic pattern. Including those with increased security capabilities, reducing the likelihood of password compromise when using devices in public places (for example, equipped with CCTV cameras). Thus, there is a known method for authenticating the user of an electronic device with a touch screen, limiting access to it by unauthorized users (patent US 6549194 B1 “Method for securely entering a PIN code on a touch screen”, application US09/410397 dated 10/01/1999), based on the introduction of the well-known to an authorized user of a PIN code using a randomly assigned numeric keypad layout each time an electronic device with a touch screen is entered, which reduces the likelihood of compromising the PIN code (by an outside observer remembering the trajectory of the finger as it is inserted and/or recording biological traces of the finger interaction with the screen of an electronic device). However, to reduce the likelihood of compromise, the numeric or symbolic password used must be quite complex and therefore may be difficult to remember, especially considering that the touch point of each character in this method, each time the device is unlocked, due to the formation of the numeric keypad, also changes randomly. , which makes memorization even more difficult (muscle memory does not work). The proposed method also provides for the possibility of using a PIN code as one of the options for the primary identifier, but without using a virtual keyboard that makes it difficult to enter.

The method of user authentication by applying graphic designs to the touch screen of a device, proposed in patent RU 2589397 C2 “Authentication of graphic gestures” (application No. 2013152162/08 dated 10/09/2011), has great security capabilities and ease of use, in which a personalized digital image ( on top of it) on the touch screen of the display, a custom pattern is applied (using the user’s finger, stylus, or computer mouse) - a sequence of touch points or a user gesture that combines the selection of a set of positions with movement (in the latter case, the entire trajectory of movement is recorded) with subsequent comparison of the applied pattern with drawings stored in the device memory. However, remembering touch points (trajectories of movement), despite the presence of some tolerances for the accuracy of reproduction of drawings, can be difficult for users, especially the elderly. However, the use of a primary identifier in the form of a graphic design is also possible in the claimed method.

There are known methods for user authentication using biometric user characteristics. In the invention “Method and device for unlocking with a fingerprint” (patent RU 2721361 C1, application 2019104951 dated August 16, 2018), unlocking an electronic device is carried out based on obtaining and comparing the user’s fingerprint with its reference fingerprint stored in the memory of the electronic device. In this case, to reduce energy consumption due to the user’s unintentional touch of the device’s touch screen, the pressure sensor is activated when the user’s finger is in the established touch zone and the force of his pressing on the screen exceeds a specified threshold. Using a fingerprint for user authentication (if its reference image is stored only in a protected electronic device) is also possible in the claimed method.

Another invention using biometric characteristics of the user for authentication is the invention “User authentication by combining speaker identification and reverse Turing test” (patent RU 2406163-C2, application 20005133725/08 dated November 1, 2005), in which to increase protection against unauthorized access It is proposed in the process of user authentication by voice to analyze not only the belonging of a voice fragment to an authorized user (in this case, an unauthorized user can gain access using a previously copied voice fragment of an authorized user), but also the correspondence of the content of this voice fragment as a response to a randomly given one question regarding user-related confidential information. Both of these features - user identification by voice and knowledge of confidential information - are also used in the claimed method.

Methods of biometric user authentication using several characteristics, including dynamic ones (in the form of a process changing in space-time), are becoming increasingly widespread. The invention “Access control and management system based on biometric technologies for personal authentication by voice and face” (patent RU 2638775 C2, application 2013111904 dated March 18, 2013) proposes using voice phonograms and facial video recordings together for identity authentication. The use of dynamic features, as opposed to static ones, makes it possible to increase the accuracy of recognizing an authorized user. The idea of using several and also dynamic features is important and is implemented in the proposed method.

The invention “Interactive method of biometric user authentication” (patent RU 2670648 C1, application 2017144202 dated December 18, 2017) proposes for authentication to use an interactive method for recording the dynamics of changes in the user’s biometric characteristics (heart rate, pulse wave, motility of the pupils of the eyes, diameter of the pupils of the eyes ) in response to the arrival of a randomly generated sensory stimulus from the outside, which at an unconscious level causes an individual physiological reaction leading to a change in the above biometric characteristics of the user, the dynamics of which are recorded, which, in the author’s opinion, increases the reliability of authentication in the presence of attacks by intruders who have become aware of certain “static” biometric characteristics of the user. However, it should be noted that the complexity and low reliability of receiving and, most importantly, processing such sensory signals. The use of individual physiological characteristics of the user (for example, pulse parameters, heart rate) for user authentication is also provided in the proposed method, however, it is them, and not the dynamics of their changes as a result of a specially created unconscious sensory influence. The dynamics of the electrical activity of the heart and the parameters of the user’s pulse are used in the inventive method, not only to authenticate the user, but also to recognize his physiological state for the purpose of subsequent blocking of the electronic device (to prevent the possibility of its use by others) if these indicators exceed acceptable limits , indicating the loss (possibility of loss) by the user of control over ownership of the device: fainting (pre-fainting) state, state of severe stress (including in connection with threats against oneself, etc.).

In the invention “Device and method for establishing or removing protection on content” (patent RU 2660617 C1, application 2017106247 dated July 28, 2015), various (at least two) are used to increase data security (to protect content stored on an electronic device from unauthorized users) biometric features of an authorized user (for example, fingerprint and iris pattern), samples of which are pre-recorded in an electronic device and read through additional devices worn by the user (for example, watches with a fingerprint sensor and glasses for recording the iris pattern) , and these features can be used alternately (one to protect content, the other to remove protection) or simultaneously with the possibility of replacing (if necessary) one feature with another (if it is impossible to use any of them). However, the actions described in the patent, when protection is placed on the content using one biometric feature, is removed using another, or when, after successful authentication using a biometric feature, the user installs additional protection on the content in the form of a general password or a password in relation to a specific type of used biometric feature, encrypts the content using the first encryption key obtained after successful authentication using the first biometric feature, and then encrypts the first encryption key using the second encryption key based on the use of the second biometric feature, etc., of course, increases data security , but their implementation requires additional actions distracting from the use of the electronic device for its intended purpose, which are not always convenient given the current environment. However, the method we claim also provides for the possibility of using additional electronic devices located with the user (for example, built into the user’s accessories: watches, glasses, bracelet, etc.) to register and process various biometric characteristics of the user with transmission of the result via a radio channel to a protected electronic device. device (if possible, processing can be carried out directly in the protected electronic device).

However, the user authentication used in all of the above known methods (including using his biometric characteristics) is used only at the beginning of working with a mobile device (we will further consider a smartphone as an example). After successfully passing such initial authentication, the user becomes available to most of the functions and applications implemented by the smartphone, he gets access to information stored in it, for example, phone numbers (with the ability to make calls), notes in the calendar, photos, correspondence in instant messengers. If the smartphone, after the user has successfully completed the initial authentication, is lost or stolen (stolen or taken away), then the unauthorized user gains access to this information and functions (for example, they can make a call to a paid number). Of course, access to the most dangerous functions, in particular those related to financial transactions, the use of personal data to receive or send official documents in modern smartphones, as a rule, is closed by a mandatory additional authentication process (including two-factor, using feedback) . At the user's request, it is possible to independently set up additional protection (for example, password protection) for other functions (applications). However, traditional additional authentication not only requires time to set it up (for which reason, if it depends on the will of the user, it is often postponed until “later”, which, unfortunately, most often comes too late) and subsequent passage, but also (and this most importantly) does not solve the problem of unauthorized access by an attacker to a smartphone after it has been passed by an authorized user (in particular, in the event of an unlocked smartphone being taken away). In addition, known methods of authentication based on biometric characteristics do not take into account the conditions of use of an electronic device (external environmental parameters: temperature, illumination, noise, vibration, etc.), which significantly affect the quality (probability) of recognition of a particular characteristic.

The possibility of password compromise should also be taken into account. The PIN may become known to another person as a result of theft or brute force. It's the same with a graphic design. Moreover, both of them, as an analysis of publications on this topic shows, are partially predictable. Biometric parameters can be stolen from various external databases. And fingerprints can be obtained by removing them from objects held in the hands of a legitimate user.

With biometric authentication, the presence of limiting conditions should also be taken into account. For example, the success of using a fingerprint depends on its cleanliness and degree of dryness for any of the removal methods (optical, capacitive, ultrasonic), as well as on the cleanliness of the sensor screen, which makes it inappropriate to rely only on using a fingerprint. Using the iris pattern is technically difficult to implement. Facial recognition requires the use of a 3D version, since the flat version can easily be faked using photography. So it is no coincidence that information security experts advise against trying to universally use biometric parameters stored in external databases. Biometric features cannot be lost physically (they are always with you), but they can be lost actually (they will become known to unauthorized persons), and, unlike the PIN code, which can be changed if desired, a fingerprint and other biometric indicators, as a rule, remain unchanged in course of life or weakly changeable (age-related changes). In addition, an attacker can use some of the authorized user's biometric attributes (for example, fingerprints) by taking advantage of the user's unconscious state or inducing him into this state.

Note that additional traditional authentication as a protection mechanism, in addition to the above-mentioned inconvenience and loss of time, may also carry an additional risk of compromise. Most likely, in order not to forget or get confused, the user will use the same identification sign, for example, the same PIN code, on all services (applications). Otherwise, he will have to store their list and information about which code belongs to what, in memory (which is clearly difficult and unlikely for most users) or on a separate medium (for example, in a notebook - paper or smartphone), which is more likely, but and more dangerous (the likelihood of compromise increases) and inconvenient (waste of time to find and enter).

The current use as a security measure of a time interval after which an unused smartphone is locked, so that to unlock it again you need to go through primary authentication, does not completely solve the problem of unauthorized access. It is hardly advisable to set a very short period of non-use time required to reduce the risk of use of a smartphone by an unauthorized person to an acceptable level, for example, one minute or even less (with a longer period of time, the risk of unauthorized use if the device is lost, stolen or taken away increases significantly), since this extremely inconvenient. Known options for using “smart locks” (for example, taking into account the statistics of user visits to certain places, disabling temporary blocking when the smartphone user is at home) do not provide the required high level of protection against unauthorized use.

Therefore, the most promising direction for protecting electronic devices from unauthorized access is the use of continuous authentication - authentication carried out not only at the beginning, but throughout the entire time of user interaction with the electronic device. Obviously, the term “continuous authentication” in technical terms is very conditional, because any “continuity” in technical execution is always discrete, since it is realized through periodic (cyclic) authentication. But since this is carried out with such a short cycle duration that it is impossible to perform any significant actions with the EMMU during this period (for example, every few seconds), then in a consumer sense we can talk about the continuity of authentication. Such authentication, on the one hand, must be effective (should reduce to a safe level the probability of using the electronic device by an unauthorized user), and on the other hand, be as secretive as possible and not require the user to perform inconvenient actions, and not distract him from using the electronic device for its intended purpose. It is hardly advisable, for example, every 15...20 seconds to force the user to enter a PIN code or fingerprint, pronounce a code word, draw something or sign on the screen, etc. It is also desirable that the implementation of such continuous authentication would not require significant design modifications EMMU, i.e. would be based on a standard set of sensors (sensors), which are found, for example, in modern smartphones and would not significantly increase the power consumption of the smartphone and would not increase the requirements for its computing resources. In addition, when implementing such continuous secretive authentication, it is advisable to minimize (even to the point of eliminating) access to external (in relation to the EMMU and the user) databases for checking (comparing) the user’s identification characteristics and, if possible, refuse to use those characteristics of the EMMU user , which are already stored in certain external databases and can be compromised. In addition, access to external databases can be blocked intentionally or unintentionally (due to a targeted attack on the communication channel or a failure in its operation), which precludes the use of the electronic control unit for its intended purpose. It is advisable to use features that are formed during the user’s interaction with the EMMU and are not sent for comparison and confirmation to external databases (for example, in some cases it may be useful to use the above-mentioned individual dynamics of the user’s unconscious reactions to a particular irritating sensory signal, but this signal is more appropriate receive not from an external authentication center, as proposed in patent RU 2670648 C1, but from an additional device located with the user or directly from the EMMU).

It should also be noted that in changing environmental conditions (lighting, temperature, precipitation, noise from equipment, the presence of other people nearby, etc.), taking into account the current tasks and actions of the EMMU user, it is advisable to make continuous authentication multifactorial with the possibility of adaptive use taking into account the current situation the most convenient and informative sensory signals of various physical natures, characterizing the user and the process of his interaction with the EMMU. These sensory signals can reflect the biometric characteristics of the user (including dynamic ones: electrocardiogram signals, electroencephalograms, keyboard handwriting, gait, etc.) or characteristics of an object owned by the legitimate user, which plays the role of a special key. Such a key can be used, for example, an RFID tag or any object that is in the user’s possession and a priori converted into a password symbol. A photograph of such an item in the user’s possession, for example, a wristwatch, taken at a certain angle from a certain distance using the camera of a protected smartphone can serve as confirmation of the legality of the user’s ownership of the smartphone. Signs can also be knowledge that the user has (this could be, for example, a password, PIN code). A combination of features is possible (for example, when using a voice password, not only the spoken password word can be recognized, but also the user can be identified by voice). Moreover, it is advisable to select the characteristics used (one or more) and assign the threshold value of the probability of their belonging to an authorized user of the electronic control device for authentication, as noted above, taking into account the actions performed by the user and the current environment, i.e. adaptively. Moreover, to obtain these signs, in addition to the standard set of EMMU sensors (for a smartphone this is: accelerometer, gyroscope, magnetometer, camera, microphone, barometer, hygrometer, positioning system (GPS), light and proximity sensors, etc.), sensors associated with EMMU can be used (for example, via radio channel) additional electronic devices worn by an authorized user (in clothing pockets, attached to clothing or sewn into clothing, built into personal items: watches, glasses, bracelets, jewelry and other accessories) and allowing, for example, to control the current distance to the EMMU, as well as register and recognize the user’s biometric signs or transfer these signs to the EMMU for recognition (pulse parameters, movement dynamics, etc.). It is precisely this integrated approach, described in the current and previous paragraphs, that is implemented in the proposed method.

The patent “Method for protecting a mobile phone and other items from theft” (utility model patent RU 67306 U1, application No. 2006106043/09 dated February 26, 2006) attempts to implement the idea of continuously monitoring the ownership of an authorized user’s mobile phone using an RFID tag embedded into items worn by the user along with the phone (clothes, bracelet, watch strap, ring, purse, bag, ...). This tag periodically independently emits electromagnetic signals into the surrounding space (active tag) or re-radiates (reflects) electromagnetic signals received from the mobile phone emitting them (passive tag). The transceiver device of the mobile phone searches and detects electromagnetic signals emitted or re-emitted by the tag. If signals emanating from the tag are detected, a decision is made that the mobile phone is at an acceptable distance from the user (3...10 meters) and continues to operate in normal mode while receiving signals from the tag. If signals from the tag are not detected due to the phone being located further than the permissible distance from the user, then a decision is made that the mobile phone is stolen, and the operation goes into emergency mode. In this mode, the phone sends a command to turn on the light and (or) sound alarm and vibration alarm on it and on the electronic device located next to the tag, which attracts the attention of the user and people nearby, and a signal about the incident is sent to the telecom operator to block the phone. To prevent an attacker from disabling a stolen phone that is still in the area of the RFID tag, a PIN code is provided to disable the phone. The function of an RFID tag can be performed by a wireless Bluetooth headset, which sends a signal to turn on the alarm when the range of the headset is exceeded. However, recognizing the advisability of using the idea of electronic communication (for example, via a radio channel) of a mobile phone with other electronic items located on the user’s person to control the distance between the mobile phone and the user, it should be noted that only this method of protection against unauthorized use of a mobile phone is insufficient. The described method, in particular, does not protect against the use of a mobile phone by an attacker who has stolen the phone, but is secretly close to its owner (for example, when, when stealing a phone in a cafe, the attacker is seated at the next table). Access to the phone and its functions is not blocked even when the phone is taken away using force (threat of force), when the attacker, having taken the phone, places himself next to the owner (within the “visibility” of the electronic RFID tag) and uses the phone and its functions in his own way. discretion. However, this method partially reflects the features present in the proposed method. The common features of the proposed method with this known method are the placement on the user (on the body, on (in) clothes, in accessories) of a special electronic device (in particular, RFID tags), via a radio channel connected to the EMMU (in analogy - with a mobile phone) , to control the distance between the user and the EMMU and is capable of emitting signals that attract the attention of the user and other persons when the distance between the user and the EMMU exceeds a predetermined threshold value.

The closest to the claimed method is the method of electronically linking a user to a mobile device, described in the patent “BIO binding for user authentication” (RU 2690214 C2, application 2016143181 dated March 30, 2015). In this case, electronic binding means the biological presence of the user within a certain distance from the mobile device, which is taken to mean that the mobile device is in the physical possession of the user (from the case when the mobile device is directly in the hands of the user, in the pocket of his clothing, in the user’s bag, etc. until the mobile device is within “line of sight”). As a technical result, the invention claims to increase the efficiency of authentication of the user of a mobile device - minimizing the need for re-authentication of the user of a mobile device remaining in his possession, using a primary identifier, which can be used: password, passphrase, gesture, digital certificate, PIN code and/or biometric identifier (for example, retinal pattern, fingerprints, iris pattern, etc.), in case of exit from the authentication state (for example, when the period of non-use of the mobile device exceeds a specified period of time or if you press the shutdown button or switch to another application). This is achieved through periodic monitoring (after the user has successfully completed initial authentication to use the mobile device) of the user's biological presence within a certain distance from the mobile device and, based on this monitoring, calculating the probability of the mobile device being in the user's physical possession and comparing it with a given threshold value, exceeding which confirms that the mobile device is in the physical possession of the user. If the calculated probability is less than the threshold value, then the user must re-authenticate using the primary identifier to further interact with the mobile device or specific functions and services of the mobile device. The magnitude of the physical distance between the user and the mobile device is determined, judging by the description of the patent, by processing information from the accelerometer and gyroscope (apparently, we are talking about solving the classical problem of inertial navigation) or from a positioning sensor, as well as by finding the difference between the values of the emitted powers and received after reflection from the user’s body of radio signals (radar) and (or) sound signals (acoustic location) and its comparison with a given threshold value. To make a decision about whether a mobile device is in the physical possession of the user, this difference obviously must be less than a given threshold value, which indicates the stability of the bio-reference. In addition, as follows from the description, after passing the initial authentication, in some cases, control of the presence of the required bio-link can be carried out by registering the user’s image signal (face, appearance, clothing), “registered at the time the bio-link was created.” In this case, it is possible to a priori introduce not one, but several threshold values for the probability of a mobile device being in the physical possession of the user, exceeding the calculated probability of each of which allows the user access only to a certain set of actions and functions (obviously, the larger the threshold value exceeds the calculated probability , the more and more important actions (functions) are allowed access). It is also possible for the bio-link to gradually (over time) weaken in the event of “a break or weakening over a period of time” of the “line of sight of radio communications”. However, "at any time, successful primary identification can restore the bio-reference to its original maximum stability and state."

In accordance with the claims, this method contains the steps of:

“- authenticate the user on the mobile device at the first time and, based on this, authenticate the user to use the set of functions on the mobile device;

- capturing first measurement data regarding device ownership through the mobile device and, based on an evaluation of the first measurement data regarding device ownership, determining that the mobile device is in physical possession of the user;

- capture movement data using a movement sensor or positioning sensor of a mobile device;

- determining that the movement data exceeds a threshold and has resulted in a triggering event, and, in response thereto, causing the mobile device to capture second device ownership measurement data containing radio signal data, audio wave data, image data or video data;

- determining, based at least in part on an assessment of the second device ownership measurement data captured by the mobile device, a value representing the probability that the device remains in the physical possession of the user;

- at a second, later point in time, while the user is still authenticated on the mobile device, upon determining that said value is greater than the first threshold value and less than the second threshold value, disabling a subset of said functions on the mobile device in respect of which the user was previously authenticated at the first time;

- and upon determining that said value is less than the first threshold value, requiring the user to re-authenticate on the mobile device.”

In this case: - “the user is authenticated based on at least one of a password, a passphrase, a gesture, a digital certificate and a biometric identifier”; - the first and second measurement data “contain a plurality of radio signal readings”, based on the difference between which the fact of movement of the mobile device is determined; - based on the readings of the movement or positioning sensor, the correspondence of the movement of the mobile device to the known movement pattern of the user is determined; - capture the audio signal to determine, based on a combination of audio location and voice recognition, whether the user is moving and “whether the user is within a threshold distance from the mobile device”; - the first and second measurement data comprise images obtained by a camera associated with the mobile device, at least one of which indicates that the camera has a direct line of sight to the user.

The actions proposed in the prototype method are, in general, appropriate, since, firstly, they can significantly reduce the likelihood of unauthorized access to a mobile device compared to other methods discussed above, in which only primary authentication is carried out at the very beginning of the user’s interaction with the mobile device, and, secondly, they eliminate the need to periodically re-enter the primary identifier (for example, after a specified period of non-use of a mobile device carried by the user). However, these actions are not enough, and their sequence is not optimal to achieve the greatest protection against unauthorized use of a mobile device. In addition, judging by the description, when “capturing measurement data”, environmental conditions that affect the quality of recognition are not taken into account. In general, the method has the following disadvantages:

- firstly, the disadvantage is the fact that “the capture of the first measurement data regarding ownership of the device through a mobile device”, based on the assessment of which “it is established that the mobile device is in the physical possession of the user”, occurs not before, but after the start of using the mobile devices. Meanwhile, it is almost obvious that this process of “capturing” the user’s biometric data - in fact, the process of teaching a mobile device to recognize a specific user, should be carried out as a separate stage before using the mobile device, using a special method and in a safe place, obtaining the required volume statistical data on various reliable sensory biometric features: voice (timbre, including in different emotional states of the user), photos and videos (face, clothing, accessories, gestures), biometrics (ECG, pulse, pattern of the iris or retina, ... ), behavior (user's gait, the sequence of usual user actions after passing the initial authentication, taking into account their temporal characteristics, the dynamics of using the touch screen when typing, moving from one application to another, etc.). This early learning process before practical use is the most effective and is currently carried out on all image recognition systems (in particular, using deep learning neural networks). At the same time, the sample on which training takes place must be complete and representative, otherwise gross errors are possible during the work process, so the learning process - a rather long and responsible process - cannot begin and take place during the actual use of the mobile device for its intended purpose with support only for those learning examples that the sensors can record. While using a mobile device for its intended purpose, it can (and should!) only undergo additional training (with the addition of recognition experience) based on registration and, if necessary, correction of information about certain biometric, behavioral characteristics of the user (due to age-related changes, changes due to injuries, illness, etc.), on which the mobile device was trained. While using the device during additional training, new features can also be registered that were not used during initial training, but belong to the authorized user and can be useful for subsequent user authentication. That is, the required recognition efficiency can be achieved only after a certain training time, but not at the beginning of the process of using a mobile device. But not only the ability to make gross mistakes is a consequence of such learning “on the fly.” More importantly, for some time after the start of such training in the process of collecting necessary-sufficient statistics (“capturing the first measurement data regarding ownership of a mobile device”), after passing the initial authentication, the mobile device is practically not protected from the possibility of its unauthorized use by another user in the event loss of a mobile device, its theft or seizure. Moreover, an unauthorized user can not only use a mobile device, but also further train (essentially, train) it to suit himself, thus becoming an “authorized user” with all the ensuing possible consequences;

- secondly, using the movement of a mobile device as a triggering event for “capturing the second data of changes regarding the ownership of a mobile device” is not sufficient to protect the mobile device from unauthorized use, because the mobile device may not move anywhere, but may already be in the hands of the person who stole it (or who took it away) and an unauthorized user nearby. The sound or radar sensors used in the prototype, capable of determining the proximity of a human biological body to a mobile device, are unlikely (judging by the description) to distinguish the body of an authorized user from an unauthorized one. Capturing a user's video image or voice to recognize identifying characteristics may also be insufficient if an authorized user is forced to carry out the commands of an unauthorized user under the threat of force or as a result of the use of force. It seems that in this case the mobile device should respond not only to the proximity of the biological body or the clothing, hairstyle, skin tone of the user (as proposed in the prototype), but also to the psychophysiological state of the user (for example, pulse parameters, blood pressure, ...), indicating the criticality of the situation in which he is located, with automatic activation of blocking of the mobile device and notification of persons (organizations) associated with the user. Blocking in this situation can also be carried out by a predetermined code word or gesture of an authorized user;

- thirdly, the movement of a mobile device exceeding a threshold distance, which should lead, judging by the description of the invention, to “the capture of the second measurement data regarding ownership of the device by the mobile device,” can occur immediately after the user has passed the initial authentication, due to which “the capture of the first measurement data" may not take place, and the movement itself may be a consequence of the theft of a mobile device, so the "capture of second measurement data" in this case will most likely be associated with the legalization of the use of a mobile device by an attacker;

- fourthly, the use of sound location and (or) radar to determine the degree of physical proximity of the mobile device and the user leads not only to additional energy consumption of the power source, which is important, but also (and this is the main thing) to the need to have separate devices - generators and receivers of sound waves and (or) radio waves for the purpose of monitoring the location of a biological body near a mobile device (but it should be noted that this body may not be the body of an authorized user);

- fifthly, it is almost obvious that the quality of recognition of a particular feature depends on the environmental conditions in which this recognition occurs (lighting, noise, vibration, precipitation,...), and on the technical capabilities of the mobile device used. However, the invention does not take this into account;

- sixthly, the desire to cover as many cases as possible with the invention leads to very original, but hardly interpreted as the author of the invention wants, situations, for example, when the presence of a mobile device “within line of sight” is declared to be a situation where the mobile device is “in user’s possession” (as a comment, let’s imagine a situation that is not very good for an authorized user, but a very real one, when a person who has stolen a mobile device, having run away from the authorized user of the mobile device at a sufficiently large, safe distance, turns to him and, laughing, shows the stolen mobile device , which is “in line of sight” for an authorized user).

We note that, unfortunately, in the text of the description there are ideas that are not reflected in the claims. For example, a very useful idea is to introduce several threshold levels for the probability of a mobile device being in the user’s physical possession, differing in the level of access. Or another idea of weakening the bio-link over time (for example, according to a linear law) in the event of a break or weakening of the “line of sight of the bio-link radio communication”. True, what “direct visibility of bio-linking radio communication” is, as well as the “degree of its disruption” indicated in the description, is unclear. Nevertheless, we took into account the idea of multi-level thresholds as a feature available in the prototype and is also present in the claimed method.

What the claimed method has in common with the prototype is the presence of stages in which:

- setting two threshold values for the probability of an electronic multifunctional mobile device being in the physical possession of an authorized user and a threshold value of the distance between the electronic multifunctional mobile device and the user, the excess of which means that the electronic multifunctional mobile device is no longer in the physical possession of an authorized user;

- authenticate the user using a primary identifier (password, key phrase, PIN code, gesture) and, if successful, consider the user authorized and allow him to access all functions (applications, services) of the electronic multifunctional mobile device;

In the process of using an electronic multifunctional mobile device by the user for its intended purpose, the user is continuously authenticated, for which, through sensors of the electronic multifunctional mobile device and sensors of additional electronic devices located with the user, signs associated with the user are periodically recorded (photo and video images of the user and voice signals emitted by him, as well as signals carrying information: about the dynamics of user behavior, about the movements of an electronic multifunctional mobile device, about the current distance between an electronic multifunctional mobile device and the user), based on the processing of information about which the value of the current probability of finding an electronic multifunctional mobile device in the physical possession of an authorized person is determined user;

- compare the obtained value of the current probability of finding an electronic multifunctional mobile device in the physical possession of an authorized user with two specified threshold probability values and, depending on the result of the comparison, make one of three decisions: firstly, the decision that the electronic multifunctional mobile device is in in the physical possession of an authorized user while maintaining the user’s access to all functions (applications, services) of an electronic multifunctional mobile device, if the resulting value of the current probability is greater than the first (larger) specified threshold probability value; secondly, a decision that the electronic multifunctional mobile device is no longer in the physical possession of the authorized user, blocking the electronic multifunctional mobile device and informing the user of the need to re-authenticate in order to gain access to all functions (applications, services) of the electronic multifunctional mobile device by means of the primary identifier, if the received current probability value is less than the second (smaller) specified probability threshold value; thirdly, a decision that the electronic multifunctional mobile device is in the physical possession of an authorized user, but with a probability insufficient for access to all functions (applications, services) of the electronic multifunctional mobile device, on the basis of which the user is prohibited, informing him about this , access to the most important (responsible, critical) functions (applications, services) of an electronic multifunctional mobile device, if the received current probability value is less than the first (larger) specified threshold probability value, but greater than the second (smaller) specified threshold probability value;

- at any time, through a new successful authentication by the user using the primary identifier, the user’s access to closed (due to the insufficient value of the current probability) functions (applications, services) of the electronic multifunctional mobile device is restored.

DISCLOSURE OF THE INVENTION

The technical result (goal) of the invention is to reduce the likelihood of unauthorized access to the contents and functions (applications, services) of the electronic device, including in the event of loss of the electronic device or its theft. An additional effect is to increase the likelihood of returning a lost (stolen, taken) electronic device to its authorized user or, if the device does not belong to an authorized user, to the owner (in particular, the organization) that issued the device to the user.

The technical result is achieved through the implementation, in the process of using the EMMU for its intended purpose, of continuous multi-factor authentication of the user in whose physical possession the used EMMU is located, in order to confirm that the EMMU is in the physical possession of an authorized (legal) user to whom the EMMU was previously linked, trained to recognize signs associated with that user. If the legality of the physical possession of the electronic device is not confirmed by the user, the electronic device is blocked with the issuance of warning signals (light, sound, vibration signals) to the authorized user on an additional electronic device located with him, connected via a radio channel to the electronic device, and also, if the user is not the owner of the electronic device, radio signals notifications to the owner (organization) that issued the EMMU to the user.

To do this, before using the electronic multifunctional device for its intended purpose in conditions that exclude access to the electronic device by unauthorized (illegal) users, set two threshold values for the probability of finding an electronic multifunctional mobile device in the physical possession of an authorized user P ₁ and P ₂ (P ₁ >P ₂ ) and threshold distance S _add. between the EMMU and the user, the excess of which means that the EMMU is no longer in the physical possession of the authorized user (see Fig. 1), after which the user is authenticated through a primary identifier, which is used as what was transferred to the user in the form of an item or knowledge along with the EMMU upon its receipt (purchase), for example, a password, PIN code, graphic design, hardware token. If authentication is successful, the user is considered authorized, is allowed to access all functions (applications, services) of the EMMU and the EMMU binding mode to him is activated (EMMU training mode). Further, in conditions that still exclude access to the electronic device by unauthorized users and are most consistent with the various real conditions of use of the electronic device for its intended purpose (external environmental conditions: illumination, precipitation, temperature, vibration, noise (wind, people, animals, machines), electromagnetic interference ), carry out this linking by teaching the EMMU to recognize an authorized user based on characteristics inherent to him (associated with him), for which purpose these signs are periodically recorded using EMMU sensors and sensors of additional electronic devices located with the user (on the arm, on the body, on the neck, in clothes, in clothing pockets, in accessories) and connected to the EMMU via a radio channel, and then, using representative training and test samples, through artificial intelligence technologies (in particular, using artificial neural networks of deep learning or using a more advanced, but also requiring large computing and energy resources, technology that combines the capabilities of modern deep learning in relation to artificial neural networks and the capabilities of an approach based on symbolic description, rules and Bayesian inference), implemented directly on the EMMU or on those located with the user and connected to the EMMU (via radio channel) separate additional electronic devices, process the registered characteristics, summarizing and storing them.

The binding is carried out for each user characteristic. The following may be recorded as signs:

- static and dynamic biometric signs of an authorized user (timbre and spectrum of voice, photo and video images of the face or its parts (ear, nose, lips), fingerprint, pattern of the iris and/or retina, pattern of blood vessels of the face (its parts) or vessels of the hand (palm), geometry of the hand, pulse parameters (frequency, rhythm, voltage), size (shape) and dynamics of the heart, electrical activity of the heart (electrocardiogram parameters), parameters of the electroencephalogram or its part, odor emitted by the user, chemical composition of the user’s exhaled breath air, dynamics of body movement when walking (gait), dynamics of hand movement with an electronic multifunctional mobile device, dynamics of a signature, dynamics of typing letters on the virtual keyboard of an electronic multifunctional mobile device, microvibration of fingers);

- secondly, signs of unique images specially created by the user, in which there are biometric characteristics of the user and (or) his knowledge of confidential information (in particular, gestures, poses, sequence of actions, code words or phrases, tune motifs);

- thirdly, signs of unique images of objects carried by the user: visual images, in particular, photographs of any object carried by the user (for example, his wristwatch), taken by the camera of an electronic multifunctional mobile device from a certain distance and angle. Since only the user knows about this (object, distance, angle), receiving such a photograph with an EMMU camera during its further use in real conditions of use means that the EMMU is in the possession of an authorized user. Signs of olfactory images (for example, a specific, not perceptible by a person, smell of any object located on the user) and sound images (for example, infra- or ultrasonic, inaudible to humans, low-intensity signals periodically emitted by electronic devices located on the user) can also be used. .

The number of examples in the training set for each feature is set taking into account the variability of the parameters of the registered feature and the possible presence of various types of interference in the conditions of further real use of the EMMU for its intended purpose (for example, when recognizing a user’s face using an EMMU camera, the training sample should contain examples obtained from different angles, and when recognizing voices using an EMMU microphone, they take into account the presence of noise interference: the voices of other people nearby, noise from wind (rain) or equipment).

In the process of binding the EMMU, the user uses the EMMU for its intended purpose and performs all actions permitted and necessary for registering the signs associated with it.

After this, the success of linking the electronic device is checked for each criterion by conducting a test check to determine the probability of finding the device in the physical possession of an authorized user, completing the binding process for one or another criterion if, based on the results of the test test, the value of the probability of finding it is determined based on the processing of registered information about this characteristic The electronic device in the physical possession of the authorized user exceeds the first (larger) specified threshold probability value, and the process of linking by attribute is repeated with a new test check, if, based on the results of the current test check, the probability value of the electronic device being in the physical possession of the authorized user is determined based on the processing of information about this characteristic user is less than the first (larger) specified threshold probability value. Moreover, the maximum number of repetitions of training to recognize an individual feature in the process of linking the EMMU to an authorized user, if unsuccessful (the value of the probability of recognition of this feature determined in each test after each repetition of training was less than the first (larger) specified threshold probability value) excludes this feature from signs associated with an authorized user are set based on the total number of registered signs (depending on the capabilities of the EMMU sensors and additional electronic devices carried by the user) and their information content.

At the same time, for various environmental conditions and user actions in the learning process, the EMMU finds and remembers the most preferred sign (set of signs) of the user, through which the authorized user in the recognition cycle is recognized with a probability of no less than the first (larger) threshold value required for admission to all functions (applications, services) of the EMMU, which allows in the future, in real conditions of using the EMMU for its intended purpose, the selection of a registered feature (set of features) to be carried out adaptively and automatically.

After successful training of the EMMU to recognize an authorized user, the primary identifier is changed, and also, if necessary (depending on the training results and conditions for further use of the EMMU), two threshold values of the probability of the EMMU in the physical possession of the authorized user and the value of the threshold distance between the EMMU and the user are changed. Next, the EMMU is put into readiness mode for use for its intended purpose and, after the authorized user has successfully passed authentication through a modified primary identifier in real-life conditions of use, the mode of using the EMMU for its intended purpose is activated and the authorized user is allowed to use all functions (applications, services) of the EMMU.

In the process of using the EMMU by the user for its intended purpose, they continuously authenticate the user, for which, by means of EMMU sensors and sensors of additional electronic devices carried by the user, signs associated with the user are periodically recorded on which the EMMU was trained, based on the processing of information about which the value of the current probability of finding electronic multifunctional mobile device in the physical possession of an authorized user. Then the obtained value of the current probability of the electronic device being in the physical possession of an authorized user is compared with two specified threshold probability values and, depending on the result of the comparison, one of three decisions is made: firstly, the decision that the electronic device is in the physical possession of an authorized user while maintaining the user's access to all functions (applications, services) of the EMMU, if the received value of the current probability is greater than the first (larger) specified threshold probability value; secondly, a decision that the EMMU is no longer in the physical possession of the authorized user, blocking the EMMU and informing the user of the need to re-authenticate using the primary identifier to gain access to all functions (applications, services) of the EMMU, if the received value of the current probability less than the second (smaller) specified probability threshold; thirdly, the decision that the electronic device is in the physical possession of an authorized user, but with a probability insufficient for access to all functions (applications, services) of the electronic device, on the basis of which the user is prohibited, informing him about this, access to the most important ( responsible, critical) functions (applications, services) of the EMMU, if the received value of the current probability is less than the first (larger) specified threshold probability value, but greater than the second (smaller) specified threshold probability value. The ban can be introduced either immediately after receiving the result of comparing the current probability with the threshold values, or after some time sufficient for the authorized user to complete the started action with the EMMU. To restore access to all functions (applications, services) of the electronic device, the user must re-activate the electronic device using the primary identifier, after which the periodic collection of sensory information about the user’s characteristics begins again, cyclically calculating the current probability of finding the device in the physical possession of an authorized user and comparing it with threshold values, on the basis of which the user is allowed or denied (limited) access to the functions (applications, services) of the electronic device.

At any point in time, through successful authentication by the user using the primary identifier, the user's access to closed (due to the insufficient value of the current probability) functions (applications, services) of the EMMU is restored.

An important feature of the proposed method is that in the process of using the EMMU for its intended purpose in real-life conditions of use when carrying out continuous user authentication by registering user-related features, the EMMU is additionally trained to recognize an authorized user based on new data on the characteristics on which the EMMU was initially trained. As a result of additional training, the signs of an authorized user are clarified, taking into account both temporary (image or due to illness) and temporary (due to changes in habits, behavioral automatisms, aging) changes in signs associated with the user, as well as taking into account new ones not taken into account previously (during the initial EMMU training) signs. Based on additional training, the “experience” of the EMMU is replenished and thereby improves the quality of the connection between the EMMU and the authorized user.

The continuity of the authentication process both when linking the EMMU to the user and when using the EMMU for its intended purpose in real conditions of use is carried out in cycles, and the duration of the cycle, depending on the conditions of use of the EMMU, is set to vary from one to several periods of registration of signs associated with the user and such , in which an attacker, having gained unauthorized access to the EMMU, would not be able to commit any illegal action during the recognition cycle.

In the process of using the EMMU for its intended purpose in real conditions of use, by means of an RFID tag located with the user, connected via a radio channel with the EMMU, the current distance between the RFID tag and the user is periodically determined (the period is set equal to the duration of the user recognition cycle), if it exceeds a specified threshold distances, meaning that the electronic device is lost (forgotten, stolen), turn on an alert signal (light, sound, vibration signal) for the user and, if the user is not the owner of the electronic device, an alert signal to the owner (organization) that issued the electronic device to the user, block access to functions (applications) , services) EMMU, and also prohibit its shutdown. The prohibition on turning off allows you to purposefully search for a lost or stolen electronic device, and to facilitate the search, the value (amplitude) of the alert signal generated by an RFID tag located with the user is set as a function of the current distance between the electronic device and the RFID tag (in particular, inversely proportional to this distance , so that as you approach the EMMU, the signal magnitude increases), which facilitates (simplify) the procedure for searching for a lost (forgotten, stolen) EMMU.

In the case when, in the process of using the electronic device for its intended purpose, the collected data about the user’s characteristics during the recognition cycle is not enough to calculate the probability of the electronic device being in the physical possession of an authorized user, access to the functions (applications, services) of the electronic device is temporarily blocked and the user is informed (for example, by a special light and (or) sound signal) about the need to pass authentication within a certain time using a valid primary identifier or to perform one or several special registered actions, worked out for such a case during the training process of the electronic device, which have the required authentication information content in the current conditions of use of the electronic device. For example, give the EMMU the opportunity to photograph the user’s face from a certain angle and (or) with a certain facial expression, take a specific pose, answer a question that requires knowledge of confidential information by typing or, if conditions allow, by voice (in the latter case, additional authentication is carried out by voice).

In the event of a threat to access the functions (applications, services) of the electronic device of an unauthorized user (in particular, who has taken the electronic device from an authorized user through the use of force or the threat of force and is in close proximity to the authorized user, so that the user is recognized by sensors of the electronic device and additional electronic devices as authorized), provide for the possibility of emergency blocking of the EMMU by performing a predetermined and practiced action registered by the user (by sensors of the EMMU or additional electronic devices located with the user) (for example, pronouncing a code word or phrase recorded by the microphone of the EMMU).

The choice of a specific list of sensors used (EMMU and additional electronic devices connected to it, located with the user) for recording user characteristics when linking and using the EMMU for its intended purpose in real conditions of use is determined based on the purpose and technical capabilities of the EMMU and additional electronic devices, as well as the required level of protection of the electronic control unit from unauthorized access. At the same time, they also take into account the availability of available (not interfering with the use of the EMMU for its intended purpose) computing power that can be spent on recognition, as well as the amount of remaining charge of the EMMU battery and (or) batteries of additional electronic devices carried by the user that are used to register and process the user’s characteristics (with a small remaining charge, from the signs that are appropriate for the current situation, those signs are adaptively and automatically selected, the registration and processing of information about which requires minimal energy consumption).

If necessary, between the first and second specified threshold values for the probability of the electronic device being in the physical possession of an authorized user, several additional probability thresholds are set, gradually, as they decrease, increasingly limiting the user's access to the functions (applications, services) of the electronic device, i.e. The lower the value of the current probability, the fewer functions (applications, services) of the EMMU the user is allowed to access.

Information about the user’s characteristics obtained during the process of binding (training) the EMMU and refined during the process of additional training of the EMMU in real-life conditions of use is stored only in the EMMU and (or) in additional electronic devices carried by the user and used for his recognition, and is not transmitted or stored. in external devices.

Signals from some of the sensors that register parameters of processes occurring in the user’s body (in particular, electrocardiogram parameters, pulse parameters) as signs associated with the user’s authorization are used not only to authenticate the user, but also to recognize his critical psychophysiological state, indicating loss ( the possibility of loss) by the user of control over the possession of the electronic device due to being in a faint (pre-fainting) state or in a state of severe stress (for example, in the case of the use or threat of physical violence), recognizing which, they block the electronic device (to prevent unauthorized access to it) and send signals calling the necessary help (for example, an ambulance).

An authorized user in whose physical possession the EMMU is allowed to repeatedly change the primary identifier.

In general, the implementation of the proposed method, in the case of the above-mentioned integration of the capabilities of modern machine learning and a logical approach to creating artificial intelligence based on a symbolic description, rules and Bayesian inference, makes it possible to create the most effective method of user authentication, free from the significant shortcomings of modern methods used for image recognition. artificial neural networks of deep learning (the need to have a large representative training sample, the tendency to make gross errors due to excessive interpretation (taking into account unimportant factors), the difficulty of separating borderline cases, opacity, ease of targeted deception).

Claims (27)

1. A method for protecting an electronic multifunctional mobile device from unauthorized access, including the steps of: - setting two threshold values for the probability of an electronic multifunctional mobile device being in the physical possession of an authorized user and a threshold value of the distance between the electronic multifunctional mobile device and the user, the excess of which means that the electronic multifunctional mobile device is no longer in the physical possession of an authorized user; - authenticate the user using a primary identifier in the form of a password, key phrase, PIN code, gesture and, if successful, consider the user authorized and allow him to access all functions of the electronic multifunctional mobile device; - during the user’s use of an electronic multifunctional mobile device for its intended purpose, the user is continuously authenticated, for which purpose, through sensors of the electronic multifunctional mobile device and sensors located with the user of additional electronic devices, signs associated with the user are periodically recorded, including his photo and video images produced by him voice signals, as well as signals carrying information about the dynamics of user behavior, the movements of an electronic multifunctional mobile device, the current distance between the electronic multifunctional mobile device and the user, based on the processing of information about which the value of the current probability of finding an electronic multifunctional mobile device in physical possession is determined authorized user; - compare the obtained value of the current probability of finding an electronic multifunctional mobile device in the physical possession of an authorized user with two specified threshold probability values and, depending on the result of the comparison, make one of three decisions: firstly, the decision that the electronic multifunctional mobile device is in in the physical possession of an authorized user while maintaining the user’s access to all functions of an electronic multifunctional mobile device, if the received value of the current probability is greater than the first (larger) specified threshold probability value; second, a determination that the mobile electronic multifunction device is no longer in the physical possession of the authorized user, locking the mobile electronic multifunction device and informing the user of the need to re-authenticate through the primary identifier in order to gain access to all functions of the mobile electronic multifunction device, if the resulting current probability value is less than the second (smaller) specified threshold probability value; thirdly, a decision that the electronic multifunctional mobile device is in the physical possession of an authorized user, but with a probability insufficient to allow access to all functions of the electronic multifunctional mobile device, on the basis of which the user is prohibited, informing him about this, access to the most important functions of an electronic multifunctional mobile device, if the received current probability value is less than the first (larger) specified threshold probability value, but greater than the second (smaller) specified threshold probability value; - at any time, through a new successful authentication by the user using the primary identifier, the user’s access to the functions of the electronic multifunctional mobile device that are closed due to the insufficient value of the current probability is restored, characterized in that: - before starting to use the electronic multifunctional mobile device for its intended purpose in conditions that exclude access to the electronic multifunctional mobile device by unauthorized users, after setting two threshold values for the probability of finding the electronic multifunctional mobile device in the physical possession of an authorized user and the threshold value for the distance between the electronic multifunctional mobile device and by the user, exceeding which means that the electronic multifunctional mobile device is no longer in the physical possession of the user, successful authentication by the user through the primary identifier issued to him upon purchase (reception) of the electronic multifunctional mobile device with the user's access to all functions of the electronic multifunctional mobile device activates the binding mode electronic multifunctional mobile device to an authorized user (training mode of an electronic multifunctional mobile device) and bind the electronic multifunctional mobile device to an authorized user, teaching the electronic multifunctional mobile device to recognize an authorized user based on characteristics associated with him, for which purpose, in the process of further completion by the user of all authorized and actions necessary for binding to an electronic multifunctional mobile device through sensors of an electronic multifunctional mobile device and sensors of additional electronic devices located with the user, periodically register, summarize and remember the static and dynamic biometric characteristics inherent in the authorized user, signs of unique images created by the user in the form of gestures, poses , sequences of movements, phrases, as well as signs of unique images of objects carried by the user, check the success of the binding for each sign by conducting a test to determine the probability of finding an electronic multifunctional mobile device in the physical possession of an authorized user, completing the binding process for one or another sign, if based on the results of the test check, the value of the probability of finding an electronic multifunctional mobile device in the physical possession of an authorized user, determined based on the processing of registered information about this attribute, exceeds the first (larger) specified threshold probability value, and the process of linking by attribute is repeated with a new test check, if according to the results of the current test check, the value of the probability of finding an electronic multifunctional mobile device in the physical possession of an authorized user, determined on the basis of processing information about this attribute, is less than the first (larger) specified threshold probability value; - after successful training of the electronic multifunctional mobile device to recognize an authorized user using the characteristics selected for recognition, the primary identifier is changed, the electronic multifunctional mobile device is switched to readiness mode for its intended use, and after the authorized user has successfully passed authentication through the changed primary identifier in real conditions of use, the use mode is activated electronic multifunctional mobile device for its intended purpose and allow an authorized user to use all functions of the electronic multifunctional mobile device; - user recognition with calculation of the current probability that the electronic multifunctional mobile device is physically owned by an authorized user, both during training of the electronic multifunctional mobile device and during its subsequent use for its intended purpose in real conditions of use, is carried out in cycles, the duration of each of which is, depending on the conditions of use of the electronic multifunctional mobile device, they are set to vary from one to several periods of registration of user characteristics, and the maximum duration of the recognition cycle is chosen in such a way as to ensure continuity of the recognition process; - in the process of using an electronic multifunctional mobile device for its intended purpose in real conditions of use when carrying out continuous user authentication by registering user-related features, the electronic multifunctional mobile device is additionally trained to recognize an authorized user based on new data about the features associated with it, which makes it possible to clarify the associated features signs authorized by the user and, as a result, improve the quality of binding an electronic multifunctional mobile device to it; - in the process of using an electronic multifunctional mobile device for its intended purpose in real conditions of use, using an additional electronic device located with the user in the form of an RFID tag, connected via a radio channel with the electronic multifunctional mobile device, the current distance between the RFID tag and the user is periodically determined, if exceeded which of a given threshold distance, meaning that an electronic multifunctional mobile device is lost (forgotten, stolen), include a light, sound, vibration alert signal to the user and, if the user is not the owner of the electronic multifunctional mobile device, an alert signal to the owner who issued the electronic multifunctional mobile device to the user , block access to the functions of the electronic multifunctional mobile device, and also prohibit it from being turned off. 2. The method according to claim 1, characterized in that training an electronic multifunctional mobile device to recognize an authorized user based on registered characteristics is carried out using representative training and test samples using artificial intelligence technologies using deep learning artificial neural networks implemented directly on the electronic multifunctional mobile device or on individual additional electronic devices located with the user and connected via a radio channel to the electronic multifunctional mobile device. 3. The method according to claim 1, characterized in that the signs associated with the authorized user are periodically recorded by sensors of the electronic multifunctional mobile device and sensors of additional electronic devices located with the user and connected to the electronic multifunctional mobile device via a radio channel, as in in the process of training an electronic multifunctional mobile device, and in the process of using it for its intended purpose in real conditions, the following are used: firstly, the biometric characteristics of an authorized user, including: the timbre and spectrum of his voice; photo and video images of the face or its parts in the form of ear, nose, lips; fingerprint; drawing of the iris and/or retina; drawing of facial vessels; pattern of palm vessels; hand geometry; pulse parameters in the form of its frequency, rhythm, tension; size and shape of the heart, its dynamics, electrical activity in the form of electrocardiogram parameters; electroencephalogram parameters; odor emitted by the user; the chemical composition of the air exhaled by the user; dynamics of body movement when walking - gait; dynamics of hand movement with an electronic multifunctional mobile device; signature dynamics; dynamics of typing letters on the virtual keyboard of an electronic multifunctional mobile device; microvibration of fingers; secondly, signs of unique images created by the user in the form of: gestures, poses, sequences of actions, code words or phrases, melodic motifs, in which the user’s biometric characteristics and (or) knowledge of confidential information are present; thirdly, signs of unique images of objects carried by the user: visual images in the form of a photograph of an object carried by the user, taken by the camera of an electronic multifunctional mobile device from a certain distance and angle; olfactory images in the form of a specific, not perceptible by a person, smell of any object in the user’s possession; sound images in the form of infrasonic or ultrasonic low-intensity signals periodically emitted by additional electronic devices located with the user. 4. Method according to any one of paragraphs. 1, 3, characterized in that in the process of linking an electronic multifunctional mobile device to an authorized user, the number of examples in the training set for each attribute is set taking into account the variability of the parameters of the recorded attribute and the possible presence of various types of interference in the conditions of real use of an electronic multifunctional mobile device for its intended purpose . 5. Method according to any one of paragraphs. 1, 3, 4, characterized in that the maximum number of repetitions of training to recognize an individual feature in the process of linking an electronic multifunctional mobile device to an authorized user, in the event of unsuccessful completion of which (the value of the probability of recognition of this feature determined in each test after each repetition of training) turned out to be less than the first (greater than) a given threshold probability value) exclude this feature from the features associated with the authorized user, set based on the total number of registered features and their informativeness. 6. The method according to claim 1, characterized in that for various environmental conditions and user actions during the learning process of the electronic multifunctional mobile device, the most preferred sign (set of signs) of the user is found and stored, through which the authorized user is recognized with probability in the recognition cycle, not less than the first (higher) threshold value required for access to all functions of an electronic multifunctional mobile device, which allows later, in real conditions of using an electronic multifunctional mobile device for its intended purpose, the selection of a registered feature (set of features) to be carried out adaptively and automatically. 7. The method according to claim 1, characterized in that information about the user’s characteristics, obtained during the implementation of the mode of binding an electronic multifunctional mobile device to an authorized user and refined in the process of additional training of the electronic multifunctional mobile device in real conditions of use, is stored only in the electronic multifunctional mobile device and (or) in additional electronic devices located with the user and used to recognize him, is not transmitted and is not stored in external devices. 8. The method according to claim 1, characterized in that, if necessary, between the first and second specified threshold values of the probability of finding an electronic multifunctional mobile device in the physical possession of an authorized user, several additional threshold probability values are set, gradually, as they decrease, increasingly limiting access user to the functions of an electronic multifunctional mobile device. 9. The method according to claim 1, characterized in that the value (amplitude) of a light, sound, vibration alert signal generated by the user’s RFID tag when the distance between the electronic multifunctional mobile device and the user exceeds a specified threshold distance, meaning that the electronic multifunctional mobile device mobile device is lost (forgotten, stolen), is set inversely proportional to this distance, which facilitates the procedure for finding a lost (forgotten, stolen) electronic multifunctional mobile device. 10. The method according to claim 1, characterized in that in the case when, in the process of using an electronic multifunctional mobile device for its intended purpose, the collected data about the characteristics of the user during the recognition cycle is not enough to calculate the probability of finding the electronic multifunctional mobile device in the physical possession of an authorized user, temporarily block access to the functions of the electronic multifunctional mobile device and, through a light and (or) sound signal, inform the user about the need to pass authentication within a certain time using a valid primary identifier or perform one or more actions worked out for such a case during the training process of the electronic multifunctional mobile device, having, in the current conditions of use of an electronic multifunctional mobile device, the required authentication information content. 11. The method according to claim 1, characterized in that if there is a threat to access the functions of an electronic multifunctional mobile device by an unauthorized user who has taken the electronic multifunctional mobile device from an authorized user through the use of force or the threat of force and is in close proximity to the authorized user, so that the sensors of the electronic multifunctional mobile device and additional electronic devices recognize the user as authorized, provide for the possibility of emergency blocking of the electronic multifunctional mobile device by means of an authorized user performing a predetermined and when binding the electronic device registered by the sensors of the electronic multifunctional mobile device or additional electronic devices located with the authorized user a multifunctional mobile device of a proven action in the form of an authorized user pronouncing a code word or phrase registered by the microphone of an electronic multifunctional mobile device. 12. Method according to any one of paragraphs. 1, 3, characterized in that the choice of a specific list of used sensors of an electronic multifunctional mobile device and additional electronic devices connected to it, located with the user, for recording the user’s characteristics when binding and using the electronic multifunctional mobile device for its intended purpose in real conditions of use is determined based on from the purpose and technical capabilities of the electronic multifunctional mobile device and additional electronic devices, as well as the required level of protection of the electronic multifunctional mobile device from unauthorized access. 13. Method according to any one of paragraphs. 1, 3, 6, characterized in that the selection of registered user characteristics for recognition when using an electronic multifunctional mobile device for its intended purpose in real conditions of use is carried out taking into account the available computing power that does not interfere with the use of the electronic multifunctional mobile device for its intended purpose, which can be spent on recognizing, as well as the amount of remaining battery charge of an electronic multifunctional mobile device and (or) batteries of additional electronic devices located with the user, used to register and process the user’s signs, and when the remaining charge is low, signs that are appropriate for the current situation are adaptively and automatically selected those signs, registration and processing of information about which requires minimal energy consumption. 14. Method according to any one of paragraphs. 1, 9, characterized in that the period of determining the current distance between the RFID tag and the user’s body, when exceeding a given threshold distance, a decision is made that the electronic multifunctional mobile device is lost (forgotten, stolen), and the electronic multifunctional mobile device is blocked, is set equal to the duration of the user recognition cycle based on the registered characteristics. 15. Method according to any one of paragraphs. 1, 3, characterized in that the signals of some of the sensors that register parameters of processes occurring in the user’s body, in the form of electrocardiogram parameters, pulse parameters, as signs associated with an authorized user, are used not only to authenticate the user, but also to recognize his critical psychophysiological a condition indicating the loss or possibility of loss by the user of control over the possession of an electronic multifunctional mobile device due to being in a faint or near-fainting state or in a state of severe stress, recognizing which, they block access to the functions of the electronic multifunctional mobile device and send signals to call the necessary help. 16. The method according to claim 1, characterized in that the authorized user, in whose physical possession the electronic multifunctional mobile device is, is allowed to repeatedly change the primary identifier.

Images