WO2019013667A1

WO2019013667A1 - Method of implementing automated user authentication on the basis of his/her signature

Info

Publication number: WO2019013667A1
Application number: PCT/RU2017/000541
Authority: WO
Inventors: Евгений Борисович ЮГАЙ; Станислав Игоревич АШМАНОВ
Original assignee: Евгений Борисович ЮГАЙ
Priority date: 2017-07-11
Filing date: 2017-07-21
Publication date: 2019-01-17
Also published as: RU2671305C1

Abstract

A method is claimed for automated user authentication on the basis of his/her signature made by moving a mobile device in the air. A method comprises steps in which: a first set of acceleration values of motion of a user's mobile device are obtained at each point in time by means of an accelerometer of the mobile device in the form of a digital signal; the digital signal is processed for the first set of acceleration values of motion of the mobile device; a second set of acceleration values of motion of the user's mobile device is obtained at each point in time by means of an accelerometer of the mobile device in the form of a digital signal; the digital signal is processed for the second set of acceleration values of motion of the mobile device; the magnitude of the distance between the first set of acceleration values of motion of the mobile device and the second set of acceleration values of motion of the mobile device is determined; the magnitude of the distance determined in the preceding step is compared with a predetermined value, wherein if the distance is less than said value, the user is identified and granted access to the mobile device; if the distance is greater than said value, the user is not identified and not granted access to the mobile device.

Description

METHOD OF IMPLEMENTING AUTOMATED USER AUTHENTICATION ON THE BASIS OF ITS SIGNATURE

TECHNICAL FIELD

[1] This technical solution generally relates to the field of computing technology, and in particular to systems and methods for performing automated user authentication based on his signature, carried out by air movement of a mobile device.

BACKGROUND

[2] Currently, the term “authentication” in information systems refers to checking personal data of an object (for example, a person, computer, information process) in order to authorize access of this object to resources (for example, services, networks, systems, applications).

[3] In the prior art, there are various ways to ensure the authenticity of a user working with a mobile device and requesting access to it. Authentication can occur based on several factors:

[4] - data known to the user, for example, entering a password on a mobile device;

[5] - an item possessed by the user, for example, a magnetic card or a chip card that is attached to a mobile device;

[6] - an action performed only by the user, for example, a live signature on the touch screen;

[7] - physical parameter inherent in the user, for example, a fingerprint.

[8] Currently, the basis of the most widely used method of authenticating a user on a mobile device is the use of a static password together with an identifier (login). The advantage of this method is that it can be fully implemented using software, thereby avoiding additional costs (hardware). However, network espionage to collect personal data of legitimate users is a hindrance to the implementation of this method. Personal data obtained in this way can be used further. Hashing or encrypting passwords cannot fully solve this problem, since they only protect communication lines, leaving the user terminal (mobile device) in which Now you can often find a lot of malicious applications (hacker programs).

[9] Authentication methods based on user-provided tools, such as an authentication token, chip card, smart card, or flash drive, are undesirable because they cause additional hardware costs.

[10] At the same time, the currently used biometric authentication methods are mostly complex, and they require the implementation of the tedious preliminary determination of parameters. SUMMARY OF INVENTION

[11] This technical solution is aimed at eliminating the shortcomings inherent in existing solutions, known from the prior art.

[12] A technical problem (in other words, a technical problem) in this technical solution is to authenticate the user on the basis of his signature, carried out by the movement of the mobile device in the user's hand.

[13] The technical result manifested in solving the above technical problem is to simplify the authentication of the user of the mobile device.

[14] Also, an additional technical result in relation to the prior art is to increase the accuracy of authentication of a mobile device user by comparing the accelerations of the mobile device.

[15] This technical result is achieved due to the implementation of the method of automated user authentication based on his signature, carried out by air movement of a mobile device, in which a first set of acceleration values of a mobile device of a user is obtained at each time using an accelerometer of a mobile device in the form of a digital signal; processing the digital signal for the first set of acceleration values of the mobile device obtained in the previous step; get the second set of values of the acceleration of the mobile device of the user at each point in time through the accelerometer of the mobile device in the form of a digital signal; processing the digital signal for the second set of values of the acceleration of the mobile device, obtained on previous step; determine the distance between the first set of values of the accelerations of the mobile device and the second set of values of the accelerations of the mobile device; compare the distance value determined in the previous step with a predetermined value, and if the distance is less than this value, identify the user and allow access to the resource; and if the distance is greater than this value, they do not identify the user and do not allow access to the resource.

[16] In some embodiments, the implementation of the technical solution when receiving the first set of values of the accelerations of the mobile device, the accelerometer converts it into an electrical digital signal that is conditioned.

[17] In some embodiments of the technical solution, a gyroscope is used additionally with an accelerometer.

[18] In some embodiments of the technical solution, the digital signal is processed for the first set of acceleration values of the mobile device by smoothing and filtering.

[19] In some embodiments of the technical solution, when a signal is smoothed, a low-pass filter and / or a Kalman filter are used.

[20] In some embodiments, the implementation of the technical solution determines the distance between the first set of values of the accelerations of the mobile device and the second set of values of the accelerations of the mobile device by using the Euclidean metric.

BRIEF DESCRIPTION OF THE DRAWINGS

[21] The features and advantages of the present invention will become apparent from the following detailed description of the invention and the accompanying drawings, in which:

[22] Fig. 1 shows an example of the implementation of a variant of a technical solution according to the method of automated user authentication based on his signature, carried out by air movement of a mobile device;

[23] Figure 2 shows a conditioned ADC input signal, where the samples are represented by red dots; [24] FIG. 3 shows an enlarged scale of the conditioned input signal of the ADC shown in FIG. 2, where the samples are represented by red dots;

[25] FIG. 4 shows the recovery of the signal from samples, in which the samples on the graph are connected by straight lines;

[26] FIG. 5 shows an example of signal processing by means of a low-pass filter;

[27] FIG. 6 shows the direction of the axes of the accelerometer;

[28] FIG. 7 shows a graph of the display of accelerometer acceleration values when the mobile device moves sharply along the horizontal line from left to right;

[29] FIG. 8 shows a graph of the display of accelerometer acceleration values when a mobile device moves around a circle;

[30] FIG. 9 shows an example of the implementation of a technical solution, according to which the user forms a given set of values by writing his signature by the movement of a mobile device in the air;

[31] FIG. 10 shows an automated user authentication system based on his signature, carried out by airborne movement of a mobile device. DETAILED DESCRIPTION OF THE INVENTION

[32] This technical solution can be implemented on a computer, in the form of a system or machine-readable medium containing instructions for performing the above method.

[33] The technical solution can be implemented as a distributed computer system.

[34] In this solution, a system is a computer system, a computer (electronic computer), a CNC (numerical control), a PLC (programmable logic controller), computerized control systems, and any other devices capable of performing a predetermined, well-defined sequence of operations. (actions, instructions).

[35] A command processing device is an electronic unit or an integrated circuit (microprocessor) that executes machine instructions (programs). [36] The command processing device reads and executes machine instructions (programs) from one or more data storage devices. In the role of a storage device can act, but not limited to, hard drives (HDD), flash memory, ROM (read-only memory), solid-state drives (SSD), optical drives.

[37] Program - a sequence of instructions intended for execution by a computer control device or command processing device.

[38] The following will describe the terms and concepts necessary to implement this technical solution.

[39] Authentication is a security feature that authenticates a person who gains access to an automated system by comparing the identifier given to them and the presented confirming factor.

[40] A digital signal is a discrete signal that gives an idea of the value of the monitored quantity in digital form.

[41] Accelerometer - a measuring device designed to measure accelerations.

[42] Distance - in this technical solution, the content of the term "distance", as in mathematics, is associated with the concept of a metric and a metric space.

[43] A metric is a function that determines distances in a metric space.

[44] Analog-to-digital converter (ADC) is a device for automatically converting analog (continuous in time) signals to their equivalent discrete signals represented by a digital code. Analogous values are most often electrical voltage or current, oscillation frequency, phase shifts, and angles of rotation. Numeric codes are represented mainly in binary, binary-decimal or decimal number systems. A / D converters are widely used for interfacing analog signal sources (eg, meters, converters, AVMs) with digital recorders and digital ones. devices (eg, with a microprocessor, computer).

[45] A numeric code is a sequence of numbers (signals) that obeys a certain law, by means of which the conditional representation of a numerical value of a quantity is realized. [46] Training sample (training sample) - a sample that is used for tuning (optimization of parameters) of a dependency model.

[47] Test (or control) sample (test sample) - a sample by which the quality of the constructed model is evaluated.

5 [48] FIG. 1 is a flowchart showing a method for performing automated user authentication based on his signature, carried out by air movement of a mobile device, which comprises the following steps:

[49] Step 101: get the first set of values of the accelerations of the mobile device of the user at each time using the accelerometer of the mobile device in the form of a digital signal.

[50] In this step, a first set of acceleration values of the mobile device is obtained on the mobile device, which is stored in the memory 904 (FIG. 10) of the mobile device. This set of values is formed on the mobile

15 by writing the user a movement of his signature in the air in the air (as shown in FIG. 9). To ensure the required accuracy, it is recommended to perform a signature, for example, the upper left corner of the mobile device. This first set of acceleration values of a mobile device is called a reference, which is

20 analog password. Also, the above set of values may be part of the training sample for the operation of the INS (artificial neural network) in the future when recognizing the user's signature. A training sample is a finite set of objects for which their class affiliation is known, which is used to train (train) a neural network. With

25 learning a neural network with a teacher the elements of the training sample are fed to the input of the neural network. The output vector is adjusted to some predetermined target vector by changing the weights of the neural network connections. The training sample should not contain contradictions, since the neural network unambiguously compares the output values with the input. In some embodiments, the implementation of the technical solution uses a multilayer perceptron, which is a neural network with direct signal propagation (without feedback), trained with the teacher. For learning, the INS can be used without limitation, the back propagation method (Backpropagation) or the elastic propagation method (Resilient propagation or Rprop), or genetic

35 algorithm (Genetic Algorithm), etc. [51] In some embodiments, to help isolate a user’s trajectory, use a button on a mobile device to distinguish between intentional gesture movement and unintentional movement or deflection. The button can be pressed during the execution of the signature, or it can be pressed once at the beginning of the gesture and once at the end. In another embodiment, the trajectory speed can be used to determine the beginning and end of gestures. For example, when a user starts a long, fast motion, it is assumed that the gesture has begun. When some time passes, and the total amount or speed of movement decreases to a level below a certain predetermined threshold, it is assumed that the gesture is complete. In yet another embodiment, any movement is considered a potential gesture movement, and a moving data window is used to capture gestures.

[52] Since the user may not be familiar with the experience of using hand movement to draw a given trajectory of his signature in the air, feedback can be used to help the user learn how to control the trajectory. In this case, the image (scanned user's signature) corresponding to the trajectory of points X and Y can be displayed, for example, on the display of the mobile device, in order to notify the user about the trajectory that will be used in the gesture recognition algorithm.

[53] The effective acceleration obtained as a set of values (called sampling) at each time point at each point in space, the accelerometer 903 (Fig. 10) converts into an electrical digital signal that is conditioned (amplified or attenuated, and if necessary to him a constant offset is added) so that its possible levels correspond to the input range of the analog-to-digital converter (ADC). The ADC periodically samples the signal, that is, it remembers the current value of the signal and converts it into a digital code proportional to this value. Depending on the number of ADC bits, the code takes positive integer values in the range from 0 to 2 ^N - 1, where N is the number of ADC bits. For example, for a 10-bit ADC, the code can take values from 0 to 1023, for a 12-bit one, from 0 to 4095. The values of the ADC codes are dimensionless, they are also called ADC units. Samples are performed by the ADC at regular time intervals T. In FIG. Figure 2 shows an air-conditioned ADC input signal; the samples are represented by red dots. The set of acceleration values can be received in parallel or sequentially in approximately the following form: 519 535 530 ... (these are samples NQ1, 2.3 ...). IF viewed on a larger scale in FIG. 2 (see Fig. 3), it can be seen that the signal and sample values are different: sampling in ADC units is the nearest integer number corresponding to the value of the selected signal. Data processing component 902 (FIG. 10) restores the signal by 5 samples, in the simplest case, the samples on the graph are connected by straight lines, and the degree of distortion of the graph is the smaller, the greater the number of ADC bits and the higher the sampling rate (the higher these parameters, the more expensive accelerometer), as shown in FIG. 4. In some embodiments, the implementation of the technical solution uses linear interpolation, for example, splines. In some embodiments, the implementation of the technical solution uses Bezier curves for interpolation.

[54] As an accelerometer 903 (FIG. 10), depending on the architecture at the moment, several types of devices can be used, for example, based on MEMS. The main purpose of the accelerometer is to provide

15 information about the current acceleration of the device, more precisely, the difference between the acceleration of the device and the acceleration of gravity. At rest, the sensor readings coincide with the gravitational acceleration vector. In some embodiments, the implementation of the accelerometer may be based on the capacitor principle. In such a system, the movable part is made in the form of a normal weight,

20 displaced depending on which way the device is tilted. As it shifts, there is a change in the capacitance of the capacitor, and more specifically, the voltage. These data allow us to obtain the displacement of the sinker, and with it the desired acceleration. This is how the accelerometer works. The most common type are piezoelectric systems. They are based on

25 is a weight acting under pressure on the piezocrystal. In response, it produces an electric current, so that you can calculate the desired acceleration, if you know the parameters of the entire system. In some embodiments, a gyroscope is used with the accelerometer. Its main purpose is to measure the angular velocity zo with respect to one or several axes. Actually, the combination of an accelerometer and a gyroscope allows you to track and record movement in three-dimensional space. The direction of the axes of the accelerometer is shown in FIG. 6: x, y and z, which are recorded acceleration values that are displayed on the graph. For example, if there is a rather sharp movement of a mobile device.

35 along the horizontal line from left to right, the graph looks like it is shown in FIG. 7 The graph clearly highlights two bursts along the X axis, the first movement start, the second - the end. The values on the other axes do not change significantly. Similarly, bursts of values on the Y and Z axes are formed when moving in the corresponding direction. The circular motion shown in FIG. 8, was held counterclockwise, the starting point of the movement was the lower point of the circle. Graphs of changes in the values along the X and Z axes practically coincide; it is along these axes that the circle was drawn. A phase shift naturally takes place, since the horizontal change began earlier than the vertical one. Let's take a closer look at the accelerometer readings on the vertical axis Z. The graph has a sinusoidal view, since the beginning of the movement was at the lowest point of the circle, where sin takes the value - 1, then the full phase of the sinusoid is expected from the first lower point to the second, which can be observed chart. The decline to the first low point and the rise from the second, characterize the beginning of the movement (acceleration) and completion (deceleration). In fact, the first set of mobile device acceleration values is a reference.

[55] It will be clear to a person skilled in the art that the present invention is not limited to MEMS-based devices, and that the MEMS-based embodiments described herein are illustrative and that the present invention can be implemented using any accelerometer, compass and gyroscope, which may be located in a mobile device. Those skilled in the art will appreciate that the present invention may also use other types of inertial sensors that may be included in a mobile device, for example, quartz sensors. Other types of inertial sensors, which include mechanical components on a micro or millimeter scale and can be combined with an electronic circuit, can also be used in the present invention.

[56] Step 102: processing of the digital signal for the first set of acceleration values of the mobile device obtained in the previous step.

[57] Accelerometer readings on mobile devices are subject to quite a lot of noise. Noise sometimes reaches 0.08d, as a result, there is an urgent need to combat noise. Below we will consider several approaches to smoothing and filtering accelerometer data.

[58] In some embodiments, the averages method is used, which is one of the simplest noise filtering methods. On each step k, the value of v _{k is} calculated as the average of the n previous accelerometer values, that is, v _k = ^Σί = ° ^ - This method gives a good smoothing for average values of n, but when using this method there may be a large delay in the values that need to be taken into account .

5 [59] In some embodiments of the technical solution, one of the ways to combat noisy data is to use a filter. The task of motion recognition by a mobile device imposes one essential requirement on the filter — a performance requirement sufficient to use a real-time filter with minimal delay. Of course, a big plus of the filter is the proximity of the values to the initial ones. In some embodiments, the implementation of the technical solution can use a low-pass filter that has the ability to filter signals above a specified frequency, that is, the filter passes low-frequency signals, which allows you to get rid of noise

15 signal interference, in this case, the accelerometer readings. The simplest low-pass filter is described by the following formula: О _п = 0 _п _ _г + a (J _n - 0 _η _ι), where O is the output signal value (filtered), 1 _n is the input values (unfiltered), is the filter coefficient, taking values from 0 to 1. When 1 is equal, the output values coincide with the input. Work example

20 of the low pass filter is shown in FIG. 5. When processing the result is smoother.

[60] In some embodiments, a Kalman filter is used. The Kalman filter is often used to filter the values of various kinds of signals, it can be found in GPS receivers, handlers of readings

25 sensors, etc. Kalman filter is a type of recursive filter.

To calculate the state of the system for the current work step, it needs a state estimate (in the form of a system state estimate and an estimate of the error in determining this state) at the previous work step and measurement at the current step. Iterations of the Kalman filter are divided into two phases: zo extrapolation and correction. During the extrapolation filter receives a preliminary assessment of the state of the system _to x \ _k - _g on the current step of the final assessment of the state from the previous step. This preliminary assessment is also called an a priori assessment of the state, since no observations of the corresponding step are used to obtain it. In the correction phase, a priori extrapolation supplemented by appropriate current measurements to correct the estimate. Adjusted score is also called a posteriori state estimate, a estimate of the state vector x _k. Usually, these two phases alternate: extrapolation is made on the basis of the results of the correction to the next observation, and the correction is made together with the observations available in the next step, and so on.

[61] Additional filtering to clean the signal can eliminate involuntary hand shake and other unwanted high-frequency components. Jitter is a small and fast vibration inherent in the human hand.

[62] Step 103: get the second set of values of the accelerations of the user's mobile device at each point in time through the accelerometer of the mobile device in the form of a digital signal.

[63] At this step, a second set of acceleration values of the user's mobile device at each time point to the data processing device, which is stored in memory 904 (Fig. 100) of the mobile device, is received on the mobile device of the user, when the user performs authentication directly at the mobile device using the run-in-the-air signature (as shown in Fig. 9).

[64] In some embodiments of the technical solution, the second set of acceleration values of the mobile device is used either as part of a training set for training the ANN, or as a test set to test the technical solution, i.e. for the task of classifying a user's signature.

[65] Step 104: processing of the digital signal for the second set of acceleration values of the mobile device obtained in the previous step.

[66] In this step, as shown above, the accelerometer data is smoothed and filtered. Similarly, processing algorithms, known from the prior art to any specialist, whose application is also obvious, can be used.

[67] Having on the component 902 of data processing two sets of values of the accelerations of the movement of the mobile device, obtained earlier, it is necessary and to compare, carrying out before this additional preprocessing. For correct comparison, first set the same duration of execution user of the signature in the air by stretching or narrowing the graph, getting the same length. In fact, at this stage, the same size of the sets of values of the accelerations of the mobile device is established, since it is incorrect to compare, for example, the signature that lasts 5 seconds and 10 seconds. It is also necessary to take into account that the accelerometer has some delay in determining the values of accelerations, which can be very significant, and therefore set the same coinciding time points for sets of values of accelerations by interpolating the values of accelerations. There may also be a problem in which the user first forms a signature by moving slowly in the air, and a second time and then at a different speed, for example, more intensively. To eliminate this problem, compress the motion path graph (by multiplying all points of two compared sets of mobile device acceleration values), for example, to the range [0; 1], resulting in two sets of mobile device acceleration values having the same minimum and maximum, and the problem is is decided.

[68] In some embodiments, a naive Bayes classifier is used to determine if the second set of acceleration values of a mobile device belongs to the same class as the first set of acceleration values.

[69] Step 105: Determine the distance between the first set of acceleration values of the mobile device and the second set of acceleration values of the mobile device.

[70] At this step, two sets of acceleration values are compared to determine the distance between them, for example, using the Euclidean metric (or Euclidean distance)

[71] For the two sets of values of acceleration of the mobile device movement a = _(g, ..., α _η) and b = (b _x, ..., b _n) obtained previously, the Euclidean distance is determined as follows: d { a, b) = V ("i - hi ² + (α ₂ - b ₂ ) ² + - + (α _η - b _n Y =

[72] In some embodiments of the technical solution, a Manhattan metric, a Gromov-Hausdorff metric, etc. may be used, not limited to, which is obvious to an expert in the level of technology. [73] Step 106: compare the distance value determined in the previous step with a predetermined value.

[74] After determining the distance between the first set of acceleration values of the mobile device and the second set of acceleration values of the mobile device using the metric, this value is compared with a predetermined value of ε (for example, a threshold value).

[75] If the previously determined distance d is less than a given value of ε, identify the user and allow access to the mobile device or resource residing on the mobile device;

[76] If a previously determined distance d is greater than a given value of ε, do not identify the user and do not allow access to the mobile device or resource residing on the mobile device.

[77] A user or developer of a mobile device can also adjust recognition thresholds in various ways. Each individual signature may contain independent thresholds and may also contain a finite probabilistic scale factor that transfers the signature relative to other signatures, allowing the application developer or user to choose a signature from two similar signatures that are easier to run.

[78] In some embodiments, mobile device 900 may be a mobile phone, a messaging device, a tablet and a personal digital assistant, etc.

[79] Referring to FIG. 10, system 900 may include one or more of the following components: a data processing component 902, an accelerometer 903, a memory 904, a power component 906, a multimedia component 908, an input / output (I / O) interface 912, a touch component 914, data transfer component 916.

[80] In some embodiments, the data processing component 902 mainly manages all operations of the mobile device 900, for example, a display, data transmission, an accelerometer operation, and a data recording operation. Data processing component 902 may include one or more processors 918 that implement instructions for completing all or part of the steps of the above methods. In addition, the data processing component 902 may include one or more modules for a convenient process of interaction between the processing component 902 and other components. For example, component 902 The data processing may include a multimedia module for conveniently facilitating interaction between the multimedia component 908 and the processing component 902.

[81] The memory 904 is configured to store various types of data to support the operation of the mobile device 900. Examples of such data include instructions from any application or method, image, video, etc. Memory 904 can be implemented as cloud storage data, any type of volatile storage device, non-volatile storage device, or a combination of the two, such as Static Random Access Memory (ELR), Electrically Erasable Programmable Permanent Memory downgrading device (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic memory, flash memory, magnetic or optical disk.

[82] In some embodiments, the power component 906 provides electricity to various components of the mobile device 900. The power component 906 may include a power management system, one or more power sources, and other nodes for generating, controlling, and distributing electricity to the mobile device 900.

[83] In some embodiments, the multimedia component 908 includes a screen that provides an output interface between the mobile device 900 and the user. In some embodiments, the implementation of the screen can be a liquid crystal display (LCD) or a touch panel (SP). If the screen includes a touch pad, the screen may be implemented as a touch screen to receive input from a user. The touchpad includes one or more touch sensors in the sense of gestures, touch, and sliding of the touch pad. The touch sensor can not only feel the touch or gesture of turning, but also determine the duration of time and pressure associated with the mode of operation of the touch and slip.

[84] The I / O interface 912 provides an interface between the data processing component 902 and the peripheral interface module.

[85] The sensor component 914 comprises one or more sensors and is configured to provide various aspects of assessing the state of the mobile device 900. For example, a touch component 914 can detect on / off states of a mobile device 900, the relative arrangement of components, such as the display and keypad of a mobile device 900, a change in the position of the mobile device 900 or one component of the mobile 5 device 900, the presence or absence of contact between the user and the mobile device 900, as well as the orientation or acceleration / deceleration and temperature change of the mobile device 900. The touch component 914 comprises a proximity sensor, nny to detect the presence of the object located nearby when there is no physical contact. The sensor component 914 comprises an optical sensor (for example, a CMOS or CCD image sensor) adapted to be used in the visualization of the application. In some embodiments, the sensor component 914 includes an acceleration sensor, a gyroscope, a magnetic sensor, a pressure sensor, or a temperature sensor.

[86] The communication component 916 is configured to facilitate

15 wired or wireless communication between a mobile device 900 and other devices. Mobile device 900 may access a wireless network based on a communications standard, such as WiFi, 2G or 3G, or a combination thereof. In one exemplary embodiment, data transfer component 916 receives a broadcast signal or a broadcast, associated information from

20 external broadcast control system through a broadcast channel. In one embodiment, the data transfer component 916 comprises a near field communication (NFC) module to facilitate near communication. For example, an NFC module can be based on radio frequency identification (RFID) technology, a data transmission association technology in infrared

25 band (IrDA), ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[87] In an exemplary embodiment, the memory 904 includes instructions that are executed by the processor 918 of the mobile device 900 for implementing the automated authentication methods described above for the user based on his signature. For example, a non-volatile computer-readable medium may be a ROM, a random access memory (RAM), a compact disk, a magnetic tape, floppy disks, optical data storage devices, and the like.

Claims

FORMULA

1. The method of automated user authentication based on his signature, carried out through the movement in the air of a mobile device, comprising the following steps:

5 · get the first set of values of the acceleration of the mobile device of the user at each point in time through the accelerometer of the mobile device in the form of a digital signal;

• they process the digital signal for the first set of acceleration values of the mobile device obtained in the previous step;

• get the second set of values of the acceleration of the user's mobile device at each point in time through the accelerometer of the mobile device in the form of a digital signal;

• process the digital signal for the second set of 15 values of the acceleration of the mobile device obtained in the previous step;

• determine the distance between the first set of values of the accelerations of the mobile device and the second set of values of the accelerations of the mobile device;

20 · compare the distance value determined at the previous step with a predetermined value, and

o if the distance is less than this value, identify the user and allow access to the mobile device; o If the distance is greater than this value, do not 25 identify the user and do not allow access to the mobile device.

2. A method according to claim 1, characterized in that upon receipt of the first set of acceleration values of the mobile device, an accelerometer converts it into an electrical digital signal, which is conditioned.

3. The method according to p. 1, characterized in that it additionally uses a gyroscope with an accelerometer.

4. The method according to claim 1, characterized in that they process the digital signal for the first set of values of the accelerations of the mobile device by means of smoothing and filtering.

5. The method according to p. 4, characterized in that when smoothing the signal using a low pass filter and / or Kalman filter.

6. The method according to claim 4, characterized in that determine the distance between the first set of values of the accelerations of the mobile device and the second set of values of the accelerations of the mobile device through the use of the Euclidean metric.