KR102532516B1 - Person authentication device and person authentication method based on brain-machine interface - Google Patents

Abstract

A personal authentication device based on a brain-machine interface according to an embodiment of the present invention includes a signal acquisition unit that repeatedly acquires EEG signals from a user who has received cognitive information, processes the EEG signals provided from the signal acquisition unit, and processes the EEG signals. A signal processing unit that determines whether the deviations of the EEG signals are within the error range, and repeatedly provides cognitive information to the user until the deviations of the EEG signals converge within the error range, and the user's registered EEG signal converged by the signal processing unit. It may include a database that stores registration information.

Description

Brain-machine interface based personal authentication device and personal authentication method

The present invention relates to a personal authentication device and method, and more particularly, to a brain-machine interface based personal authentication device and method.

Personal authentication using biometric information is attracting attention instead of conventional personal authentication using passwords, security keys, and the like. Biometric information does not have the risk of being used for hacking, copying, or theft like passwords, and there is no risk of being altered or lost. Accordingly, biometric information can be usefully used for personal authentication. Among biometric information, a brainwave signal may be used for personal authentication.

EEG signals may be generated when signals are transmitted between nerve cells in the brain and may represent electrical signals generated according to brain activity. Characteristics of EEG signals, such as frequency, amplitude, and phase, can be changed according to brain activity. Individuals may have unique EEG signal characteristics. The characteristics of these EEG signals can be used to authenticate individuals.

The present invention is to solve the above technical problem, and the present invention can provide a brain-machine interface based personal authentication device and personal authentication method.

A personal authentication device based on a brain-machine interface according to an embodiment of the present invention includes a signal acquisition unit that repeatedly acquires EEG signals from a user who has received cognitive information, processes the EEG signals provided from the signal acquisition unit, and processes the EEG signals. A signal processing unit that determines whether the deviations of the EEG signals are within the error range, and repeatedly provides cognitive information to the user until the deviations of the EEG signals converge within the error range, and the user's registered EEG signal converged by the signal processing unit. It may include a database that stores registration information.

A brain-machine interface-based personal authentication method according to another embodiment of the present invention includes the steps of repeatedly acquiring brainwave signals from a user who has received cognitive information by a signal acquisition unit, by a signal processing unit, from the signal acquisition unit. Processing the provided EEG signals, determining, by the signal processing unit, whether the deviations of the EEG signals are within the error range, by the signal processing unit, providing cognitive information to the user until the deviations of the EEG signals converge within the error range. It may include repeatedly providing information to the user, and storing registration information of the user's registered EEG signal converged by the signal processing unit in a database.

According to an embodiment of the present invention, a personal authentication device based on a brain-machine interface may authenticate a user using an EEG signal of the user. Thus, the user does not have to remember passwords or security keys.

1 is a diagram showing an example of a brain-machine interface based personal authentication device and a user according to an embodiment of the present invention.
FIG. 2 is a waveform showing EEG signals generated in the brain of the user of FIG. 1 as an example.
3 and 4 are diagrams showing results of each user registered in the personal authentication device of FIG. 1 by way of example.
FIG. 5 is a diagram exemplarily illustrating a case in which a user who is not registered in the personal authentication device of FIG. 1 requests authentication.
6 is a flowchart illustratively illustrating a process of registering a user in a personal authentication device according to an embodiment of the present invention.
7 is a flowchart illustratively illustrating a process of registering a user in a personal authentication device according to another embodiment of the present invention.
8 is a flowchart illustratively illustrating a process of authenticating a user by a personal authentication device according to an embodiment of the present invention.

In the following, embodiments of the present invention will be described clearly and in detail to the extent that a person skilled in the art can easily practice the present invention.

A brain-machine interface (BMI) or brain-computer interface (BCI) can connect a human or animal brain and an external device such as a machine or computer. The brain-machine interface can be used to directly or indirectly acquire an EEG signal generated according to a user's intention, extract features or characteristics of the acquired EEG signal, and control a device based on the extracted characteristics or characteristics. there is. The present invention relates to a brain-machine interface based personal authentication device and personal authentication method. The personal authentication device according to an embodiment of the present invention may authenticate the user using the user's EEG signal. Thus, the user does not have to remember passwords or security keys.

1 is a diagram showing an example of a brain-machine interface based personal authentication device and a user according to an embodiment of the present invention. The user 10 may receive cognitive information from the personal authentication device 100 and generate an EEG signal. The personal authentication device 100 may acquire the user's EEG signal and authenticate the user 10 using the EEG signal. The personal authentication device 100 may include a signal acquisition unit 110 , a signal processing unit 120 , and a database 130 .

The signal acquisition unit 110 may acquire an EEG signal from the user 10 receiving cognitive information. In an embodiment, the user 10 may repeatedly receive cognitive information from the personal authentication device 100 and repeatedly generate brain wave signals. Cognitive information may also be referred to as input stimuli to the user 10 . The signal acquiring unit 110 may repeatedly acquire EEG signals repeatedly generated in the brain of the user 10 . For example, deviations of EEG signals repeatedly generated in the brain of the user 10 may gradually decrease.

Types of EEG signals may be classified according to how close the distance between the brain of the user 10 and electrodes receiving the EEG signals is. For example, the EEG signal may include electrodes implanted in the brain of the user 10, electrodes attached to the epidermis of the brain of the user 10 for electrocorticography (ECOG) signal acquisition, and electroencephalography (EEG) signal acquisition. This may be generated by any one of the electrodes attached to the scalp of the user 10 (eg, the electrodes of the cap-shaped EEG collection device). As the distance between the user 10's brain and the electrodes is closer, the reliability or resolution of the EEG signal may increase. For example, the signal acquiring unit 110 may acquire an EEG signal from any one of the electrodes described above. In an embodiment, although not shown in FIG. 1 , the personal authentication device 100 may include any one of the electrodes described above and may include an EEG collection device capable of receiving EEG of the user 10. . Alternatively, the personal authentication device 100 may communicate with an external EEG collection device.

The signal processor 120 may process EEG signals provided from the signal acquisition unit 110 . The signal processing unit 120 may include a preprocessing unit 121 , an extraction unit 122 , a classification unit 123 , and a comparison unit 124 .

The preprocessor 121 may receive the EEG signals from the signal acquisition unit 110 and may preprocess them. The preprocessor 121 may amplify the received EEG signals. The preprocessor 121 may remove noise from the received EEG signals. The pre-processor 121 may convert the amplified or filtered EEG signals into digital signals. For example, although not shown in FIG. 1 , the preprocessor 121 may include an amplifier, a filter, an analog-to-digital converter (ADC), and the like for preprocessing the received EEG signals.

The extractor 122 may extract features or characteristics of the EEG signals preprocessed by the preprocessor 121 . For example, characteristics of EEG signals may represent changes in voltage, current, amplitude, magnitude, frequency, phase, and the like over time. The extractor 122 may periodically or non-periodically extract features of EEG signals. The characteristics of the EEG signals extracted by the extractor 122 may be unique information for each individual. For example, the extractor 122 may extract at least one value among voltage, current, amplitude, amplitude, frequency, and phase of the EEG signal at points in time. Values extracted by the extractor 122 may be digital values of at least one bit or more. Intervals of viewpoints may or may not be constant.

The classification unit 123 may classify EEG signals based on the extraction result of the extraction unit 122 and the cognitive information. For example, one or more users 10 may be registered in the personal authentication device 100, and the classification unit 123 may classify brainwave signals according to the users 10. The classification unit 123 may classify EEG signals according to characteristics of EEG signals, such as voltage, current, amplitude, magnitude, frequency, and phase. The classification unit 123 may classify EEG signals according to cognitive information provided to the user 10 . Cognitive information and EEG signals may be mapped by the classification unit 123 . The classification unit 123 may classify EEG signals according to conditions other than the above-described exemplary conditions.

The extraction result of the extraction unit 122 and the classification result of the classification unit 123 may be stored in the database 130 . The brain wave signal of the user 10 may be sequentially processed by the pre-processing unit 121, the extraction unit 122, and the classifying unit 123 of the signal processing unit 120, and the information of the processed brain wave signal (i.e., extraction results and classification results) may be stored in the database 130 .

For example, EEG signals generated from the user 10 receiving cognitive information (or input stimuli) for the first time or several times may have relatively low consistency or convergence. If the consistency or convergence of EEG signals is low, these EEG signals may not be suitable for authenticating the user 10 . Accordingly, the signal processor 120 may repeatedly provide or transmit cognitive information to the user 10 in order to improve consistency or convergence of the EEG signals. The personal authentication device 100 may further include a device, a circuit, and the like for providing recognition information to the user 10 .

The signal processing unit 120 may determine whether deviations of EEG signals are within an error range. The signal processor 120 may repeatedly provide cognitive information to the user 10 until the deviations of the EEG signals converge within an error range. For example, the signal processing unit 120 may provide the same recognition information to the user 10 dozens or hundreds of times. In summary, the signal processing unit 120 may repeatedly provide cognitive information to the user 10 and train the user 10 so that EEG signals having relatively high consistency or convergence are generated in the brain of the user 10 . The user 10 trained by the personal authentication device 100 may generate an EEG signal identical to or similar to an EEG signal converged through training in a future authentication process.

The signal processing unit 120 may repeatedly receive and process EEG signals from the user 10 who repeatedly receives cognitive information. The signal processing unit 120 may converge deviations of EEG signals within an error range and store information of the converged EEG signals in the database 130 . Here, the EEG signal converged by the signal processing unit 120 may be a unique registered EEG signal of the user 10, and the information of the converged EEG signal may be registered EEG information. The margin of error may be determined in advance and stored in the database 130 . The error range may be changed by the signal processor 120. As the error range is smaller, the number of times the signal processing unit 120 provides the recognition information to the user 10 may increase, but the accuracy or reliability of authentication may increase.

The comparator 124 includes registration information of the registered EEG signal of the user 10 generated by the preprocessor 121, the extractor 122, and the classifier 123 and other users previously stored in the database 130. It is possible to compare registration information of registered EEG signals. The comparison operation of the comparator 124 may be performed before registration information of the registered brain wave signal of the user 10 is finally stored in the database 130 . If the registration information of the user 10's registered EEG signal is stored in the database 130 even though the deviation between the registered EEG signal of the user 10 and the registered EEG signal of another user is within the error range, the signal processing unit 120 may not properly perform authentication of the user 10 in the future. Therefore, the signal processing unit 120 can prevent deviations of registered EEG signals of users registered in the personal authentication device 100 from being within an error range based on the comparison result of the comparison unit 124 .

The database 130 includes the cognitive information provided to the user 10, the extraction result of the extractor 122, the classification result of the classifier 123, the comparison result of the comparator 124, and the registered EEG of the user 10. The registration information of the signal can be stored. The database 130 may store the above-described information generated inside the personal authentication device 100 to register the user 10 . For example, the database 130 may store EEG signal information extracted by the extractor 122 . Information stored in the database 130 may be classified or mapped by the classification unit 123 according to various conditions described above. Information stored in the database 130 may be used to authenticate the user 10 in the future. When registration information of the user 10 is stored in the database 130 , the user 10 may be registered in the personal authentication device 100 .

In an embodiment, the signal acquisition unit 110, the signal processing unit 120, and the database 130 may be integrated in one semiconductor device or independently implemented in several semiconductor devices. For example, the semiconductor device may include a system on chip (SoC), a field-programmable gate array (FPGA), an application specific integrated circuit (ASIC), and the like. For example, the personal authentication device 100 may include a transceiver including a transmitter for providing cognitive information to the user 10 and a receiver for receiving an EEG signal from the user 10 . The receiver and transmitter may be implemented in the form of hardware in the signal acquisition unit 110 and the signal processing unit 120 or in the personal authentication device 100, respectively.

In an embodiment, the signal processing unit 120 may include a central processing unit (CPU), a graphics processing unit (GPU), a vision processing unit (VPU), a neural processing unit (NPU), a digital signal processor (DSP), a SoC, and an ASIC. , FPGA, etc. may be implemented using various semiconductor devices. In particular, the signal processor 120 may include an artificial neural network capable of performing operations of the extractor 122 and the classifier 123 . An artificial neural network may be implemented using the above-described semiconductor devices. The signal processing unit 120 may extract features of EEG signals and classify EEG signals using an artificial neural network.

In an embodiment, the database 130 may be implemented using various storage devices such as a hard disk drive (HDD), a solid state drive (SSD), a server, RAM, and memory capable of storing digital information. can The database 130 may be implemented separately from the signal processing unit 120 in terms of hardware. Of course, the database 130 may also be implemented within the signal processing unit 120 .

FIG. 2 is a waveform showing EEG signals generated in the brain of the user of FIG. 1 as an example. FIG. 2 will be described with reference to FIG. 1 . In the waveform of FIG. 2 , a horizontal axis may represent time, and a vertical axis may represent voltage of an EEG signal. However, the vertical axis may represent current, magnitude, phase, frequency, etc. of the EEG signal. The unit of the vertical axis may be determined according to the type of EEG signal provided to the personal authentication device 100 . In FIG. 2 , it is assumed that the first to fifth EEG signals 21 to 25 are sequentially provided to the personal authentication device 100 . The number of first to fifth EEG signals 21 to 25 generated in the brain of the user 10 is not limited to that shown. For convenience of description and illustration, the first to fifth EEG signals 21 to 25 are illustrated to overlap each other.

As described above, the signal processing unit 120 may provide recognition information for registration to the user 10 . The user 10 may recognize cognitive information and generate a first EEG signal 21 according to the cognitive information. The signal processing unit 120 may receive and process the first EEG signal 21 . Thereafter, the signal processor 120 may provide the same recognition information to the user 10 again. The user 10 may recognize the cognitive information again and generate the second EEG signal 22 according to the cognitive information. The signal processor 120 may receive and process the second EEG signal 22 .

The signal processor 120 may determine whether a deviation between the first EEG signal 21 and the second EEG signal 22 is within an error range. For example, the first EEG signal 21 and the second EEG signal 22 may be generated from the user 10 receiving cognitive information once or twice. Since the first EEG signal 21 and the second EEG signal 22 are generated from the user 10 who has received cognitive information several times, the difference between the first EEG signal 21 and the second EEG signal 22 is an error. May be out of range (dotted line shown). The signal processing unit 120 may repeatedly provide the same cognitive information to the user 10 until the deviation of the EEG signals is within an error range.

The signal processing unit 120 may receive and process the third to fifth EEG signals 23 to 25 . For example, the third to fifth EEG signals 23 to 25 may be generated from the user 10 having sufficiently received cognitive information until the EEG signals converge. A deviation of the third to fifth EEG signals 23 to 25 may be within an error range (shown by a solid line). In an embodiment, the signal processing unit 120 may select one of the third to fifth EEG signals 23 to 25 as the registered EEG signal of the user 10 . In another embodiment, the signal processing unit 120 may select the fifth EEG signal received last as the registered EEG signal of the user 10 . In another embodiment, the signal processing unit 120 may calculate or calculate an average feature of features of the third to fifth EEG signals 23 to 25 . The signal processing unit 120 may select the average EEG signal having the above-described average characteristics as the registered EEG signal.

3 and 4 are diagrams showing the result of each user being registered in the personal authentication device of FIG. 1 as an example. 3 and 4 will be described with reference to FIG.

The signal processing unit 120 of the personal authentication device 100 may repeatedly provide the same recognition information to the users A and B. The signal processor 120 may generate cognitive information including stimulus information for at least one of sight, hearing, and smell. For example, the cognitive information may include images having shapes of numbers 1 to 3. However, the number of images provided to the users A and B, the shape of the images, and the order of the images are not limited to those shown in FIGS. 3 and 4 . Authentication accuracy or reliability may improve as the number of images provided to the users A and B increases.

Registration information of registered EEG signals of users A and B for cognitive information may be stored in the database 130 of the personal authentication device 100 . Users A and B may sequentially recognize images having shapes of numbers 1 to 3 and repeatedly provide EEG signals for each image to the personal authentication device 100 .

Referring to FIG. 3 , the signal processing unit 120 uses registration information of registered EEG signals for the images, based on EEG signals generated when the user A recognizes images having the shapes of numbers 1 to 3, respectively. Can be stored in the database 130, respectively. The signal processing unit 120 may store information on error ranges of user A's registered EEG signals in the database 130, respectively. The registered EEG signals of the user A stored in the database 130 may be classified by the signal processing unit 120 to correspond to any one of the images having the shapes of numbers 1 to 3.

Similarly, referring to FIG. 4 , the signal processing unit 120 generates registered EEG signals for the images based on EEG signals generated when the user B recognizes images having the shapes of numbers 1 to 3, respectively. Registration information may be respectively stored in the database 130 . The signal processing unit 120 may store information on error ranges of the registered EEG signals of the user B in the database 130, respectively. The registered EEG signals of the user B stored in the database 130 may be classified by the signal processing unit 120 to correspond to any one of the images having the shapes of numbers 1 to 3.

Referring to FIGS. 3 and 4 , users A and B recognize the same cognitive information, but EEG signals generated by users A and B may be different from each other. That is, registration information for the user A who recognized the same recognition information and registration information for the user B who recognized the same recognition information may be different from each other. More specifically, a deviation between the registered EEG signal of user A and the registered EEG signal of user B may be outside the error range. The error range may be the error range of the user A's registered EEG signal or the error range of the user B's registered EEG signal. The personal authentication device 100 may authenticate users A and B, respectively, based on differences in registration information.

FIG. 5 is a diagram exemplarily illustrating a case in which a user who is not registered in the personal authentication device of FIG. 1 requests authentication. FIG. 5 will be described with reference to FIGS. 3 and 4 . In FIG. 5 , it is assumed that users A and B are each registered with the personal authentication device 100 and user C is not registered with the personal authentication device 100 . That is, registration information of registered EEG signals for users A and B is stored in the database 130, but registration information of registered EEG signals for user C is not stored.

The user C may request authentication from the personal authentication device 100 . The personal authentication device 100 may provide recognition information to the user C who has requested authentication. Here, the recognition information provided to the user C may be the same as the recognition information provided to the users A and B to register the users A and B. More specifically, the number, shape, and order of images provided to user C are the number, shape, and order of images provided to users A and B for registering users A and B, respectively. same.

Illustratively, in FIG. 5 , the personal authentication device 100 first provides an image having a shape of the number 1 to the user C. If the authentication brain wave signal of the user C for the image having the shape of number 1 is valid, the personal authentication device 100 sequentially sends the image having the shape of number 2 and the image having the shape of number 3 to the user (C). ) will be provided.

Referring to FIG. 5 , the user C may recognize an image having the shape of the number 1. The signal processing unit 120 of the personal authentication device 100 may receive an authentication brainwave signal from the user C. The signal processing unit 120 may receive registration information about the registered EEG signals of the users A and B and the error range from the database 130 . The signal processing unit 120 may determine whether a deviation between the authentication EEG signal of the user C and the registered EEG signal of the user A is within an error range. The signal processing unit 120 may also determine whether a deviation between the user C's authentication EEG signal and the user B's registered EEG signal is within an error range. Of course, the signal processing unit 120 may further perform a comparison operation between the registered EEG signals of other users registered in the database and the authentication EEG signal of the user C.

Illustratively, in FIG. 5 , the user C's authentication EEG signal and the user A's registered EEG signal are shown. Since the registration information of the registered EEG signal for the user C is not stored in the database 130, the deviation between the user C's authentication EEG signal and any registered user's registered EEG signal will be outside the margin of error. The authentication EEG signal of user C and the registered EEG signal of any registered user do not match each other. Accordingly, the signal processing unit 120 may provide the user C with an authentication result indicating authentication failure and reject or ignore the user C's authentication request. If the users A and B request authentication, the signal processing unit 120 may provide authentication results indicating successful authentication to the users A and B, respectively.

6 is a flowchart illustratively illustrating a process of registering a user in a personal authentication device according to an embodiment of the present invention. FIG. 6 will be described with reference to FIG. 1 .

In step S110, the user 10 may request registration from the personal authentication device 100. The signal processor 120 of the personal authentication device 100 may provide recognition information for registration to the user 10 . As described above, the cognitive information may include stimulus information for at least one of visual, auditory, and olfactory senses. If another user requests registration with the personal authentication device 100, the signal processing unit 120 may provide the same recognition information as that provided to the user 10 to the other user.

In step S120, the signal acquisition unit 110 may acquire an EEG signal generated from the user 10 who has recognized the cognitive information. The signal processor 120 may amplify or filter the EEG signal acquired by the signal acquisition unit 110 . The signal processing unit 120 may extract and classify features of the preprocessed EEG signal. The signal processing unit 120 may store the EEG signal extraction result and classification result in the database 130 .

In step S130, the signal processing unit 120 may determine whether deviations of EEG signals are within an error range. Here, the brain wave signals may include the brain wave signal acquired in step S120 and the brain wave signal previously acquired. The error range is a predetermined value and may be previously stored in the database 130 . If the deviations of the EEG signals are within the error range, step S140 proceeds. If the deviations of the EEG signals are outside the error range, step S110 is performed again. That is, the signal processor 120 may repeatedly provide cognitive information to the user 10 until the deviations of the EEG signals converge within an error range. Through this, the user 10 may be trained by the personal authentication device 100 to generate EEG signals whose deviations are within an error range.

In step S140 , the signal processing unit 120 may not immediately store the registration information of the user 10's registered EEG signal in the database 130 . The registered EEG signal is an EEG signal converged by the signal processing unit 120 . The signal processor 120 may determine whether a deviation between the registered EEG signal of another user and the registered EEG signal of the user 10 previously stored in the database 130 is outside the error range. If the above deviation is outside the error range, step S150 may proceed. Otherwise, step S160 may proceed.

In step S150 , the signal processing unit 120 may store registration information of the registered EEG signal of the user 10 in the database 130 . For example, the registration information may include arbitrary values such as voltage, current, amplitude, amplitude, frequency, and phase of the registered EEG signal and a value of an error range for the registered EEG signal.

In step S160, the signal processing unit 120 may change or reduce the error range for the user 10. The signal processing unit 120 may change or reduce the value of the error range stored in the database 130 . After step S160 (ie, after the error range is adjusted), steps S110 to S140 may be performed again. A deviation between the EEG signal of the user 10 newly converged through steps S110 to S140, that is, the new registered EEG signal and the previously registered EEG signal of another user stored in the database 130 may be outside the error range. Thereafter, in step S150 , the signal processing unit 120 may store new registration information of the user 10's new registered EEG signal in the database 130 .

7 is a flowchart illustratively illustrating a process of registering a user in a personal authentication device according to another embodiment of the present invention. FIG. 7 will be described with reference to FIGS. 1 and 6 . Steps S210 to S250 are substantially the same as steps S110 to S150.

In step S260, the signal processing unit 120 may change recognition information about the user 10. The signal processing unit 120 may change recognition information stored in the database 130 . In a future authentication process, the recognition information provided to the user 10 and the recognition information provided to other users may be different. After step S260 (ie, after the recognition information is changed), steps S210 to S240 may proceed again. A deviation between the EEG signal of the user 10 newly converged through steps S110 to S140, that is, the new registered EEG signal and the previously registered EEG signal of another user stored in the database 130 may be outside the error range. Thereafter, in step S250 , the signal processing unit 120 may store new registration information of the user 10's new registered EEG signal in the database 130 . In this case, recognition information provided to users registered in the personal authentication device 100 may be different from each other.

8 is a flowchart illustratively illustrating a process of authenticating a user by a personal authentication device according to an embodiment of the present invention. FIG. 8 will be described with reference to FIG. 1 .

In step S310, the user 10 may request authentication from the personal authentication device 100. The signal processing unit 120 of the personal authentication device 100 may provide recognition information for authentication to the user 10 . The recognition information provided to the user 10 in step S310 and the recognition information provided to the user 10 in steps S110 and S210 may be the same. The recognition information provided to the user 10 and the recognition information provided to other users may be the same. If the recognition information provided to the user 10 is changed in step S260, the recognition information provided to the user 10 and the recognition information provided to other users may be different.

In step S320, the signal acquisition unit 110 may acquire an authentication EEG signal generated from the user 10 who has recognized the cognitive information. The signal processor 120 may amplify or filter the authentication EEG signal acquired by the signal acquisition unit 110 . The signal processing unit 120 may extract and classify features of the preprocessed authentication EEG signal.

In step S330, the signal processor 120 may compare authentication information of the authentication EEG signal with registration information of the registered EEG signal. Authentication information of the authentication EEG signal may be generated by the extraction unit 122 and the classification unit 123 of the signal processing unit 120 . Registration information of the registration brain wave signal was stored in the database 130 in steps S150 and S250. The signal processing unit 120 may determine whether a deviation between the registered EEG signal and the certified EEG signal stored in the database 130 is within an error range in steps S150 or S250. Here, registration information of registered EEG signals stored in the database 130 is for at least one user.

In step S340, the signal processing unit 120 may provide or feedback an authentication result indicating successful authentication to the user 10 when the difference between the authentication EEG signal and the registered EEG signal is within an error range. In step S350, the signal processing unit 120 may provide or feedback an authentication result indicating authentication failure to the user 10 when the deviation between the authentication EEG signal and the registered EEG signal is outside the error range.

In an embodiment, the personal authentication device 100 may include a feedback device for providing an authentication result to the user 10 . For example, the feedback device may be a display device. In another embodiment, the personal authentication device 100 may further include an application interface unit for providing authentication results to other electronic devices. Upon receiving the authentication result, the electronic device may provide the authentication result to the user 10 . Alternatively, the electronic device may determine whether to process commands requested by the user 10 based on the authentication result.

What has been described above are specific examples for carrying out the present invention. The present invention will include not only the above-described embodiments, but also embodiments that can be simply or easily changed in design. In addition, the present invention will also include techniques that can be easily modified and implemented in the future using the above-described embodiments.

10: user;
100: personal authentication device;
110: signal acquisition unit;
120: signal processing unit;
130: database;

Claims (18)

In the personal authentication device including a signal acquisition unit, a signal processing unit, and a database,
The signal processing unit repeatedly provides first cognitive information to the user until first EEG signals converged within a first error range are obtained from the user,
The signal acquisition unit receives the first EEG signals from the user generating the first EEG signals converged within the first error range in response to the first cognitive information;
The signal acquisition unit selects a first registered EEG signal based on the first EEG signals;
When the difference between the second registered EEG signal and the first registered EEG signal stored in the database is outside the first error range, the signal processing unit stores registration information of the first registered EEG signal in the database, and
When the difference between the second registered brain wave signal and the first registered brain wave signal stored in the database is within the first error range, the signal processing unit changes the first error range into a second error range,
A deviation between the first registered brain wave signal and the second registered brain wave signal is outside the second error range. According to claim 1,
The signal acquisition unit is implanted in the user's brain, an electrode attached to the epidermis of the user's brain for electrocorticography (ECOG) signal acquisition, and an electrode attached to the user's scalp for EEG (electroencephalography) signal acquisition. A personal authentication device that repeatedly acquires the first brain wave signals from one of the attached electrodes. According to claim 1,
The signal processing unit:
a pre-processing unit receiving the first EEG signals from the signal acquisition unit, amplifying the first EEG signals, and removing noise from the EEG signals; and
and an artificial neural network extracting features of the first brain wave signals and classifying the first brain wave signals based on the first cognitive information. delete delete delete delete According to claim 1,
The recognition information includes stimulus information for at least one of visual, auditory, and olfactory senses. According to claim 1,
When the registration information of the first registered brain wave signal of the user is stored in the database, the signal processing unit provides the first recognition information to the user requesting authentication, and authentication of the authentication brain wave signal provided from the user. A personal authentication device that compares information with the registered information of the first registered brain wave signal, and provides an authentication result to the user based on the comparison result. According to claim 9,
The signal processing unit determines whether a deviation between the authentication brain wave signal and the first registered brain wave signal is within the first error range.
delete delete delete delete delete delete delete delete

Images