KR102298699B1 - Method and apparatus for denoising sleep signal using signal block - Google Patents

Abstract

A technique for a sleep signal noise removal method and apparatus is provided. A sleep signal noise removal method according to an embodiment of the present invention includes: obtaining a sleep signal provided in time series by a sensor unit; dividing into signal blocks, generating first filter information corresponding to the first sleep stage based on a feature of the first sleep stage, and using the first filter information to correspond to the first sleep stage generating second filter information corresponding to the second sleep stage based on a feature of a second sleep stage different from the first sleep stage and removing the noise of the first signal block; The method may include removing noise of a second signal block corresponding to the second sleep phase by using filter information.

Description

Method and device for removing noise from sleep signal using signal block

The present invention relates to a method and apparatus for removing noise present in a sleep signal. More particularly, it relates to a method and apparatus for removing noise from a sleep signal using a plurality of blocks.

A biosignal obtained using a non-invasive method such as a wearable device includes various noises. In particular, since EEG used for sleep signal analysis is generated deep within the body, the probability of attenuation and deformation is very high. However, since the sleep signal may be obtained in various forms according to the sleep stage, there is a limitation in that accurate noise removal on the sleep signal cannot be performed by a uniform noise removal method using a single filter.

Accordingly, there is a need to provide a technology capable of accurately removing only noise included in a sleep signal.
The technology underlying the present invention includes: i) Korean Patent Publication No. 10-1235441 (Invention Title: Bio-signal-based automatic sleep stage classification system), ii) Korean Patent Application Laid-Open No. 10-2017-0142227 ( Title of the invention: Sleep state monitoring method using biosignals) and iii) Korean Patent Publication No. 10-1917855 (Title of the invention: Method of collecting biosignals to determine sleep status and deriving sleep factors and a server using the same) there is

The technical problem to be solved by the present invention relates to a method and apparatus for accurately removing noise from sleep signals obtained in various sleep stages.

The technical problems of the present invention are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art from the following description.

In order to solve the above technical problem, the sleep signal noise removal method according to an embodiment of the present invention includes: acquiring a sleep signal provided in time series by a sensor unit; a time determined based on the user's sleep phase information Splitting the sleep signal into a plurality of signal blocks according to intervals, generating first filter information corresponding to the first sleep stage based on a feature of the first sleep stage, and using the first filter information removing the noise of the first signal block corresponding to the first sleep stage by using generating filter information and removing noise of a second signal block corresponding to the second sleep phase by using the second filter information.

In an embodiment, the step of removing the noise of the third signal block corresponding to the first sleep stage by using the first filter information includes the first signal block and the third signal block corresponding to the first sleep stage. The method may further include updating the first filter information using at least one of and removing noise of a fourth signal block corresponding to the first sleep phase by using the updated first filter information. .

In an embodiment, updating the first filter information includes updating the first filter information by stacking the first signal block and the third signal block according to a specified weight, wherein the weight is Each of the plurality of signal blocks may be determined based on the measured position of the electrode.

In an embodiment, the step of updating the first filter information by stacking the first signal block and the third signal block includes performing frequency conversion of a signal obtained by stacking the first signal block and the third signal block, and , removing noise from the transformed data, inversely transforming the noise-removed data, and updating the first filter information using the inversely transformed data.

In an embodiment, the step of dividing the sleep signal into a plurality of signal blocks may include dividing the sleep signal into a plurality of signal blocks according to a time interval determined based on a period of a sleep signal pattern included in each sleep phase. may include.

In an embodiment, the step of dividing the sleep signal into a plurality of signal blocks may include dividing the sleep signal into a plurality of signal blocks according to a time interval longer than the minimum period of the sleep signal pattern included in the first sleep phase. may include.

In an embodiment, the first filter information includes pattern information and frequency information of a sleep signal included in the first sleep phase, and the second filter information includes a pattern of a sleep signal included in the second sleep phase. information and frequency information.

In an embodiment, the step of removing the noise of the first signal block corresponding to the first sleep phase by using the first filter information includes removing a signal above a threshold frequency from the first signal block, The threshold frequency may be the highest frequency of the sleep signal pattern included in the first sleep phase.

In an embodiment, the step of removing the signal above the threshold frequency from the first signal block includes performing frequency conversion of the first signal block and attenuating a high frequency region above the threshold frequency among the converted data, and the threshold frequency The method may include inversely transforming data in which the above high-frequency region is attenuated.

In order to solve the above technical problem, a sleep signal noise removal device according to another embodiment of the present invention is determined based on a sleep signal acquisition unit that acquires a sleep signal provided in time series by a sensor unit, and sleep phase information of a user A sleep signal division unit that divides the sleep signal into a plurality of signal blocks according to a time interval to be used, and generates first filter information corresponding to the first sleep stage based on a feature of the first sleep stage, and A first noise removing unit that removes noise of a first signal block corresponding to the first sleep stage using first filter information and a feature of a second sleep stage different from the first sleep stage and a second noise removing unit that generates second filter information corresponding to the second sleep stage and removes noise of a second signal block corresponding to the second sleep stage by using the second filter information.

In an embodiment, the first noise removing unit removes noise of a third signal block corresponding to the first sleep stage by using the first filter information, and the first signal block corresponding to the first sleep stage and updating the first filter information using at least one of the third signal blocks, and removing the noise of the fourth signal block corresponding to the first sleep phase using the updated first filter information.

In an embodiment, the first noise removing unit updates the first filter information by stacking the first signal block and the third signal block according to a specified weight, and the weight is measured by each of the plurality of signal blocks It can be determined based on the position of the electrode.

In an embodiment, the first noise removing unit performs frequency conversion of a signal obtained by stacking the first signal block and the third signal block, removes noise of the converted data, and inversely transforms the data from which the noise has been removed and update the first filter information using the inversely transformed data.

In an embodiment, the sleep signal divider may divide the sleep signal into a plurality of signal blocks according to a time interval determined based on a period of a sleep signal pattern included in each sleep phase.

In an embodiment, the sleep signal divider may divide the sleep signal into a plurality of signal blocks according to a time interval longer than the minimum period of the sleep signal pattern included in the first sleep phase.

In an embodiment, the first filter information includes pattern information and frequency information of a sleep signal included in the first sleep phase, and the second filter information includes a pattern of a sleep signal included in the second sleep phase. information and frequency information.

In an embodiment, the first noise removing unit may include removing a signal above a threshold frequency from the first signal block, wherein the threshold frequency may be the highest frequency of a sleep signal pattern included in the first sleep phase. .

In an embodiment, the first noise removing unit is configured to perform frequency conversion of the first signal block, attenuate a high frequency region above the threshold frequency among the converted data, and inversely transform data in which a high frequency region above the threshold frequency is attenuated. can

In order to solve the above technical problem, a recording medium according to another embodiment of the present invention may be a computer-readable program in which a program for performing the noise removal method according to any one of claims 1 to 9 is recorded.

1 is a diagram illustrating an example of a network environment according to an embodiment of the present invention.
2 is a block diagram of an apparatus for removing noise from a sleep signal according to an embodiment of the present invention.
3 is a block diagram for explaining in detail a processor operation of an apparatus for removing noise from a sleep signal according to an embodiment of the present invention.
4 is a flowchart of a sleep signal noise removal method according to an embodiment of the present invention.
5 is a conceptual diagram of a sleep signal noise removal method according to an embodiment of the present invention.
6 is a diagram for explaining in detail a signal block in a sleep signal noise removal method according to an embodiment of the present invention.
7 is a diagram for explaining a method of updating filter information according to an embodiment of the present invention.
8 is a diagram for explaining a method of updating filter information using a signal block having a specified weight according to an embodiment of the present invention.
9 is a diagram for explaining in detail a method of removing a specific frequency region in a method for removing noise from a sleep signal according to an embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. Advantages and features of the present invention, and a method for achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments published below, but may be implemented in various different forms, only these embodiments make the publication of the present invention complete, and common knowledge in the technical field to which the present invention pertains It is provided to fully inform the possessor of the scope of the invention, and the present invention is only defined by the scope of the claims. Like reference numerals refer to like elements throughout.

Unless otherwise defined, all terms (including technical and scientific terms) used herein may be used with the meaning commonly understood by those of ordinary skill in the art to which the present invention belongs. In addition, terms defined in a commonly used dictionary are not to be interpreted ideally or excessively unless clearly defined in particular. The terminology used herein is for the purpose of describing the embodiments and is not intended to limit the present invention. As used herein, the singular also includes the plural unless specifically stated otherwise in the phrase.

Hereinafter, some embodiments of the present invention will be described with reference to the drawings.

1 is a diagram illustrating an example of a network environment according to an embodiment of the present invention.

The network environment of FIG. 1 shows an example including a sleep signal noise removal device 10 , a sleep signal providing server 20 , sleep signal generating terminals 41 , 42 , 43 , 44 , 45 , and a communication network 30 . . FIG. 1 is an example for the description of the invention, and the number of sleep signal acquisition terminals or the number of servers is not limited as in FIG. 1 .

In the sleep signal noise removal system according to an embodiment of the present invention, the sleep signal noise removal device 10 is time-series from the sleep signal providing server 20 or the sleep signal generating terminals 41 , 42 , 43 , 44 , 45 . Receive the generated sleep signal, divide the sleep signal into a plurality of signal blocks based on each sleep stage information, and respond to each sleep stage using filter information generated based on the characteristics of each sleep stage It is possible to remove the noise of the signal block.

The sleep signal providing server 20 may receive the sleep signal generated by the sleep signal generating terminals 41 , 42 , 43 , 44 , 45 through the communication network 30 , and transmit the received sleep signal to the communication network 30 . may be transmitted to the sleep signal noise removal device 10 through the

The sleep signal generating terminals 41 , 42 , 43 , 44 and 45 may sense various biosignals related to the user's sleep signal. Specifically, the sleep signal generating terminals 41, 42, 43, 44, 45 according to some embodiments of the present invention may be a measuring device capable of measuring multiple bio-signals such as a user's pulse, blood pressure, electrocardiogram and brain wave, It may be a stimulation device that transmits a stimulation signal to the user's sleep control brain region. According to an embodiment, the sleep signal generating terminals 41 , 42 , 43 , 44 , and 45 may be wearable devices worn by the user, or devices having a sensor mounted in a shallow area of the body.

As described above, by measuring a signal of a deep area within the body or a shallow area of the body or outside the body, various noises including physical noise may be included in the obtained biosignal. Accordingly, using the sleep signal denoising method according to some embodiments of the present invention, noise removal may be performed on the entire obtained biosignal first, or noise removal may be performed on a partial area requiring noise removal. .

In some embodiments of the present invention, the sleep signal generated by the sleep signal generating terminals 41 , 42 , 43 , 44 and 45 may be directly transmitted to the sleep signal noise removing device 10 through the communication network 30 , but other In the embodiment, the sleep signal is transmitted from the sleep signal generating terminals 41, 42, 43, 44, 45 to the sleep signal providing server 20 through the communication network 30, and after being pre-processed through a predefined algorithm, The sleep signal may be transmitted to the noise canceling device 10 .

The communication method is not limited, and not only a communication method using the communication network 30 (eg, a mobile communication network, a wired Internet, a wireless Internet, a broadcasting network), but also short-range wireless communication between devices may be included. For example, the communication network 30 may include a personal area network (PAN), a local area network (LAN), a campus area network (CAN), a metropolitan area network (MAN), a wide area network (WAN), and a broadband network (BBN). , the Internet, and the like. In addition, the communication network 30 may include any one or more of a network topology including a bus network, a star network, a ring network, a mesh network, a star-bus network, a tree or a hierarchical network, etc. not limited

The sleep signals obtained from the various sleep signal generating terminals 41, 42, 43, 44, 45 include various types of noise, and the noise removed according to a plurality of sleep stages may be different due to the nature of the sleep signal. According to the sleep signal noise removal method according to some embodiments of the present invention, filter information for removing noise may be generated based on each sleep stage information.

Hereinafter, the internal configuration of the sleep signal noise removal apparatus and the sleep signal providing server according to an embodiment of the present invention will be described in detail with reference to FIG. 2 .

The sleep signal noise removal apparatus 10 according to an embodiment may include a memory 11 , a processor 12 , a communication module 13 , and an input/output interface 14 . The memory 11 is a computer-readable recording medium and may include a random access memory (RAM), a read only memory (ROM), and a permanent mass storage device such as a disk drive. Also, a sleep signal to be noise removed, a signal block into which the sleep signal is divided, and filter information for each sleep stage may be temporarily or permanently stored in the memory 11 .

The processor 12 may be configured to process instructions of a computer program by performing basic arithmetic, logic, and input/output operations. The instructions may be provided to the processor 12 by the memory 11 or the communication module 13 . For example, the processor 12 may be configured to execute instructions received according to program code stored in a recording device such as the memory 11 . A detailed description of the processor 12 will be described later with reference to FIG. 3 .

The communication module 13 may provide a function for communicating with the sleep signal providing server 20 through a communication network. For example, a request generated by the processor 12 of the sleep signal noise removal device 10 according to a program code stored in a recording device such as the memory 11 provides a sleep signal through a communication network under the control of the communication module 13 It may be transmitted to the server 20 . Conversely, signals, commands, contents, files, etc. provided under the control of the processor 22 of the sleep signal providing server 20 may be transmitted to the sleep signal noise removal device 10 through the communication module 13 through the communication network. have. For example, a control signal or command of the sleep signal providing server 20 received through the communication module 13 may be transmitted to the processor 12 or the memory 11 , and contents or files may be removed from sleep signal noise. It may be stored in a storage medium that the device 10 may further include.

In addition, the input/output interface 14 may receive a user's input and display output data. The input/output interface 14 according to an embodiment may receive sleep phase information from a user and may output a sleep signal from which noise has been removed.

Also, in other embodiments, the sleep signal noise removal apparatus 10 may include more components than those of FIG. 2 . However, there is no need to clearly show most of the prior art components. For example, the sleep signal noise removal device 10 may include a battery and a charging device for supplying power to internal components of the user terminal, and is implemented to include at least some of the above-described input/output devices or a transceiver ( transceiver), a Global Positioning System (GPS) module, and other components such as a database may be further included.

In an embodiment, the sensor unit 24 included in the sleep signal providing server 20 may sense a biosignal related to the sleep signal, and may receive the biosignal from external devices that sense the biosignal. In addition, in order to clearly explain the features of the present invention, block diagrams of the sleep signal noise removal device 10 and the sleep signal providing server 20 are respectively shown, but the sleep signal noise removal system according to an embodiment of the present invention is shown in Fig. It should be noted that the apparatus for removing noise from the sleep signal 10 and the server for providing the sleep signal 20 may be provided in one physical device without being limited to that shown in FIG. 2 .

Hereinafter, with reference to FIG. 3, the internal configuration of the processor of the terminal or server performing the method for removing noise of a sleep signal according to an embodiment of the present invention will be reviewed in detail. For ease of understanding, a processor to be described below is assumed to be the processor 12 of the sleep signal noise removing device 10 shown in FIG. 2 for ease of understanding. Note that the processor may be the processor 12 of the sleep signal noise providing server 20 shown in FIG. 2 .

The processor 12 of the sleep signal noise providing apparatus 10 according to an embodiment of the present invention includes a sleep signal obtaining unit 51 , a sleep signal dividing unit 52 , a first noise removing unit 53 , and a second noise. and a removal unit 54 . According to some embodiments, components of the processor 12 may be selectively included or excluded from the processor 12 . In addition, according to some embodiments, the components of the processor 12 may be separated or combined to express the functions of the processor 12 .

The processor 12 and components of the processors 12 may control the sleep signal noise removal apparatus 10 to perform steps included in the sleep signal noise removal method of FIG. 4 . For example, the processor 12 and the components of the processor 12 may be implemented to execute instructions according to the code of the operating system included in the memory 11 and the code of at least one program. Here, the components of the processor 12 are functions of different functions of the processor 12 performed by the processor 12 according to instructions provided by the program code stored in the sleep signal noise removal device 10 . can be expressions. The internal configuration and specific operation of the processor 12 will be described with reference to the sleep signal noise removal method of FIG. 4 and the embodiments of FIGS. 5 to 9 .

4 is a flowchart illustrating a method for removing noise from a sleep signal in a time series according to an embodiment of the present invention.

Referring to FIG. 4 , the sleep signal obtaining unit 51 may obtain a sleep signal provided in time series by the sensor unit ( S110 ). The sleep signal may be a signal including a user's biosignal related to sleep and various noises. In particular, since biometric information obtained through the wearable device may be obtained through a sensor located in a shallow place or outside the body, the biosignal obtained through the wearable device may further include physical noise in addition to the sleep signal.

Thereafter, the sleep signal dividing unit 52 may divide the sleep signal into a plurality of signal blocks according to a time interval determined based on the user's sleep stage information. The sleep signal includes a similar sleep signal pattern for each cycle according to the sleep stage. The sleep signal pattern is a characteristic pattern that appears repeatedly according to the sleep stage, and may be detected using an algorithm that detects a change in the size and frequency of the sleep signal. For example, the sleep signal pattern may include a sleep spindle or a K-complex.

Since the characteristics of the sleep signal are different for each sleep phase, the sleep signal noise removal method according to some embodiments of the present invention may improve accuracy by removing noise from the sleep signal based on the sleep phase information. For example, in the case of a sleep phase including a sleep spindle in a sleep signal, a signal block may be generated at a time interval longer than the period of the sleep spindle. Sleep stage can be divided into non-REM sleep stage and REM sleep stage, and non-REM sleep is further divided into three sleep stages (N1 sleep, stage N1; N2 sleep, stage N2; N3 sleep, stage N3) according to the depth of sleep. can A method of generating a signal block according to the steps will be described in detail below with reference to FIG. 6 .

Thereafter, in step S130, a sleep phase corresponding to the noise removal target signal block may be identified.

In an embodiment, when the sleep stage to which the noise removal target signal block corresponds is the first sleep stage, the first noise remover 53 is configured to perform a first sleep stage corresponding to the first sleep stage based on the characteristics of the first sleep stage. Filter information may be generated (S140), and then, noise of the first signal block corresponding to the first sleep phase may be removed using the first filter information (S150).

The first filter information may include information on a sleep signal pattern and frequency information of a sleep signal included in the first sleep phase. In this case, the highest frequency of the sleep signal pattern included in the first sleep phase may be designated as a threshold frequency, and high frequencies above the threshold frequency included in the signal block may be removed as noise. For example, in the case of a sleep phase including sleep spindles in the sleep signal, the threshold frequency may be higher than the highest frequency of the sleep spindles. A method of removing noise from a signal block will be described in detail below with reference to FIG. 9 .

In an embodiment, when the sleep stage to which the noise removal target signal block corresponds is the second sleep stage, the second noise remover 54 is configured to perform a second sleep stage corresponding to the second sleep stage based on the characteristics of the second sleep stage. Filter information may be generated (S160), and then, noise of the second signal block corresponding to the second sleep stage may be removed using the second filter information (S170).

Like the first filter information, the second filter information may include information on a sleep signal pattern and frequency information of a sleep signal included in a second sleep phase different from the first sleep phase. In this case, the highest frequency of the sleep signal pattern included in the second sleep phase may be designated as a threshold frequency, and high frequencies above the threshold frequency included in the signal block may be removed as noise.

In addition, in an embodiment, when the lowest frequency of the sleep signal pattern is designated as the threshold frequency, low frequencies less than the threshold frequency included in the signal block may be removed as noise. For example, in the case of a sleep phase in which a sleep spindle is included in the sleep signal, the threshold frequency may be higher than the highest frequency of the sleep spindle or lower than the lowest frequency of the sleep spindle. A detailed method for removing noise from a signal block will be described in detail below with reference to FIG. 9 .

Hereinafter, a method for removing noise from a sleep signal according to an embodiment of the present invention will be described in detail with reference to FIG. 5 .

First, the sleep signal provided through the sensor unit 24 may be divided into a plurality of signal blocks. In an embodiment, with respect to the sleep signal that is time series data, a sleep signal for an adjacent time of about 30 seconds may be a sleep signal corresponding to the same sleep stage. Accordingly, the sleep signal obtained in some embodiments of the present invention may be divided into a plurality of consecutive signal blocks having a size of 30 seconds.

Thereafter, according to some embodiments of the present invention, a plurality of signal blocks corresponding to the same sleep stage may be grouped. For example, a plurality of signal blocks corresponding to the N1 sleep stage may be grouped into one group, and a plurality of signal blocks corresponding to the N2 sleep stage may be grouped into another group.

In addition, noise of a plurality of signal blocks may be removed by using filter information generated for each sleep stage. According to an embodiment of the present invention, the filter information generated for each sleep stage may be updated using a signal block from which noise has been removed.

Finally, the noise-removed sleep signal may be generated by recombining the noise-removed signal blocks.

Hereinafter, a method of removing noise using different filter information for each sleep stage will be described in detail with reference to FIG. 6 .

First, according to some embodiments of the present invention, a signal block may be divided into different time intervals according to sleep phases. Specifically, for a sleep signal obtained as one time series data, the size of the signal block corresponding to the sleep signal 101 of the first sleep stage, the size of the signal block corresponding to the sleep signal 102 of the second sleep stage, The size of the signal block corresponding to the sleep signal 103 of the third sleep stage and the size of the signal block corresponding to the sleep signal 104 of the fourth sleep stage may be different from each other.

For example, the sleep signal corresponding to the N1 sleep stage may be divided into 30 seconds, which is a time interval in which the same sleep phase is usually continued, and the sleep signal corresponding to the N2 sleep stage is 0.6 seconds longer than the duration of the sleep spindle. The sleep signal corresponding to the N3 sleep stage can be divided into a minimum size of 6 seconds to check whether the low-frequency signal continues, and the REM sleep stage is randomly generated by a low-amplitude sawtooth wave. Since it also appears, the sleep signal corresponding to the REM sleep stage may be divided into 5 seconds.

In this way, when the sleep signal is divided into different time intervals according to the sleep phase, a plurality of signal blocks corresponding to each sleep phase may be generated. Thereafter, according to some embodiments of the present invention, noise may be removed from the plurality of signal blocks by using filter information corresponding to each sleep stage.

In more detail, noise may be removed from the plurality of signal blocks corresponding to the first sleep stage by using the first filter information corresponding to the first sleep stage (111), and a plurality of signal blocks corresponding to the second sleep stage may be removed. In the signal block, noise may be removed using second filter information corresponding to the second sleep stage ( 112 ), and the plurality of signal blocks corresponding to the third sleep stage is a third signal block corresponding to the third sleep stage. Noise may be removed using the filter information (113), and a plurality of signal blocks corresponding to the fourth sleep stage may be noise removed using the fourth filter information corresponding to the fourth sleep stage (114). ).

In an embodiment, in the case of the N1 sleep phase, the brain waves appear within the range of alpha waves (8-13 Hz) to theta waves (4-7 Hz). Accordingly, by setting the threshold frequency to 14 Hz, high frequencies of 14 Hz or higher included in the signal block may be removed.

In an embodiment, in the case of the N2 sleep stage, the EEG includes a sleep spindle and a K complex wave in the range of 11 to 16 Hz. Therefore, by setting the threshold frequency to 17 Hz, high frequencies of 17 Hz or more included in the signal block can be removed.

In an embodiment, in the case of the N3 sleep phase, since the EEG mainly includes low-frequency signals, the high frequency of 3 Hz or more included in the signal block may be removed by setting the threshold frequency to 3 Hz.

In an embodiment, in the case of the REM sleep phase, a high frequency signal similar to the N1 sleep phase and similar to the N1 sleep phase is observed, so that the threshold frequency is set to 14 Hz, so that high frequencies of 14 Hz or more included in the signal block may be removed.

7 is a diagram for explaining a method of updating filter information according to an embodiment of the present invention.

In an embodiment of the present invention, the first noise removing unit 53 may first remove the noise of the third signal block corresponding to the first sleep stage by using the first filter information. In an embodiment, the third signal block may be a signal block obtained after the first signal block corresponding to the first sleep phase.

Thereafter, the first noise removing unit 53 may update the first filter information by stacking the first signal block and the third signal block. In one embodiment, the first noise removing unit 53 corresponds to the first sleep phase by stacking the first signal block and the third signal block according to a weight specified based on the position of the measuring electrode measuring the sleep signal. The first filter information to be used may be updated. A detailed description will be given later with reference to FIG. 8 .

Then, the first noise removing unit 53 may remove the noise of the fourth signal block corresponding to the first sleep stage by using the updated first filter information.

Hereinafter, a method of updating filter information using a signal block having a specified weight according to an embodiment of the present invention will be described in detail with reference to FIG. 8 .

The sleep signal 101 corresponding to the first sleep stage obtained according to an embodiment of the present invention may be a signal measured using one or more electrode signals.

When the sleep signal is measured using a plurality of electrodes, the magnitude of the sleep signal varies according to the position of the measuring electrode. Therefore, according to some embodiments of the present invention, a process of normalizing the magnitudes of the plurality of sleep signals by giving different weights to each of the plurality of sleep signals obtained using the plurality of electrodes is required.

For example, when the first signal block 121 and the third signal block 122 corresponding to the first sleep phase are measured using different electrodes, the first signal block and the third signal block are determined by a specified weight. can be stacked accordingly.

After performing frequency conversion, noise may be removed from the stacked signal blocks, and the noise-removed frequency data may be inversely transformed to finally obtain a noise-removed sleep signal.

Hereinafter, a method for removing a specific frequency region in the sleep signal noise removal method according to an embodiment of the present invention will be described in detail with reference to FIG. 9 .

A plurality of signal blocks 210 corresponding to a specific sleep phase may be stacked according to weights designated based on the positions of the measurement electrodes. Thereafter, frequency conversion may be performed on the data 220 obtained by stacking signal blocks. The frequency transform may be, for example, a Discrete Fourier Transform (DFT) transform.

The noise removal method according to some embodiments of the present invention may be performed on the data 230 on which the frequency conversion of the signal block is performed. In an embodiment, attenuation or removal may be performed for a high frequency region greater than or equal to a maximum threshold frequency, and attenuation or removal may be performed for a low frequency region less than a minimum threshold frequency with respect to the data 230 on which the frequency conversion has been performed.

As described above, in the case of the N1 sleep phase, in the case of the N1 sleep phase, the high frequency of 14 Hz or more included in the signal block may be removed by designating the threshold frequency as 14 Hz, and in the case of the N2 sleep phase, the critical frequency is designated as 17 Hz and included in the signal block High frequencies above 17 Hz can be removed, and in the case of the N3 sleep stage, by designating the threshold frequency as 3 Hz, high frequencies above 3 Hz included in the signal block can be removed. High frequencies of 14 Hz or higher included in the .

Thereafter, after noise in the high frequency region greater than the maximum threshold frequency and the low frequency region less than the minimum threshold frequency are removed, inverse frequency transformation may be performed on the frequency-converted data 230 .

The sleep signal noise removal method according to another embodiment of the present invention may be performed using a global denoising method. In the present embodiment, the noise-removing record in the same sleep phase among the previous sleep signals on the same day of the user or the sleep signals on different days in the past among a plurality of signal block groups corresponding to a specific sleep phase may be used.

More specifically, the sleep signal noise removal method according to another embodiment of the present invention uses the Gaussian mixture model (GMM) learned from the noise-removed sleep signal patch as a prior probability, and EPLL ( Noise can be removed by selecting a component having the maximum Expected Patch Log Likelihood and projecting a signal block to be noise-removed in the direction of the eigenvalue of the maximum component. In this case, since it is necessary to store the previous sleep signal information, a separate storage device may be further used.

Since the time period in which a specific sleep stage appears among the total sleep time of a person is uniform on average, the sleep signal noise removal method according to the embodiments of the present invention can reduce the amount of computation by selectively searching for prior information of past sleep records, and The accuracy of noise removal can be improved by removing noise from the sleep signal using sleep information of a wider period.

The specific implementations described in the present invention are only examples, and do not limit the scope of the present invention in any way. For brevity of the specification, descriptions of conventional electronic components, control systems, software, and other functional aspects of the systems may be omitted. In addition, the connection or connection members of the lines between the components shown in the drawings exemplarily represent functional connections and/or physical or circuit connections, and in an actual device, various functional connections, physical connections that are replaceable or additional may be referred to as connections, or circuit connections. In addition, unless there is a specific reference such as “essential” or “importantly”, it may not be a necessary component for the application of the present invention.

In the specification of the present invention (especially in the claims), the use of the term “above” and similar referential terms may be used in both the singular and the plural. In addition, when a range is described in the present invention, each individual value constituting the range is described in the detailed description of the invention as including the invention to which individual values belonging to the range are applied (unless there is a description to the contrary). same as Finally, the steps constituting the method according to the present invention may be performed in an appropriate order unless the order is explicitly stated or there is no description to the contrary. The present invention is not necessarily limited to the order in which the steps are described. The use of all examples or exemplary terms (eg, etc.) in the present invention is merely for the purpose of describing the present invention in detail, and unless defined by the claims, the scope of the present invention is limited by the examples or exemplary terminology. it's not going to be In addition, those skilled in the art will recognize that various modifications, combinations, and changes may be made in accordance with design conditions and factors within the scope of the appended claims or their equivalents.

The embodiments according to the present invention described above may be implemented in the form of program instructions that can be executed through various computer components and recorded in a computer-readable recording medium. The computer-readable recording medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded on the computer-readable recording medium may be specially designed and configured for the present invention, or may be known and available to those skilled in the art of computer software. Examples of computer-readable recording media include hard disks, magnetic media such as floppy disks and magnetic tapes, optical recording media such as CD-ROMs and DVDs, and magneto-optical media such as floppy disks. medium), and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. A hardware device may be converted into one or more software modules to perform processing in accordance with the present invention, and vice versa.

In the above, the present invention has been described with reference to specific matters, such as specific components, and limited embodiments and drawings, but these are only provided to help a more general understanding of the present invention, and the present invention is not limited to the above embodiments. Those of ordinary skill in the art to which the invention pertains can make various modifications and changes from these descriptions.

Therefore, the spirit of the present invention should not be limited to the above-described embodiments, and the scope of the spirit of the present invention is not limited to the scope of the scope of the present invention. will be said to belong to

Claims (19)

A sleep signal noise removal method performed by a computing device, the method comprising:
acquiring a sleep signal provided in time series by a sensor unit;
dividing the sleep signal into a plurality of signal blocks according to a time interval determined based on the user's sleep phase information;
First filter information corresponding to the first sleep stage is generated based on a feature of the first sleep stage, and noise of a first signal block corresponding to the first sleep stage using the first filter information removing the; and
Generates second filter information corresponding to the second sleep stage based on a feature of a second sleep stage different from the first sleep stage, and corresponds to the second sleep stage using the second filter information removing the noise of the second signal block to
How to remove sleep signal noise. According to claim 1,
removing noise of a third signal block corresponding to the first sleep phase by using the first filter information;
updating the first filter information by using at least one of the first signal block and the third signal block corresponding to the first sleep phase; and
The method further comprising: removing noise of a fourth signal block corresponding to the first sleep phase by using the updated first filter information;
How to remove sleep signal noise. 3. The method of claim 2,
Updating the first filter information includes:
and updating the first filter information by stacking the first signal block and the third signal block according to a specified weight,
The weight is determined based on the position of the electrode at which each of the plurality of signal blocks is measured,
How to remove sleep signal noise. 4. The method of claim 3,
Updating the first signal block and the third signal block to update the first filter information,
performing frequency conversion on a signal obtained by stacking the first signal block and the third signal block, and removing noise from the converted data; and
Inversely transforming the noise-removed data, and updating the first filter information using the inverse-transformed data.
How to remove sleep signal noise. According to claim 1,
The step of dividing the sleep signal into a plurality of signal blocks comprises:
Comprising the step of dividing the sleep signal into a plurality of signal blocks according to a time interval determined based on the period of the sleep signal pattern included in each sleep phase,
How to remove sleep signal noise. 6. The method of claim 5,
The step of dividing the sleep signal into a plurality of signal blocks comprises:
Comprising the step of dividing the sleep signal into a plurality of signal blocks according to a time interval longer than the minimum period of the sleep signal pattern included in the first sleep phase,
How to remove sleep signal noise. According to claim 1,
The first filter information,
Includes pattern information and frequency information of the sleep signal included in the first sleep phase,
The second filter information,
Containing pattern information and frequency information of the sleep signal included in the second sleep phase,
How to remove sleep signal noise. 8. The method of claim 7,
The step of removing the noise of the first signal block corresponding to the first sleep phase using the first filter information includes:
removing a signal above a threshold frequency from the first signal block;
The threshold frequency is the highest frequency of the sleep signal pattern included in the first sleep phase,
How to remove sleep signal noise. 9. The method of claim 8,
The step of removing a signal above a threshold frequency from the first signal block comprises:
performing frequency conversion of the first signal block and attenuating a high-frequency region above the threshold frequency among the converted data; and
Inversely transforming data in which a high frequency region above the threshold frequency is attenuated,
How to remove sleep signal noise. a sleep signal acquisition unit for acquiring a sleep signal provided in time series by the sensor unit;
a sleep signal division unit that divides the sleep signal into a plurality of signal blocks according to a time interval determined based on the user's sleep phase information;
First filter information corresponding to the first sleep stage is generated based on a feature of the first sleep stage, and noise of a first signal block corresponding to the first sleep stage using the first filter information a first noise removing unit to remove and
Generates second filter information corresponding to the second sleep stage based on a feature of a second sleep stage different from the first sleep stage, and corresponds to the second sleep stage using the second filter information and a second noise removing unit for removing the noise of the second signal block,
Sleep signal noise canceling device. 11. The method of claim 10,
The first noise removing unit,
The noise of the third signal block corresponding to the first sleep stage is removed using the first filter information, and at least one of the first signal block and the third signal block corresponding to the first sleep stage is used to remove the noise. updating the first filter information and removing noise of a fourth signal block corresponding to the first sleep phase by using the updated first filter information;
Sleep signal noise canceling device. 12. The method of claim 11,
The first noise removing unit,
The first filter information is updated by stacking the first signal block and the third signal block according to a specified weight, and the weight is determined based on the position of the electrode at which each of the plurality of signal blocks is measured.
Sleep signal noise canceling device. 13. The method of claim 12,
The first noise removing unit,
A signal obtained by stacking the first signal block and the third signal block is subjected to frequency conversion, noise of the converted data is removed, the data from which the noise has been removed is inversely transformed, and the second signal is converted using the inversely transformed data. 1 to update filter information,
Sleep signal noise canceling device. 11. The method of claim 10,
The sleep signal dividing unit,
Splitting the sleep signal into a plurality of signal blocks according to a time interval determined based on the period of the sleep signal pattern included in each sleep stage,
Sleep signal noise canceling device. 15. The method of claim 14,
The sleep signal dividing unit,
dividing the sleep signal into a plurality of signal blocks according to a time interval longer than the minimum period of the sleep signal pattern included in the first sleep phase;
Sleep signal noise canceling device. 11. The method of claim 10,
The first filter information,
Includes pattern information and frequency information of the sleep signal included in the first sleep phase,
The second filter information,
Containing pattern information and frequency information of the sleep signal included in the second sleep phase,
Sleep signal noise canceling device. 17. The method of claim 16,
The first noise removing unit,
removing a signal above a threshold frequency from the first signal block;
The threshold frequency is the highest frequency of the sleep signal pattern included in the first sleep phase,
Sleep signal noise canceling device. 18. The method of claim 17,
The first noise removing unit,
performing frequency conversion of the first signal block, attenuating a high frequency region equal to or greater than the threshold frequency among the converted data, and inversely transforming data in which a high frequency region greater than or equal to the threshold frequency is attenuated;
Sleep signal noise canceling device. A computer-readable recording medium in which a program for performing the method for removing noise according to any one of claims 1 to 9 is recorded.

Images