A snoring sound detection method, device, system, and storage medium

Abstract

The application discloses a snoring sound detection method, device, system and storage medium, and belongs to the sleep monitoring field. The vibration sensor is used to collect the vibration signal of a human body, and a breathing signal and a snoring sound envelope signal are extracted from the vibration signal. The power spectrum S xx (f) of the snoring sound envelope signal is calculated yy (f), and the cross power spectrum S xy (f) of the breathing signal and the snoring sound envelope signal is calculated. The coherence function of the breathing signal and the snoring sound envelope signal is calculated, and the maximum value of the coherence function curve in the defined frequency range is compared with a threshold value. When the maximum value is greater than or equal to the threshold value, it is determined that the user snores. When the maximum value is less than the threshold value, it is determined that the user does not snore. Through the above steps, the interference of other snoring sounds and environmental noise is reduced, and it is determined whether the user snores or not.

Description

Technical Field

[0001] This invention relates to the field of sleep monitoring, and in particular to a snoring detection method, device, system, and storage medium. Background Technology

[0002] Snoring is a common sleep phenomenon. During sleep, airflow through the upper respiratory tract impacts the edges of the pharyngeal mucosa and secretions on its surface, causing vibrations and producing snoring sounds. The snoring area extends from the nasopharynx to the epiglottis, including the soft palate, uvula, tonsils, palatopharyngeal arch, palatoglossal arch, base of the tongue, pharyngeal muscles, and pharyngeal mucosa. Snoring can cause repeated pauses in breathing during sleep, leading to cerebral hypoxia and potentially triggering various cardiovascular and cerebrovascular diseases. Due to repeated episodes of intermittent nocturnal hypoxia and disruption of sleep structure, it can cause damage to a range of target organs, including hypertension, coronary heart disease, arrhythmia, pulmonary hypertension and pulmonary heart disease, ischemic or hemorrhagic stroke, metabolic syndrome, psychological abnormalities, and mood disorders. Furthermore, it can cause left ventricular failure, recurrent nocturnal asthma attacks, and in children, OSAHS can lead to developmental delays and intellectual impairment. Some diseases can cause snoring, such as mandibular retrusion, nasal polyps, adenoid hypertrophy, and chronic rhinitis; snoring can also occur after drinking alcohol or taking sedatives and sleeping pills; obesity can cause fat accumulation in the neck, leading to airway narrowing and thus snoring.

[0003] Currently, there are many algorithms and devices for detecting snoring, such as wearable sleep monitoring headphones and polysomnography (PSG). However, wearable devices may have issues such as forgetting to wear them or charging them, while PSG and other devices require a professional physician to wear them in a specific ward for sleep monitoring, which involves problems such as complicated operation, high cost, discomfort affecting sleep, and limited experimental conditions.

[0004] Current methods for determining snoring involve collecting sound through microphones and then comparing the waveforms. However, this method has low accuracy and is easily affected by the snoring of others and environmental noise, leading to measurement errors for the subjects. Furthermore, collecting snoring sounds with microphones involves an intrusion into personal privacy and is not very user-friendly. Summary of the Invention

[0005] In order to overcome the shortcomings of the prior art, one of the objectives of the present invention is to provide a snoring detection method that is easy to use and is not affected by the snoring of other people or environmental noise.

[0006] In order to overcome the shortcomings of the prior art, the second objective of this invention is to provide a snoring detection device that is easy to use and is not affected by the snoring of other people or environmental noise.

[0007] To overcome the shortcomings of the prior art, the third objective of this invention is to provide a snoring detection system that is easy to use and unaffected by the snoring of others or environmental noise.

[0008] To overcome the shortcomings of the prior art, the fourth objective of this invention is to provide a storage medium that is easy to use and unaffected by other people's snoring or environmental noise.

[0009] One of the objectives of this invention is achieved through the following technical solution:

[0010] A snoring detection method includes the following steps:

[0011] Signal acquisition: Vibration signals of the human body are acquired using vibration sensors, and respiratory signals and snoring envelope signals are extracted from the vibration signals;

[0012] Calculate the power spectrum of the signal: Calculate the power spectrum S of the respiratory signal respectively. xx (f) Power spectrum S of snoring envelope signal yy (f) and the cross-power spectrum S of the respiratory signal and the snoring envelope signal. xy (f) Calculate the coherence function of the respiratory signal and the snoring envelope signal based on the power spectrum of the respiratory signal, the power spectrum of the snoring envelope signal, and the cross-power spectrum of the respiratory signal and the snoring envelope signal.

[0013] To determine if someone snores: compare the maximum value of the coherence function curve within a defined frequency range with a threshold. If the maximum value is greater than or equal to the threshold, snoring is determined; if the maximum value is less than the threshold, no snoring is determined.

[0014] Furthermore, in the signal acquisition step, extracting the respiratory signal from the vibration signal specifically involves: smoothing the vibration signal, and then downsampling the signal after smoothing to obtain the respiratory signal.

[0015] Furthermore, in the signal acquisition step, extracting the snoring envelope signal from the vibration signal specifically involves: filtering the vibration signal through a high-pass filter, taking the envelope of the filtered signal using short-time average amplitude or the Hilbert method, and filtering the enveloped signal to make the envelope features more obvious.

[0016] Furthermore, in the step of calculating the signal power spectrum, the power spectrum S of the respiratory signal... xx (f)=X(f)·X * X(f) is the Fourier transform of the respiratory signal x(t), and * denotes complex conjugate.

[0017] Furthermore, in the step of calculating the signal power spectrum, the power spectrum S of the snoring envelope signal... yy(f)=Y(f)·Y * Y(f) is the Fourier transform of the snoring envelope signal y(t), and * denotes complex conjugate.

[0018] Furthermore, in the step of determining whether snoring occurs, the threshold is a value between 0.65 and 0.75.

[0019] Furthermore, considering the periodicity of the respiratory signal and the snoring envelope signal, the defined frequency range is 0.1Hz to 0.55Hz.

[0020] The second objective of this invention is achieved by the following technical solution:

[0021] A snoring detection device is provided for implementing any of the above-described snoring detection methods. The snoring detection device includes a piezoelectric sensor and a processor. The piezoelectric sensor is used for non-contact acquisition of vibration signals from the human body. The processor is communicatively connected to the piezoelectric sensor and calculates the power spectrum S of the respiratory signal. xx (f) Power spectrum S of snoring envelope signal yy (f) and the cross-power spectrum S of the respiratory signal and the snoring envelope signal. xy (f) Calculate the coherence function of the respiratory signal and the snoring envelope signal, and the processor determines whether snoring occurs based on the coherence function.

[0022] The third objective of this invention is achieved by the following technical solution:

[0023] A snoring detection system, used to implement any of the above-described snoring detection methods, includes a signal acquisition module and a signal processing module, wherein the signal processing module is communicatively connected to the signal acquisition module, the signal acquisition module acquires vibration signals of the human body, and the signal processing module calculates the power spectrum S of the respiratory signal based on the vibration signals. xx (f) Power spectrum S of snoring envelope signal yy (f) and the cross-power spectrum S of the respiratory signal and the snoring envelope signal. xy (f) Calculate the coherence function of the respiratory signal and the snoring envelope signal, and the processor determines whether snoring occurs based on the coherence function.

[0024] The fourth objective of this invention is achieved by the following technical solution:

[0025] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is any of the above-described snoring detection methods.

[0026] Compared to existing technologies, the snoring detection method of the present invention utilizes a vibration sensor to collect vibration signals from the human body, extracts respiratory signals and snoring envelope signals from the vibration signals, and calculates the power spectrum S of the respiratory signals respectively. xx (f) Power spectrum S of snoring envelope signal yy (f) and the cross-power spectrum S of the respiratory signal and the snoring envelope signal. xy (f) Calculate the coherence function of the respiratory signal and the snoring envelope signal. Find the maximum value of the coherence function curve within a defined frequency range, compare this maximum value with a threshold, and determine whether snoring occurs when the maximum value is greater than or equal to the threshold; otherwise, determine whether snoring occurs when the maximum value is less than the threshold. By utilizing the synchronous correlation between breathing and snoring, the interference of others' snoring and environmental noise can be reduced, and the user's snoring can be accurately determined. Attached Figure Description

[0027] Figure 1 This is a flowchart of the snoring detection method of the present invention;

[0028] Figure 2 The raw vibration signal detected by the vibration sensor;

[0029] Figure 3 The original snoring signal separated from the original signal;

[0030] Figure 4 The envelope of the snoring signal;

[0031] Figure 5 The smoothed respiratory signal;

[0032] Figure 6 This is the respiratory signal after 10-fold downsampling;

[0033] Figure 7 The processed respiratory signal;

[0034] Figure 8 This is the processed snoring envelope signal;

[0035] Figure 9 A schematic diagram comparing respiratory signals and snoring envelope signals;

[0036] Figure 10 The power spectrum of the respiratory signal;

[0037] Figure 11 The power spectrum of the snoring envelope signal;

[0038] Figure 12 A schematic diagram of the coherence coefficients of the cross-power spectrum of the respiratory signal and the snoring envelope signal;

[0039] Figure 13 for Figure 12 A schematic diagram of the maximum value of the coherent function;

[0040] Figure 14 for Figure 12 A schematic diagram showing the frequency of the maximum value of the coherence function;

[0041] Figure 15 This is a diagram illustrating the results of a data set on whether or not someone snores.

[0042] Figure 16 A schematic diagram of the subject's respiratory signals;

[0043] Figure 17 A schematic diagram of the snoring envelope signal of the interfering party;

[0044] Figure 18 A schematic diagram comparing respiratory signals and the envelope signal of snoring from the interfering party;

[0045] Figure 19 A schematic diagram of the coherence coefficients of the cross-power spectrum of the subject's respiratory signal and the snoring envelope signal of the interfering party;

[0046] Figure 20 This is a schematic diagram for judging snoring under interference conditions. Detailed Implementation

[0047] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0048] It should be noted that when a component is said to be "fixed to" another component, it can be directly on the other component or it can be fixed through another intermediate component. When a component is said to be "connected to" another component, it can be directly connected to the other component or it may be fixed through another intermediate component. When a component is said to be "set on" another component, it can be set directly on the other component or it may be set through another intermediate component. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0049] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0050] Figure 1 The snoring detection method of this application includes the following steps:

[0051] Signal acquisition: Vibration signals of the human body are acquired using vibration sensors, and respiratory signals and snoring envelope signals are extracted from the vibration signals;

[0052] Calculate the power spectrum of the signal: Calculate the power spectrum S of the respiratory signal respectively. xx (f) Power spectrum S of snoring envelope signal yy (f) and the cross-power spectrum S of the respiratory signal and the snoring envelope signal. xy (f) Calculate the coherence function of the respiratory signal and the snoring envelope signal based on the power spectrum of the respiratory signal, the power spectrum of the snoring envelope signal, and the cross-power spectrum of the respiratory signal and the snoring envelope signal.

[0053] To determine if someone snores: compare the maximum value of the coherence function curve within a defined frequency range with a threshold. If the maximum value is greater than or equal to the threshold, snoring is determined; if the maximum value is less than the threshold, no snoring is determined.

[0054] In the signal acquisition step, a vibration sensor is used to achieve non-contact data acquisition. Specifically, in this embodiment, the vibration sensor is a piezoelectric sensor. The vibration sensor can be used independently or in conjunction with a pillow or mattress. The acquired raw signal is as follows: Figure 2 As shown, it includes snoring signals and breathing signals.

[0055] When extracting snoring signals from vibration signals, snoring is mainly concentrated in the high-frequency range. The signal is then passed through a high-pass filter to obtain... Figure 3 The snoring signal shown is then used to obtain its envelope using short-time average amplitude or the Hilbert method, which includes two 10x downsampling processes, to obtain... Figure 4 The envelope of the snoring signal shown is subsequently subjected to a series of filtering operations to make the envelope features more obvious, such as... Figure 8 As shown.

[0056] When extracting respiratory signals from vibration signals, considering that the normal respiratory rate is 12-20 breaths per minute, corresponding to a frequency range of 0.2Hz-0.33Hz, vibration signals within this range can be directly selected. Then, a smoothing process can be performed on the vibration signals to obtain the desired signal. Figure 5 The respiratory signal shown was then downsampled twice by a factor of 10 to obtain the signal shown below. Figure 6 The respiratory signal shown is then stored after being filtered by high-pass and low-pass filters, such as... Figure 7 As shown, the comparison between the respiratory signal and the snoring envelope signal at this time is as follows: Figure 9 As shown.

[0057] In the step of calculating the signal power spectrum, the signal power spectrum is the energy distribution of the signal in the frequency domain. The power spectral density S(f) of the signal is defined as the ratio of the power of the signal at frequency f to the frequency interval Δf, that is:

[0058]

[0059] Here, p(f,Δf) is the power of the signal within the frequency range [f,f+Δf]. In practical calculations, methods such as the periodogram method, Welch method, and AR model can be used to estimate the power spectral density of the signal.

[0060] The respiratory signal is x(t). The Fourier transform of the respiratory signal is X(f), and the power spectrum S of the respiratory signal is... xx (f)=X(f)·X * (f), * indicates complex conjugation. The power spectrum calculation results of the respiratory signal are as follows: Figure 10 As shown.

[0061] The snoring envelope signal is y(t). The Fourier transform of the snoring envelope signal is Y(f), and the power spectrum S of the snoring envelope signal is... yy (f)=Y(f)·Y * (f), * indicates complex conjugation. The power spectrum calculation results of the respiratory signal are as follows: Figure 11 As shown.

[0062] The cross-power spectrum of the respiratory signal and the snoring envelope signal describes the degree of mutual influence between the two signals in the frequency domain. Their cross-power spectrum is expressed as:

[0063] Sxy(f)=lim T→∞ E{X(f)Y * (f)}

[0064] Among them, G xy(f) is the cross power spectral density of signals x(t) and y(t), E{·} denotes the expectation operation, X(f) and Y(f) are the Fourier transforms of signals x(t) and y(t), and * denotes complex conjugate.

[0065] The specific calculation formula is the cross-power spectrum of the two signals: S xy (f)=X(f)·Y * (f) Finally, the coherence function is calculated. coherence function such as Figure 12 As shown. The maximum value of the coherence function and the frequency corresponding to the maximum value are as follows. Figure 13 as well as Figure 14 As shown.

[0066] Because the respiratory signal and the snoring envelope signal have a periodic relationship in the time domain, meaning their peak periods are almost synchronous, snoring often occurs during inhalation. The cross-power spectral amplitude coherence estimate C... xy The range is [0, 1], and the larger the value, the stronger the coherence. The cross-power spectrum amplitude coherence threshold is set to a value within the range of 0.65-0.75. In this embodiment, the cross-power spectrum amplitude coherence threshold is set to 0.7. When the cross-power spectrum amplitude coherence is greater than or equal to 0.7, the subject is judged to be snoring. Figure 15 As shown, when the cross-power spectrum amplitude coherence is less than 0.7, the subject is judged not to snore.

[0067] In testing the snoring detection method of this application, data from two individuals were collected. The subjects' data were preprocessed to obtain respiratory signals, such as... Figure 16 The data from the interfering party is preprocessed to obtain the snoring envelope signal, such as... Figure 17 Plot the time-domain diagrams of the two signals, as shown below. Figure 18 As can be seen, the two signals do not show a clear relationship in terms of period; that is, the periods of their peaks are asynchronous. The cross-power spectra of the two signals are then plotted, as shown below. Figure 19 It can be observed that the cross-power spectrum does not show a peak corresponding to the period, indicating that the correlation between the two signals is not strong. Finally, this set of data was used to determine whether snoring occurs, and the results are as follows: Figure 20 As can be seen, no snoring warning was detected. This demonstrates that everyone's breathing rate differs, and this invention effectively prevents the snoring of others from affecting the snoring detection of the test subjects.

[0068] This application's snoring detection method combines the physiological characteristics of the subject's breathing and snoring. While there is a synchronous correlation between each person's breathing and snoring, the likelihood of others' snoring, one's own or others' voices, or environmental noise being synchronized with one's own breathing is very low. Therefore, it can reduce interference from others' snoring and environmental noise, improving the accuracy of snoring detection. Furthermore, the vibration sensor is a non-contact signal acquisition method, making it convenient to use.

[0069] This application also relates to a snoring detection device for implementing the above-described snoring detection method. The snoring detection device includes a piezoelectric sensor and a processor. The piezoelectric sensor is used for non-contact acquisition of vibration signals from the human body. The processor is communicatively connected to the piezoelectric sensor and calculates the power spectrum S of the respiratory signal. xx (f) Power spectrum S of snoring envelope signal yy (f) and the cross-power spectrum S of the respiratory signal and the snoring envelope signal. xy (f) Calculate the coherence function of the respiratory signal and the snoring envelope signal. The processor determines whether snoring occurs based on the coherence function.

[0070] This application also relates to a snoring detection system for implementing the above-mentioned snoring detection method, including a signal acquisition module and a signal processing module. The signal processing module is communicatively connected to the signal acquisition module. The signal acquisition module acquires vibration signals from the human body, and the signal processing module calculates the power spectrum S of the respiratory signal based on the vibration signals. xx (f) Power spectrum S of snoring envelope signal yy (f) and the cross-power spectrum S of the respiratory signal and the snoring envelope signal. xy (f) Calculate the coherence function of the respiratory signal and the snoring envelope signal. The processor determines whether snoring occurs based on the coherence function.

[0071] This application also relates to a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, describes the above-described snoring detection method.

[0072] Compared to existing technologies, the snoring detection method of this invention utilizes a vibration sensor to collect vibration signals from the human body, extracts respiratory signals and snoring envelope signals from the vibration signals, and calculates the power spectrum S of the respiratory signals respectively. xx (f) Power spectrum S of snoring envelope signal yy (f) and the cross-power spectrum S of the respiratory signal and the snoring envelope signal. xy (f) Calculate the coherence function of the respiratory signal and the snoring envelope signal. The coherence function is compared with a threshold. If the coherence function is greater than or equal to the threshold, snoring is determined; if the coherence function is less than the threshold, no snoring is determined. By taking advantage of the synchronous correlation between breathing and snoring, the interference of other people's snoring and environmental noise can be reduced, and the user's snoring can be accurately determined.

[0073] The above embodiments merely illustrate several implementation methods of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the invention patent. It should be noted that, for those skilled in the art, several modifications and improvements can be made without departing from the concept of the present invention. These are all equivalent modifications and improvements made to the above embodiments based on the essential technology of the present invention, and all of these fall within the protection scope of the present invention.

Claims

1. A method for detecting snoring, characterized in that, Includes the following steps: Signal acquisition: Vibration signals of the human body are acquired using vibration sensors, and respiratory signals and snoring envelope signals are extracted from the vibration signals; Calculate the power spectrum of the signals: Calculate the power spectrum of the respiratory signals respectively. Power spectrum of snoring envelope signal and the cross-power spectrum of respiratory signals and snoring envelope signals. The coherence function curves of the respiratory signal and the snoring envelope signal are calculated based on the power spectrum of the respiratory signal, the power spectrum of the snoring envelope signal, and the cross-power spectrum of the respiratory signal and the snoring envelope signal. ; To determine if someone snores: Find the maximum value of the coherence function curve within a defined frequency range, compare this maximum value with a threshold, and if the maximum value is greater than or equal to the threshold, then snoring is determined. If the maximum value is less than the threshold, it is determined that there is no snoring.

2. The snoring detection method according to claim 1, characterized in that: In the signal acquisition step, extracting the respiratory signal from the vibration signal specifically involves: smoothing the vibration signal, and then downsampling the smoothed signal to obtain the respiratory signal.

3. The snoring detection method according to claim 1, characterized in that: In the signal acquisition step, extracting the snoring envelope signal from the vibration signal specifically involves: filtering the vibration signal through a high-pass filter, taking the envelope of the filtered signal using short-time average amplitude or the Hilbert method, and filtering the signal with the taken envelope to make the envelope characteristics more obvious.

4. The snoring detection method according to claim 1, characterized in that: In the step of calculating the power spectrum of the signal, the power spectrum of the respiratory signal... , Respiratory signals The Fourier transform of , where * denotes complex conjugate.

5. The snoring detection method according to claim 1, characterized in that: In the step of calculating the signal power spectrum, the power spectrum of the snoring envelope signal is... , The envelope signal of snoring The Fourier transform of , where * denotes complex conjugate.

6. The snoring detection method according to claim 1, characterized in that: In the step of determining whether snoring occurs, the threshold is a value within the range of 0.65-0.

75.

7. The snoring detection method according to claim 1, characterized in that: The defined frequency range is 0.1Hz to 0.55Hz.

8. A snoring detection device for implementing the snoring detection method as described in any one of claims 1-7, characterized in that: The snoring detection device includes a piezoelectric sensor and a processor. The piezoelectric sensor is used for non-contact acquisition of vibration signals from the human body. The processor is communicatively connected to the piezoelectric sensor and calculates the power spectrum of the respiratory signal. Power spectrum of snoring envelope signal and the cross-power spectrum of respiratory signals and snoring envelope signals. The processor calculates the coherence function of the respiratory signal and the snoring envelope signal, and determines whether snoring occurs based on the coherence function.

9. A snoring detection system for implementing the snoring detection method as described in any one of claims 1-7, characterized in that: The system includes a signal acquisition module and a signal processing module, which are communicatively connected to the signal acquisition module. The signal acquisition module acquires vibration signals from the human body, and the signal processing module calculates the power spectrum of the respiratory signal based on the vibration signals. Power spectrum of snoring envelope signal and the cross-power spectrum of respiratory signals and snoring envelope signals. The signal processing module calculates the coherence function of the respiratory signal and the snoring envelope signal, and determines whether snoring occurs based on the coherence function.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the snoring detection method as described in any one of claims 1-7.

Images