A snoring monitoring method and system, storage medium
By combining the identification results of vibration signals and sound pressure level sequences, snoring events are comprehensively judged, solving the misjudgment problem of snoring monitoring in existing technologies and achieving highly accurate and privacy-protected snoring monitoring.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- DONGGUAN DERUCCI BEDDING CO LTD
- Filing Date
- 2026-06-26
- Publication Date
- 2026-07-31
AI Technical Summary
Existing snoring monitoring methods have a high rate of misjudgment, and improving monitoring accuracy is an urgent problem to be solved.
Combining the identification results of vibration signals and sound pressure level sequences, vibration signals are collected by piezoelectric sensors and sound pressure level sequences are collected by decibel meters. By utilizing the time-domain and frequency-domain characteristics of vibration signals and the fluctuation characteristics and differences of sound pressure level sequences, a comprehensive judgment is made on whether snoring events exist, excluding environmental noise and body movement interference.
It improves the accuracy of snoring detection, reduces the false alarm rate, and avoids the risk of privacy leakage by discarding waveform phase and fine frequency information.
Smart Images
Figure CN122490451A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of smart home technology, and in particular to a snoring monitoring method and system, and a storage medium. Background Technology
[0002] As we all know, sleep quality directly affects an individual's mental state and even their quality of life, and is of great importance to human health.
[0003] However, due to various factors such as physical health, environmental pollution, and work stress, people generally experience a decline in sleep quality. For example, snoring can significantly impact sleep quality. Severe snoring can even cause sleep apnea, leading to oxygen deprivation, demonstrating its serious harm.
[0004] Therefore, some smart products have emerged on the market that can monitor a user's snoring during sleep and adjust the product's status based on the snoring events to achieve the effect of stopping snoring.
[0005] However, current snoring monitoring methods have a high rate of misjudgment, and improving the accuracy of monitoring is an urgent problem to be solved.
[0006] The above information is provided as background information only to aid in understanding this application and does not constitute an assertion or admission that any of the above content can be used as prior art relative to this application. Summary of the Invention
[0007] This application provides a snoring monitoring method and system, and a storage medium, to improve the accuracy of snoring monitoring results.
[0008] To achieve the above objectives, this application provides the following technical solution: In a first aspect, embodiments of this application provide a snoring monitoring method, including: Vibration signals within the target area, a first sound pressure level sequence within the target area, and a second sound pressure level sequence within a reference area are collected; wherein, the target area includes a preset user head area and / or back area, and the reference area is farther away from the user head and / or back compared to the target area; Snoring is identified based on the time-domain and / or frequency-domain characteristics of the vibration signal; Based on the fluctuation characteristics of the first sound pressure level sequence and the difference in sound pressure level characteristics between the first sound pressure level sequence and the second sound pressure level sequence, snoring is identified; By combining the snoring recognition results based on the vibration signal and the snoring recognition results based on the first sound pressure level sequence and the second sound pressure level sequence, it is determined whether a snoring event is currently occurring.
[0009] Optionally, the snoring recognition based on the time-domain and / or frequency-domain characteristics of the vibration signal includes: Extract the time-domain envelope of the vibration signal, and determine whether the time interval between several consecutive adjacent peaks of the time-domain envelope is within a preset time interval range, and whether the difference between several consecutive time intervals is within a preset difference range. And / or, perform spectral analysis on the vibration signal to calculate whether the ratio of energy within a preset snoring frequency band to the total energy of the vibration signal exceeds a preset energy ratio threshold.
[0010] Optionally, the method for determining the time interval range, the difference range, and the energy percentage threshold includes: Collect the raw vibration signals of the current user during at least one sleep cycle; Multiple candidate signal segments are extracted from the original vibration signal. The candidate signal segments satisfy the following conditions: the interval between adjacent peaks of their time domain envelope is within the initial estimated range, and the energy proportion of their spectrum in the preset snoring frequency band exceeds the initial estimated threshold. From the plurality of candidate signal segments, select a plurality of clean signal segments whose motion interference is less than a preset threshold; The time interval range is adaptively set based on the statistical distribution of the time intervals between several consecutive adjacent peaks in the time domain envelope of each pure signal segment. The difference range is adaptively set based on the statistical distribution of the differences between several time intervals in the time domain envelope of each pure signal segment; Based on the statistical distribution of the ratio of energy of the preset snoring frequency band to total energy in the spectrum of each pure signal segment, the energy ratio threshold is adaptively set.
[0011] Optionally, the step of selecting multiple clean signal segments with motion interference less than a preset threshold from the plurality of candidate signal segments includes: for each candidate signal segment, Acquire auxiliary sensing data of the current user during the same time period, wherein the auxiliary sensing data includes at least one of the following: image data, infrared thermal imaging data, pressure distribution data, acceleration data, and gyroscope data; The image data is acquired by a camera device placed in the current sleep environment, the infrared thermal imaging data is acquired by an infrared thermal imager placed at the head of the bed, the pressure distribution data is acquired by a pressure sensor array placed on the mattress, the acceleration data is acquired by the accelerometer of the wearable device currently worn by the user, and the gyroscope data is acquired by the gyroscope of the wearable device. Based on the auxiliary sensing data, it is determined whether the user has any physical movement events in the current time period. If not, the current signal segment is identified as the pure signal segment.
[0012] 5. The snoring monitoring method according to claim 1, characterized in that, the step of identifying snoring based on the fluctuation characteristics of the first sound pressure level sequence and the difference in sound pressure level characteristics between the first sound pressure level sequence and the second sound pressure level sequence includes: Within a preset time window, the peak and trough values of the first sound pressure level sequence are extracted, and the number of fluctuations in which the difference between the peak and trough values within the time window is not less than a preset fluctuation amplitude threshold is counted. When the number of fluctuations is not less than a preset fluctuation number threshold, the difference between the peak values of the first sound pressure level sequence and the second sound pressure level sequence within the time window is compared; if the difference is not less than a preset difference threshold, it is identified as snoring, otherwise it is identified as no snoring.
[0013] Optionally, the step of combining the snoring recognition results based on the vibration signal and the snoring recognition results based on the first sound pressure level sequence and the second sound pressure level sequence to determine whether a snoring event currently exists includes: When both the snoring recognition result based on the vibration signal and the snoring recognition result based on the first sound pressure level sequence and the second sound pressure level sequence are identified as snoring, it is finally determined that a snoring event exists; otherwise, it is determined that no snoring event exists.
[0014] Optional, also includes: When a snoring event is detected, an anti-snoring control signal is generated; the anti-snoring control signal is used to trigger at least one action among pillow height adjustment, mattress tilt angle adjustment, vibration reminder, or audio playback.
[0015] Optionally, the reference area includes the foot of the bed area.
[0016] Secondly, embodiments of this application provide a snoring monitoring system for implementing any of the snoring monitoring methods described above, the snoring monitoring system comprising: A piezoelectric sensor is disposed in the target area to collect the vibration signal; A first decibel meter is installed in the target area to collect the first sound pressure level sequence; A second decibel meter is set in the reference area to collect the second sound pressure level sequence; A snoring recognition unit is used to recognize snoring based on the time-domain and / or frequency-domain characteristics of the vibration signal; to recognize snoring based on the fluctuation characteristics of the first sound pressure level sequence and the difference in sound pressure level characteristics between the first sound pressure level sequence and the second sound pressure level sequence; and to comprehensively determine whether a snoring event exists by combining the snoring recognition results based on the vibration signal and the snoring recognition results based on the first sound pressure level sequence and the second sound pressure level sequence.
[0017] Thirdly, embodiments of this application provide a computer-readable storage medium having computer-executable instructions stored thereon, which are executed by a computer processor to implement the snoring monitoring method as described in any of the preceding claims.
[0018] Compared with the prior art, this application has the following beneficial effects: This application combines vibration signal recognition results and sound pressure level sequence recognition results to comprehensively determine whether a snoring event is currently occurring. It fully utilizes the complementary advantages of the two recognition methods. The vibration signal recognition method can effectively eliminate interference from non-vibration sources (such as environmental noise, other people's voices) and large-amplitude body movements. The sound pressure level sequence recognition method can effectively eliminate interference from non-sound sources (such as small-amplitude body movements) and most environmental noise far away from the head and / or back, thereby improving the accuracy of snoring detection results and reducing the false alarm rate.
[0019] Meanwhile, the sound pressure level sequence collected in this application only retains the amplitude envelope of the sound, discarding the waveform phase and fine frequency information, making it impossible to reconstruct the speech content, thus fundamentally avoiding the risk of privacy leakage.
[0020] This application has other features and advantages that will be apparent from or will be set forth in detail in the accompanying drawings and following detailed description, which together serve to explain the particular principles of this application. Attached Figure Description
[0021] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0022] Figure 1 This is a flowchart of the snoring monitoring method provided in the embodiments of this application; Figure 2 This is a schematic diagram of the snoring recognition principle based on two sound pressure level sequences provided in the embodiments of this application. Detailed Implementation
[0023] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0024] Firstly, please refer to Figure 1 This application provides a snoring monitoring method, including: S10. Acquire vibration signals within the target area, a first sound pressure level sequence within the target area, and a second sound pressure level sequence within the reference area; wherein, the target area includes a preset user head area and / or back area, and the reference area is farther away from the user head and / or back compared to the target area.
[0025] When a user snores, the soft tissue vibrations in the throat and chest cavity are transmitted through the bones and muscles to the bed or pillow, where they are sensed by piezoelectric sensors and converted into electrical signals. Therefore, the vibration signal in this step can refer to the mechanical vibration signal collected by piezoelectric sensors (such as piezoelectric ceramic sensors or piezoelectric thin-film sensors) and transmitted through the bed or pillow. It is understood that this signal includes vibration components caused by snoring, as well as vibration components caused by other body movements such as turning over or kicking during the same period.
[0026] First Sound Pressure Level Sequence: This refers to a data sequence of sound pressure level changes over time, acquired and output in real time by a first decibel meter positioned in the target area (the user's head area and / or back area). This sequence reflects the changes in sound intensity near the user's head and / or back, but discards the waveform phase and fine frequency information of the original sound pressure, making it impossible to reconstruct the speech content, thus protecting user privacy.
[0027] The second sound pressure level sequence refers to the data sequence of sound pressure level changes over time, which is collected and output in real time by a second decibel meter placed in a reference area (such as the foot of the bed, under the bed, or on the ground, which is farther away from the user's head and / or back than the target area). This second sound pressure level sequence will be used to compare with the first sound pressure level sequence to determine whether the sound source is from near the user's head and / or back.
[0028] S20. Snoring recognition based on the time-domain and / or frequency-domain characteristics of vibration signals.
[0029] Among them, time-domain characteristics refer to the morphological characteristics of the vibration signal as it changes over time, which may include, but are not limited to, the peak interval of the signal envelope, the stability of the difference between multiple consecutive time intervals, and the variation law of the peak amplitude. Frequency-domain characteristics refer to the energy distribution characteristics exhibited after the vibration signal is transformed into the frequency domain through Fourier transform.
[0030] The inventors discovered that the vibration signals caused by snoring differ from those caused by body movement in both the time and frequency domains.
[0031] For example, the vibration signal caused by snoring exhibits periodic fluctuations in the time domain. The time interval between adjacent peaks is relatively consistent with the human respiratory cycle, and the differences between multiple consecutive intervals are small (i.e., good stability). In contrast, the vibration signal caused by body movement usually presents as isolated spikes or chaotic pulse trains in the time domain, lacking periodicity.
[0032] For example, the vibration signal energy caused by snoring is mainly concentrated in the relatively low-frequency band, where the energy proportion is usually high. In contrast, the vibration signal spectrum caused by body movements (such as turning over or kicking) is wider, containing a relatively high-frequency component, and the energy proportion of the high-frequency band is lower.
[0033] Therefore, based on the time-domain and / or frequency-domain characteristics of vibration signals, snoring can be initially identified, and snoring events that are significantly different from body movement events can be distinguished.
[0034] S30. Based on the fluctuation characteristics of the first sound pressure level sequence and the difference in sound pressure level characteristics between the first and second sound pressure level sequences, snoring is identified.
[0035] Fluctuation characteristics: This refers to the fluctuations in the first sound pressure level sequence over time. When a user snores, the snoring sound exhibits periodic, large fluctuations: the sound pressure level rises rapidly to a peak (snoring burst), then slowly decreases to a trough (breathing interval); the difference between each peak and trough is usually large (e.g., more than 6 dB). Therefore, by counting the number of fluctuations that meet the fluctuation amplitude threshold within a preset time window, snoring can be distinguished from steady sounds (such as ambient noise) or random fluctuations (such as brief coughs).
[0036] Sound pressure level characteristic difference: This refers to the contrast between the first and second sound pressure level sequences. Because sound waves attenuate with distance in the air, the sound pressure level of a user's snoring is significantly higher in the target area (e.g., the head of the bed, near the user's head) than in the reference area (e.g., the foot of the bed, away from the user's head). However, ambient noise, television sounds, or other people's voices from outside the sleeping room show smaller differences in sound pressure levels between these two areas. Therefore, by comparing the peak differences between the two sequences, it is possible to determine whether the sound source is near the user's head and / or back, thus eliminating environmental interference.
[0037] S40. Combining the snoring recognition results based on vibration signals and the snoring recognition results based on the first sound pressure level sequence and the second sound pressure level sequence, determine whether a snoring event is currently occurring.
[0038] As mentioned earlier, preliminary snoring identification can be performed based on vibration signals, as well as based on the first and second sound pressure level sequences. However, If only the snoring recognition result based on vibration signal is used: Although the piezoelectric sensor is sensitive to vibration and can capture the weak vibration caused by snoring, the user's body movements such as turning over, kicking, and slight limb twitching will also generate vibration signals, and their amplitude and frequency may be similar to the vibration signals caused by snoring. As a result, the piezoelectric sensor is prone to misjudging body movement events as snoring events, resulting in a high false alarm rate. If the snoring recognition results based solely on the first and second sound pressure level sequences are used: although environmental noise can be effectively eliminated through fluctuation characteristics and spatial attenuation differences, it cannot distinguish snoring from the user's own chatter, sleep talking, etc., because these sounds also meet the conditions of "the sound source is located near the head and / or back" and "there is a relatively similar attenuation trend in the target area and the reference area", which leads to non-snoring human voices being misjudged as snoring.
[0039] Therefore, this application adopts a combined strategy: combining vibration signal recognition results and sound pressure level sequence recognition results to ultimately determine whether a snoring event exists. This fully utilizes the complementary advantages of the two recognition methods. The vibration signal recognition method can effectively eliminate interference from non-vibration sources (such as environmental noise, other people's voices) and large-amplitude body movements, while the sound pressure level sequence recognition method can effectively eliminate interference from non-sound sources (such as small-amplitude body movements) and most environmental noise far away from the head and / or back, thereby improving the accuracy of snoring detection results and reducing the false alarm rate.
[0040] Meanwhile, the sound pressure level sequence collected in this application only retains the amplitude envelope of the sound, discarding the waveform phase and fine frequency information, making it impossible to reconstruct the speech content, thus fundamentally avoiding the risk of privacy leakage.
[0041] Furthermore, in an optional implementation, step S20, which involves identifying snoring based on the time-domain and / or frequency-domain characteristics of the vibration signal, may include: Extract the time-domain envelope of the vibration signal, and determine whether the time interval between several consecutive adjacent peaks of the time-domain envelope is within a preset time interval range, and whether the difference between several consecutive time intervals is within a preset difference range; where the time-domain envelope can refer to the outer contour line obtained after smoothing the vibration signal waveform, which reflects the macroscopic change trend of the peak amplitude of the vibration signal. And / or, perform spectral analysis on the vibration signal to calculate whether the ratio of energy within the preset snoring frequency band to the total energy of the vibration signal exceeds a preset energy ratio threshold.
[0042] Understandably, the preset time interval range, difference range, and energy percentage threshold are matched with the vibration signal characteristics caused by snoring.
[0043] To improve adaptability to the current user, in one optional implementation, the time interval range, difference range, and energy percentage threshold can be determined using an adaptive method, including: S21. Collect the raw vibration signal of the current user during at least one sleep cycle.
[0044] S22. Extract multiple candidate signal segments from the original vibration signal. The candidate signal segments satisfy the following conditions: the interval between adjacent peaks of their time domain envelope is within the initial estimated range, and the energy proportion of their spectrum in the preset snoring frequency band exceeds the initial estimated threshold.
[0045] It should be noted that the initial estimation range and initial estimation threshold in this step can be set to a relatively large range in advance, without any specific limitation.
[0046] S23. Select multiple clean signal segments from multiple candidate signal segments whose motion interference is less than a preset interference threshold.
[0047] S24. Based on the statistical distribution of the time intervals between several consecutive adjacent peaks in the time domain envelope of each pure signal segment, adaptively set the time interval range.
[0048] S25. Based on the statistical distribution of the differences between several time intervals in the time domain envelope of each pure signal segment, adaptively set the difference range.
[0049] S26. Based on the statistical distribution of the ratio of energy of the preset snoring frequency band to the total energy in the spectrum of each pure signal segment, adaptively set the energy ratio threshold.
[0050] In other words, this embodiment first extracts multiple candidate signal segments from the original vibration signal using a wide range of initial parameters based on the current user's historical sleep vibration data. Then, it eliminates body motion interference to retain pure signal segments. Finally, based on the statistical distribution of the time-domain interval, interval difference, and frequency-domain energy ratio of the pure signal segments, it adaptively sets three threshold ranges required for identification.
[0051] This makes this embodiment highly adaptable: by collecting users' historical sleep data, reasonable interval range, difference range and energy ratio threshold parameters for vibration recognition can be adaptively set to adapt to individual differences among different users.
[0052] Furthermore, multiple clean signal segments with motion interference less than a preset threshold are selected from multiple candidate signal segments, including: for each candidate signal segment, Acquire auxiliary sensing data of the current user during the same time period. The auxiliary sensing data includes at least one of the following: image data, infrared thermal imaging data, pressure distribution data, acceleration data, and gyroscope data. Among them, image data is collected by a camera device placed in the current sleep environment, infrared thermal imaging data is collected by an infrared thermal imager placed at the head of the bed, pressure distribution data is collected by a pressure sensor array placed on the mattress, acceleration data is collected by the accelerometer of the wearable device currently worn by the user, and gyroscope data is collected by the gyroscope of the wearable device. The system determines whether the user has any physical activity during the current time period based on auxiliary sensor data. If not, the current signal segment is labeled as a clean signal segment.
[0053] Considering the significant overlap between vibration signals caused by snoring and those caused by minor body movements in both the time and frequency domains, making effective differentiation difficult using only a single vibration signal characteristic, this embodiment utilizes auxiliary sensor data to directly detect body movement events from other dimensions, rather than relying on the characteristics of the vibration signal itself. This effectively eliminates body movement interference, ensuring that the selected pure signal segments accurately represent the user's snoring vibrations, thus making the settings of various parameters more accurate and reliable. Furthermore, different types of auxiliary sensor data can be used individually or in combination depending on the actual application scenario. When used in combination, the judgment results of body movement events can be mutually verified, further improving the accuracy of the detection results.
[0054] In an optional implementation, step S3 involves identifying snoring based on the fluctuation characteristics of the first sound pressure level sequence and the difference in sound pressure level characteristics between the first and second sound pressure level sequences. Specifically, this may include: Within a preset time window, extract the peak and trough values of the first sound pressure level sequence, and count the number of fluctuations within the time window where the difference between the peak and trough values is not less than a preset fluctuation amplitude threshold. When the number of fluctuations is not less than the preset fluctuation threshold, the difference between the peak values of the first sound pressure level sequence and the second sound pressure level sequence within the comparison time window is used; if the difference is not less than the preset difference threshold, it is identified as snoring, otherwise it is identified as no snoring.
[0055] For example, such as Figure 2 As shown, if within the same 20-second time window, the first sound pressure level sequence exhibits at least three sets of fluctuations where the difference between the peak and trough values is greater than 6 dB, and the peak value of the first sound pressure level sequence is at least 4 dB higher than the peak value of the second sound pressure level sequence, then snoring is identified.
[0056] This embodiment employs a two-stage screening process: first, "screening for the number of large fluctuations," and then "screening for the peak difference at different locations." This ensures that the identified sound not only possesses the periodic, large fluctuation characteristic of snoring but also confirms that the sound source is located near the user's head and / or back. This effectively eliminates interference from environmental noise, occasional sounds, other people's snoring, and the user's own speech, achieving highly accurate acoustic snoring recognition.
[0057] In an optional implementation, step S4, combining the snoring recognition results based on vibration signals and the snoring recognition results based on the first sound pressure level sequence and the second sound pressure level sequence, to determine whether a snoring event currently exists, may include: If both the snoring recognition results based on vibration signals and the snoring recognition results based on the first sound pressure level sequence and the second sound pressure level sequence identify snoring, then it is determined that a snoring event exists; otherwise, it is determined that no snoring event exists.
[0058] In addition, the snoring monitoring method of this embodiment may also include: When a snoring event is finally determined to exist, an anti-snoring control signal is generated; this anti-snoring control signal is used to trigger at least one of the following intervention actions: Pillow height adjustment: By inflating or mechanically raising and lowering the pillow, the user's head is slowly raised, opening the airway, reducing airway obstruction, and physically suppressing snoring.
[0059] Mattress tilt angle adjustment: Adjust the height of the mattress airbags, or adjust the height of the mattress via an electric bed, so that the user's upper body is slightly tilted, using gravity to help keep the airway open and reduce tongue fall back.
[0060] Vibration alert: Gentle vibrations (such as those from a vibrator inside the pillow or a wearable device) prompt the user to adjust their sleeping position without fully waking them, thereby stopping snoring.
[0061] Audio playback: Play sound waves or white noise of a specific frequency to guide the user to change their breathing rhythm or sleeping position, or wake the user with a gradual sound to interrupt snoring.
[0062] The above intervention actions can be used individually or in combination, and the intensity of the intervention can be adaptively adjusted according to the user's historical snoring severity.
[0063] Secondly, embodiments of this application provide a snoring monitoring system for implementing the snoring monitoring method described in any of the above embodiments. The snoring monitoring system includes: A piezoelectric sensor is disposed in the target area to collect the vibration signal; A first decibel meter is installed in the target area to collect the first sound pressure level sequence; A second decibel meter is set in the reference area to collect the second sound pressure level sequence; A snoring recognition unit is used to recognize snoring based on the time-domain and / or frequency-domain characteristics of the vibration signal; to recognize snoring based on the fluctuation characteristics of the first sound pressure level sequence and the difference in sound pressure level characteristics between the first sound pressure level sequence and the second sound pressure level sequence; and to comprehensively determine whether a snoring event exists by combining the snoring recognition results based on the vibration signal and the snoring recognition results based on the first sound pressure level sequence and the second sound pressure level sequence.
[0064] The above system can execute the methods provided in any embodiment of this application, and has the corresponding functional modules and beneficial effects for executing the methods. Furthermore, based on the foregoing embodiments, features not explained in this embodiment are explained using the methods described in the foregoing embodiments, and will not be repeated here.
[0065] Thirdly, embodiments of this application provide a computer-readable storage medium having computer-executable instructions stored thereon, which, when executed by a processor, implement the snoring monitoring method provided in all embodiments of this application.
[0066] Any combination of one or more computer-readable media may be used. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device.
[0067] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of sending, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.
[0068] Program code contained on a computer-readable medium may be transmitted using any suitable medium, including—but not limited to—wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.
[0069] Computer program code for performing the operations of this application can be written in one or more programming languages or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, and C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).
[0070] Finally, it should be noted that although the above embodiments have been described in the text and drawings of this application, this should not limit the scope of patent protection of this application. Any technical solutions that are based on the essential concept of this application and utilize the content described in the text and drawings of this application, resulting in equivalent structural or procedural substitutions or modifications, as well as the direct or indirect application of the technical solutions of the above embodiments to other related technical fields, are all included within the scope of patent protection of this application.
Claims
1. A method for monitoring snoring, characterized in that, include: Vibration signals within the target area, a first sound pressure level sequence within the target area, and a second sound pressure level sequence within a reference area are collected; wherein, the target area includes a preset user head area and / or back area, and the reference area is farther away from the user head and / or back compared to the target area; Snoring is identified based on the time-domain and / or frequency-domain characteristics of the vibration signal; Based on the fluctuation characteristics of the first sound pressure level sequence and the difference in sound pressure level characteristics between the first sound pressure level sequence and the second sound pressure level sequence, snoring is identified; By combining the snoring recognition results based on the vibration signal and the snoring recognition results based on the first sound pressure level sequence and the second sound pressure level sequence, it is determined whether a snoring event is currently occurring.
2. The snoring monitoring method according to claim 1, characterized in that, The snoring identification based on the time-domain and / or frequency-domain characteristics of the vibration signal includes: Extract the time-domain envelope of the vibration signal, and determine whether the time interval between several consecutive adjacent peaks of the time-domain envelope is within a preset time interval range, and whether the difference between several consecutive time intervals is within a preset difference range. And / or, perform spectral analysis on the vibration signal to calculate whether the ratio of energy within a preset snoring frequency band to the total energy of the vibration signal exceeds a preset energy ratio threshold.
3. The snoring monitoring method according to claim 2, characterized in that, The method for determining the time interval range, the difference range, and the energy percentage threshold includes: Collect the raw vibration signals of the current user during at least one sleep cycle; Multiple candidate signal segments are extracted from the original vibration signal. The candidate signal segments satisfy the following conditions: the interval between adjacent peaks of their time domain envelope is within the initial estimated range, and the energy proportion of their spectrum in the preset snoring frequency band exceeds the initial estimated threshold. From the plurality of candidate signal segments, select a plurality of clean signal segments whose motion interference is less than a preset threshold; The time interval range is adaptively set based on the statistical distribution of the time intervals between several consecutive adjacent peaks in the time domain envelope of each pure signal segment. The difference range is adaptively set based on the statistical distribution of the differences between several time intervals in the time domain envelope of each pure signal segment; Based on the statistical distribution of the ratio of energy of the preset snoring frequency band to total energy in the spectrum of each pure signal segment, the energy ratio threshold is adaptively set.
4. The snoring monitoring method according to claim 3, characterized in that, The step of selecting multiple clean signal segments with motion interference less than a preset threshold from the multiple candidate signal segments includes: for each candidate signal segment, Acquire auxiliary sensing data of the current user during the same time period, wherein the auxiliary sensing data includes at least one of the following: image data, infrared thermal imaging data, pressure distribution data, acceleration data, and gyroscope data; The image data is acquired by a camera device placed in the current sleep environment, the infrared thermal imaging data is acquired by an infrared thermal imager placed at the head of the bed, the pressure distribution data is acquired by a pressure sensor array placed on the mattress, the acceleration data is acquired by the accelerometer of the wearable device currently worn by the user, and the gyroscope data is acquired by the gyroscope of the wearable device. Based on the auxiliary sensing data, it is determined whether the user has any physical movement events in the current time period. If not, the current signal segment is identified as the pure signal segment.
5. The snoring monitoring method according to claim 1, characterized in that, The method of identifying snoring based on the fluctuation characteristics of the first sound pressure level sequence and the difference in sound pressure level characteristics between the first and second sound pressure level sequences includes: Within a preset time window, the peak and trough values of the first sound pressure level sequence are extracted, and the number of fluctuations in which the difference between the peak and trough values within the time window is not less than a preset fluctuation amplitude threshold is counted. When the number of fluctuations is not less than a preset fluctuation number threshold, the difference between the peak values of the first sound pressure level sequence and the second sound pressure level sequence within the time window is compared; if the difference is not less than a preset difference threshold, it is identified as snoring, otherwise it is identified as no snoring.
6. The snoring monitoring method according to claim 1, characterized in that, The determination of whether a snoring event exists by combining the snoring recognition results based on the vibration signal and the snoring recognition results based on the first sound pressure level sequence and the second sound pressure level sequence includes: When both the snoring recognition result based on the vibration signal and the snoring recognition result based on the first sound pressure level sequence and the second sound pressure level sequence are identified as snoring, it is finally determined that a snoring event exists; otherwise, it is determined that no snoring event exists.
7. The snoring monitoring method according to claim 1, characterized in that, Also includes: When a snoring event is detected, an anti-snoring control signal is generated; the anti-snoring control signal is used to trigger at least one of the following actions: pillow height adjustment, mattress tilt angle adjustment, vibration reminder, or audio playback.
8. The snoring monitoring method according to claim 1, characterized in that, The reference area includes the area at the foot of the bed.
9. A snoring monitoring system, characterized in that, For implementing the snoring monitoring method according to any one of claims 1 to 8, the snoring monitoring system comprises: A piezoelectric sensor is disposed in the target area to collect the vibration signal; A first decibel meter is set in the target area to collect the first sound pressure level sequence; A second decibel meter is set in the reference area to collect the second sound pressure level sequence; A snoring recognition unit is used to recognize snoring based on the time-domain and / or frequency-domain characteristics of the vibration signal; to recognize snoring based on the fluctuation characteristics of the first sound pressure level sequence and the difference in sound pressure level characteristics between the first sound pressure level sequence and the second sound pressure level sequence; and to determine whether a snoring event exists by combining the snoring recognition results based on the vibration signal and the snoring recognition results based on the first sound pressure level sequence and the second sound pressure level sequence.
10. A computer-readable storage medium having computer-executable instructions stored thereon, characterized in that, The computer-executable instructions are executed by a computer processor to implement the snoring monitoring method as described in any one of claims 1 to 8.