A wearable smart snore stopper based on snoring frequency band analysis and intelligent intervention

Abstract

The application discloses a wearable smart snore stopper based on snoring frequency band analysis and intelligent intervention, aiming at solving the problems of existing snore stoppers, such as being easily interfered by environmental noise, lacking intelligent sleep posture monitoring and personalized intervention. The snore stopper comprises a snore stopper body and a wearable structure, the snore stopper body is provided with a sound detection module, an intelligent control module, a vibration control module and a vibration intervention module; the wearable structure realizes the fixation of the body, the sound detection module collects and amplifies the sleep environment sound signal, the intelligent control module performs digital processing, FFT frequency domain analysis, adaptive noise filtering and snoring feature extraction on the signal, generates a hierarchical vibration control instruction after determining the effective snoring, the vibration control module generates a driving signal according to the instruction, and the vibration intervention module executes hierarchical vibration intervention and returns the working state. The application improves the snoring recognition accuracy, realizes precise and personalized vibration intervention, improves the sleep quality and optimizes the user experience.

Description

Technical Field

[0001] This invention belongs to the field of intelligent medical device technology, specifically to a wearable intelligent anti-snoring device based on snoring frequency band analysis and intelligent intervention. Background Technology

[0002] Sleep is an indispensable physiological state for the human body, playing a vital role in maintaining health and improving quality of life. However, with the accelerated pace of life, increased stress, and the rise of unhealthy lifestyle habits, snoring, a sleep disorder, is becoming increasingly common. Snoring not only affects the sleep quality of oneself and others but can also negatively impact health, causing hypoxia, lethargy, hypoxemia, and even a series of complications such as sleep apnea syndrome, cardiovascular and cerebrovascular diseases, and even sudden death. To address this issue, anti-snoring devices have been widely used in the field of sleep health. Currently, there are many types of anti-snoring devices on the market, including liquid anti-snoring devices, anti-snoring mouthguards, anti-snoring pillows, anti-snoring nose clips, anti-snoring products that assist in side sleeping, and CPAP machines. These products achieve the purpose of stopping snoring through different principles, such as moisturizing the airway, fixing the jaw, physically adjusting sleeping posture, expanding the nasal passage, or providing positive air pressure.

[0003] Although anti-snoring devices are widely used in the field of sleep health, existing technologies still have some problems. Most anti-snoring devices use methods based on volume thresholds or single characteristics to identify snoring, which are easily affected by environmental noise in complex sleep environments, resulting in low accuracy in snoring identification. Furthermore, existing devices generally lack intelligent monitoring and adjustment functions for sleeping posture, failing to precisely adjust according to the user's personalized needs. Different users have different reasons for snoring and different sleeping posture habits; therefore, anti-snoring devices lacking personalized adjustment often fail to achieve the desired anti-snoring effect.

[0004] To address these issues, existing technologies have undergone several improvements. For example, some anti-snoring devices are beginning to employ more advanced snoring recognition algorithms to improve accuracy in complex environments. These algorithms may combine multiple feature parameters, such as the frequency, amplitude, and duration of snoring, to more comprehensively describe the characteristics of snoring. Simultaneously, some devices are introducing sleep posture monitoring functions, using built-in sensors or external devices to monitor the user's sleeping posture and make corresponding adjustments based on changes in posture. For instance, anti-snoring pillows can detect snoring sounds to trigger air mattress inflation, causing the user to sleep on their side, thereby reducing snoring.

[0005] However, despite improvements in existing technologies, several issues remain. While snoring recognition devices based on more advanced algorithms have improved accuracy, they often involve high computational complexity and demanding device performance, leading to increased costs. Furthermore, sleep posture monitoring and adjustment functions still have limitations in practical applications. For example, some devices may not monitor sleep posture accurately or promptly enough, failing to adjust quickly based on real-time changes in the user's sleeping position. Additionally, existing devices often lack tiered vibration intervention mechanisms, unable to provide personalized vibration intervention based on the severity of snoring, thus affecting anti-snoring effectiveness. Therefore, there is an urgent need for anti-snoring devices capable of accurately identifying snoring characteristics and providing personalized adjustments through tiered vibration intervention and sleep posture monitoring mechanisms to effectively improve sleep quality. Summary of the Invention

[0006] To address the issues of existing anti-snoring devices being susceptible to environmental noise interference in snoring recognition and lacking intelligent sleep posture monitoring and personalized intervention, this invention provides a wearable intelligent anti-snoring device based on snoring frequency band analysis and intelligent intervention. It improves recognition accuracy by eliminating background noise through adaptive noise filtering, and dynamically adjusts the vibration feedback mode by combining intelligent control and sleep posture monitoring. With real-time feedback from an OLED display, it achieves precise and comfortable personalized intervention, improves sleep quality, and optimizes the user experience.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a wearable intelligent anti-snoring device based on snoring frequency band analysis and intelligent intervention, comprising an anti-snoring device body and a wearable structure; the anti-snoring device body includes a sound detection module, an intelligent control module, a vibration control module and a vibration intervention module, and the wearable structure is used to fix the anti-snoring device body. The sound detection module is used to collect sound signals from the sleep environment, amplify them, and output them. The intelligent control module is used to perform digital processing, FFT frequency domain analysis, adaptive noise filtering and snoring feature extraction on the received sound signal. When a valid snoring event is determined, a graded vibration control command is generated and output. The vibration control module is used to receive graded vibration control commands, generate drive signals, and output them. The vibration intervention module is used to receive drive signals, perform graded vibration intervention actions, and output its own working status to the intelligent control module.

[0008] Furthermore, the sound detection module includes a MEMS microphone and an analog preamplifier; the MEMS microphone has a pickup frequency range of 200~800Hz; the amplification factor of the analog preamplifier is adjustable, with an adjustment range of 10~100 times.

[0009] Furthermore, the intelligent control module performs digital processing, FFT frequency domain analysis, adaptive noise filtering, and snoring feature extraction on the received sound signal. Based on the snoring intensity and duration, it generates graded vibration control commands, with the specific constraints being: Level 1 graded vibration control command: Snoring intensity is below the low threshold and duration is <5s; Level 2 graded vibration control command: Snoring intensity is between the low and medium thresholds and the duration is ≥5s and <10s; Level 3 vibration control command: Snoring intensity is higher than the high threshold and the duration is ≥10s.

[0010] Furthermore, the intelligent control module includes a main processing unit, a clock circuit, and a reset circuit; the main processing unit uses an STM32F103C8T6 microcontroller; the clock circuit adopts a dual crystal oscillator design; the reset circuit includes a button reset submodule, supporting level reset and manual button reset.

[0011] Furthermore, the vibration control module includes a PWM signal generation submodule and a motor drive unit; the PWM signal generation submodule is embedded in the intelligent control module and outputs a PWM signal with an adjustable duty cycle, the duty cycle adjustment range being 0~100%; the motor drive unit is an L298N or a motor driver with equivalent driving capability.

[0012] Furthermore, the vibration intervention module includes a DC vibration motor and a status detection submodule; the rated voltage of the DC vibration motor is 3.7~5V, and the vibration intensity is divided into 3 levels, corresponding to speeds of 3000~5000r / min, 5000~8000r / min, and 8000~12000r / min respectively; the status detection submodule is implemented by the current detection circuit embedded in the motor drive unit, which is used to detect the working status of the DC vibration motor and transmit it back to the intelligent control module.

[0013] Furthermore, it also includes a sleeping posture monitoring module, which includes an accelerometer, a pressure sensor, and a data preprocessing submodule; the accelerometer is a triaxial accelerometer; the pressure sensor is a flexible pressure sensor; the data preprocessing submodule is implemented by the ADC module built into the sensor, which is used to convert the analog acquisition signal into a 16-bit digital signal and transmit it to the intelligent control module.

[0014] Furthermore, the intelligent control module performs digital processing, FFT frequency domain analysis, adaptive noise filtering, and snoring feature extraction on the received sound signal to obtain snoring intensity and duration. Simultaneously, it acquires sleeping posture data and generates graded vibration control commands based on the snoring intensity, duration, and sleeping posture data, with specific limitations as follows: When the user is sleeping in a supine position: Level 1 graded vibration control command: Snoring intensity is below the low threshold and duration is <5s; Level 2 graded vibration control command: Snoring intensity is between the low and medium thresholds and the duration is ≥5s and <10s; Level 3 vibration control command: Snoring intensity is higher than the high threshold and the duration is ≥10s; If snoring continues for ≥3 seconds and the sleeping position remains supine after implementing the corresponding level of vibration intervention, the current vibration level will be increased by one level. When the user is sleeping on their side or stomach, regardless of the intensity and duration of snoring, a first-level graded vibration control command is generated.

[0015] Furthermore, it also includes a display module, which receives display control commands from the intelligent control module, displays the device's working status and sleep-related data in real time, and receives user operation signals and transmits them to the intelligent control module. The display module includes an OLED display screen and a button input submodule. The OLED display screen displays the detection status, vibration countdown, snoring frequency, snoring intensity curve, sleep quality score, and wearing tightness indication. The button input submodule has 3-5 physical buttons with functions for power on / off, threshold adjustment, mode switching, and data clearing. It also includes a power supply module, which comprises a lithium battery, a low-power power management chip, and a sleep control submodule; the lithium battery is a polymer lithium battery with a rated voltage of 3.7V and a capacity of 500~1000mAh; the low-power power management chip supports 5V / 1A charging and 3.7V regulated output, and has overcharge, over-discharge, and short-circuit protection functions; the sleep control submodule is used to receive sleep / wake-up commands from the intelligent control module and control the system to enter a low-power sleep mode.

[0016] Furthermore, the wearable structure features a highly elastic adjustable strap with a built-in flexible pressure sensor to detect the tightness of the fit and prompt the user to adjust it via a display module.

[0017] Compared with the prior art, the present invention has at least the following beneficial effects: This invention provides a wearable smart anti-snoring device based on snoring frequency band analysis and intelligent intervention. By integrating a sound detection module, an intelligent control module, a vibration control module, and a vibration intervention module, it achieves integrated control of snoring signal acquisition and processing with graded vibration intervention. The intelligent control module uses digital processing FFT frequency domain analysis and adaptive noise filtering to accurately extract snoring features, effectively eliminating environmental noise interference and significantly improving the accuracy of snoring recognition. This solves the problem of misjudgment and missed judgment in existing devices that rely on single feature recognition. At the same time, it achieves targeted vibration intervention through graded vibration control commands, balancing anti-snoring effect with sleep comfort, thus enhancing the intelligence and practicality of the device.

[0018] This invention configures the sound detection module as a MEMS microphone and an adjustable analog preamplifier. The MEMS microphone's pickup frequency band covers the core snoring frequency band of 200~800Hz, enabling accurate acquisition of snoring characteristic signals and adapting to the acquisition needs of snoring signals of different intensities. The adjustable amplification of the analog preamplifier from 10 to 100 times can flexibly adapt to the sound signal intensity of different sleep environments, further improving the accuracy of snoring signal acquisition. This provides a high-quality signal foundation for subsequent snoring feature extraction and recognition, ensuring a high accuracy rate for snoring recognition.

[0019] This invention achieves precise generation of graded vibration control commands by clearly defining graded and quantitative standards for snoring intensity and duration. This ensures that vibration intervention strategies match the actual severity of snoring, avoiding over- or under-intervention. It enables personalized and dynamic intervention for snoring, abandoning the traditional one-size-fits-all approach. While effectively stopping snoring, it maximizes the user's sleep comfort and improves the adaptability and rationality of device intervention.

[0020] This invention limits the hardware configuration of the intelligent control module, selecting an STM32F103C8T6 microcontroller as the main processing unit, coupled with a clock circuit with a dual crystal oscillator design and a reset circuit supporting both level reset and manual button reset. This ensures the high efficiency of data processing, the stability of the clock signal, and the reliability of circuit operation in the intelligent control module. The dual crystal oscillator design can adapt to the clock requirements of different processing scenarios, improving data processing efficiency. The button reset submodule enables the device to have a manual reset function, enhancing the convenience of device operation and fault recovery capability, and providing hardware support for the stable implementation of core functions such as snoring signal processing and instruction generation.

[0021] This invention configures the vibration control module as a PWM signal generation submodule and a motor drive unit, with the PWM signal duty cycle adjustable from 0 to 100%. This allows for flexible adjustment of the drive signal output power, precisely matching the power requirements of different levels of vibration intervention. By selecting an L298N or equivalent motor driver, the driving stability and power output reliability of the vibration intervention module are ensured, enabling precise execution of graded vibration intervention actions. This ensures a high degree of consistency between vibration intensity and graded control commands, improving the accuracy and effectiveness of the intervention actions.

[0022] This invention defines the hardware parameters and functions of the vibration intervention module. The rated voltage of the DC vibration motor (3.7~5V) is compatible with the equipment's power supply standard. The three vibration intensity levels correspond to specific speed ranges, achieving precise grading of vibration intensity and forming a precise match with the graded vibration control commands. The status detection submodule detects the motor's operating status in real time through a current detection circuit and transmits the data back, allowing the intelligent control module to monitor the execution of intervention actions in real time. This facilitates timely adjustments to the control strategy, avoids intervention failures caused by motor malfunctions, and improves the reliability and dynamic adjustment capability of the equipment intervention.

[0023] This invention adds a sleep posture monitoring module consisting of a triaxial accelerometer, a flexible pressure sensor, and a data preprocessing submodule. This module can accurately collect users' sleep posture data. The triaxial accelerometer can comprehensively capture changes in body posture, while the flexible pressure sensor further assists in verifying the sleep posture status. The data preprocessing submodule converts analog signals into 16-bit digital signals, improving the accuracy and transmission efficiency of sleep posture data collection. This provides precise data support for the intelligent control module to formulate personalized intervention strategies based on sleep posture data, making vibration intervention more closely match the user's actual sleep state and further enhancing the personalization and effectiveness of the intervention.

[0024] This invention generates graded vibration control commands by combining snoring intensity, duration, and sleeping posture data through an intelligent control module. It then formulates differentiated intervention rules, implementing graded interventions based on snoring severity for supine users and setting up a grade-up mechanism to ensure anti-snoring effectiveness. For side-lying or prone users, only a first-level mild intervention is implemented, minimizing disruption to sleep. This achieves precise, personalized, and dynamic intervention based on both sleeping posture and snoring characteristics, significantly improving the adaptability of the intervention strategy. While ensuring anti-snoring effectiveness, it also significantly improves user sleep comfort, addressing the lack of specificity in existing intervention methods.

[0025] This invention adds a display module and a power supply module. The OLED display screen of the display module can provide real-time feedback on the device's working status, snoring data, sleep quality scores, and other information. The button input submodule allows users to manually adjust parameters, improving the device's interactivity and ease of operation, allowing users to monitor the device and sleep status in real time and adjust usage parameters independently. The power supply module uses a 500~1000mAh polymer lithium battery, paired with a low-power power management chip with protection functions and a sleep control submodule, to achieve low-power operation of the device. A single full charge provides long battery life, meeting users' long-term usage needs. At the same time, overcharge, over-discharge, and short-circuit protection functions improve the safety of device use, and the sleep mode further reduces energy consumption, balancing the device's functionality, interactivity, and battery life.

[0026] This invention features a wearable structure with a highly elastic adjustable strap and a built-in flexible pressure sensor. The highly elastic adjustable design allows the strap to adapt to different users' head shapes or wearing area sizes, improving fit and comfort. The flexible pressure sensor can detect the tightness of the fit in real time and prompt the user to adjust it via a display module. This avoids discomfort caused by wearing it too tightly or device displacement and inaccurate signal acquisition caused by wearing it too loosely, ensuring the comfort and stability of the device while ensuring the working accuracy of modules such as sound detection and sleep posture monitoring, thus providing wear protection for the effective realization of the device's core functions. Attached Figure Description

[0027] Figure 1 This is a schematic diagram of the overall hardware framework of an embodiment of the present invention, showing the basic hardware structure of a wearable smart anti-snoring device, including a sound detection module, an intelligent control module, a vibration intervention module and a display module. Each module works together to realize the functions of snoring sound collection, analysis and vibration feedback. Figure 2 This is a schematic diagram of the overall hardware framework of another embodiment of the present invention. Figure 1 Based on the embodiment shown, a sleeping posture monitoring module is further introduced to collect user posture information and make a joint judgment with the snoring recognition result in order to achieve more precise vibration intervention control. Figure 3 This is a schematic diagram of the snoring detection and signal processing flow in this invention, showing the process of sound signal acquisition, frequency domain analysis, and noise filtering. Figure 4 This is a schematic diagram of the vibration intervention control process based on snoring characteristics in the present invention, illustrating the process of generating graded vibration intervention commands according to snoring intensity and duration; Figure 5 This is a schematic diagram of the wearing structure of the wearable smart anti-snoring device of the present invention, showing how the anti-snoring device body is fixed to the user's body through an adjustable wearing structure. Detailed Implementation

[0028] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0029] like Figures 1-5 As shown, this invention provides a wearable smart anti-snoring device, comprising an anti-snoring device body and a wearable structure. The anti-snoring device body includes a sound detection module, an intelligent control module, a vibration control module, a vibration intervention module, a sleep posture monitoring module, a display module, and a power supply module. Each module uses the intelligent control module as the core for data processing and decision-making. The other modules establish bidirectional or unidirectional data connections with the intelligent control module. The power supply module provides stable power to all electrical modules, forming a complete closed loop for snoring recognition, analysis, intervention, and feedback. The wearable structure uses a highly elastic, adjustable strap structure to comfortably and securely fix the anti-snoring device body to the user's body (e.g., upper arm) or pajamas, ensuring the effectiveness of sound collection and wearing comfort. Specifically: 1. Sound detection module The sound detection module is used to collect sound signals in the sleep environment, complete signal amplification, digital processing, snoring frequency band feature extraction and background noise filtering, and output effective snoring feature data to the intelligent control module; The sound detection module includes two sub-modules: a MEMS microphone and an analog preamplifier. The MEMS microphone is adapted for miniaturized wearable devices and has a pickup frequency range of 200~800Hz (the core frequency band of snoring) to collect raw sound electrical signals. The analog preamplifier has an adjustable amplification factor, ranging from 10 to 100 times, and is used to linearly amplify the weak electrical signals collected by the MEMS microphone to meet the signal strength requirements of subsequent digital processing.

[0030] 2. Intelligent control module The intelligent control module, as the core control unit of the equipment, is used to complete the digitization of all collected signals, algorithm calculation, and logical decision-making, generate control commands for vibration intervention and display output, and receive data from each module to achieve closed-loop control. The intelligent control module comprises three sub-modules: a main processing unit, a clock circuit, and a reset circuit. The main processing unit uses an STM32F103C8T6, or an MCU with equivalent processing capabilities (such as an STM32F401CCU6 or AT32F415C8T7), with a main frequency ≥72MHz, flash memory ≥64KB, and RAM ≥20KB. It is used to perform signal digitization, Fast Fourier Transform (FFT), adaptive noise filtering algorithm, snoring level determination, sleep posture joint analysis, and control command generation. The clock circuit adopts a dual crystal oscillator design with crystal parameters of 32.768KHz (real-time clock) + 8MHz (system main clock), providing a stable and accurate clock source for the MCU and ensuring the timing synchronization of algorithm operation and signal acquisition. The reset circuit includes a button reset sub-module, supporting level reset / button manual reset. When the system experiences program crashes or signal abnormalities, it restores the device to its initial working state, improving stability.

[0031] 3. Vibration control module The vibration control module receives vibration grading instructions from the intelligent control module, converts digital control signals into motor drive signals, and precisely adjusts the speed and working duration of the vibration motor to achieve graded vibration drive control. The vibration control module includes two sub-modules: a PWM signal generation sub-module and a motor drive unit. The PWM signal generation sub-module is embedded in the intelligent control module and outputs a PWM signal with an adjustable duty cycle, ranging from 0 to 100%, used to adjust the operating power of the vibration motor. The motor drive unit uses an L298N or an equivalent motor driver (such as L9110S or DRV8833) to adapt to the DC vibration motor, amplify the PWM control signal, and drive the vibration motor to work.

[0032] 4. Vibration Intervention Module The vibration intervention module is used to perform vibration intervention actions, reminding users to adjust their sleeping posture through vibrations of different intensities and durations, while simultaneously transmitting the motor's operating status back to the intelligent control module to achieve intervention status feedback; The vibration intervention module consists of two sub-modules: a DC vibration motor and a status detection sub-module. The DC vibration motor is adapted to the miniaturization requirements of wearable devices, with a rated voltage of 3.7~5V and three levels of vibration intensity adjustment (low / medium / high, corresponding to speeds of 3000~5000r / min, 5000~8000r / min, and 8000~12000r / min, respectively). Different vibration modes are achieved based on the drive signal. The status detection sub-module is implemented by the current detection circuit embedded in the motor drive unit, which detects the motor's operating current and determines whether the motor is performing the vibration action normally.

[0033] 5. Sleep posture monitoring module The sleep posture monitoring module is used to collect the user's sleep posture information in real time, determine whether the user is in a snoring position (such as lying on their back), and transmit the sleep posture data to the intelligent control module to realize the joint analysis of snoring and sleep posture, thereby improving the accuracy of intervention. The sleep posture monitoring module comprises three sub-modules: an accelerometer, a pressure sensor, and a data preprocessing sub-module. The accelerometer is a triaxial accelerometer with a measurement range of ±2g / ±4g and a resolution of ≥12 bits. It collects the user's body posture angles and movement states to determine supine / side-lying / prone sleeping postures. The pressure sensor is a flexible pressure sensor with a range of 0~10N and a sensitivity of ≥0.1V / N. It is designed to fit the wearable structure and assists in detecting the contact state between the user's body and the wearing structure to verify the sleep posture judgment results. The data preprocessing sub-module is implemented by the ADC module built into the sensor, which converts the analog acquisition signal into a 16-bit digital signal, reducing the computational load on the intelligent control module.

[0034] 6. Display module The display module provides real-time visual feedback on the device's operating status and sleep-related data, and allows users to adjust device parameters via buttons, enhancing the user experience. The display module consists of two sub-modules: an OLED display and a button input sub-module. The OLED display uses a monochrome / color OLED, with a size of 0.96~1.3 inches and a resolution of ≥128×64. The displayed content includes detection status, vibration countdown, snoring frequency, snoring intensity curve, sleep quality score, and wearing tightness indication. The button input sub-module is equipped with 3~5 physical buttons, with functions including power on / off, threshold adjustment, mode switching, and data clearing. It is designed for miniaturization and converts the user's manual operation signals into digital signals for transmission to the intelligent control module.

[0035] 7. Power supply module The power supply module provides a stable and compatible operating voltage for all electrical modules, supports low-power sleep mode, and extends the device's battery life. The power supply module consists of three sub-modules: a lithium battery, a low-power power management chip, and a sleep control sub-module. The lithium battery is a polymer lithium battery with a rated voltage of 3.7V and a capacity of 500~1000mAh, which is suitable for the miniaturization and battery life requirements of wearable devices. The low-power power management chip supports 5V / 1A charging and 3.7V regulated output, and has overcharge, over-discharge, and short-circuit protection functions. The sleep control sub-module receives sleep / wake commands from the intelligent control module and controls the system to enter sleep mode during non-snoring periods to reduce power consumption and extend battery life.

[0036] 8. Wearable structure The wearable structure incorporates a flexible pressure sensor to detect the tightness of the fit and prompts the user to adjust it via an OLED display, ensuring the accuracy of sound collection.

[0037] Furthermore, such as Figure 3 , Figure 4 As shown, when the wearable smart anti-snoring device of the present invention is used, the specific data flow is as follows: Step 1: Data Acquisition The MEMS microphone of the sound detection module collects the raw sound signal and transmits it to the analog preamplifier for amplification. The amplified analog sound signal is then transmitted to the intelligent control module, which performs digital processing, FFT frequency domain analysis, and adaptive noise filtering to extract effective snoring features. The accelerometer and pressure sensor in the sleeping posture monitoring module collect user posture signals, which are then converted into digital signals by the data preprocessing submodule and transmitted to the intelligent control module. Step 2, Data Processing: The intelligent control module receives valid snoring characteristic data from the sound detection module and digital sleeping posture data from the sleeping posture monitoring module. After algorithmic calculation and logical decision-making, it sends graded vibration control commands to the vibration control module, display control commands to the display module, and sleep / wake-up commands to the power supply module. Simultaneously, it receives motor operating status data from the vibration intervention module and user operation signals from the display module to achieve closed-loop regulation. The power supply module receives the sleep / wake-up commands from the intelligent control module and provides a stable operating voltage to all electrical modules to ensure normal operation of the equipment.

[0038] Step 3: Vibration Intervention The vibration control module receives the graded vibration control command from the intelligent control module, generates a PWM drive signal, and transmits it to the DC vibration motor of the vibration intervention module. The vibration intervention module receives the drive signal from the vibration control module, executes the vibration intervention action, and at the same time detects the motor's working status through the status detection submodule and sends the status data back to the intelligent control module.

[0039] Step 4: Data Display: The display module receives display control commands from the intelligent control module and synchronously displays the device's working status and sleep-related data; the button input submodule transmits user operation signals to the intelligent control module to achieve parameter adjustment.

[0040] This invention accurately acquires the core frequency band signal of snoring through a sound detection module, and effectively solves the problem of snoring misidentification under environmental noise interference by combining FFT analysis and adaptive filtering algorithms of the intelligent control module. Simultaneously, relying on a sleep posture monitoring module to acquire user posture data in real time, and in conjunction with a graded feedback mechanism constructed by a vibration control module and a vibration intervention module, it achieves personalized intelligent intervention based on snoring intensity, duration, and sleep posture. Clinically validated, this invention not only improves the snoring recognition accuracy to 96.3%, but also significantly optimizes the user experience and ease of use while ensuring effective snoring reduction through three-level vibration adjustment and a maximum battery life of up to 72 hours. The overall technical effect is superior to existing mainstream anti-snoring products, and it has good potential for widespread application.

[0041] Example 1: Sound Detection and Processing Based on Snoring Frequency Band Characteristics This embodiment provides a sound detection and processing method based on snoring frequency band characteristics to achieve accurate identification of snoring signals. The specific steps are as follows: 1. Wearing device: Secure the wearable smart anti-snoring device to the user's upper arm or pajamas with the adjustable strap, view the tightness prompts on the OLED display, adjust the strap to the appropriate tightness, and ensure that the MEMS microphone can effectively collect sound signals; 2. System Initialization: After the device is powered on, the intelligent control module completes system initialization, the clock circuit provides a stable clock source, the reset circuit is in standby mode, the sound detection module and display module are started, and the power supply module enters normal power supply mode. 3. Sound Acquisition and Amplification: The MEMS microphone acquires sound signals from the sleep environment in real time. The weak electrical signals acquired are transmitted to an analog preamplifier, which amplifies the signals by 10 to 100 times to meet the needs of subsequent digital processing. 4. Signal Processing and Feature Extraction: The amplified analog sound signal is transmitted to the main processing unit (STM32F103C8T6) of the intelligent control module. The main processing unit performs digital processing on the signal, converts the time domain signal into a frequency domain signal through the FFT algorithm, and extracts the energy features in the 200~800Hz frequency band. 5. Noise filtering: The main processing unit uses an adaptive noise filtering algorithm to dynamically adjust the filtering parameters according to the ambient noise level, suppressing interference signals in non-snoring frequency bands (such as ambient noise and bed friction noise), and retaining only the effective features related to snoring. 6. Snoring detection: When the snoring frequency band energy exceeds the preset threshold within 3 consecutive detection cycles (each detection cycle is 1s), the system determines it as a valid snoring event and sends the determination result to the intelligent control module for subsequent vibration intervention control; if it does not exceed the threshold, it is determined as environmental noise and no intervention action is initiated.

[0042] Example 2: Graded vibration intervention control based on snoring intensity and duration This embodiment provides a graded vibration intervention control method based on snoring intensity and duration, which effectively suppresses snoring without affecting the user's sleep comfort. The specific steps are as follows: 1. Parameter preset: Users can preset the snoring intensity threshold (low threshold, medium threshold, high threshold) and duration threshold (first threshold, second threshold) through the button input submodule on the display module, where the first threshold is 5s and the second threshold is 10s; 2. Valid snoring reception: The intelligent control module receives the valid snoring events determined in Example 1, and simultaneously collects data on the duration and energy intensity of the snoring. 3. Grading and Implementation of Interventions: (1) When the snoring intensity is lower than the low threshold and the duration is less than the first threshold (5s), the intelligent control module generates a first-level vibration intervention command and sends it to the vibration control module; the vibration control module outputs a PWM signal with a duty cycle of 20% to drive the DC vibration motor at a speed of 3000~5000r / min to perform low-intensity, short-duration vibration (vibration duration 1s, interval 5s) to remind the user to adjust their sleeping posture slightly; (2) When the snoring intensity is between the low threshold and the medium threshold, and the duration is ≥ the first threshold (5s) and < the second threshold (10s), the intelligent control module generates a second-level vibration intervention command and sends it to the vibration control module; the vibration control module outputs a PWM signal with a duty cycle of 50% to drive the DC vibration motor to perform medium-intensity, intermittent vibration at a speed of 5000~8000r / min (vibration duration 2s, interval 3s). (3) When the snoring intensity is higher than the high threshold and the duration is greater than or equal to the second threshold (10s), the intelligent control module generates a third-level vibration intervention command and sends it to the vibration control module; the vibration control module outputs a PWM signal with a duty cycle of 80% to drive the DC vibration motor to perform high-intensity vibration (vibration duration 3s, interval 2s) at a speed of 8000~12000r / min until the snoring disappears; 4. Intervention Feedback: During vibration intervention, the vibration intervention module monitors the motor's operating status in real time through the status detection submodule and sends the status data back to the intelligent control module. If a motor abnormality is detected, the intelligent control module controls the display module to output a fault prompt and stops the vibration intervention. 5. Intervention Stop: When the intelligent control module detects that the snoring intensity is below the low threshold and the duration is ≥3s, it determines that the snoring has disappeared, sends a stop vibration command, the vibration control module stops driving the motor, and the intervention process ends.

[0043] Example 3: Intelligent Joint Intervention and Control Combining Sleep Posture Monitoring like Figure 2 As shown, this embodiment, based on Embodiments 1 and 2, further introduces a sleep posture monitoring mechanism to improve the effectiveness of anti-snoring intervention. The specific steps are as follows: 1. Sleep posture data acquisition: The three-axis accelerometer (measurement range ±2g, resolution 12-bit) and flexible pressure sensor (range 0~10N, sensitivity 0.1V / N) of the sleep posture monitoring module acquire the user's posture change information in real time. After being converted into a 16-bit digital signal by the data preprocessing submodule, the data is transmitted to the intelligent control module. 2. Sleeping posture judgment: The intelligent control module determines the user's current sleeping posture (supine, side, or prone) based on the received posture data. Supine is a sleeping posture that is more likely to induce snoring, while side and prone are sleeping postures that are less likely to induce snoring. 3. Joint intervention: (1) If the user is determined to be in a supine sleeping position and the intelligent control module receives a valid snoring event, the vibration intervention is performed according to the graded vibration intervention strategy of Example 2, and the current sleeping position and intervention level are displayed in real time through the display module. (2) If the smart control module still detects effective snoring (duration ≥ 3s) after the vibration intervention is performed, and the sleeping position is still supine, the smart control module will adjust the vibration intervention strategy, increase the current vibration level by one level, extend the vibration duration by 0.5s, and shorten the interval duration by 0.5s until the user adjusts to a side or prone sleeping position, or the snoring disappears. (3) If the user is determined to be sleeping on their side or stomach and a valid snoring sound is detected, the intelligent control module determines that the snoring is not caused by sleeping posture factors, executes the first level of vibration intervention, and prompts the user to check the tightness of the fitting or adjust the sleeping posture through the display module. 4. Intervention Optimization: The intelligent control module records the sleeping posture data, snoring data, and intervention effect of each intervention, forming a personalized intervention file for the user. Subsequently, based on the user's sleep habits, the snoring threshold and vibration level parameters are dynamically adjusted to improve the accuracy and comfort of the intervention.

[0044] Example 4: Clinical application data and comparison with existing technologies To verify the practicality and superiority of this invention, 100 snoring patients (40 with mild snoring, 40 with moderate snoring, and 20 with severe snoring) were selected for a clinical trial for 30 days. Simultaneously, existing mainstream anti-snoring devices (anti-snoring pillows, anti-snoring mouthguards, and ordinary wearable anti-snoring devices) were selected as a control group, with 25 patients in each group. The same trial period was observed, and various performance indicators were compared. Specific data are shown in the table below.

[0045] Explanation of testing indicators: Snoring recognition accuracy: The ratio of the number of times a valid snoring sound is correctly identified to the total number of detections; Anti-snoring effectiveness rate: The ratio of the number of patients whose snoring frequency decreased by ≥50% after trial to the total number of patients in the group; Environmental noise immunity: The accuracy of correctly identifying valid snoring sounds in an environment with a signal-to-noise ratio ≥40dB; Apnea-hypopnea Index (AHI): The total number of apnea and hypopnea events that occur per hour of sleep on average.

[0046] The clinical data above demonstrates that the device of this invention significantly outperforms existing mainstream anti-snoring devices in various indicators, including snoring recognition accuracy, anti-snoring effectiveness, adaptability to complex environments, and user comfort. Specifically, the snoring recognition accuracy of this invention reaches 96.3%, a maximum improvement of 17.8 percentage points compared to the control group; the anti-snoring effectiveness for severe snoring reaches 85.0%, an improvement of 27 percentage points compared to ordinary vibration intervention products; and the recognition retention rate reaches 94.1% in a background noise environment of 40dB, effectively addressing the pain point of existing devices being easily interfered with by environmental noise. Furthermore, this invention boasts a long battery life of 72 hours and a wearing comfort score as high as 8.9. For position-related snoring and the resulting mild nocturnal hypopnea symptoms, this invention shows significant improvement effects through a precise sleep posture-linked intervention strategy, demonstrating high value for daily health management and promising clinical application prospects.

Claims

1. A wearable intelligent anti-snoring device based on snoring frequency band analysis and intelligent intervention, characterized in that, It includes an anti-snoring device body and a wearable structure; the anti-snoring device body includes a sound detection module, an intelligent control module, a vibration control module and a vibration intervention module, and the wearable structure is used to fix the anti-snoring device body; The sound detection module is used to collect sound signals from the sleep environment, amplify them, and output them. The intelligent control module is used to perform digital processing, FFT frequency domain analysis, adaptive noise filtering and snoring feature extraction on the received sound signal. When a valid snoring event is determined, a graded vibration control command is generated and output. The vibration control module is used to receive graded vibration control commands, generate drive signals, and output them. The vibration intervention module is used to receive drive signals, perform graded vibration intervention actions, and output its own working status to the intelligent control module.

2. The wearable intelligent anti-snoring device based on snoring frequency band analysis and intelligent intervention according to claim 1, characterized in that, The sound detection module includes a MEMS microphone and an analog preamplifier; the MEMS microphone has a pickup frequency range of 200~800Hz; the amplification factor of the analog preamplifier is adjustable, with an adjustment range of 10~100 times.

3. A wearable intelligent anti-snoring device based on snoring frequency band analysis and intelligent intervention according to claim 1, characterized in that, The intelligent control module performs digital processing, FFT frequency domain analysis, adaptive noise filtering, and snoring feature extraction on the received sound signal. Based on the snoring intensity and duration, it generates graded vibration control commands with the following specific constraints: Level 1 graded vibration control command: Snoring intensity is below the low threshold and duration is <5s; Level 2 graded vibration control command: Snoring intensity is between the low and medium thresholds and the duration is ≥5s and <10s; Level 3 vibration control command: Snoring intensity is higher than the high threshold and the duration is ≥10s.

4. A wearable intelligent anti-snoring device based on snoring frequency band analysis and intelligent intervention according to claim 1, characterized in that, The intelligent control module includes a main processing unit, a clock circuit, and a reset circuit; the main processing unit uses an STM32F103C8T6 microcontroller; the clock circuit adopts a dual crystal oscillator design; the reset circuit includes a button reset submodule, which supports level reset and manual button reset.

5. A wearable intelligent anti-snoring device based on snoring frequency band analysis and intelligent intervention according to claim 1, characterized in that, The vibration control module includes a PWM signal generation submodule and a motor drive unit; the PWM signal generation submodule is embedded in the intelligent control module and outputs a PWM signal with an adjustable duty cycle, the duty cycle adjustment range being 0~100%; the motor drive unit uses an L298N or a motor driver with equivalent driving capability.

6. A wearable intelligent anti-snoring device based on snoring frequency band analysis and intelligent intervention according to claim 1, characterized in that, The vibration intervention module includes a DC vibration motor and a status detection submodule. The rated voltage of the DC vibration motor is 3.7~5V, and the vibration intensity is divided into 3 levels, corresponding to speeds of 3000~5000r / min, 5000~8000r / min, and 8000~12000r / min, respectively. The status detection submodule is implemented by the current detection circuit embedded in the motor drive unit, which is used to detect the working status of the DC vibration motor and transmit it back to the intelligent control module.

7. A wearable intelligent anti-snoring device based on snoring frequency band analysis and intelligent intervention according to claim 1, characterized in that, It also includes a sleeping posture monitoring module, which includes an accelerometer, a pressure sensor, and a data preprocessing submodule; the accelerometer is a triaxial accelerometer; the pressure sensor is a flexible pressure sensor; the data preprocessing submodule is implemented by the ADC module built into the sensor, which is used to convert the analog acquisition signal into a 16-bit digital signal and transmit it to the intelligent control module.

8. A wearable intelligent anti-snoring device based on snoring frequency band analysis and intelligent intervention according to claim 7, characterized in that, The intelligent control module performs digital processing, FFT frequency domain analysis, adaptive noise filtering, and snoring feature extraction on the received sound signal to obtain snoring intensity and duration. Simultaneously, it acquires sleeping posture data and generates graded vibration control commands based on the snoring intensity, duration, and sleeping posture data, with specific limitations as follows: When the user is sleeping in a supine position: Level 1 graded vibration control command: Snoring intensity is below the low threshold and duration is <5s; Level 2 graded vibration control command: Snoring intensity is between the low and medium thresholds and the duration is ≥5s and <10s; Level 3 vibration control command: Snoring intensity is higher than the high threshold and the duration is ≥10s; If snoring continues for ≥3 seconds and the sleeping position remains supine after implementing the corresponding level of vibration intervention, the current vibration level will be increased by one level. When the user is sleeping on their side or stomach, regardless of the intensity and duration of snoring, a first-level graded vibration control command is generated.

9. A wearable intelligent anti-snoring device based on snoring frequency band analysis and intelligent intervention according to claim 1, characterized in that, It also includes a display module, which receives display control commands from the intelligent control module, displays the device's working status and sleep-related data in real time, and receives user operation signals and transmits them to the intelligent control module. The display module includes an OLED display screen and a button input submodule. The OLED display screen displays the detection status, vibration countdown, snoring frequency, snoring intensity curve, sleep quality score, and wearing tightness indication. The button input submodule has 3-5 physical buttons with functions for power on / off, threshold adjustment, mode switching, and data clearing. It also includes a power supply module, which comprises a lithium battery, a low-power power management chip, and a sleep control submodule; the lithium battery is a polymer lithium battery with a rated voltage of 3.7V and a capacity of 500~1000mAh; the low-power power management chip supports 5V / 1A charging and 3.7V regulated output, and has overcharge, over-discharge, and short-circuit protection functions; the sleep control submodule is used to receive sleep / wake-up commands from the intelligent control module and control the system to enter a low-power sleep mode.

10. A wearable intelligent anti-snoring device based on snoring frequency band analysis and intelligent intervention according to claim 1, characterized in that, The wearable structure features a highly elastic adjustable strap with a built-in flexible pressure sensor to detect the tightness of the fit and prompt the user to adjust it via a display module.

Images