Audio sleep-aiding system

Abstract

The invention belongs to the technical field of sleep aiding systems, and particularly discloses an audio sleep aiding system which is characterized in that brain waves of a user are collected through a sleep instrument and converted into corresponding brain wave digital signals to be uploaded to a cloud server, and the brain wave digital signals are subjected to sleep staging through the cloud server; according to the sleep staging method, a sleep staging result is obtained, highly-adaptive customized sleep-aiding audios are generated according to the sleep staging result, then the customized sleep-aiding audios are issued, power amplification and playing of the sleep-aiding audios are carried out through a sleep instrument, and a user is enabled to quickly relax and enter a sleep state. According to the method, through efficient automatic sleep staging, accurate recognition and classification of sleep stages can be achieved, the sleep state of the user can be effectively improved through adaptation of customized audio, the audio sleep aiding technology is guided to deeply develop towards individuation and intelligentization, brand new vitality is injected for improving the sleep quality and life quality of the user, and the method is worthy of popularization and application. The method has a wide application prospect and remarkable social and economic benefits.

Description

Technical Field

[0001] This invention belongs to the technical field of sleep aid systems, and specifically relates to an audio sleep aid system. Background Technology

[0002] Music-assisted sleep refers to the method of improving sleep quality by listening to specific types of music to regulate emotions and relieve stress. The main functions of music-assisted sleep include the following:

[0003] 1. Affects the autonomic nervous system. Music can affect the autonomic nervous system, especially the parasympathetic nervous system. Soothing music can stimulate the activity of the parasympathetic nervous system, reduce the excitability of the sympathetic nervous system, and put the body into a relaxed state, preparing it for sleep.

[0004] 2. Changes in brainwave activity. Different types of music can cause different changes in brainwaves. When people listen to soothing music, the brain produces more alpha waves and theta waves. Alpha waves usually appear when people are relaxed, meditating, or resting with their eyes closed, while theta waves are associated with deep relaxation and light sleep.

[0005] 3. Relieve stress and anxiety. In modern life, people face various stresses and anxieties, and these negative emotions can affect sleep quality. Listening to music can be an effective form of psychotherapy to help people relieve stress and anxiety. Music can touch people's emotions, evoke resonance, and allow people to temporarily forget their troubles, relax, and thus help them fall asleep.

[0006] 4. Create a sleep environment. Music can create a quiet and comfortable sleep environment. Listening to music before bed can block out external noise, making it easier for people to concentrate and relax.

[0007] 5. Distract yourself. For people prone to overthinking, listening to music before bed can distract them and reduce distractions. When people focus on the melody and rhythm of music, brain activity shifts from thinking about problems to appreciating the music, thereby reducing anxiety and tension and helping them fall asleep.

[0008] 6. Create sleep cues. If a person develops the habit of listening to music before bed, then the music will become a sleep cue. When specific music is heard, the brain will automatically enter a sleep preparation state, and the body will gradually relax, preparing to fall asleep.

[0009] However, most existing audio sleep aid products require users to select the corresponding type of music according to their own preferences. In other words, the audio can only be subjectively selected by the user and cannot be automatically adapted to the user's sleep stage, nor can personalized sleep aid audio customization be achieved. Summary of the Invention

[0010] The purpose of this invention is to provide an audio-based sleep aid system to solve the aforementioned problems existing in the prior art.

[0011] To achieve the above objectives, the present invention adopts the following technical solution:

[0012] This invention provides an audio-assisted sleep system, including a head-mounted sleep aid, a mobile terminal, and a cloud server. The head-mounted sleep aid is equipped with an EEG sensor, a motherboard, and a speaker. The motherboard integrates an EEG pre-amplification circuit, a microcontroller (MCU), a Bluetooth module, and an audio amplifier module. The MCU is connected to both the EEG pre-amplification circuit and the Bluetooth module. The audio amplifier module is connected to both the Bluetooth module and the speaker. The EEG pre-amplification circuit and the EEG sensor are connected via a magnetic connector. The Bluetooth module establishes a Bluetooth connection with the mobile terminal, and the mobile terminal establishes a communication connection with the cloud server. The EEG sensor collects the user's raw EEG signal and transmits it to the EEG pre-amplification circuit via the magnetic connector. The EEG pre-amplification circuit amplifies and denoises the raw EEG signal to obtain a denoised EEG signal. The received EEG signal is converted into a digital brainwave signal and transmitted to a Bluetooth module. The Bluetooth module then transmits the digital brainwave signal to a mobile terminal, which uploads it to a cloud server. The cloud server classifies and identifies the digital brainwave signal to obtain corresponding sleep stage results. Based on the sleep stage results, it matches corresponding customized sleep aid audio and transmits the customized sleep aid audio to the mobile terminal. The mobile terminal then transmits the customized sleep aid audio to the Bluetooth module, which in turn transmits it to an audio amplifier module. The audio amplifier module amplifies the customized sleep aid audio to obtain amplified sleep aid audio, which is then transmitted to a speaker. The speaker plays the amplified sleep aid audio to the user.

[0013] In one possible design, the EEG sensor is used for single-channel raw EEG signal acquisition and transmission.

[0014] In one possible design, when the cloud server classifies and identifies EEG digital signals, it first performs time-domain and frequency-domain decomposition on the EEG digital signals, and extracts multivariate statistical features based on the decomposed components obtained from the time-domain and frequency-domain decomposition; then, it inputs the multivariate statistical features into a pre-trained and tested random forest classifier for classification prediction to obtain the classification prediction result; and then uses a hidden Markov model to perform deep optimization on the classification prediction result to obtain the final sleep staging result.

[0015] In one possible design, the microcontroller (MCU) is an EPC001 chip.

[0016] In one possible design, the microcontroller (MCU) is connected to a power on / off button circuit, which is used to control the power on / off of the MCU.

[0017] In one possible design, the microcontroller (MCU) is connected to a status indicator light, which is used to indicate the operating status of the microcontroller (MCU).

[0018] In one possible design, the head-mounted sleep aid is equipped with a battery, and the motherboard integrates a battery management system and a power module. The battery management system is used to manage the power path of the battery and transmit the battery power to the power module. The power module is used to regulate the battery power and output regulated power to power the microcontroller (MCU) and Bluetooth module.

[0019] In one possible design, the power module employs an ME6212 series low dropout linear regulator.

[0020] In one possible design, the battery management system includes a BQ24076 linear charger connected to a magnetic connector for receiving external magnetic charging power and charging the battery using the magnetic charging power.

[0021] In one possible design, the audio amplifier module includes an LM4674 audio power amplifier.

[0022] Beneficial Effects: This invention collects the user's brainwaves using a sleep aid device, converts them into corresponding digital brainwave signals, and uploads them to a cloud server. The cloud server then stages the sleep patterns of the brainwave signals and generates highly tailored, customized sleep-aid audio based on the staged sleep results. This customized sleep-aid audio is then distributed and played through the sleep aid device, allowing the user to quickly relax and fall asleep. This invention, through efficient automatic sleep staging, achieves accurate identification and classification of sleep stages. By adapting customized audio, it effectively improves the user's sleep state, leading audio-assisted sleep technology towards personalization and intelligence. It injects new vitality into improving users' sleep quality and quality of life, possessing broad application prospects and significant socio-economic benefits. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention 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 the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of the system architecture of the present invention;

[0025] Figure 2 This is a circuit diagram of the EPC001 chip;

[0026] Figure 3 This is a circuit diagram of the power button circuit;

[0027] Figure 4 This is a circuit diagram for a status indicator light.

[0028] Figure 5 This is a circuit diagram of the BQ24076 linear charger.

[0029] Figure 6 This is a circuit diagram of the power module;

[0030] Figure 7 This is a circuit diagram of the EEG pre-amplifier acquisition circuit.

[0031] Figure 8 This is a circuit diagram of an audio power amplifier module. Detailed Implementation

[0032] It should be noted that the descriptions of these embodiments are intended to aid in understanding the invention and do not constitute a limitation thereof. The specific structural and functional details disclosed herein are merely for describing exemplary embodiments of the invention. However, the invention may be embodied in many alternative forms and should not be construed as being limited to the embodiments described herein.

[0033] It should be understood that, unless otherwise explicitly specified and limited, the corresponding terms should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be an electrical connection, a direct connection, or an indirect connection through an intermediate medium; it can also refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in the embodiments according to the specific circumstances.

[0034] Specific details are provided in the following description to provide a complete understanding of the exemplary embodiments. However, those skilled in the art will understand that the exemplary embodiments can be implemented without these specific details. For example, the system may be shown in block diagrams to avoid obscuring the example with unnecessary details. In other embodiments, well-known processes, structures, and techniques may be shown without non-essential details to avoid obscuring the embodiments.

[0035] Example:

[0036] This embodiment provides an audio-based sleep aid system, such as... Figure 1 As shown, the device includes a head-mounted sleep aid, a mobile terminal, and a cloud server. The head-mounted sleep aid is equipped with an EEG sensor, a motherboard, and a speaker. The motherboard integrates an EEG pre-amplification circuit, a microcontroller (MCU), a Bluetooth module, and an audio amplifier module. The MCU is connected to both the EEG pre-amplification circuit and the Bluetooth module. The audio amplifier module is connected to both the Bluetooth module and the speaker. The EEG pre-amplification circuit and the EEG sensor are connected via a magnetic connector. The Bluetooth module establishes a Bluetooth connection with the mobile terminal, and the mobile terminal establishes a communication connection with the cloud server. The EEG sensor collects the user's raw EEG signal and transmits it to the EEG pre-amplification circuit via the magnetic connector. The EEG pre-amplification circuit amplifies and de-noises the raw EEG signal to obtain a de-noised EEG signal. The noise-reduced EEG signal is transmitted to a microcontroller (MCU). The MCU converts the received EEG signal into a digital brainwave signal and transmits it to a Bluetooth module. The Bluetooth module then transmits the digital brainwave signal to a mobile terminal. The mobile terminal uploads the digital brainwave signal to a cloud server. The cloud server classifies and identifies the digital brainwave signal to obtain corresponding sleep stage results. Based on the sleep stage results, it matches a customized sleep aid audio and transmits the customized sleep aid audio to the mobile terminal. The mobile terminal then transmits the customized sleep aid audio to the Bluetooth module, which in turn transmits it to an audio amplifier module. The audio amplifier module amplifies the customized sleep aid audio to obtain amplified sleep aid audio, which is then transmitted to a speaker. The speaker plays the amplified sleep aid audio to the user.

[0037] In practice, users can wear a head-mounted sleep aid and connect it to a mobile device (such as a mobile phone or computer) via Bluetooth. During the user's sleep, the EEG (Electroencephalogram) sensor on the head-mounted sleep aid can collect the user's raw EEG signals in real time. Human brainwaves generally include the following types:

[0038] 1. Alpha waves: When alpha waves are present, the human brain enters a right-brain activity state, subconscious activity masks conscious activity, resulting in wakefulness and relaxation, concentration, quick thinking, faster information reception and transmission, improved work and study efficiency, enhanced immunity, creativity, and emotional stability.

[0039] 2. Beta waves: When beta waves are present, the human brain enters a left-brain activity state, conscious activity masks subconscious activity, leading to mental tension, distraction, low work efficiency, and easy fatigue. In everyday life (especially in the work environment), the brain is most active in the beta wave band. For example, during study and work, or in conscious thinking activities such as reasoning, calculation, and driving, beta waves are the most common high-frequency waves in a conscious state.

[0040] 3. Theta waves appear in the moments before falling asleep and upon waking in the morning. It is a very brief brainwave state, occurring when a person is in a light sleep or semi-conscious state, with increased unconscious activity and receptiveness to suggestion. It is the deepest level of trance and is used in regression hypnosis. These brainwaves are related to our deepest feelings and emotions. Prolonged exposure to theta waves (i.e., light sleep) can negatively impact sleep quality.

[0041] 4. Delta waves appear in deep sleep, when there is no feedback from brain activity. They are brain waves at the unconscious level and are the slowest brain waves.

[0042] The EEG sensor collects the raw EEG signal and transmits it to the EEG pre-amplifier circuit. This circuit amplifies and reduces the noise of the raw EEG signal, resulting in a denoised EEG signal. The microcontroller (MCU) then converts the EEG signal into a digital brainwave signal, which is transmitted to a mobile terminal via Bluetooth. The mobile terminal, using an application such as a mobile app, uploads the brainwave signal to a cloud server. The cloud server classifies and identifies the brainwave signal, staging sleep patterns and matching customized sleep-aid audio based on the sleep stage. This customized sleep-aid audio is then transmitted to the mobile terminal. The mobile terminal then transmits the customized sleep-aid audio to the Bluetooth module, which in turn transmits it to the audio amplifier module. Finally, the audio amplifier module amplifies the customized sleep-aid audio signal, producing amplified sleep-aid audio that is played to the user through a speaker, helping them relax and fall asleep quickly.

[0043] Furthermore, the EEG sensor is used for single-channel raw EEG signal acquisition and transmission. Using a single-channel EEG sensor to acquire and transmit EEG signals ensures data validity and reliability while avoiding the increased number of electrodes and connecting wires associated with multi-channel signal acquisition. This effectively improves the user experience, reduces interference with sleep comfort, and achieves a perfect balance between technical performance and user experience. Users can enjoy personalized audio services while maintaining a comfortable sleep state, thus enhancing the system's practical application value.

[0044] Furthermore, when the cloud server classifies and identifies digital brainwave signals:

[0045] First, multimodal decomposition is performed on the EEG digital signal, including time-domain decomposition and frequency-domain decomposition. Frequency-domain decomposition, through the design of a finite impulse response (FIR) bandpass filter, accurately decomposes the EEG digital signal into characteristic waves of different frequency bands, deeply mining the potential information of the EEG signal at the frequency domain level, providing crucial data support for subsequent feature extraction and analysis. Time-domain decomposition employs Empirical Mode Decomposition (EMD) technology to extract the intrinsic patterns of the EEG digital signal, comprehensively analyzing the subtle changes of the EEG signal in the time domain, laying a solid foundation for feature extraction and in-depth analysis, and ensuring that the system can capture the characteristic information of the EEG signal from all angles. The frequency-domain and time-domain decompositions complement each other and synergistically enhance each other, providing the system with richer and more comprehensive EEG signal characteristic information, significantly improving the overall performance and classification accuracy of the system. This enables the system to more accurately identify the user's sleep state and emotional changes, providing solid data support for personalized audio generation.

[0046] After obtaining the corresponding decomposed components through time-domain and frequency-domain decomposition, multivariate statistical features can be extracted based on these components. These multivariate statistical features can cover mean, maximum, minimum, energy, standard deviation, skewness, kurtosis, fractal dimension, etc., to comprehensively characterize the features of EEG signals, providing detailed and rich data support for subsequent classification and ensuring that the system can accurately identify the user's sleep state.

[0047] After extracting multivariate statistical features, these features are input into a pre-trained and tested random forest classifier for classification prediction, yielding the prediction results. The optimal number of trees in the random forest classifier is 120. This precise selection of the parameter provides crucial support for the model's efficient operation and accurate classification, ensuring the system maintains high accuracy and efficiency when processing complex EEG data. During the pre-training and testing of the random forest classifier, single-channel EEG data from the Sleep-EDF and Sleep-EDF Expanded databases provided by PhysioNet can be selected as the training and testing sets. These datasets cover rich sleep EEG information, providing high-quality and diverse data resources for system training and validation, ensuring the system's robustness and generalization ability. With its powerful classification performance and excellent anti-overfitting ability, the random forest classifier can accurately classify sleep stages, laying a solid foundation for subsequent HMM optimization and ensuring the initial accuracy of sleep staging.

[0048] After initial classification using a random forest classifier, a Hidden Markov Model (HMM) can be used to deeply optimize the classification prediction results of the random forest classifier, yielding the final sleep stage results. The HMM utilizes the classification prediction results of the random forest classifier as observations to deeply optimize the classification prediction results, further improving the accuracy and reliability of sleep stage classification and achieving refined identification of sleep stages. The HMM does not require predefined, cumbersome rules and can automatically optimize the classification results, exhibiting high flexibility and adaptability, and is applicable to any machine learning-based sleep stage classification method. Independent subject testing on the Sleep-EDF Expanded database shows that the HMM refinement process is significantly effective, greatly improving classification accuracy, especially in the identification rate of sleep stage 1 (N1), highlighting the unique advantages of this optimization method in identifying key sleep stages and providing strong experimental evidence for the practical application of the system.

[0049] Furthermore, such as Figure 2The microcontroller (MCU) described above can be the EPC001 chip. The EPC001 chip integrates a high-precision analog front-end for EEG, ECG, EMG, and PPG, a 64MHz RISC-V microcontroller (MCU), powerful ECG and PPG feature extraction capabilities (Heart APP), and a 24-bit high-precision ADC, making it a highly integrated wearable health device solution chip. Pulse waves and ECGs can be sampled synchronously at independent sampling rates; the maximum configurable sampling rate for pulse waves is 4kHz, and for ECGs, it is 32kHz. The EPC001 chip features ultra-low noise programmable amplifiers for EEG, ECG, and EMG, and is equipped with a 24-bit high-precision ADC, enabling weak bioelectrical signals to be converted into digital signals with maximum accuracy, ensuring the precision of measurement results. Meanwhile, the EPC001 chip is equipped with four programmable LED drivers for the pulse wave PPG acquisition system, compatible with various LED specifications. In addition, its low-power LED analog front-end circuit and 24-bit high-precision ADC are used to acquire pulse wave PPG signals, capturing complete pulse wave PPG signals with high dynamic range and high precision. The EPC001 chip also integrates a 64MHz 32-bit RISC-V microcontroller (MCU) and various peripherals, supporting complex signal processing algorithms and offering strong scalability. Its diverse peripheral interfaces allow the SoC to externally expand with multiple different sensors to extract signal features such as ECG and PPG. The powerful RISC-V core MCU processing and its memory space facilitate the integration of different wireless transceiver protocol stacks for transmitting measurement data.

[0050] The microcontroller (MCU) can be connected to a corresponding power on / off button circuit, which enables power on / off control of the MCU. For example... Figure 3As shown, the power on / off button circuit can be controlled by an AO3401A PMOS transistor. The PMOS transistor's gate has two paths that can be grounded. When the button is pressed, the PMOS transistor's gate is grounded through the diode and the button, turning on the PMOS transistor and supplying power to the system. When the subsequent circuit pulls the GPIO8_KEY1 pin high, the transistor Q2 turns on, grounding the PMOS transistor's gate, and the PMOS transistor continues to conduct. Power-on procedure: In the absence of power, the system can be started using button SW1. When button SW1 is pressed, the gate of PMOS transistor Q1 is grounded, turning on Q1 and supplying power to the subsequent system through Q1. When pin GPIO5_KEY1 is low, pressing the button pulls GPIO8_KEY2 high; releasing the button turns on Q1. After power-on, the MCU can detect button presses via GPIO5_KEY1. When a button is pressed, GPIO5_KEY1 is grounded and at a low level. When the button is released, GPIO5_KEY1 is pulled up to a high level through a pull-up resistor. At this time, the button can be used as a function key for long press, short press, and double press to achieve the desired function. System shutdown: When the MCU detects a long press or double press, it pulls GPIO8_KEY2 low, Q2 is turned off, Q1 is turned off, and the system shuts down. Diodes D1 and D2 prevent the VCC terminal from being connected to the 3.3V power supply terminal through the R2, R11 circuit, preventing the 3.3V power supply voltage from rising and damaging the system. The microcontroller MCU can be connected to... Figure 4 The status indicator lights shown indicate the operating status of the microcontroller (MCU).

[0051] Furthermore, the head-mounted sleep aid is equipped with a battery, and the motherboard integrates a battery management system and a power module. The battery management system is used to manage the power path of the battery and transmit the battery power to the power module. The power module is used to regulate the voltage of the battery power and output regulated power to power the microcontroller MCU and Bluetooth module.

[0052] like Figure 5 As shown, the battery management system includes a BQ24076 linear charger. The BQ24076 linear charger connects to a magnetic connector for receiving external magnetic charging power and using it to charge the battery. The BQ24076 features system power path management suitable for space-constrained portable applications, input voltage protection, and support for unregulated adapters. The BQ24076 also features Dynamic Power Path Management (DPPM), allowing it to charge the battery independently while providing power.

[0053] like Figure 6As shown, the power module uses the ME6212 series low-dropout linear regulator. The ME6212 series regulator is a high-precision, high-ripple rejection ratio, low-noise, ultra-fast response, and low-dropout linear regulator manufactured using CMOS technology. The ME6212 series regulator integrates a fixed reference voltage source, error correction circuit, current limiting circuit, phase compensation circuit, and a low-resistance MOSFET, achieving high ripple rejection, low output noise, ultra-fast response, and low dropout performance. The ME6212 series regulator is compatible with ceramic capacitors, which are smaller than tantalum capacitors, and eliminates the need for a 0.1μF bypass capacitor, saving space. The high-speed response of the ME6212 series regulator can handle fluctuations in load current, making it particularly suitable for handheld and RF products. The output can be turned off via the CE pin on the control chip, with power consumption below 1μA after shutdown.

[0054] Furthermore, the EEG pre-acquisition circuit is as follows: Figure 7 As shown, RLD_OUT has a common-mode electrode bias of Vref and feeds back an inverted common-mode noise signal (enoise_cm) to reduce the total noise at the input of the measurement amplifier gain stage. ECG_P and ECG_N are separated to indicate that RLD_OUT provides a common-mode reference point for a portion of the EEG signal. Assuming the common-mode noise signal enoise is parasiticly coupled to the input, the feedback of the inverted common-mode noise signal enoise_cm reduces the total noise at each input and filters out any remaining noise using external methods or by utilizing the common-mode rejection ratio (CMRR) of the measurement amplifier.

[0055] Furthermore, the audio amplifier module includes, as follows: Figure 8 The LM4674 audio power amplifier is shown. The LM4674 is a single-supply, high-efficiency, 2.5W / channel, filterless switching audio amplifier. Its low-noise PWM architecture eliminates the output filter, thereby reducing the number of external components, board space consumption, and system cost, and simplifying the design. Compared to traditional Class AB amplifiers, the LM4674 has higher efficiency. The Bluetooth module uses dual-mode Bluetooth technology, compatible with Bluetooth V5.0 and V2.1+EDR, and supports Bluetooth Piconet and Scatternet networking protocols.

[0056] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An audio-based sleep aid system, characterized in that, The device includes a head-mounted sleep aid, a mobile terminal, and a cloud server. The head-mounted sleep aid is equipped with an EEG sensor, a motherboard, and a speaker. The motherboard integrates an EEG pre-amplification circuit, a microcontroller (MCU), a Bluetooth module, and an audio amplifier module. The MCU is connected to both the EEG pre-amplification circuit and the Bluetooth module. The audio amplifier module is connected to both the Bluetooth module and the speaker. The EEG pre-amplification circuit and the EEG sensor are connected via a magnetic connector. The Bluetooth module establishes a Bluetooth connection with the mobile terminal, and the mobile terminal establishes a communication connection with the cloud server. The EEG sensor collects the user's raw EEG signal and transmits it to the EEG pre-amplification circuit via the magnetic connector. The EEG pre-amplification circuit amplifies and denoises the raw EEG signal to obtain a denoised EEG signal. The noise-reduced EEG signal is transmitted to a microcontroller (MCU). The MCU converts the received EEG signal into a digital brainwave signal and transmits it to a Bluetooth module. The Bluetooth module then transmits the digital brainwave signal to a mobile terminal. The mobile terminal uploads the digital brainwave signal to a cloud server. The cloud server classifies and identifies the digital brainwave signal to obtain corresponding sleep stage results. Based on the sleep stage results, it matches a customized sleep aid audio and transmits the customized sleep aid audio to the mobile terminal. The mobile terminal then transmits the customized sleep aid audio to the Bluetooth module, which in turn transmits it to an audio amplifier module. The audio amplifier module amplifies the customized sleep aid audio to obtain amplified sleep aid audio, which is then transmitted to a speaker. The speaker plays the amplified sleep aid audio to the user.

2. The audio sleep aid system according to claim 1, characterized in that, The EEG sensor is used for single-channel raw EEG signal acquisition and transmission.

3. The audio sleep aid system according to claim 1, characterized in that, When classifying and recognizing digital brainwave signals, the cloud server first performs time-domain decomposition and frequency-domain decomposition on the digital brainwave signals, and then extracts multivariate statistical features based on the decomposed components obtained from the time-domain decomposition and frequency-domain decomposition. Then, the multivariate statistical features are input into a pre-trained and tested random forest classifier for classification prediction to obtain the classification prediction results; then, a hidden Markov model is used to perform deep optimization on the classification prediction results to obtain the final sleep staging results.

4. The audio sleep aid system according to claim 1, characterized in that, The microcontroller (MCU) uses the EPC001 chip.

5. An audio sleep aid system according to claim 1, characterized in that, The microcontroller (MCU) is connected to a power on / off button circuit, which is used to control the power on / off of the microcontroller (MCU).

6. The audio sleep aid system according to claim 1, characterized in that, The microcontroller (MCU) is connected to a status indicator light, which is used to indicate the working status of the microcontroller (MCU).

7. An audio sleep aid system according to claim 1, characterized in that, The head-mounted sleep aid is equipped with a battery. The motherboard integrates a battery management system and a power module. The battery management system is used to manage the power path of the battery and transmit the battery power to the power module. The power module is used to regulate the voltage of the battery power and output regulated power to power the microcontroller (MCU) and Bluetooth module.

8. An audio sleep aid system according to claim 7, characterized in that, The power module uses the ME6212 series low dropout linear regulator.

9. An audio sleep aid system according to claim 7, characterized in that, The battery management system includes a BQ24076 linear charger, which is connected to a magnetic connector for receiving external magnetic charging power and charging the battery using the magnetic charging power.

10. An audio sleep aid system according to claim 1, characterized in that, The audio power amplifier module includes the LM4674 audio power amplifier.

Images