Earphone wearing fitting degree self-checking and sound leakage compensation system

Abstract

The invention discloses an earphone wearing fit degree self-inspection and sound leakage compensation system, which comprises a sensor module, which comprises a multi-mode distributed sensor array composed of a contact pressure sensor, a non-contact distance sensor and an inertial measurement unit, and is used for collecting wearing state physical signals in real time, wherein the contact type pressure sensor adopts a flexible piezoelectric film array and is in grid layout along an auricle contact surface; the non-contact distance sensor adopts an infrared TOF sensor and is arranged in an epitaxial area of the sound production unit; and the inertial measurement unit comprises a three-axis gyroscope and an accelerometer.

Description

Technical Field

[0001] The present invention relates to the technical field of earphone systems, and in particular to an earphone wearing fit self-detection and sound leakage compensation system. Background Art

[0002] Current mainstream headphone products have significant technical bottlenecks in wearing status monitoring and acoustic compensation:

[0003] Single detection dimension: Existing technologies (such as CN110913313A) mostly use a single pressure sensor or accelerometer to determine the wearing status, and cannot simultaneously sense multi-dimensional parameters such as ear canal gap, contact pressure distribution, and motion disturbance;

[0004] Compensation mechanism split: The traditional solution (US20200162874A1) only compensates for sound leakage by adjusting the EQ curve, but ignores the problem of aggravated physical leakage caused by the deformation of the earphone structure. Actual measurements show that when the earplug is offset by 0.5mm, the low-frequency attenuation can reach 12dB;

[0005] Poor environmental adaptability: The existing active noise reduction system (refer to Bose QC45) has not established a multi-parameter coupling compensation model of temperature-noise-motion. The softening of the memory foam ear pads in a high temperature environment will cause up to 40% deformation compensation failure;

[0006] Lack of predictive capabilities: Mainstream products (such as AirPods Pro) can only passively respond to immediate leaks and are unable to predict the trend of fit degradation through historical wearing data, resulting in compensation delays of more than 500ms. Summary of the invention

[0007] The present invention aims to solve one of the technical problems in the related art at least to a certain extent. To this end, one object of the present invention is to provide an earphone wearing fit self-checking and sound leakage compensation system, comprising:

[0008] The sensor module includes a multi-modal distributed sensor array consisting of a contact pressure sensor, a non-contact distance sensor, and an inertial measurement unit to collect physical signals of the wearing status in real time, including:

[0009] The contact pressure sensor adopts a flexible piezoelectric film array, which is arranged in a grid along the auricle contact surface, has a thickness of 10 μm and a pressure resolution of 0.1 kPa;

[0010] The non-contact distance sensor adopts an infrared TOF sensor, which is arranged in the outer area of ​​the sound unit, has an operating frequency of 10MHz and a detection resolution of 0.05mm;

[0011] The inertial measurement unit includes a three-axis gyroscope (range ±2000dps) and an accelerometer (sampling rate 200Hz);

[0012] The central processing module performs feature fusion on heterogeneous sensor data based on a multi-stage Bayesian network to construct a dynamic wearing model, and the model determines the fitting level by analyzing three elements: the pressure distribution gradient curve, the residual space ratio of the concha, and the motion stability parameter;

[0013] The sound leakage compensation execution module includes an acoustic parameter compensation unit and a structural deformation compensation unit, where:

[0014] The acoustic parameter compensation unit performs adaptive ANC algorithm updates for multi-band dynamic EQ adjustment and leakage sound field analysis;

[0015] The structural deformation compensation unit uses a shape memory alloy drive mechanism to adjust the curvature of the headphone structure;

[0016] The user adaptive feedback module includes a bio-impedance sensor (operating frequency 10 kHz, measurement accuracy ±5 Ω) and a mobile terminal interaction interface;

[0017] The environment compensation module integrates a MEMS microphone array and a surface temperature sensor, where:

[0018] The MEMS microphone array uses 4-microphone beamforming technology to monitor environmental noise;

[0019] The surface temperature sensor is an NTC thermistor with an accuracy of ±0.5 °C;

[0020] The system realizes real-time signal processing and control logic through an STM32F7 series DSP chip.

[0021] Preferably, the method for establishing the dynamic wearing model includes:

[0022] Construct a three-dimensional auricle contact feature matrix F c =(P aug ,σ p ,D gap ), where:

[0023] P aug is the average pressure value of the auricle contact surface, in kPa;

[0024] σ p is the standard deviation of the pressure distribution;

[0025] D gap is the gap distance between the sound generating unit and the concha, in mm;

[0026] Use a support vector machine classifier (kernel function is RBF, penalty factor C = 2.5) to output four levels of fitting degrees (A - D levels), where:

[0027] C-level trigger acoustic parameter compensation;

[0028] D-level synchronous activation of acoustic parameter compensation and structural deformation compensation.

[0029] Preferably, the calculation of the motion stability parameter of the inertial measurement unit adopts the formula:

[0030]

[0031] where a rea1 is the measured acceleration value, a ref is the static reference value, a max is the preset maximum threshold;

[0032] When K m < 0.7, the compensation activation condition is triggered.

[0033] Preferably, the acoustic parameter compensation performs the following operations:

[0034] The dynamic EQ adjusts the center frequencies of three frequency bands: 150 Hz, 500 Hz, and 2 kHz;

[0035] The compensation amount of the ANC algorithm is calculated as ΔG = ΔG0×(1 + 0.3×L d ), where L d is the relative value of the detected leakage amount, and the convergence speed is increased by 40% compared to the baseline.

[0036] Preferably, the working logic of the user adaptive feedback module includes:

[0037] When the detected wearing pressure offset > 15% and lasts for 20 seconds, a three-dimensional graphical wearing guidance animation is provided through the mobile terminal APP;

[0038] The bioimpedance sensor data is used to correct the misjudgment of the skin contact state of the pressure sensor.

[0039] Preferably, the control strategy of the environmental compensation module includes:

[0040] When the environmental noise > 65 dB SPL (A-weighted), the low-frequency compensation weight is increased by 30%;

[0041] When the temperature sensor detects that the contact surface temperature > 35 °C, the structural deformation compensation amplitude is limited within 80% of the calibrated value.

[0042] Preferably, it also includes a predictive fitting maintenance function, which is achieved in the following way:

[0043] Establish a user-exclusive wearing mode database in the embedded memory, storing ≥ 50 historical wearing records;

[0044] Use an LSTM neural network (deployed in the TensorFlowLite framework) to predict the change trend of the fitting degree in the next 15 minutes;

[0045] When the predicted trend slope k < -0.05 / min, activate the compensation subsystem 200 ms in advance.

[0046] Preferably, it supports multi-device networking and collaboration functions, including:

[0047] Realize data synchronization of bilateral earphones through a 2.4GHz band wireless communication module (transmission delay < 5ms);

[0048] When the detected bilateral fitting degree difference > 2 levels, start the group equalization compensation mode and preferentially adjust the consistency of the compensation phase in the low frequency band.

[0049] Preferably, its industrial design includes:

[0050] A hidden layout structure for sensors, with the housing opening rate < 3%;

[0051] A composite heat conduction structure (thermal conductivity ≥ 5W / mK tested according to ASTM D5470 standard);

[0052] A replaceable earplug contact surface component, which can be manually replaced within 30 seconds using a quick-release buckle structure.

[0053] Preferably, the deformation response time of the shape memory alloy drive mechanism ≤ 150ms, the curvature adjustment range is ±15°, and the drive accuracy is 0.5°.

[0054] The above solutions of the present invention have at least the following beneficial effects:

[0055] Precise perception and decision-making:

[0056] Multi-source data fusion: Through the heterogeneous data fusion of a contact pressure sensor (0.1kPa resolution), a TOF distance sensor (0.05mm accuracy) and an IMU (200Hz sampling), the misjudgment rate of the wearing state is reduced to less than 5% (compared with 30% of the traditional solution);

[0057] Advantages of dynamic modeling: Use a three-dimensional feature matrix F constructed by a Bayesian network + SVM classifier c , to achieve four-level fitting degree classification (A-D level), and the classification accuracy rate reaches 92.3% (measured data set N = 1000);

[0058] Synergistic compensation efficiency

[0059] Acoustic-Structural Dual Compensation: When D-level fit is detected, the frequency response curve correction (tri-band compensation at 150 / 500 / 2000 Hz) and shape memory alloy drive (±15° curvature adjustment) are synchronously activated, reducing the low-frequency leakage by 18 dB;

[0060] Environment Adaptation: Combining a temperature sensor (±0.5℃ accuracy) and a microphone array (65 dB SPL threshold), dynamically adjusts the compensation strategy, improving the sound quality stability by 40% in high-temperature and high-noise environments;

[0061] Forward-looking Maintenance Ability

[0062] LSTM Prediction Module: By analyzing more than 50 historical wearing data, predicts the change trend of the fit within 15 minutes (MAE < 0.8%), triggers compensation 200 ms in advance, and avoids perceptible sound quality mutations;

[0063] Group Cooperative Compensation: The bilateral earphones achieve phase synchronization through 2.4GHz low-latency communication (<5ms), eliminating the sound field imbalance problem caused by unilateral leakage.

[0064] Additional aspects and advantages of the present invention will be given in part in the following description, become apparent in part from the following description, or be learned through the practice of the present invention. Brief Description of the Drawings

[0065] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on the structures shown in these drawings without creative efforts.

[0066] Figure 1 is the flowchart of the earphone wearing fit self-check and sound leakage compensation system provided in the embodiments of the present invention;

[0067] Figure 2 is the verification data table of the technical effects provided in the embodiments of the present invention;

[0068] Figure 3 is the experimental data table of multi-modal compensation provided in the embodiments of the present invention;

[0069] Figure 4 is the multi-scenario experimental data table of acoustic parameter compensation provided in the embodiments of the present invention;

[0070] Figure 5 is the data table of the contact state discrimination matrix provided in the embodiments of the present invention;

[0071] Figure 6It is an experimental data table for judging the calibration effect provided in the embodiments of the present invention;

[0072] Figure 7 It is a data table for wearing improvement of different user groups provided in the embodiments of the present invention;

[0073] Figure 8 It is an experimental data table for noise compensation performance provided in the embodiments of the present invention;

[0074] Figure 9 It is a data table for different temperature control effects provided in the embodiments of the present invention;

[0075] Figure 10 It is an experimental data table for key feature extraction provided in the embodiments of the present invention;

[0076] Figure 11 It is a data table for prediction accuracy test provided in the embodiments of the present invention;

[0077] Figure 12 It is an experimental data table for compensation effect comparison provided in the embodiments of the present invention;

[0078] Figure 13 It is a data table for energy consumption optimization provided in the embodiments of the present invention;

[0079] Figure 14 It is an experimental data table for transmission data of different working modes provided in the embodiments of the present invention;

[0080] Figure 15 It is an experimental data table for synchronous control provided in the embodiments of the present invention;

[0081] Figure 16 It is an experimental data table for sound field equalization provided in the embodiments of the present invention;

[0082] Figure 17 It is an experimental data table for endurance impact test provided in the embodiments of the present invention;

[0083] Figure 18 It is an experimental data table for dynamic performance test provided in the embodiments of the present invention;

[0084] Figure 19 It is an experimental data table for fatigue life test provided in the embodiments of the present invention

[0085] Figure 20 It is an experimental data table for acoustic performance improvement provided in the embodiments of the present invention.

[0086] The realization, functional features and advantages of the object of the present invention will be further described with reference to the embodiments and the accompanying drawings. Detailed implementation manners

[0087] Embodiments of the present invention will be described in detail below. Examples of the embodiments are shown in the accompanying drawings, where like or similar reference numerals designate like or similar elements or elements having like or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to explain the present invention, and should not be construed as limiting the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0088] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "circumferential", "radial", etc. are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention.

[0089] In addition, the terms "first" and "second" are only used for descriptive purposes and should not be construed as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present invention, "a plurality" means two or more unless otherwise specifically defined.

[0090] In the present invention, unless otherwise clearly specified and defined, the terms "mounted", "connected", "coupled", "fixed", etc. should be construed in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral connection; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0091] In the present invention, unless otherwise clearly specified and defined, the first feature being "on" or "under" the second feature may include direct contact between the first and second features, or may include contact between the first and second features not directly but through additional features therebetween. Moreover, the first feature being "above", "over" and "on top of" the second feature includes the first feature being directly above and obliquely above the second feature, or merely indicating that the horizontal height of the first feature is higher than that of the second feature. The first feature being "under", "below" and "beneath" the second feature includes the first feature being directly below and obliquely below the second feature, or merely indicating that the horizontal height of the first feature is less than that of the second feature.

[0092] The headphone wearing fit self-checking and sound leakage compensation system according to the embodiment of the present invention will be described in detail below with reference to the accompanying drawings.

[0093] Please refer to Figure 1 , in this embodiment, it includes: a sensor module, which includes a multi-modal distributed sensor array composed of a contact pressure sensor, a non-contact distance sensor, and an inertial measurement unit, and real-time collects physical signals of the wearing state, where: the contact pressure sensor uses a flexible piezoelectric film array, which is arranged in a grid layout along the auricle contact surface, with a thickness of 10 μm and a pressure resolution of 0.1 kPa; the non-contact distance sensor uses an infrared TOF sensor, which is arranged in the extension area of the sound generating unit, with a working frequency of 10 MHz and a detection resolution of 0.05 mm; the inertial measurement unit includes a three-axis gyroscope (range ±2000 dps) and an accelerometer (sampling rate 200 Hz); a central processing module, which performs feature fusion on heterogeneous sensor data based on a multi-stage Bayesian network to construct a dynamic wearing model, and the model determines the fitting level by analyzing three elements: the pressure distribution gradient curve, the residual space ratio of the concha, and the motion stability parameter; a sound leakage compensation execution module, which includes an acoustic parameter compensation unit and a structural deformation compensation unit, where: the acoustic parameter compensation unit executes an adaptive ANC algorithm update for multi-band dynamic EQ adjustment and leakage sound field analysis; the structural deformation compensation unit uses a shape memory alloy drive mechanism to adjust the curvature of the headphone structure; a user adaptive feedback module, which includes a bio-impedance sensor (working frequency 10 kHz, measurement accuracy ±5 Ω) and a mobile terminal interaction interface; an environment compensation module, which integrates a MEMS microphone array and a surface temperature sensor, where: the MEMS microphone array uses a 4-microphone beamforming technology to monitor environmental noise; the surface temperature sensor is an NTC thermistor with an accuracy of ±0.5 °C; the system realizes real-time signal processing and control logic through an STM32F7 series DSP chip.

[0094] Precise perception and decision-making:

[0095] Multi-source data fusion: By fusing heterogeneous data from a contact pressure sensor (0.1 kPa resolution), a TOF distance sensor (0.05 mm accuracy), and an IMU (sampling at 200 Hz), the misjudgment rate of the wearing state is reduced to less than 5% (compared with 30% of the traditional solution); Advantages of dynamic modeling: A three-dimensional feature matrix F is constructed using a Bayesian network + SVM classifier c , achieving a four-level fitting degree classification (A - D level), with a classification accuracy of 92.3% (measured dataset N = 1000);

[0096] Synergistic compensation efficiency

[0097] Acoustic-structure dual compensation: When a D-level fitting degree is detected, the frequency response curve correction (three-band compensation at 150 / 500 / 2000 Hz) and the shape memory alloy drive (±15° curvature adjustment) are synchronously activated, reducing the low-frequency leakage by 18 dB; Environmental adaptability: Combining a temperature sensor (±0.5 °C accuracy) and a microphone array (65 dB SPL threshold), the compensation strategy is dynamically adjusted, and the sound quality stability is improved by 40% in high-temperature and high-noise environments;

[0098] Forward-looking maintenance ability

[0099] LSTM prediction module: By analyzing more than 50 historical wearing data, the changing trend of the fitting degree within 15 minutes is predicted (MAE < 0.8%), and the compensation is triggered 200 ms in advance to avoid perceivable sound quality mutations; Group collaborative compensation: The bilateral earphones achieve phase synchronization through 2.4 GHz low-latency communication (< 5 ms), eliminating the sound field imbalance problem caused by unilateral leakage.

[0100] In this embodiment, the method for establishing the dynamic wearing model includes:

[0101] Construct a three-dimensional auricle contact feature matrix F c =(P aug ,σ p ,D gap ), where:

[0102] P aug is the average pressure value of the auricle contact surface, with the unit of kPa;

[0103] σp is the standard deviation of the pressure distribution;

[0104] D gap is the distance between the sound generating unit and the concha cavity, with the unit of mm;

[0105] Using a support vector machine classifier (with the kernel function as RBF and the penalty factor C = 2.5) to output the four-level fitting degree (A - D level), where:

[0106] Level C triggers acoustic parameter compensation;

[0107] D - level synchronous activation of acoustic parameter compensation and structural deformation compensation.

[0108] Implementation process of dynamic wearing model:

[0109] S1. Multi - modal data acquisition

[0110] Obtaining pressure distribution data: The flexible piezoelectric film array generates a 32×32 grid pressure distribution map with a sampling period of 20 ms (for example, the peak pressure on the auricle contact surface is 1.8 kPa, and the standard deviation σ_p = 0.3 kPa);

[0111] Measuring the spatial gap: The infrared TOF sensor detects the distance between the extension of the sound - generating unit and the concha cavity at a pulse frequency of 10 MHz, and the measured gap value d = 0.8 mm (when the standard volume of the concha cavity is 150 mm 3 , the residual space ratio R = 0.8 / 1.5 = 53.3%);

[0112] Extracting motion parameters: The IMU accelerometer collects 200 Hz three - axis acceleration data, and after sliding window filtering, calculates the motion stability index S = Σ(|a_i - aref|) / a_max, where a_ref = 9.8 m / s 2 (vertical static state reference value), a_max = 15 m / s 2 .

[0113] S2. Construction of three - dimensional feature matrix

[0114] Construct the auricle contact feature matrix M as follows:

[0115] (example measured value)

[0116] Matrix normalization processing: Map the μ_p range [0, 5 kPa] to [0, 1], the σ_p range [0, 2 kPa] to [0, 1], and the R range [0%, 100%] to [0, 1];

[0117] Feature vector generation: X = [0.36, 0.15, 0.533] (corresponding to the above example values).

[0118] S3. Support vector machine classification decision

[0119] Kernel function calculation: Adopt the RBF kernel K(X_i,X_j) = exp(-γ||X_i - X_i||2);

[0120] where γ = 1 (feature dimension) = 1 / 3;

[0121] Classification process:

[0122] Load the pre-trained model parameters (penalty factor C = 2.5, support vector library contains 200 sets of labeled data);

[0123] Calculate the decision function f(X)=sign(Σα_iy_iK(X,X_i)+b),

[0124] Where α_i is the Lagrange multiplier;

[0125] Output four-level classification results:

[0126] Grade A (f(X)>0.5): No compensation required;

[0127] B level (0 <f(X)≤0.5):记录数据;

[0128] C level (-0.5 <f(X)≤0):激活声学补偿;

[0129] Class D (f(X)≤-0.5): acoustic + structural double compensation.

[0130] S4. Compensation Execution Verification Experiment

[0131] C-level triggering scenario:

[0132] When μ_p is 1.2 kPa (lower than the threshold value of 1.5 kPa) and R=65%:

[0133]

[0134] D-Class Trigger Scenario:

[0135] When intense motion is detected (S=0.85>threshold 0.8) and R=72%:

[0136]

[0137]

[0138] Technical effect verification data: (such as Figure 2 shown).

[0139] Multi-level compensation collaborative workflow:

[0140] Initial wearing test:

[0141] The pressure array detects that the contact area coverage is <85% → triggers TOF precise scanning;

[0142] The measured R=58% (C-level threshold 60%) → the classifier outputs C-level;

[0143] Acoustic compensation activation:

[0144] The dynamic EQ boosts +4 dB in the 150 Hz frequency band (calculated according to the formula Δ = 1.5 ^ (0.58 / 0.6) = 1.47 times);

[0145] The ANC algorithm adjusts the step parameter from 0.02 to 0.028 (a 40% increase);

[0146] Continuous monitoring and upgrade:

[0147] The accelerometer detects continuous vibration (the average value of the S index for 10 seconds > 0.7);

[0148] Reclassified as D level → Activate the SMA drive, and the curvature increases by 3°;

[0149] After structural compensation, R drops to 52% → The system returns to the C-level compensation mode.

[0150] In this embodiment, the calculation of the motion stability parameter of the inertial measurement unit adopts the formula:

[0151]

[0152] Where a rea1 is the measured acceleration value, a ref is the static reference value, and a max is the preset maximum threshold;

[0153] When K m < 0.7, the compensation activation condition is triggered.

[0154] Implementation process of motion stability compensation:

[0155] S1. IMU data acquisition and preprocessing

[0156] Hardware configuration

[0157] Three-axis gyroscope: range ±2000 dps, zero-bias stability 0.05° / s / √Hz;

[0158] Accelerometer: range ±16 g, sampling rate 200 Hz, noise density 100 μg / √Hz;

[0159] Data synchronization:

[0160] / / STM32F7 code snippet DMA double-buffer mode

[0161] HAL_IMU_Start_DMA(&hIMU, imu_buffer, 3); / / Three-axis data synchronization capture osDelay(5); / / 5ms timing synchronization for other sensor data

[0162] S2. Calculation of motion stability parameters

[0163] Original data processing:

[0164] Acceleration data filtering: A 4th-order Butterworth low-pass filter (cutoff frequency 25 Hz) is adopted;

[0165] Gyroscope data calibration: Automatically calibrate the zero bias based on the stationary state in the first 3 seconds after startup;

[0166] Parameter calculation example:

[0167] Measured acceleration data: a_x = 1.2g, a_y = 0.8g, a_z = 10.3g (vertical direction);

[0168] Static reference value: a_ref = [0,01]g (normalized gravity vector);

[0169] Maximum threshold setting: a_max = 2g (empirical value);

[0170] Calculation of motion stability index:

[0171] S = (|1.2 - 0| + |0.8 - 0| + |10.3 - 1|) / 2

[0172] = (1.2 + 0.8 + 9.3) / 2

[0173] = 11.3 / 2 = 5.65 > 1 (trigger the trigger compensation bar

[0174] S3. Dynamic threshold adjustment strategy

[0175] Adaptive threshold algorithm:

[0176]

[0177] Environmental compensation coupling:

[0178] When the temperature sensor detects > 35°C, automatically increase a_max by 15% (to prevent false triggering caused by sensor drift at high temperatures).

[0179] S4. Compensation collaborative execution

[0180] Trigger logic:

[0181] graph LR

[0182] A[IMU data update] --> B{Is S > 1?}

[0183] B -- Yes --> C[Send interrupt signal to DSP]

[0184] C --> D[Read the current fitting degree level]

[0185] D --> E{Level = C / D?}

[0186] E--Yes-->F[Enhance ANC algorithm gain]

[0187] E--No-->G[Only record logs]

[0188] F-->H[Limit SMA drive amplitude]

[0189] Multi-modal compensation example:

[0190] Scenario 1: Subway commute (continuous vibration S = 0.9);

[0191] Activate high-frequency ANC enhancement: Increase the noise reduction depth by +6 dB in the 2 kHz frequency band;

[0192] Structural compensation: Fine-tune the SMA by 0.3° curvature to improve the sealing performance.

[0193] Scenario 2: Basketball exercise (impact peak S = 2.1)

[0194] Start the emergency mode: EQ low-frequency boost +8 dB to compensate for sound leakage;

[0195] Structural compensation: Increase the curvature by 2° and lock it for 10 seconds;

[0196] Bio-impedance monitoring: Real-time detection of whether headphone displacement occurs.

[0197] Experimental verification data: (as Figure 3 shown).

[0198] Technical details implementation:

[0199] Real-time guarantee:

[0200] Interrupt response time: <2 ms (from IMU trigger to compensation start)

[0201] Data pipeline processing:

[0202]

[0203] Safety protection mechanism:

[0204] Overload detection: When S > 2.5 continuously for 3 times, automatically reduce the ANC intensity to prevent howling;

[0205] Thermal protection: Trigger forced cooling (duty cycle reduced to 30%) when the temperature of the SMA driver > 60°C; User interaction feedback:

[0206]

[0207] Verification of typical application scenarios:

[0208] Gym strength training:

[0209] The impact peak during barbell squats is detected (S = 1.5);

[0210] Compensation system response:

[0211] The ANC convergence speed is increased to 180 ms (52% faster than the conventional mode);

[0212] The dynamic EQ enhances +5 dB in the 150 Hz frequency band to compensate for low-frequency leakage;

[0213] The SMA driver fine-tunes the curvature by 0.1° every 2 seconds to track muscle movement;

[0214] Ride for commuting:

[0215] Continuous vibration environment (average S value is 0.7);

[0216] The system automatically switches to the motion optimization mode:

[0217] The weight of the pressure sensor is reduced by 30% (to avoid misjudgment of wind noise);

[0218] The TOF scanning frequency is increased to 2 times per second (to strengthen gap monitoring);

[0219] By quantifying the mapping relationship between motion parameters and compensation amounts, the acoustic performance of the system is maintained stable in dynamic scenarios.

[0220] In this embodiment, the acoustic parameter compensation performs the following operations:

[0221] The dynamic EQ adjusts the center frequencies of three frequency bands: 150 Hz, 500 Hz, and 2 kHz;

[0222] The compensation amount of the ANC algorithm is calculated as ΔG = ΔG0×(1 + 0.3×L d ) where L d is the relative value of the detected leakage amount, and the convergence speed is increased by 40% compared to the baseline.

[0223] Implementation process of acoustic parameter compensation:

[0224] I. Hardware configuration of the compensation system

[0225]

[0226] II. Implementation process of dynamic EQ compensation

[0227] S1. Frequency band feature extraction

[0228] Sensor data mapping:

[0229] Calculation of the compensation weight for the low-frequency band (150 Hz) percentage

[0230] LF_weight = 0.6 * (1 - μ_p / 2.5) + 0.4 * (R / 100);

[0231] % Medium - frequency (500Hz) compensation factor

[0232] MF_factor = clamp(S * 2, 0.5, 1.8);

[0233] % High - frequency (2kHz) sensitivity adjustment

[0234] HF_sensitivity = 1.2 ^ (d / 0.5); / / d is the TOF detection gap

[0235] S2. Multi - band dynamic adjustment

[0236] 150Hz band compensation:

[0237]

[0238] Example: When the detected low - frequency leakage amount L = 0.8:

[0239] Δ = 1.5 ^ (0.8 / 0.6) = 1.5 ^ 1.33 ≈ 1.77

[0240] Final increase = - 3.0 * 1.77 = - 5.3dB

[0241] 500Hz band compensation logic:

[0242]

[0243] III. ANC algorithm enhancement implementation

[0244] S1. Leakage amount detection and parameter update

[0245] Leakage amount calculation model:

[0246] L = (α * (1 - μ_p / 2) + β * (d / 1.2)) / (α + β) % α = 0.7 (pressure weight), β = 0.3 (gap weight) Step - size dynamic adjustment:

[0247]

[0248] S2. Convergence acceleration mechanism

[0249] Improved LMS algorithm:

[0250]

[0251] IV. Multi - scenario verification data (as Figure 4 shown)

[0252] V. Dynamic Compensation Logic Tree

[0253] graph TD

[0254] A[Sensor fusion data] --> B{Calculation of leakage volume L}

[0255] B --> C[Generate dynamic EQ parameters]

[0256] C --> D1[150Hz: Seal repair]

[0257] C --> D2[500Hz: Motion compensation]

[0258] C --> D3[2kHz: Environmental compensation]

[0259] B --> E[Calculation of ANC step size Δ = 1.5 ^ (L / 0.6)]

[0260] E --> F[Update the filter]

[0261] F --> G{Convergence speed monitoring}

[0262] G --> H[Is the speed insufficient?] --> Yes --> I[Increase the step by 10%]

[0263] G --> No --> J[Maintain parameters]

[0264] VI. Anti-interference Design

[0265] Cross-band protection:

[0266] When the 150Hz gain > +6dB, automatically limit the 500Hz adjustment range within ±3dB;

[0267] High-frequency compensation is dynamically coupled with the ambient noise level:

[0268] if MEMS_noise > 70dB:

[0269] hf_gain = min(hf_gain, 4.0) # Prevent howling

[0270] Transient response optimization:

[0271]

[0272] Temperature compensation mechanism:

[0273]

[0274] VII. Analysis of typical compensation waveforms Before compensation: There is an obvious depression (-8dB) in the low-frequency band (100 - 300Hz), and a resonance peak at 1kHz; After compensation:

[0275] 150 Hz boost + 5.3 dB to fill the leakage;

[0276] The Q value at 500 Hz is extended from 1.2 to 0.8 to broaden the compensation bandwidth;

[0277] The ANC depth is optimized from -25 dB to -38 dB.

[0278] VIII. User Scenario Verification

[0279] Scenario 1: Office work in a coffee shop (mid-frequency noise):

[0280] TOF detection R = 55% (Class C)

[0281] Dynamic EQ activation:

[0282]

[0283] The ANC convergence is accelerated to 85 ms, and the ambient noise is attenuated by 12 dB.

[0284] Scenario 2: Cycling (wind noise + vibration):

[0285] IMU detection S = 1.1 (Class D);

[0286] System response:

[0287] % Structural compensation

[0288] SMA_curvature += 2.3? / / Enhance strong density

[0289] % Acoustic compensation

[0290] ANC_step = 0.028; / / 40% improvement

[0291] EQ_gain(150 Hz) = 1.5^(1.1 / 0.6)*base_gain = 2.05 × reference value

[0292] Through the quantization formula and adaptive algorithm, the sensor network is deeply integrated with the compensation strategy. In a typical leakage scenario, the system can restore the acoustic performance to more than 90% of the optimal state within 150 ms, with an efficiency improvement of 2.3 times compared to the traditional fixed compensation scheme

[0293] In this embodiment, the working logic of the user adaptive feedback module includes:

[0294] When the detected wearing pressure offset > 15% and lasts for 20 seconds, provide a three-dimensional graphical wearing guidance animation through the mobile terminal APP;

[0295] Bio-impedance sensor data is used to correct the misjudgment of the skin contact state of the pressure sensor.

[0296] I. Hardware System Architecture

[0297]

[0298]

[0299] II. Implementation of Pressure Offset Detection Algorithm S1. Establishment of Benchmark Pressure Mode

[0300] S2. Real-time Offset Calculation

[0301] S3. Continuous Judgment Logic

[0302] III. 3D Graphical Guidance System

[0303] Ear Canal Modeling and Animation Generation

[0304]

[0305]

[0306] IV. Bio-impedance Correction Mechanism

[0307] Contact State Discrimination Matrix (as Figure 5 shown)

[0308] Dynamic Compensation Case Scenario: Sweating during exercise causes the pressure sensor to falsely report poor contact

[0309] V. Performance Verification Data

[0310] Judging the Calibration Effect (as Figure 6 shown)

[0311] User Wearing Improvement (as Figure 7 shown)

[0312] VI. Technical Implementation Details

[0313] Low-latency Transmission Optimization

[0314]

[0315] Edge Computing Strategy

[0316]

[0317] Privacy and Security Design

[0318] Bio-impedance data encryption: AES-256-CTR mode;

[0319] Local data processing: Sensitive data does not leave the device;

[0320] User authorization mechanism: OAuth 2.0 device binding.

[0321] VII. Typical application scenarios

[0322] Scenario 1: Novice wears for the first time

[0323] Multiple pressure nodes detected < 0.5 kPa;

[0324] Bio-impedance shows a sudden increase in impedance in Area 3 (not in contact);

[0325] APP starts a 3D animation:

[0326] Highlight the cymba conchae area;

[0327] Generate a composite adjustment arrow of rotation + forward push;

[0328] After the user adjusts according to the guidance:

[0329] The number of pressure distribution compliance nodes increases from 5 / 64 to 58 / 64;

[0330] The sound leakage level is optimized from -28 dB to -43 dB.

[0331] Scenario 2: Micro-displacement during long-term work

[0332] A continuous 20-minute low-frequency shift (9 - 12%) is detected;

[0333] The system automatically starts a preventive prompt:

[0334] A mild reminder pops up: "Wearing looseness detected, it is recommended to make fine adjustments";

[0335] Show the pressure distribution comparison heat map;

[0336] After bio-impedance verification of effective contact:

[0337] Only activate the EQ compensation (without triggering structural adjustment).

[0338] Deeply integrate the sensor network with the intelligent feedback mechanism, and transform the passive adaptation of traditional headphones into active intelligent adjustment through bio-impedance-pressure data fusion and three-dimensional interactive guidance.

[0339] In this embodiment, the control strategy of the environmental compensation module includes: when the environmental noise > 65 dB SPL (A-weighted), the low-frequency compensation weight is increased by 30%; when the temperature sensor detects that the contact surface temperature > 35 °C, the structural deformation compensation amplitude is limited within 80% of the calibration value.

[0340] Environmental Compensation Implementation Process

[0341] I. Environmental Noise Compensation System Architecture

[0342]

[0343] II. Noise Triggered Compensation Implementation Process

[0344] S1. Sound Field Feature Extraction

[0345]

[0346] S2. Dynamic Weight Adjustment

[0347]

[0348]

[0349] S3. Compensation Cooperative Control

[0350] graph TD

[0351] A[Four-Microphone Array] --> B{A-weighted Noise > 65 dB?}

[0352] B -- Yes --> C[Calculate Low-Frequency Compensation Amount]

[0353] C --> D[Update EQ Coefficient]

[0354] D --> E[ANC Beam Direction Adjustment]

[0355] B -- No --> F[Maintain Reference Parameters]

[0356] E --> G{Is it in operation?}

[0357] G -- Yes --> H[Improve Convergence by 40%]

[0358] G -- No --> I[Standard Convergence Mode]

[0359] III. Temperature Compensation Execution Mechanism

[0360] Temperature Detection and Thermal Model

[0361]

[0362] Structural Compensation Limiting Algorithm

[0363]

[0364]

[0365] Thermal coupling control strategy

[0366]

[0367] IV. Verifying the performance of data noise compensation in multiple scenarios (as Figure 8 shown) and the temperature control effect (as Figure 9 shown) V. Safety protection design

[0368] Emergency cooling strategy

[0369]

[0370]

[0371] VI. Typical application scenarios

[0372] Scenario 1: Outdoor sports in summer

[0373] Environmental parameters: noise 72 dB(A), temperature 38 °C;

[0374] System response

[0375] % Noise compensation

[0376] low_freq_weight = 1.3+(72 - 65) / 10*0.5 = 1.65% 65% boost set_EQ_gain(150 Hz, +6.2 dB);

[0377] % Temperature protection

[0378] target_curvature = 12?*0.8 - (38 - 35)*0.2 = 9.6? - 0.6? = 9.0? Effect verification:

[0379] The low - frequency leakage is optimized from - 18 dB to - 32 dB;

[0380] The temperature of the SMA driver is stabilized at 41 °C (forced cooling not triggered).

[0381] Scenario 2: Indoor HVAC environment in winter

[0382] Environmental parameters: noise 62 dB(A), temperature 28 °C

[0383] System response

[0384] # Noise not exceeding the threshold, basic compensation

[0385] low_freq_weight = 1.0

[0386] set_EQ_gain(150Hz, +3.0dB)

[0387] # Temperature compensation term

[0388] thermal_expansion = (28 - 25) * 0.03 = 0.09?

[0389] drive_SMA(base_curvature + 0.09?

[0390] Effect verification:

[0391] The structural deformation accuracy is maintained within ±0.4°;

[0392] The energy consumption is reduced to 85% of the normal temperature mode.

[0393] VII. Core algorithm innovation

[0394] Environment - physiological parameter coupling model

[0395] Compensation weight = α * (noise level / 65) + β * (motion index) + γ * (temperature coefficient) where: α = 0.6, β = 0.3, γ = 0.1 (feature weight analysis based on PCA analysis)

[0396] Thermal inertia prediction algorithm

[0397]

[0398] Deeply integrate the environmental perception module with the intelligent compensation strategy, and achieve the optimal balance between acoustic performance and device reliability through noise - temperature two - factor coupling control.

[0399] In this embodiment, it also includes a predictive fitting maintenance function, which is implemented in the following way: establish a user - specific wearing mode database in the embedded memory, store ≥50 historical wearing records; use an LSTM neural network (deployed in the TensorFlow Lite framework) to predict the change trend of the fitting degree in the next 15 minutes; when the predicted trend slope k < -0.05 / min, activate the compensation subsystem 200ms in advance.

[0400] Process of predictive fitting maintenance

[0401] I. Hardware architecture of the prediction system

[0402]

[0403] II. Historical data management mechanism

[0404] Data storage structure

[0405]

[0406] Data preprocessing process

[0407] graph TD

[0408] A[Original sensor] --> B[Outlier filtering]

[0409] B --> C[Feature engineering]

[0410] C --> D[Time series alignment]

[0411] D --> E[Normalization processing]

[0412] E --> F[Sliding window division]

[0413] F --> G[LSTM training set]

[0414] Key feature extraction (as Figure 10 shown)

[0415] III. Implementation of LSTM prediction model

[0416] Network structure parameters

[0417]

[0418] Embedded deployment optimization

[0419]

[0420] Trend prediction algorithm

[0421] IV. Predictive compensation control logic early trigger mechanism

[0422]

[0423] Multi-mode compensation strategy

[0424] Prediction accuracy test (as Figure 11 shown)

[0425] Compensation effect comparison (as Figure 12 shown)

[0426] VI. Typical application scenarios

[0427] Scenario 1: Vigorous aerobics exercise

[0428] Prediction stage

[0429] % LSTM input features (after normalization)

[0430] input = [0.67, 0.82, 0.91, 0.55, 0.72, 0.63, 0.88, 0.79, 0.68];

[0431] % Predicted output

[0432] predicted_slope = 0.23 # Exceeded threshold 0.15

[0433] System response

[0434] / / Activate compensation 200ms in advance

[0435] osTimerStart(compensationTimer, 200);

[0436] / / Pre - adjustment parameters

[0437] pre_set_anc_gain(1.2);

[0438] pre_adjust_SMA_curvature(+0.7?);

[0439] Scenario 2: Micro - displacement in the office scenario

[0440] Predictive analysis

[0441] # Historical data pattern recognition

[0442] similar_patterns = find_similar_records(current_state)

[0443] # Predicted slope

[0444] k = 0.09 # Below the trigger threshold

[0445] System behavior:

[0446] Maintain the reference compensation parameters;

[0447] Record the current state in the historical database;

[0448] Update the prediction model every 5 minutes.

[0449] VII. Key technologies

[0450] Lightweight time - series modeling

[0451] Input dimension compression: Original 64 - dimensional pressure data → 3 - dimensional PCA. Processing efficiency improvement: The number of model parameters reduced from the original 256K → 98K

[0452] The inference speed is excellent, with a speed of 8 ms per time (meeting the real-time requirement).

[0453] Incremental learning mechanism

[0454]

[0455] Energy consumption optimization design (such as Figure 13 shown)

[0456] Deeply integrate multi-modal sensing data with the prediction engine, and achieve intelligent compensation 200 ms ahead through LSTM time series modeling.

[0457] In this embodiment, multi-device networking and collaborative functions are supported, including: realizing bilateral earphone data synchronization through a 2.4 GHz band wireless communication module (transmission delay < 5 ms); when the bilateral fitting degree difference > 2 levels is detected, start the group equalization compensation mode and preferentially adjust the low-frequency band compensation phase consistency.

[0458] Process of multi-device networking and collaboration

[0459] I. Wireless networking system architecture

[0460]

[0461] II. Bilateral data synchronization mechanism

[0462] Timestamp alignment protocol

[0463] Sensor data fusion

[0464]

[0465] Guarantee of transmission reliability (such as Figure 14 shown) III. Group equalization compensation strategy

[0466] Difference detection algorithm

[0467]

[0468] Phase consistency control

[0469] graph TD

[0470] A[Detect bilateral difference > 2 levels] --> B[Read current acoustic parameters]

[0471] B --> C{Is the low-frequency band phase > 90?}

[0472] C -- Yes --> D[Adjust slave device DSP delay]

[0473] C -- No -- > E [Main device reset reference phase]

[0474] D -- > F [Verify frequency response consistency]

[0475] E -- > F

[0476] F -- > G {Compliance?}

[0477] G -- No -- > H [Enable hardware clock alignment]

[0478] Dynamic delay compensation

[0479]

[0480] Cooperative compensation implementation example

[0481] Scenario 1: Unilateral looseness (Grade D)

[0482] Left ear fit: Grade C (trigger acoustic compensation);

[0483] Right ear fit: Grade A (normal state);

[0484] System response:

[0485]

[0486] Scenario 2: Bilateral instability during intense exercise

[0487] Initial state: Left ear Grade B, right ear Grade C (difference of 1 grade);

[0488] State after exercise: Left ear Grade D, right ear Grade B (difference of 2 grades);

[0489] Dynamic compensation process:

[0490] Wireless transmission layer starts fast retransmission mode (interval 2ms);

[0491] Main processor executes joint Kalman filter:

[0492] fused_state = 0.6 * left_state + 0.4 * right_state Structural compensation symmetric adjustment:

[0493] / / Bilateral SMA drive synchronization

[0494] drive_SMA(left, target_curve - 1.5?;

[0495] drive_SMA(right, target_curve + 0.8?;

[0496] V. Performance Verification: Data Synchronization Control Performance (as Figure 15 shown)

[0497] Sound Field Equalization Effect (as Figure 16 shown)

[0498] Battery Life Impact Test (as Figure 17 shown)

[0499] VI. Fault Tolerance and Recovery Mechanisms

[0500] Disconnection Emergency Strategy

[0501] graph LR

[0502] A[Wireless Disconnection Detection] --> B{When Disconnected}

[0503] B -- <100ms --> C[Cached Data Compensation]

[0504] B -- 100 - 500ms --> D[Inertial Prediction Compensation]

[0505] B --> 500ms --> E[Switch to Standalone Mode]

[0506] Data Integrity Guarantee

[0507]

[0508] Dynamic Power Consumption Regulation

[0509]

[0510] Integrate the sensing network and collaborative architecture, and through innovative design of the delay compensation algorithm and dynamic topology management, achieve sub-millisecond synchronization control for both earbuds.

[0511] In this embodiment, its industrial design includes: a hidden layout structure for sensors, with a housing aperture ratio < 3%; a composite heat conduction structure (thermal conductivity ≥ 5 W / mK tested according to ASTM D5470 standard); a replaceable earplug contact surface component, which can be manually replaced within 30 seconds using a quick-release buckle structure.

[0512] Industrial Design Implementation Process:

[0513] Pressure Sensor: An 8×8 piezoelectric film array can be embedded in the inner layer of the earplug silicone sleeve, with a 0.3mm corrugated skin-friendly silicone on the surface;

[0514] TOF Sensor: An invisible window (aperture 0.1mm, density 200 holes / cm 2 ) is formed under the sound-emitting mesh cloth through laser drilling technology;

[0515] Bioimpedance electrode: The annular nano silver wire is printed on the edge of the earplug contact surface (line width 15μm).

[0516] In this embodiment, the deformation response time of the shape memory alloy driving mechanism is ≤150ms, the curvature adjustment range is ±15°, and the driving accuracy is 0.5°.

[0517] Implementation process of the shape memory alloy driving mechanism:

[0518] I. SMA drive system architecture design

[0519] A three-level drive architecture is adopted to achieve micron-level deformation control, and the specific configuration is as follows:

[0520] Mechanical execution layer

[0521] A net-shaped actuator woven from Ni-Ti-Cu shape memory alloy wires, with a single wire diameter of 80μm and a weaving density of 120 meshes / cm 2 , pre-deformation treatment to form a two-way memory effect. The drive unit is arranged in a bionic muscle bundle shape, and 6 groups of independent control units are arranged along the curvature radius direction of the earphone housing. Each group includes:

[0522] Core drive wire: length 15mm, phase change temperature 35℃±2℃;

[0523] Auxiliary reset spring: made of 316L stainless steel, elastic coefficient 0.8N / mm;

[0524] Thermal interface material: graphene-silicone grease composite material, thermal conductivity 8.5W / mK.

[0525] Drive control layer

[0526] Dual closed-loop control system based on STM32F7:

[0527] Inner loop temperature control: PWM pulse width modulation is adopted, frequency 20kHz, resolution 0.1℃;

[0528] Outer loop position control: Integrated 0.5μm precision laser displacement sensor, sampling rate 1kHz;

[0529] Safety protection module: Real-time monitoring of three parameters of current / temperature / strain, and start the fusing mechanism in case of abnormality.

[0530] Intelligent decision-making layer:

[0531] Fuse real-time sensing data to establish a deformation demand model:

[0532]

[0533]

[0534] II. Deformation Compensation Implementation Process

[0535] S1. Triggering of Multi-Source Data Fusion

[0536] When the central processing module detects that the sound leakage level ≥ Class C, start the SMA drive decision-making process:

[0537] The pressure sensor array generates an 8×8 contact force distribution cloud map and calculates the auricle contact pressure gradient ΔP;

[0538] The TOF sensor measures the distance d of the cymba conchae gap and converts the residual space ratio R = (d / 1.5mm) × 100%;

[0539] The IMU collects the variance σ of the three-axis acceleration 2 , and generates a motion stability index S = log(1 + σ 2 ).

[0540] S2. Dynamic Generation of Drive Parameters

[0541]

[0542] S3. Phase Change Driving Process

[0543] Heating stage (0 - 120ms):

[0544] Apply a 12V PWM voltage, the temperature of the alloy wire rises from 25°C to the phase change point of 38°C, the resistivity suddenly changes from 80 μΩ·cm to 95 μΩ·cm, and a 4.2% contraction strain is generated;

[0545] Holding stage (120 - 140ms):

[0546] Switch to the constant current mode (350mA) to maintain the phase change state;

[0547] Cooling stage (140 - 150ms):

[0548] Cut off the power supply, and cool down to 32°C within 3ms through the heat conduction structure, and the shape is fixed.

[0549] S4. Closed-Loop Precision Calibration

[0550] The laser displacement sensor provides real-time feedback on the deformation amount and corrects it using the incremental PID algorithm:

[0551]

[0552] III. Technical Verification Data

[0553] Dynamic performance test (as Figure 18 shown)

[0554] Fatigue life test (asFigure 19 as shown

[0555] Improved acoustic performance (such as Figure 20 as shown

[0556] IV. Application Scenarios

[0557] Scenario 1: Dynamic fitting in sports scenarios

[0558] Initial state:

[0559] When running, the earplug has a displacement of 0.8 mm due to muscle movement, the pressure gradient ΔP = 0.4 kPa / mm, and the residual space ratio R = 68%;

[0560] System response:

[0561] The central processing module calculates the target curvature of +9.5°;

[0562] Drive units B, C, and E groups are activated, applying a 12V@120ms pulse;

[0563] The curvature of the earplug contact surface increases, and the pressure distribution uniformity is improved by 37%;

[0564] Effect verification:

[0565] The sound leakage level improves from -22 dB to -36 dB, and the EQ compensation requirement is reduced by 60%.

[0566] Scenario 2: Thermal compensation for long-term wearing

[0567] Temperature change:

[0568] After continuous use for 2 hours, the contact surface temperature rises to 37°C, and the SMA phase change point drifts by +1.5°C;

[0569] Adaptive adjustment:

[0570] The temperature sensor triggers the compensation algorithm;

[0571] The PWM duty cycle automatically increases by 8% to maintain the driving force;

[0572] Laser displacement feedback corrects the deformation error in real time;

[0573] Performance maintenance:

[0574] The curvature control accuracy still reaches ±0.6°, which is 3 times better than traditional mechanical structures.

[0575] V. Innovative Technologies

[0576] Composite drive topology

[0577] Spatial layout optimization: Adopting a hexagonal honeycomb arrangement, at 18 mm 3The space accommodates 6 groups of drive units;

[0578] Strain coupling design: The strain directions of adjacent drive wires are staggered at 30°, eliminating lateral stress concentration;

[0579] Thermal field equilibrium technology: Design of a gradient heat conduction path to make the standard deviation of the temperature field <0.8°C.

[0580] Intelligent material modification process

[0581] Element doping: Adding 0.3 at% Nb increases the fatigue life to >10 5 cycles;

[0582] Surface treatment: Micro-arc oxidation generates a 5-μm Al2O3 ceramic layer, increasing the corrosion resistance by 5 times;

[0583] Pre-deformation treatment: Stress aging at 350°C for 3 hours reduces the residual strain to 0.02%.

[0584] Cross-modal control algorithm

[0585]

[0586]

[0587] Through the collaborative innovation in the three aspects of materials, structure, and control, the technical indicators are transformed into mass-production engineering solutions.

[0588] In the description of this specification, the description referring to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0589] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention. Any equivalent structural transformation made under the inventive concept of the present invention, or direct / indirect application in other related technical fields, is included in the patent protection scope of the present invention.

Claims

1. Earphone fitting self-check and sound leakage compensation system, characterized by: include: The sensor module includes a multi-modal distributed sensor array consisting of a contact pressure sensor, a non-contact distance sensor, and an inertial measurement unit to collect physical signals of the wearing status in real time, including: The contact pressure sensor adopts a flexible piezoelectric film array and is arranged in a grid along the auricle contact surface; The non-contact distance sensor adopts an infrared TOF sensor, which is arranged in the extension area of ​​the sound unit; The inertial measurement unit includes a three-axis gyroscope and an accelerometer; The central processing module performs feature fusion on heterogeneous sensor data based on a multi-stage Bayesian network and constructs a dynamic wearing model. The model determines the fit level by analyzing the pressure distribution gradient curve, the residual space ratio of the concha cavity and the motion stability parameter. The acoustic leakage compensation execution module includes an acoustic parameter compensation unit and a structural deformation compensation unit, wherein: The acoustic parameter compensation unit performs multi-band dynamic EQ adjustment and adaptive ANC algorithm update for leakage sound field analysis; The structural deformation compensation unit uses a shape memory alloy drive mechanism to adjust the curvature of the earphone structure; User adaptive feedback module, including bioimpedance sensor and mobile terminal interaction interface; Environmental compensation module, integrating MEMS microphone array and surface temperature sensor, including: The MEMS microphone array uses 4-microphone beamforming technology to monitor ambient noise; The surface temperature sensor is an NTC thermistor; The system realizes real-time signal processing and control logic through STM32F7 series DSP chips.

2. The earphone wearing fit self-checking and sound leakage compensation system according to claim 1, characterized in that: The method for establishing the dynamic wearing model comprises: Construct the three-dimensional auricle contact feature matrix F c =(P aug ,σ p ,D gap ),in: P aug is the average pressure value of the auricle contact surface, unit is kPa; σ p is the standard deviation of pressure distribution; D gap is the distance between the sound unit and the concha cavity, in mm; The support vector machine classifier is used to output four levels of fit, where: C-level trigger acoustic parameter compensation; Class D simultaneously activates acoustic parameter compensation and structural deformation compensation.

3. The earphone wearing fit self-checking and sound leakage compensation system according to claim 1, characterized in that: The motion stability parameters of the inertial measurement unit are calculated using the formula: where a rea1 is the measured acceleration value, a ref is the static reference value, a max is the preset maximum threshold; When K m When <0.7, the compensation activation condition is triggered.

4. The earphone wearing fit self-checking and sound leakage compensation system according to claim 1, characterized in that: The acoustic parameter compensation performs the following operations: Dynamic EQ adjusts the center frequency to three frequency bands: 150Hz, 500Hz, and 2kHz; The compensation amount of the ANC algorithm is calculated as ΔG = ΔG0 × (1 + 0.3 × L d ) calculation, where L d To detect the relative value of leakage, the convergence speed is improved by 40% compared with the baseline.

5. The earphone wearing fit self-checking and sound leakage compensation system according to claim 1, characterized in that: The working logic of the user adaptive feedback module includes: When the wearing pressure deviation is detected to be greater than 15% and lasts for 20 seconds, a three-dimensional graphical wearing guidance animation is provided through the mobile terminal APP; The bioimpedance sensor data is used to correct the skin contact status misjudgment of the pressure sensor.

6. The earphone wearing fit self-checking and sound leakage compensation system according to claim 1, characterized in that: The control strategy of the environmental compensation module includes: When the ambient noise is >65dB SPL, increase the low-frequency compensation weight by 30%; When the temperature sensor detects that the contact surface temperature is greater than 35°C, the structural deformation compensation amplitude is limited to within 80% of the calibration value.

7. The earphone wearing fit self-checking and sound leakage compensation system according to claim 1, characterized in that: It also includes predictive fit maintenance capabilities, which are achieved through: Establish a user-specific wearing mode database in the embedded memory, storing ≥50 historical wearing records; Use LSTM neural network to predict the fit change trend in the next 15 minutes; When the predicted trend slope k is less than -0.05 / min, the compensation subsystem is activated 200ms in advance.

8. The earphone wearing fit self-checking and sound leakage compensation system according to claim 1, characterized in that: Supports multi-device networking and collaboration functions, including: Data synchronization of both earphones is achieved through 2.4GHz band wireless communication module; When the bilateral fit difference is detected to be >2 levels, the group equalization compensation mode is activated, and the low-frequency band compensation phase consistency is adjusted first.

9. The earphone wearing fit self-checking and sound leakage compensation system according to claim 1, characterized in that: Its industrial design includes: The sensor is hidden in the layout structure, and the shell opening rate is less than 3%; Composite heat-conducting structure; The replaceable earplug contact surface assembly uses a quick-release buckle structure to enable manual replacement within 30 seconds.

10. The earphone wearing fit self-checking and sound leakage compensation system according to any one of claims 1 to 9, characterized in that: The deformation response time of the shape memory alloy driving mechanism is ≤150ms, the curvature adjustment range is ±15°, and the driving accuracy is 0.5°.

Images