A narrow-frame phone watch wearing detection system and method

Abstract

The application relates to the computer technical field and discloses a narrow-frame phone watch wearing detection system and method, aiming to solve the problems of high power consumption, insufficient accuracy and easy interference misjudgment caused by relying on a single sensor in the prior art. The method performs initial contact monitoring through a capacitive sensing unit, triggers a thermal array and an inertial measurement unit in stages after triggering, and verifies body temperature distribution characteristics and dynamic behavior patterns in sequence; when the judgment is ambiguous, active sound vibration detection based on a linear harmonic resonator actuator is started, vibration attenuation characteristics are analyzed, and final decision is completed. The system comprises a data acquisition module, a state judgment engine and an active detection unit, and adopts a layered activation strategy to realize ultra-low power consumption operation. The application combines multi-modal sensing and physical models, significantly improves detection accuracy and anti-interference ability, and prolongs the device endurance time.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of computers, and particularly relates to a narrow-frame phone watch wearing detection system and method. BACKGROUND

[0002] With the popularity of smart wearable devices, phone watches, as typical terminals integrating communication, positioning, health monitoring and child safety protection, are widely used in low-age user groups. The effectiveness of the core functions of the phone watch is highly dependent on whether the device is correctly worn: for example, heart rate monitoring, fall detection, electronic fence alarm and remote call scenarios all require the watch to be close to the skin. However, the behavior of child users is highly uncertain, and situations such as frequent removal, loose wearing or misplacement in clothing pockets are common, leading to distorted sensor data, positioning drift and even safety function failure. In addition, the frame of the phone watch is getting narrower and narrower, which makes it more likely to cause inaccurate identification. Existing phone watches generally lack accurate discrimination ability for the real wearing state, and only rely on a single accelerometer or photoplethysmography (PPG) signal for simple threshold judgment, which is easily affected by motion artifacts, environmental light interference or poor skin contact, resulting in a large number of false positives or missed detections.

[0003] Among them, the phone watch wearing detection technology aims to accurately distinguish between "effective wearing", "ineffective wearing" and "not wearing" through multi-source sensor information fusion. The basic principle of this technology is to comprehensively utilize inertial measurement unit (IMU), optical sensor, temperature sensor and contact impedance modal data to construct a dynamic perception model of the relative position relationship between the watch and the human body. In an ideal state, the system should maintain high stability in complex daily activities (such as running, washing hands, sleeping), while taking into account the low power consumption and real-time response requirements.

[0004] The prior art has the following defects in achieving the above-mentioned objectives: first, a single sensor scheme cannot overcome the inherent defects in specific scenarios, for example, the signal-to-noise ratio of the PPG signal decreases sharply under intense exercise, and the accelerometer cannot distinguish between the watch being placed on the table and being stably worn on the wrist; second, multi-sensor fusion strategies mostly use fixed weights or rule engines, lacking the ability to adapt to individual differences (such as skin color, wrist thickness, wearing tightness) and environmental changes (such as temperature, humidity, light); third, existing algorithms usually run on application processors, with high computational overhead, making it difficult to meet the limited battery life requirements of phone watches; finally, in the context of child use, the system also needs to cope with non-standard wearing behaviors (such as reverse wearing, wearing outside the sleeve), and current methods have little recognition ability for such edge cases. The above problems collectively result in low wearing detection accuracy, high power consumption and poor user experience, and there is an urgent need for a high-precision, low-power and highly stable phone watch wearing detection system and method. SUMMARY

[0005] The present application aims to provide a narrow frame phone watch wearing detection system and method, aiming at solving the technical problems of the prior art that the wearing detection method relies on a single sensor or simple logic combination, resulting in high power consumption, insufficient accuracy, and being easily disturbed by the environment to produce misjudgment.

[0006] In order to achieve the above-mentioned purpose, the present application provides a narrow frame phone watch wearing detection method, which constructs a low-power and high-precision wearing state composite judgment model through hierarchical collection and state progressive analysis of multi-modal sensor data, combined with active acoustic vibration feedback detection. The method first uses an ultra-low-power capacitive sensing unit to continuously monitor the initial contact, and after triggering the contact event, activates the thermal array sensor and the inertial measurement unit in stages and on demand to sequentially verify the body temperature distribution characteristics and micro-motion behavior pattern of the contact object. In the case of ambiguous judgment, an active detection mechanism based on a micro-vibrator is further started to realize the final accurate discrimination of the wearing state by analyzing the attenuation characteristics of the vibration signal in the contact medium.

[0007] According to one aspect of the present application, a narrow frame phone watch wearing detection method is provided, which specifically comprises:

[0008] The capacitive sensing unit continuously monitors the capacitive value change of the phone watch shell back cover at a preset low-frequency sampling period, and generates an initial contact event signal when the capacitive value change exceeds the preset capacitive trigger threshold;

[0009] In response to the initial contact event signal, a thermal sensor array arranged on the phone watch shell back cover is activated, and the thermal sensor array collects real-time temperature data of multiple temperature measurement points covering the contact area to form a two-dimensional temperature distribution map;

[0010] The two-dimensional temperature distribution map is analyzed to calculate the temperature mean value, temperature gradient, and spatial entropy of the temperature distribution, and an effective body temperature contact signal is generated when the temperature mean value falls within the preset human body surface temperature interval, and the temperature gradient and spatial entropy conform to the characteristics of the human wrist thermodynamic model;

[0011] After receiving the effective body temperature contact signal, an inertial measurement unit is activated, which collects three-axis acceleration data and three-axis angular velocity data within a preset time window to form a time sequence dynamic data sequence;

[0012] Feature extraction is performed on the time sequence dynamic data sequence to calculate multiple dynamic feature parameters including motion intensity spectrum density, posture stability factor, and periodic motion frequency, and the dynamic feature parameters are matched with the human wearing activity feature database pre-stored in the memory to generate a dynamic feature matching degree score;

[0013] According to the dynamic characteristic matching degree score, a preliminary determination of the wearing state is performed; when the dynamic characteristic matching degree score is higher than a first preset determination threshold, it is determined that the phone watch is in a wearing state; when the dynamic characteristic matching degree score is lower than a second preset determination threshold, it is determined that the phone watch is in a non-wearing state; wherein the first preset determination threshold is greater than the second preset determination threshold;

[0014] When the dynamic characteristic matching degree score is between the first preset determination threshold and the second preset determination threshold, an active detection process is started, which specifically includes: driving a linear resonant actuator to generate a preset composite frequency vibration sequence through a control signal; while the linear resonant actuator is vibrating, the accelerometer in the inertial measurement unit is used to collect shell vibration response data generated by the vibration; the shell vibration response data is subjected to fast Fourier transform to extract its resonance peak frequency, frequency bandwidth and energy decay time constant, and a vibration decay characteristic vector is constructed; the vibration decay characteristic vector is compared with a pre-set human tissue vibration decay model to calculate an acoustic vibration coupling confidence;

[0015] Based on the acoustic vibration coupling confidence, a final decision on the wearing state is made; when the acoustic vibration coupling confidence is higher than a preset decision threshold, it is determined that the phone watch is in a wearing state, otherwise it is determined to be in a non-wearing state.

[0016] As an embodiment of the present application, the thermosensitive sensor array is composed of four negative temperature coefficient thermistors arranged in a two-by-two matrix, the negative temperature coefficient thermistors are packaged on the inner surface of the back cover of the phone watch shell, and a heat conduction path is formed between the outer surface of the back cover of the shell and the negative temperature coefficient thermistors through a medium with high thermal conductivity. The human wrist thermodynamic model features are: the temperature in the central region of the two-dimensional temperature distribution is higher than that in the edge region, and the temperature gradient decreases outward along the radial direction, and the spatial entropy of the temperature distribution is less than a preset uniform heat source entropy threshold.

[0017] Further, feature extraction is performed on the time series dynamic data sequence, specifically including: vector synthesis of three-axis acceleration data within a time window to obtain a total acceleration time series signal; short-time Fourier transform is performed on the total acceleration time series signal to obtain its energy distribution in the frequency range of zero to fifty hertz, and the energy distribution is taken as the motion intensity spectral density; the variance of the three-axis angular velocity data within the time window is calculated, and the reciprocal of the variance is taken as the attitude stability factor; autocorrelation analysis is performed on the total acceleration time series signal to extract the main period with the strongest energy in the signal, and the main period is taken as the periodic motion frequency.

[0018] As an embodiment of the present application, the linear resonant actuator is a piezoelectric ceramic vibrator, and the preset composite frequency vibration sequence comprises a base frequency sinusoidal signal with a frequency of 150 Hz and a second harmonic signal with a frequency of 300 Hz, and the duration is 200 ms. The human tissue vibration attenuation model is a database which stores the statistical distribution range of the vibration attenuation feature vector excited by the composite frequency vibration sequence when the wristwatch is worn on the wrist of a user with different ages and different body shapes. The calculation of the acoustic vibration coupling confidence is to calculate the Mahalanobis distance between the real-time extracted vibration attenuation feature vector and the statistical distribution range stored in the database, and the reciprocal of the Mahalanobis distance is normalized and used as the acoustic vibration coupling confidence.

[0019] Further, the method further comprises a wearing state transition confirmation logic. When it is determined that the wristwatch is in a wearing state transition to a non-wearing state, the method requires that the non-wearing state is determined in three consecutive detection periods, and then the state transition is finally confirmed, so as to avoid misjudgment caused by short-term intense exercise or sensor transient interference. When it is determined that the wristwatch is in a non-wearing state transition to a wearing state, the method updates the state immediately after completing a complete determination process and confirming the wearing state, so as to ensure real-time response to the wearing event.

[0020] According to another aspect of the present application, a narrow-frame wristwatch wearing detection system is provided. The system is integrated in a wristwatch, and specifically comprises:

[0021] A data acquisition module, the data acquisition module comprises a capacitive sensing unit, a thermal sensor array and an inertial measurement unit, the capacitive sensing unit, the thermal sensor array and the inertial measurement unit are electrically connected to a central processing unit; the capacitive sensing unit is used to monitor the change of the capacitance value of the back cover of the shell; the thermal sensor array is used to acquire a two-dimensional temperature distribution map of the contact area; and the inertial measurement unit is used to acquire a time sequence dynamic data sequence and a shell vibration response data;

[0022] A state determination engine, the state determination engine is solidified in the firmware of the central processing unit, and is configured to receive and process data from the data acquisition module; the state determination engine specifically comprises an initial contact analysis unit, a thermodynamic feature verification unit, a dynamic behavior matching unit and a final decision unit;

[0023] The initial contact analysis unit is used to analyze the data of the capacitive sensing unit, and when the change of the capacitance value exceeds the capacitive trigger threshold, an initial contact event signal is generated and sent to the thermodynamic feature verification unit;

[0024] A thermodynamic characteristic verification unit, which activates and processes the two-dimensional temperature distribution map collected by the thermosensitive sensor array after receiving the initial contact event signal, generates and sends an effective body temperature contact signal to the dynamic behavior matching unit after confirming that the two-dimensional temperature distribution map conforms to the wrist thermodynamic model characteristic of the human body;

[0025] A dynamic behavior matching unit, which activates and processes the time-series dynamic data sequence collected by the inertial measurement unit after receiving the effective body temperature contact signal, calculates a dynamic characteristic matching degree score, and outputs a wearing, non-wearing or pending state signal according to the relationship between the score and the first and second preset determination thresholds;

[0026] An active detection unit, which includes a linear resonant actuator and a driving circuit connected with the central processing unit; when the dynamic behavior matching unit outputs the pending state signal, the central processing unit controls the active detection unit to execute the active detection process, and the active detection unit sends the collected shell vibration response data to the final decision unit;

[0027] A final decision unit, which is used to receive the shell vibration response data, calculate the acoustic vibration coupling confidence, and output the final wearing state or non-wearing state determination result based on the comparison result between the confidence and the decision threshold.

[0028] As an embodiment of the present application, the system further comprises a non-volatile storage module connected with the central processing unit, which internally stores the capacitance trigger threshold, the human body surface temperature interval, the human wrist thermodynamic model characteristic parameter, the human wearing activity characteristic database, the first preset determination threshold, the second preset determination threshold, the human tissue vibration attenuation model and the decision threshold.

[0029] Further, the state determination engine adopts a layered activation power consumption management strategy. When no initial contact event signal is received, the entire system is in a sleep mode, and only the capacitance sensing unit works with a microampere-level current. When entering the thermodynamic characteristic verification stage, the thermosensitive sensor array is activated, and the system power consumption is increased to a milliamper level. When entering the dynamic behavior matching stage, the inertial measurement unit is activated, and the system power consumption is further increased. Only when the determination is ambiguous, the active detection unit with the highest power consumption is transiently activated, so as to maximize the average working power consumption of the system while ensuring the detection accuracy.

[0030] In summary, the present application includes at least one of the following beneficial technical effects:

[0031] (1): By adopting capacitive, thermal array, inertial measurement unit and active acoustic vibration detection four different physical principle completely different sensing modal, and carrying out hierarchical progressive fusion determination, the accuracy and anti-interference ability of wearing detection are greatly improved, and the false judgment problem caused by single sensor due to environmental changes, object camouflage and other factors is effectively overcome.

[0032] (2): By constructing a state progressive power consumption management model from ultra-low power sleep monitoring to high-precision active detection, only when necessary, the high-power sensor and processing unit are awakened step by step, so that the system runs at microamp standby power consumption in most of the time, and the battery life of the phone watch is significantly prolonged.

[0033] (3): The active acoustic vibration coupling detection mechanism based on linear resonant actuator and accelerometer is introduced, which distinguishes human tissues from non-biological objects by analyzing the attenuation characteristics of vibration in the contact medium. This method provides a difficult-to-fake, high-confidence physical criterion for wearing detection, and fundamentally solves the technical bottleneck that the traditional passive sensing method cannot accurately determine in static or ambiguous scenes.

[0034] (4): The determination logic of the present application is based on the deep analysis of physical model and data characteristics, rather than simple threshold judgment. The built-in human thermodynamic model, activity feature database and vibration attenuation model have better universality and stability, and can adapt to different users and variable use scenarios. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 is the overall technical scheme architecture schematic diagram of the present application. DETAILED DESCRIPTION

[0036] The present application provides a narrow frame phone watch wearing detection system and method, which aims to solve the problems of high power consumption, insufficient accuracy and false judgment caused by environmental interference in the prior art, which relies on single sensor or simple logic combination for wearing state judgment. Through hierarchical acquisition and state progressive analysis of multi-modal sensing data, combined with active acoustic vibration feedback detection mechanism, a low-power, high-precision wearing state composite determination model is constructed. The following will be described in detail around the specific implementation steps of the method, and the system structure supporting the operation of the wearing detection system method of the present application will be disclosed as necessary. Figure 1 , the specific implementation steps of the method will be described in detail, and the system structure supporting the operation of the wearing detection system method of the present application will be disclosed as necessary.

[0037] In order to more clearly fully disclose the technical scheme of the present application, the following is described in the form of continuous steps.

[0038] First, step S1, the narrow frame phone watch wearing detection method first uses an ultra-low power capacitive sensing unit to continuously monitor the phone watch shell back cover.

[0039] S101: configuring and starting the capacitance monitoring unit. After the system is powered on, the central processing unit initializes and configures the capacitance sensing unit to enter a continuous monitoring state. Specifically, it includes:

[0040] The capacitance sensing unit uses a mutual capacitance sensing scheme, and its core is a low-power capacitance-to-digital converter chip. The chip is connected to the central processing unit through an I2C interface. The sensing electrode is a copper foil pattern printed on a flexible circuit board, which is attached to the inner surface of the back cover of the phone watch case.

[0041] Through the I2C bus, the sampling rate register of the capacitance-to-digital converter chip is configured to 1 Hz, and its working mode is set to low-power single conversion mode. At the same time, enable the internal data ready interrupt function.

[0042] After the configuration is completed, send a start measurement command to the capacitance-to-digital converter chip. Thereafter, the chip will automatically perform capacitance measurement at a frequency of 1 Hz, and will notify the central processing unit to read the capacitance value data after each measurement through an interrupt signal. The average current consumption in this working mode can be maintained below 10 microamperes.

[0043] S102: Obtain the reference capacitance value. The system provides a calibration mechanism to obtain and store the reference capacitance value C_ref. Typical trigger occasions include: automatically executed during factory testing, or manually started by the user through the device settings menu when the watch is not explicitly worn. When the calibration process is triggered, the capacitance sensing unit performs one or more measurements, and the stable capacitance value obtained is taken as the reference capacitance value C_ref, which is permanently stored in the non-volatile storage module of the device. This value is the static reference for subsequent contact event judgment.

[0044] S103: Continuous monitoring and capacitance value comparison. When the system is running in low-power monitoring mode, the capacitance sensing unit periodically collects the current capacitance value C_current at a frequency of 1 Hz. After each sampling is completed, the system calculates the absolute difference between the current capacitance value and the reference capacitance value: ΔC = |C_current - C_ref|. This difference ΔC will be the key input to determine whether to trigger the initial contact event.

[0045] The above steps S101 to S103 constitute a complete, low-power background monitoring cycle. The system realizes continuous detection of the event "whether an object approaches or contacts the back cover of the watch" through extremely low-frequency sampling and simple comparison with the pre-stored reference value, providing an initial trigger condition for the entire wearing detection process.

[0046] S104: Determine the initial contact event triggered. The calculated capacitance change AC is compared with the preset capacitance trigger threshold C_th. The threshold C_th is a relative value determined according to experimental data, usually set to 5% to 10% of the reference capacitance value C_ref. If AC ≥ C_th, it is determined that an effective initial contact event occurs, and the system immediately generates an initial contact event signal. This signal will wake up the subsequent processing unit, marking the official start of the wearing detection process. The specific percentage of the threshold can be fine-tuned according to the shell material, structure and environmental factors to optimize the anti-interference ability.

[0047] Step S2, i.e. in response to the initial contact event signal, activate the thermal sensor array arranged on the inner surface of the back cover of the phone watch shell, specifically:

[0048] S201: Respond to the trigger and activate the sensor. When the system receives the initial contact event signal generated in step S104, the central processing unit sends a high-level enable signal to the power management module through its general input and output pin. The power management module then powers the thermal sensor array arranged on the inner surface of the back cover of the phone watch shell, activating it from the dormant state.

[0049] S202: The specific structure of the thermal sensor array is that four negative temperature coefficient thermistors are arranged in a 2-row x 2-column matrix on a printed circuit board, and the spacing between each resistor is designed according to the size of the shell back cover, usually 5 to 15 mm, to achieve effective spatial sampling of the temperature distribution of the contact area. The temperature sensing surface of each thermistor element is tightly attached to the inner surface of the shell back cover through high thermal conductivity silicone grease, which fills the small gap between the element and the shell, thereby establishing a low thermal resistance heat conduction path between the temperature sensing surface of the thermistor and the outer surface of the shell back cover. This design ensures that the temperature change of the external contact object can be quickly and accurately transmitted to the sensor.

[0050] The above steps S201 and S202 together complete the hardware preparation of the second layer of detection. S201 is a logical trigger, and S202 explicitly executes the specific structure and key installation process of the physical sensor of this layer of detection, which is the material basis for ensuring the accuracy and response speed of temperature measurement.

[0051] S203: Synchronous acquisition of temperature data. After the thermal sensor array is activated, its built-in signal conditioning and analog-to-digital conversion circuit begins to work. The array synchronously acquires real-time temperature analog signals of four temperature measurement points at a sampling frequency of 20 Hz (20 times per second), and converts them into digital temperature values T_i (where i = 1, 2, 3, 4). This sampling frequency can capture sufficient temperature change timing information in a dynamic contact scenario.

[0052] S204: Organize temperature distribution data. The system organizes the four temperature values [T1, T2, T3, T4] acquired at each sampling time according to their corresponding physical positions (2x2 matrix) to form a discrete two-dimensional temperature distribution map. This data map is the direct input for subsequent thermodynamic feature analysis, and its essence is a temperature dataset containing spatial position information.

[0053] The above steps S201 to S204 are the second-level verification after the capacitive trigger. In this phase, the temperature field raw data of the contact area is quickly acquired by activating the thermosensor array with a specific spatial layout and efficient heat channel, and is structured into a two-dimensional distribution map for analysis.

[0054] After completing the temperature data synchronous acquisition and spatial distribution mapping described in step S2, the system obtains a discrete but structured two-dimensional temperature distribution map. This map carries the original temperature field information of the contact area, but it is not sufficient to directly prove that the contact object meets the human biological characteristics. Therefore, step S3, thermodynamic feature analysis and effective contact verification, is performed. This step aims to calculate a set of carefully designed statistical quantities and spatial feature parameters, and compare them with the preset human wrist thermodynamic model, to determine whether the previous contact event originated from a human tissue with a living body temperature characteristic. The specific implementation of this analysis and verification process will be described in detail below.

[0055] Step S3: Thermodynamic feature analysis and effective contact verification. Based on the two-dimensional temperature distribution map generated in step S2, the system performs thermodynamic feature analysis to verify whether the contact object meets the human biological characteristics. This analysis is achieved by calculating and evaluating a set of key feature parameters, with the specific steps as follows:

[0056] S301: Calculate temperature distribution mean. First, calculate the temperature mean T_avg of the two-dimensional temperature distribution map. For a 2x2 matrix composed of four temperature measurement points, the temperature mean is its arithmetic mean: T_avg=(T1+T2+T3+T4) / 4, which reflects the overall temperature level of the contact area.

[0057] S302: Calculate temperature spatial gradient and verify radial trend. To quantify the change of temperature from the center to the edge and verify its radial decreasing pattern, calculate the gradient feature as follows:

[0058] Define center and edge: Approximate the temperature mean of the two temperature measurement points on the diagonal of the 2x2 array as the center temperature T_center (e.g. (T1+T4) / 2 or (T2+T3) / 2), and approximate the temperature mean of the other pair of points on the diagonal as the edge temperature T_edge.

[0059] Calculate radial temperature difference: Calculate radial temperature difference ΔT_radial = T_center - T_edge.

[0060] Calculate normalized gradient: Divide radial temperature difference by effective radial distance d_eff of sensor array (i.e. sensor spacing value set in step S202, e.g. 10mm), to get normalized temperature gradient value G: G = ΔT_radial / d_eff.

[0061] Verify trend: Check if ΔT_radial > 0 holds, as a criterion for "center temperature higher than edge temperature", i.e. a decreasing trend in radial direction.

[0062] S303: Calculate spatial entropy of temperature distribution. Further, to evaluate the uniformity (or concentration) of temperature distribution, calculate the spatial entropy H of temperature distribution. Treat the four temperature values as a discrete system, and the temperature value at each point constitutes a probability distribution. Use Shannon entropy formula to calculate:

[0063] H = -∑(p_i * log2(p_i)), where i = 1 to 4, p_i = T_i / ∑T_i (normalize temperature values to probabilities).

[0064] The smaller the entropy value H, the more concentrated the temperature distribution (e.g. heat concentrated in the center); the larger the entropy value, the more uniform the distribution.

[0065] S304: Perform feature matching and generate valid signal. Compare the three feature parameters (T_avg, ΔT_radial, H) calculated in steps S301 to S303 with the pre-set human wrist thermodynamic model decision conditions:

[0066] Temperature range condition: Temperature mean T_avg needs to fall within the pre-set human body surface temperature interval, which typically ranges from 32°C to 37°C.

[0067] Gradient trend condition: Radial temperature difference ΔT_radial (calculated in step S302) needs to be greater than zero. This directly and uniquely verifies the radial decreasing spatial distribution pattern of "higher temperature in the center region than in the edge region".

[0068] Spatial uniformity condition: Spatial entropy H needs to be less than the pre-set uniform heat source entropy threshold, which typically has a value of 0.8. This threshold is used to distinguish between human skin (relatively concentrated heat distribution) and objects with uniform environmental temperature (such as metal, plastic).

[0069] Only when the above three conditions are met simultaneously, the system determines that the current contact object has the thermodynamic characteristics of a human wrist, and generates a valid body temperature contact signal. This signal indicates that the preliminary biometric verification is passed, triggering the next stage of detection.

[0070] After the thermodynamic feature analysis and the valid body temperature contact signal generation in step S3, the system has preliminarily confirmed that the contact object has the biological heat feature of human skin. However, only the static temperature distribution cannot distinguish whether the watch is correctly worn on the active wrist or is only statically attached to the skin or other limb parts (e.g. held in the hand or placed on the arm but not fastened). In order to further verify the authenticity and effectiveness of the wearing state, the system needs to introduce the analysis of the typical wrist dynamic behavior of the user.

[0071] Based on this, the present application introduces step S4 - inertial data acquisition and dynamic behavior analysis. This step aims to collect and analyze the micro-movement or activity pattern of the user in the suspected early wearing period by activating the inertial measurement unit, so as to judge whether the current contact is accompanied by motion features consistent with human wearing characteristics.

[0072] Step S4: After receiving the valid body temperature contact signal, the system immediately enters the dynamic behavior feature acquisition stage, the core of which is to obtain sensor data reflecting the micro-movement of the user's limb in the early wearing period. This process is executed according to the following steps:

[0073] S401: The central processing unit immediately switches the inertial measurement unit from the low-power sleep mode to the working mode in response to the aforementioned valid body temperature contact signal.

[0074] S402: The system sets a data acquisition time window of a preset duration (e.g. 2 seconds) and aligns the start time of the window with the generation time of the valid body temperature contact signal. Then, the inertial measurement unit is controlled to start continuous sampling.

[0075] S403: Within the time window, the inertial measurement unit synchronously acquires three-axis acceleration data output by the three-axis accelerometer and three-axis angular velocity data output by the three-axis gyroscope. These time-sequentially arranged multi-axis data are combined to form a time-series dynamic data sequence for subsequent analysis.

[0076] Then, step S5 is performed: After obtaining the time-series dynamic data sequence, the system enters the dynamic feature extraction stage. This stage aims to quantify key behavior features from the original inertial data and evaluate the wearing possibility by comparison with known models. The specific execution steps are as follows:

[0077] S501: Acceleration vector synthesis and signal generation. The three-axis acceleration data (Ax, Ay, Az) within the time window is subjected to vector synthesis. Specifically, at each sampling time t, the modulus value of the synthesized acceleration is calculated according to the formula A_total(t) = sqrt(Ax(t)²+Ay(t)²+Az(t)²). The modulus values of all sampling times are arranged in time sequence, i.e. a total acceleration time-series signal representing the overall motion intensity is obtained.

[0078] S502: Perform short-time Fourier transform on the total acceleration time series obtained in step S501 (e.g., using Hanning window, window length 256 points, overlap rate 50%), to obtain its energy distribution spectrum (i.e., power spectral density) in the frequency range of 0 to 50 Hz. Calculate the integral value of the energy spectrum in this frequency band (i.e., sum the energy values of all frequency points in this frequency band), and define it as the motion intensity spectral density. This parameter reflects the overall intensity of the user's activity in this time window.

[0079] S503: Calculate the sample variance of each of the three-axis angular velocity data (Wx, Wy, Wz) in the preset time window. Take the arithmetic mean of the three variance values, and define the reciprocal of the mean as the attitude stability factor. The larger the factor value, the smaller the attitude angular velocity fluctuation of the wrist in the measurement period, i.e., the more stable the attitude. This feature is a typical manifestation of the device being stably worn on the wrist as part of the body, distinguishing it from being placed on a desktop or being held and shaken, etc.

[0080] The above steps S501 to S503 complete the feature quantification of motion intensity and attitude stability. Among them, the motion intensity spectral density measures the activity amount from the frequency domain, and the attitude stability factor represents the smoothness of the motion from the time domain, and the combination of the two can effectively distinguish between active wearing and device being moved randomly, etc.

[0081] S504: Periodic motion frequency calculation. Perform autocorrelation analysis on the total acceleration time series obtained in step S501. Specifically, the normalized autocorrelation function can be calculated by the conventional Fast Fourier Transform (FFT) method in the field to improve the operation efficiency and facilitate peak comparison. Subsequently, after excluding the zero delay point, in a reasonable physiological motion cycle delay range (e.g., corresponding to the frequency of 0.5 Hz to 3 Hz), the first significant peak (e.g., using local maximum value detection, or the first peak with an amplitude exceeding a preset threshold) of the autocorrelation function is found. According to the sampling frequency of the signal, convert the delay sampling point number corresponding to the peak value to the actual delay time in seconds. Calculate the reciprocal of the delay time, i.e., obtain the periodic motion frequency in Hz. This parameter is a key feature to identify regular whole-body movements such as walking and running.

[0082] S505: Combine the three parameters of motion intensity spectral density, attitude stability factor and periodic motion frequency extracted in steps S502, S503, S504 respectively in a preset order to construct a three-dimensional feature vector.

[0083] Subsequently, the feature vector is input into a human body wearing activity feature database preset in the nonvolatile storage module, and is ready for similarity matching. The database stores statistical model data of wearing state feature vectors learned from a large number of samples under various typical scenarios (such as walking, sitting, running, and sleeping). The statistical model data at least includes typical feature vectors (mean vectors) of various states and weight information (such as variance or covariance matrix) reflecting the dispersion degree of each dimension data, which provides a basis for subsequent quantitative similarity calculation.

[0084] S506: Dynamic feature matching degree score calculation. The weighted Euclidean distance algorithm is used to calculate the difference between the current feature vector and the reference feature vector (mean vector) of each state (such as typical wearing and typical non-wearing) in the database. Specifically, the weight coefficient is determined based on the variability of the variance of each dimension feature in the database (for example, taking the inverse of the variance of each dimension and normalizing), to balance the influence of different feature dimensions and dispersion degree on the total distance. The calculated weighted Euclidean distance is denoted as D. Then, the distance D is mapped to a normalized dynamic feature matching degree score by the formula score = 1 / (1+D). The score directly represents the similarity between the current dynamic feature and the target state (such as typical wearing behavior), and the higher the score value, the higher the similarity.

[0085] Step S5 finally generates a quantitative dynamic feature matching degree score. On this basis, the system enters step S6, which will perform preliminary determination and decision of the wearing state based on the score.

[0086] Step S6: Preliminary determination and active detection triggering Based on the dynamic feature matching degree score calculated in step S5, the system performs preliminary determination of the wearing state, and triggers the active detection process with high confidence when the determination is ambiguous. The process is performed according to the following steps:

[0087] S601: Threshold-based state preliminary decision.

[0088] The system compares the dynamic feature matching degree score with two preset decision thresholds:

[0089] If the score is higher than the first preset decision threshold (for example, 0.85), it is directly determined that the phone watch is in a wearing state.

[0090] If the score is lower than the second preset decision threshold (for example, 0.4), it is directly determined as a non-wearing state.

[0091] If the score is between the first and second preset decision thresholds (i.e. falls into the ambiguous interval), the determination result is pending, and the subsequent active detection process is triggered immediately.

[0092] S602: When the determination result is "pending", the central processing unit sends a control instruction to the driving circuit of the active probe unit. The driving circuit generates a preset composite frequency vibration sequence according to the instruction, which is used to drive the linear resonant actuator (for example, a piezoelectric ceramic vibrator) to generate mechanical vibration. The vibration sequence is an analog electrical signal with a duration of 200 milliseconds, and its waveform is composed of a fundamental frequency sine wave with a frequency of 150 Hz, and a second harmonic sine wave with an amplitude of 30% of the fundamental frequency and a frequency of 300 Hz.

[0093] S603: The driving circuit power amplifies the composite frequency electrical signal generated in step S602, and then drives the linear resonant actuator to work. The actuator converts the electrical signal into mechanical vibration of the same frequency, so as to excite the shell of the phone watch to produce forced vibration.

[0094] S604: At the same time when the linear resonant actuator starts to vibrate, the system controls the three-axis accelerometer in the inertial measurement unit (IMU) to work synchronously at a sampling rate of not less than 1 kHz, and collects acceleration data transmitted by the shell vibration. The collection process needs to cover the entire excitation stage (200 milliseconds) and the free decay stage after the vibration stops, which together constitute a complete vibration time sequence signal.

[0095] S605: Perform fast Fourier transform on the vibration time sequence signal collected in step S604 to obtain its frequency spectrum, and perform the following feature extraction:

[0096] In the frequency spectrum, find the peak point with the highest amplitude, and record the corresponding frequency as the resonance peak frequency.

[0097] Take the above peak frequency as the center, calculate the difference between the two frequency points corresponding to the half (-3dB) of the peak value of the resonance peak, and take this full width at half maximum value as the frequency bandwidth.

[0098] Calculate the energy decay time constant:

[0099] Intercept the time domain signal of the free decay stage after the vibration excitation ends. Calculate the envelope line of this signal (for example, by Hilbert transform or low-pass filtering after taking the absolute value). Define the envelope amplitude at the decay starting time (t=0) as A0. Find the time t_10% corresponding to the envelope line decaying to 0.1*A0. Define t_10% as the energy decay time constant.

[0100] S606: Combine the three feature parameters extracted in step S605, namely the resonance peak frequency, the frequency bandwidth, and the energy decay time constant, into a three-dimensional vibration decay feature vector. This vector comprehensively represents the response and decay characteristics of the current contact medium to a specific vibration excitation.

[0101] So far, through the preliminary judgment in step S6 and the active detection process, the system has completed the following work: first, according to the dynamic behavior matching score, make a clear judgment of "wearing", "non-wearing", or identify the "pending" state that needs further verification; second, for the "pending" state, the system performs a complete active vibration detection, including generating a specific vibration excitation, synchronously collecting the response signal, and extracting the key vibration attenuation feature vector therefrom.

[0102] The feature vector comprehensively reflects the physical interaction results of the shell and the contact medium under vibration excitation, but it cannot be directly converted into a conclusion of the wearing state. In order to convert this physical feature into the final state judgment, the system needs to perform the final quantitative comparison and decision.

[0103] Therefore, step S7: based on the vibration attenuation feature vector output by step S6, the system enters the final decision stage. This stage calculates a high-confidence decision basis by quantitatively comparing the vector with the pre-built human tissue physical model, thereby completing the final confirmation of the wearing state. The specific steps are as follows:

[0104] S701: The final decision unit receives the vibration attenuation feature vector constructed by step S606, denoted as x. At the same time, the pre-set human tissue vibration attenuation model is called from the non-volatile storage module of the system. This model is a multi-class statistical database, and its content is based on a large amount of data collected by active vibration detection on different user groups (covering different age groups such as children, teenagers, adults, and the elderly, and different body types such as thin, standard, and robust) in the wearing state.

[0105] S702: The system first selects the most matched reference group model from the database according to the basic information of the current user (such as the pre-set age group, body type category) or through preliminary screening. This model is characterized by the mean vector μ and the covariance matrix Σ of the group sample.

[0106] Then, the Mahalanobis distance between the real-time feature vector x and the group model is calculated , and the calculation formula is:

[0107]

[0108] Where x represents the vibration attenuation feature vector obtained in real time through the active detection process. This vector is a three-dimensional data point composed of the resonance peak frequency, the frequency band width, and the energy attenuation time constant extracted in step S605, which are three characteristic parameters in sequence, used to quantitatively represent the physical response characteristics of the current phone watch shell and the contact medium under a specific vibration excitation.

[0109] μ represents the reference mean vector selected from the pre-stored "human tissue vibration attenuation model" database, which matches the current user group. This vector statistically represents the average value of the vibration attenuation characteristic vector of the target user group in the correct wearing state, representing the center position of the typical characteristics of the group.

[0110] Σ represents the covariance matrix corresponding to the mean vector μ. This matrix describes the dispersion (variance) of the vibration attenuation characteristic vector of the target user group in the wearing state and the linear correlation between different dimensions, which together define the distribution pattern of the characteristic data around the center μ.

[0111] represents the inverse matrix of the covariance matrix Σ. In the calculation of Mahalanobis distance, its mathematical role is to de-correlate and standardize each dimension of the characteristic vector, so as to eliminate the influence of different characteristic dimension differences and their correlation on distance measurement, so that the calculated distance can more accurately reflect the deviation of real-time data from the reference statistical distribution.

[0112] T represents the transpose operator of the vector. It is used to convert the column vector into a row vector to meet the operation rules of matrix multiplication.

[0113] represents the calculated Mahalanobis distance. The distance is a dimensionless scalar value. The smaller the value, the closer the real-time characteristic vector x is to the center position of the target reference distribution (defined by μ and Σ) in a statistical sense, that is, the higher the possibility that x belongs to the wearing state reference group. This distance is the direct input of the subsequent calculation of the sound-vibration coupling confidence C by the mapping function .

[0114] S703: Map the calculated Mahalanobis distance to a direct and normalized confidence score. The sound-vibration coupling confidence C is calculated by the following formula: , which maps the Mahalanobis distance (range from 0 to positive infinity) to the confidence C (range from 0 to 1). When C approaches 0, C approaches 1, indicating a very high confidence; When C increases, C decreases, indicating a lower confidence.

[0115] S704: Compare the calculated sound-vibration coupling confidence C with a pre-set decision threshold (typical value is 0.7): if C is higher than the decision threshold, it is finally determined that the phone watch is in the wearing state.

[0116] If C is lower than or equal to the decision threshold, it is finally determined to be in the non-wearing state.

[0117] The step S7 constitutes the final decision loop of the whole wearing detection process. It uses the physical feature (vibration attenuation vector) obtained by the active detection in step S6, which is difficult to fake, to achieve high-precision state decision through rigorous statistical model (Mahalanobis distance) comparison and normalized confidence calculation. This step fully utilizes the essential difference between human soft tissue and rigid objects in vibration energy attenuation characteristics, fundamentally solves the problem of insufficient passive sensing ability in static, ambiguous or interference scenarios, and ensures that the final output of the whole system has high reliability and anti-interference.

[0118] In addition, the detection method disclosed in the present application also includes confirmation logic for wearing state transition. When the system determines that the phone watch is in the non-wearing state from the wearing state, it is required to output the non-wearing state determination result in the continuous three detection periods, and the state transition can be finally confirmed. The interval of each detection period is 5 seconds, which can effectively avoid the misjudgment caused by the user temporarily taking off the watch to check the information, the sensor instantaneous failure caused by the user's vigorous exercise, or the accidental obstruction of external objects. When determining the transition from the non-wearing state to the wearing state, the system completes a complete four-stage judgment process (capacitance trigger→ thermodynamic verification→ dynamic matching→ active detection when necessary) and confirms the wearing state, and immediately updates the internal state flag to ensure real-time response to the wearing event and meet the user's demand for instant function activation.

[0119] On the other hand, the implementation of the above-mentioned method of the present application depends on a highly integrated narrow-frame phone watch wearing detection system. The system is integrated in the phone watch and mainly includes a data acquisition module, a state determination engine, an active detection unit and a non-volatile storage module.

[0120] The data acquisition module includes a capacitance sensing unit, a thermal sensor array and an inertial measurement unit. The capacitance sensing unit adopts a mutual capacitance structure, and the sensing electrode is embedded in the inner side of the back cover of the shell. The detection of micro-farad level capacitance change is realized through a special low-power analog front-end chip. The four negative temperature coefficient thermistors of the thermal sensor array are packaged on a flexible printed circuit board, closely attached to the inner surface of the back cover of the shell, and fixed through thermal conductive glue. The inertial measurement unit is a commercial MEMS device, which integrates a three-axis accelerometer and a three-axis gyroscope, supports multiple range and bandwidth configurations, and its data output interface is an I2C bus connected with the central processing unit.

[0121] The state determination engine is solidified in the firmware of the central processing unit, which is implemented as an embedded program code logically divided into four functional units: the initial contact analysis unit is responsible for processing the capacitance data and generating a trigger signal; the thermodynamic feature verification unit performs temperature distribution analysis and model matching; the dynamic behavior matching unit completes the feature extraction of the timing data and database comparison; the final decision unit is dedicated to processing the vibration data analysis and confidence calculation in the active probing stage. The units are scheduled through the state machine mechanism to ensure that the subsequent modules are activated only when the preconditions are met.

[0122] The active probing unit is composed of a linear resonant actuator and its driving circuit. The linear resonant actuator is a single piezoelectric ceramic element, whose resonant frequency is factory-calibrated to 150 Hz, and is installed near the center of the shell to maximize the vibration transmission efficiency. The driving circuit includes an H-bridge power amplifier and a waveform generator, which can accurately reproduce the preset composite frequency vibration sequence.

[0123] The non-volatile storage module uses a flash memory chip to store all preset thresholds, model parameters, and feature databases. This includes capacitance trigger thresholds, upper and lower limits of human body surface temperature intervals, human wrist thermodynamic model feature parameters (such as entropy thresholds and gradient direction constraints), human body wearing activity feature databases (stored in the form of K-nearest neighbors or Gaussian mixture models), first and second preset determination thresholds, human tissue vibration attenuation models (including multiple mean vectors and covariance matrices), and decision thresholds. All parameters can be remotely updated through firmware upgrades to adapt to differences in different product batches or user groups.

[0124] The entire system adopts a layered activation power management strategy. In the non-contact state, the system is in a deep sleep mode, and only the capacitance sensing unit works continuously with a current of about 5 microamperes; once the initial contact event is triggered, the thermosensitive sensor array is awakened, and the system power consumption rises to about 200 microamperes; if the thermodynamic verification is passed, the inertial measurement unit is started, and the power consumption increases to about 1.5 milliamperes; only when the dynamic matching result is ambiguous, the active probing unit is transiently activated, and the peak power consumption can reach 10 milliamperes, but the duration is only 200 milliseconds. Through this on-demand step-by-step wake-up mechanism, the average power consumption of the system is controlled below 50 microamperes within 24 hours, significantly prolonging the battery life of the phone watch.

[0125] In summary, the present application integrates four heterogeneous sensing modalities of capacitance, thermosensitive array, inertial measurement, and active vibration, to build a wearable detection scheme with high precision, strong stability, and ultra-low power consumption. Its determination logic is based on deep modeling of human physiological features and physical interaction mechanisms, rather than simple threshold comparison, effectively overcoming the limitations of existing technologies in complex use scenarios.

[0126] It is apparent for a person skilled in the art that the present application is not limited to the details of the above-described exemplary embodiments, but that the present application can be implemented in other concrete forms without departing from the spirit or essential characteristics of the present application, and that the embodiments should therefore be considered as exemplary and not limiting in any way.

[0127] Furthermore, it should be understood that although the present specification describes exemplary embodiments, not every embodiment contains only one independent technical solution, and the present specification is described in this way only for the sake of clarity, and a person skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that a person skilled in the art can understand.

Claims

1. A method for detecting the wearing of a narrow-bezel children's smartwatch, characterized in that, include: The capacitance sensing unit continuously monitors the capacitance change of the back cover of the phone watch at a preset low-frequency sampling period. When the capacitance change exceeds the preset capacitance trigger threshold, an initial contact event signal is generated. In response to the initial contact event signal, the thermal sensor array located on the back cover of the phone watch case is activated to collect real-time temperature data of the contact area covered by the back cover, forming a two-dimensional temperature distribution map. The two-dimensional temperature distribution map is analyzed to calculate its temperature mean, temperature gradient, and spatial entropy of the temperature distribution to generate an effective body temperature contact signal. After receiving the effective body temperature contact signal, the inertial measurement unit is activated. The inertial measurement unit collects triaxial acceleration data and triaxial angular velocity data within a preset time window to form a time-series dynamic data sequence. Feature extraction is performed on the time-series dynamic data sequence to calculate dynamic feature parameters, including motion intensity spectral density, attitude stability factor, and periodic motion frequency. The dynamic feature parameters are then matched with a human wearable activity feature database pre-stored in the memory to generate a dynamic feature matching score. Based on the dynamic feature matching score, a preliminary determination of the wearing status is performed. When the dynamic feature matching score is higher than the first preset determination threshold, the phone watch is determined to be in a wearing state. When the dynamic feature matching score is lower than the second preset determination threshold, the phone watch is determined to be in a non-wearing state. The first preset determination threshold is greater than the second preset determination threshold. When the dynamic feature matching score is between the first preset determination threshold and the second preset determination threshold, an active detection process is initiated to calculate and generate an acoustic-vibration coupling confidence score. Based on the acoustic-vibration coupling confidence score, a final determination of the wearing status is made. When the acoustic-vibration coupling confidence score is higher than the preset determination threshold, the phone watch is determined to be in a wearing state; otherwise, it is determined to be in a non-wearing state. The active detection process specifically includes: driving a linear resonant actuator to generate a preset composite frequency vibration sequence through control signals; simultaneously with the vibration of the linear resonant actuator, using an accelerometer in the inertial measurement unit to collect the shell vibration response data generated by the vibration; performing a fast Fourier transform on the shell vibration response data to extract its resonant peak frequency, bandwidth, and energy decay time constant, forming a vibration decay feature vector; comparing the vibration decay feature vector with a preset human tissue vibration decay model to calculate an acoustic-vibration coupling confidence level.

2. The method for detecting the wearing of a narrow-bezel phone watch according to claim 1, characterized in that, The thermal sensor array is composed of negative temperature coefficient thermistors, which are encapsulated on the inner surface of the back cover of the phone watch case and form a heat conduction path with the outer surface of the back cover of the case through a medium with high thermal conductivity. The specific characteristics of the human wrist thermodynamic model are as follows: the temperature in the central region of the two-dimensional temperature distribution map is higher than that in the edge region, and the temperature gradient decreases radially outward. At the same time, the spatial entropy of the temperature distribution is less than the preset uniform heat source entropy threshold.

3. The method for detecting the wearing of a narrow-bezel phone watch according to claim 1, characterized in that, Feature extraction of the time-series dynamic data sequence specifically includes: vector synthesis of the three-axis acceleration data within the time window to obtain the total acceleration time-series signal; short-time Fourier transform of the total acceleration time-series signal to obtain its energy distribution in the frequency range of 0 to 50 Hz, and using it as the motion intensity spectral density; calculation of the variance of the three-axis angular velocity data within the time window, and using the reciprocal of the variance as the attitude stability factor; and autocorrelation analysis of the total acceleration time-series signal to extract the main period with the strongest energy in the signal, and using it as the periodic motion frequency.

4. The method for detecting the wearing of a narrow-bezel phone watch according to claim 1, characterized in that, It also includes a confirmation logic for the transition of the wearing state. When determining that the phone watch has changed from the wearing state to the non-wearing state, it is required that it be determined to be in the non-wearing state for three consecutive detection cycles before the state transition can be finally confirmed. When determining the transition from the non-wearing state to the wearing state, the state is updated immediately after completing a complete determination process and confirming the wearing state.

5. The method for detecting the wearing of a narrow-bezel phone watch according to claim 1, characterized in that, The capacitive trigger threshold is between 5% and 10% of the reference capacitance value, which is a reference value that is pre-calibrated and stored in the non-volatile memory module when the phone watch is in a contactless state.

6. A narrow-bezel phone watch wearing detection system, applied to the narrow-bezel phone watch wearing detection method according to any one of claims 1-5, characterized in that, include: The data acquisition module includes a capacitance sensing unit, a thermistor array, and an inertial measurement unit. All three units are electrically connected to a central processing unit. The capacitance sensing unit monitors changes in the capacitance of the housing's back cover. The thermistor array acquires a two-dimensional temperature distribution map of the contact area. The inertial measurement unit acquires time-series dynamic data sequences and housing vibration response data. The state determination engine is embedded in the firmware of the central processing unit. Its configuration is used to receive and process data from the data acquisition module. The state determination engine specifically includes an initial contact analysis unit, a thermodynamic feature verification unit, a dynamic behavior matching unit, and a final decision unit. The initial contact analysis unit is used to analyze the data of the capacitance sensing unit. When the change in capacitance value exceeds the capacitance trigger threshold, it generates and sends an initial contact event signal to the thermodynamic feature verification unit. The thermodynamic feature verification unit, upon receiving the initial contact event signal, activates and processes the two-dimensional temperature distribution map collected by the thermal sensor array. After confirming that it conforms to the thermodynamic features of the human wrist, it generates and sends an effective body temperature contact signal to the dynamic behavior matching unit. The dynamic behavior matching unit, upon receiving the effective body temperature contact signal, activates and processes the time-series dynamic data sequence collected by the inertial measurement unit, calculates the dynamic feature matching score, and outputs a wearing, non-wearing, or pending state signal based on the relationship between the score and the first and second preset judgment thresholds. The active detection unit includes a linear resonant actuator and a drive circuit, which is connected to the central processing unit. When the dynamic behavior matching unit outputs a pending state signal, the central processing unit controls the active detection unit to execute the active detection process, and the active detection unit sends the collected shell vibration response data to the final judgment unit. The final adjudication unit is used to receive the shell vibration response data, calculate the acoustic-vibration coupling confidence level, and output the final wearing state or non-wearing state determination result based on the comparison result of the confidence level and the adjudication threshold.

7. The narrow-bezel phone watch wearing detection system according to claim 6, characterized in that, The system also includes a non-volatile storage module connected to the central processing unit. The non-volatile storage module stores the capacitive trigger threshold, the human body surface temperature range, the characteristic parameters of the human wrist thermodynamic model, the human wearing activity characteristic database, the first preset judgment threshold, the second preset judgment threshold, the human tissue vibration attenuation model, and the decision threshold.

Images